Law and Governance  Law and Governance

How Doctrine Enables, Supports, and Degrades Cooperation

               Seth C. Oranburg




                   BizLaw LLC

© 2026 Seth C. Oranburg. All rights reserved.

                Published by BizLaw LLC

No part o

this book may be reproduced, stored in a retrieval system, or transmitted in any — orm or by any means without the prior written permission o — the publisher, except — or brie — quotations in reviews.  For Zeeva.

 May the institutions you inherit be susceptible to the wisdom you cultivate.

. ֹ‫ אִ ישׁ אֶ ת רֵ עֵ הוּ חַ ִיּ ים בְּ לָ עו‬,‫ שֶׁ אִ לְ מָ לֵ א מוֹ ָר אָ הּ‬,‫הֱ וֵ י מִ תְ פַּ לֵּ ל בִּ שְׁ לוֹ מָ הּ שֶׁ ל מַ לְ כוּת‬

             Pray  --- or the integrity o ---  the governing institution,

or without its awe, a man would swallow his — ellow alive.

                                        — Pirkei Avot 3:2

Contents

Introduction ……………………………………………………………………………… 1 Chapter 1. The Architecture o — Governance……………………………… 15 Chapter 2. Governance Babel …………………………………………………… 38 Chapter 3. Governance Ontology …………………………………………….. 61 Chapter 4. Law as Governance Architecture …………………………….. 83 Chapter 5. Club Good Structure …………………………………………….. 107 Chapter 6. Governance Surplus ……………………………………………… 127 Chapter 7. Law’s Governance E —

ects ……………………………………… 149 Chapter 8. Governance Evaluation Method ……………………………. 179 Chapter 9. Network Governance ……………………………………………. 228 Chapter 10. Exchange Governance …………………………………………. 259 Chapter 11. Corporate Governance ………………………………………… 285 Chapter 12. University Governance ……………………………………….. 309 Chapter 13. Knowledge Governance ………………………………………. 331 Chapter 14. Framework Limits ………………………………………………. 354 Chapter 15. Governance-Literate Law ……………………………………. 375 Conclusion ……………………………………………………………………………. 397 Notes and Re — erences …………………………………………………………….. 401 Introduction Law o — ten changes institutions without seeing what those institu- tions do. Courts see a plainti —

, a de — endant, and a dispute. Agencies see a violation and a remedy. Legislatures see a policy problem and a rule to solve it. Those — rames are o — ten use — ul. They are not complete. Groups also solve shared problems through institutions that make decisions, monitor conduct, impose sanctions, and revise their own rules over time. Those institutions govern. Law acts on them con- stantly. Yet legal analysis rarely treats governance itsel — as a distinct object o — analysis. This book starts — rom that gap. In 1959, the New York Stock Exchange ordered its member

irms to cut all private wire connections with Harold Silver, a non- member broker.1 Silver received no charges, no hearing, and no ex- planation. The Supreme Court later held that the exchange’s action could not escape antitrust scrutiny and that basic procedural protec- tions were required be — ore exclusion. The Court addressed a real abuse. But it did not ask what its intervention would do to the ex- change’s capacity to govern its own members quickly and credibly. Over time, the exchange’s disciplinary authority moved into a heav- ily supervised procedural — ramework. A legal remedy aimed at one wrong changed the institution that managed market integrity. A similar pattern appears outside securities law. A — ter the Larry Nassar scandal, — ederal en — orcement against Michigan State Univer- sity required new policies, new training, new coordinators, and ret- rospective review. 2 Those re — orms responded to visible — ailures. They did not directly repair the reporting structure that had allowed 2 Law and Governance

complaints to be suppressed inside the athletic department. The in- stitution changed its compliance sur — ace more than its governance architecture. The problem was not only misconduct. The problem was a structure that prevented accurate monitoring and accounta- bility. A third example comes — rom outside courts and agencies. In a

amous — ield experiment in Israeli daycare centers, researchers im- posed a — ine on parents who arrived late to pick up their children.3 The obvious prediction was that tardiness would — all. It rose instead, and the higher rate persisted a — ter the — ine disappeared. The price displaced a social norm. What had been governed by guilt, reputa- tion, and shared expectations was recast as a paid option. The inter- vention addressed a behavior. It also damaged the institution that had governed the behavior. These examples share a structure. In each one, an institution managed a shared problem through its own internal mechanisms. In each one, a legal or policy intervention addressed an immediate problem that observers could easily see. In each one, the interven- tion also weakened, displaced, or ignored the governance system that made cooperation possible in the — irst place. That recurring pat- tern is the subject o — this book. The central claim is simple. Governance is the organized system by which a group manages a shared problem over time. A govern- ance institution must do — our things. It must make decisions that members treat as binding. It must monitor conduct. It must impose sanctions with enough credibility to a —

ect behavior. And it must adjust when conditions change. I — any one o — those elements — ails, governance degrades. I — law changes one o — those elements, law changes governance whether it says so or not. The present book Introduction 3

supplies the missing variable: a de

inition o — governance precise enough to travel across legal — ields. Legal scholarship has not lacked interest in governance. It has lacked a de — inition that travels across — ields. Corporate lawyers speak o — governance when they discuss boards, shareholder rights, and — i- duciary duties. Administrative lawyers speak o — governance when they discuss collaborative regulation and experimentalism. Political scientists use the term — or state steering, networked authority, and international order. Commons scholars use it — or collective action and institutional design. Each usage captures something important. None supplies a general legal account o — governance as such. The same word names di —

erent objects in di —

erent literatures, which makes cross- — ield legal analysis di —

icult and sometimes impossible. This book supplies that missing account. It establishes a — unc- tional de — inition o — governance that travels across — ields and identi-

ies its minimum elements. It demonstrates the legal conditions that allow governance institutions to — orm and persist. It shows why many governance institutions have the structure o — club goods ra- ther than public goods. It distinguishes bene — its that governance in- stitutions supply to members — rom spillover bene — its they generate

or outsiders. And it o —

ers a method — or evaluating legal rules by what they do to governance. That last contribution matters most. Law is usually judged by the rights it protects, the injuries it remedies, the incentives it cre- ates, or the distributions it produces. Those are real and necessary perspectives. This book adds another one. It asks what a legal rule does to the institution through which a group manages a shared problem over time. Some legal rules enable governance. They pro- tect membership boundaries, reduce the cost o — monitoring, en — orce internal decisions, or preserve room — or institutional adaptation. 4 Law and Governance

Some legal rules degrade governance. They disable exclusion, dis- solve accountability, substitute paperwork — or institutional — unc- tion, or impose remedies that correct one abuse by damaging the whole system. Some legal rules discipline governance in a narrower and more care — ul way. They target a speci — ic — ailure while preserving the institution’s capacity to govern. Legal analysis o — ten misses that di —

erence, partly because governance has not been cognizable as a distinct legal object. This book argues that it should be. This book acknowledges that governance can produce exclu- sion, entrench insiders, mask coercion, and impose serious costs on outsiders. A governance institution is not good because it governs. It may be captured, abusive, or badly designed. The task is to judge governance with clarity and precision, not to de — er to it blindly or attack it re — lexively. Law cannot judge governance well unless it can

irst see governance clearly. A court cannot calibrate a remedy to an institution it does not recognize. An agency cannot repair a moni- toring — ailure it does not identi — y as structural. A legislature cannot weigh governance costs it cannot describe. Making governance cog- nizable—visible and analytically precise in legal analysis—is a pre- requisite to any sound assessment o — law’s e —

ects on institutions. The book proceeds in three Parts. Part I de — ines governance and identi — ies its legal conditions. It begins by o —

ering a general de — inition that can travel across corpo- rate law, administrative law, commons scholarship, nonpro — it law, network governance, and related — ields. It then surveys the existing vocabularies o — governance to show why none o — them provides a stable cross- — ield legal concept. Next it places the project in the tra- ditions that came closest to this task, especially legal institutional- ism, commons analysis, and transaction cost economics. It ends by Introduction 5

identi

ying six legal conditions that governance institutions need i —

they are to

orm, persist, and remain accountable. Part II develops the book’s original theory. It argues that gov- ernance o — ten has the structure o — a club good. That claim helps ex- plain why exclusion is central to governance, why governance is o — - ten undersupplied, and why legal rules that weaken excludability can have large institutional e —

ects. Part II then separates member bene-

its — rom spillover bene — its, develops the mechanisms by which law enables or degrades governance, and states a seven-step method — or evaluating legal rules by their governance e —

ects. The evaluative contribution is the book’s central payo —

. Part III applies the — ramework across several institutional set- tings. It examines merchant networks and private dispute resolu- tion, the New York Stock Exchange, corporate governance, univer- sities and nonpro — its, and law reviews as knowledge-producing in- stitutions. It then turns to the limits o — the — ramework itsel — . The application chapters do not try to — orce identical answers across di — -

erent domains. They test whether the same method can explain di — -

erent institutions with di —

erent problems, di —

erent legal settings, and di —

erent risks o —

ailure. The book there — ore makes both a de — initional claim and a meth- odological one. The de — initional claim is that governance is a dis- tinct institutional phenomenon with minimum elements that legal analysis can identi — y. The methodological claim is that law should be evaluated in part by what it does to governance. Once that variable becomes visible, — amiliar disputes look di —

erent. Some doctrines ap- pear more justi — ied than they — irst seemed because they subsidize the production o — valuable governance. Other doctrines appear less at- tractive because they solve a narrow problem while degrading the institution that managed a broader one. 6 Law and Governance

 A  --- inal clari --- ication is necessary at the outset. This book treats governance as one value legal analysis must weigh alongside rights, e ---

iciency, distribution, and legitimacy. Governance cannot be weighed well unless it is de — ined with precision. A legal system that cannot identi — y governance will tend to undervalue it. A legal system that romanticizes governance will overprotect it. The task is neither blindness nor de — erence. The task is clearer analysis—and recogniz- ing governance as a legally cognizable object makes that clearer anal- ysis possible. That is the aim o — this book. It names governance as a legal var- iable, de — ines the conditions under which governance exists, ex- plains how law changes governance, and o —

ers a way to judge those changes across — ields that legal scholarship usually treats separately.  Part I: De — ining Governance and Its Legal Conditions

Governance is one o

the most — amiliar words in legal scholarship and one o — the least settled. Corporate lawyers invoke it to describe the allocation o — authority within — irms. Public law scholars use it to describe the administration o — state power. Commons scholars use it to explain how communities manage shared resources. Interna- tional lawyers use it to describe coordination beyond the state. Each usage captures something real, but the concept shi — ts — rom — ield to

ield. The result is that legal analysis o — ten treats governance as im- portant while leaving its meaning unstable. Part I provides the — oundation the rest o — this book requires. Be-

ore law can be evaluated by what it does to governance, governance must be speci — ied as a legal object. That means more than collecting 8 Law and Governance

amiliar examples under a broad label. It means identi — ying what makes governance governance, what — unctional elements any gov- ernance institution must possess, and what legal conditions allow those institutions to — orm, persist, and remain accountable over time. Chapter 1 begins with the core de — initional claim. Governance is the organized system by which a group manages a shared problem over time. Each component o — that de — inition does analytical work. “Organized” distinguishes governance — rom spontaneous order. “System” emphasizes that governance is not a loose assortment o —

rules but an integrated structure whose parts depend on one an- other. “Group” and “shared problem” identi — y the constituency and the institutional task. “Over time” explains why governance must include not only present control but also the capacity to adapt as cir- cumstances change. From that de — inition — ollow — our indispensable

unctional elements: decision-making, monitoring, sanctions, and adjustment. Chapter 2 maps the competing uses o — the term. Legal literature speaks — luently about governance, but not uni — ormly. Corporate governance, administrative governance, commons governance, and global governance each rely on di —

erent assumptions about what governance is, who exercises it, and what counts as success or — ail- ure. Mapping those usages is necessary because the book’s — rame- work is not meant to ignore them; it is meant to explain why they overlap, where they diverge, and what a general legal theory o — gov- ernance must be able to accommodate. Chapter 3 places the project in its nearest intellectual lineage: legal institutionalism. Legal realists and later institutional thinkers understood that law does not merely regulate behavior at the mar- gins; it helps constitute the institutions through which social li — e is Chapter 1: What Is Governance? 9

organized. That insight is indispensable to this book. But legal in- stitutionalism stopped short o — o —

ering a portable legal ontology o —

governance itsel

, or a method — or evaluating legal rules by the e — -

ects they have on governance institutions. This chapter shows both the inheritance and the gap. Chapter 4 translates the theory into legal architecture. Govern- ance does not depend on law in every setting, but law determines whether governance can operate in the — orms legal analysis most o — - ten con — ronts. Six legal conditions matter especially: permission to organize, membership boundaries, internal decision procedures, ex- ternal en — orceability, — iduciary structure, and stakeholder standing. These conditions do not guarantee success — ul governance. Institu- tions can satis — y all six and still per — orm badly. But without them, governance either cannot — orm at all or cannot persist in a way that law can meaning — ully recognize and evaluate. Taken together, these — our chapters do three things. They estab- lish a de — inition o — governance with enough precision to travel across doctrinal — ields. They demonstrate why prior legal vocabular- ies have not — ully captured governance as a general legal phenome- non. And they identi — y the legal conditions that make governance possible without con — using those conditions with governance suc- cess. Part II builds on this — oundation by developing the book’s the- oretical account o — governance as a club good. Part III then applies that account across institutions where legal rules predictably enable, discipline, or degrade governance. Part I is there — ore the book’s vocabulary-building section, but it is not merely preliminary. The chapters in this Part establish the on- tology on which every later claim depends. I — the book is right, legal analysis has been missing a variable it regularly relies on but rarely names with precision. Part I is where that variable becomes visible. Chapter 1: What Is Governance? People say governance all the time. They might mean something

ancy, but they rarely de — ine it. Lacking shared vocabulary, it’s hard to talk about governance, per se. Instead, we speak about speci — ic applications o — governance: how corporations should be held accountable, how trading net- works sustain commercial cooperation without courts, how univer- sities exercise enormous public power without public accountabil- ity, how digital plat — orms govern hundreds o — millions o — users through terms o — service and algorithmic en — orcement. Rarely, i —

ever, has an ontology o

“governance” been used to resolve a debate about how the law allows or requires shared power. Yet there are common threads o — governance question in all o —

the above disputes. In all o

these settings, the legal question that matters most is not whether a particular rule was violated or whether a particular right was in — ringed. It is whether the institu- tion that manages the underlying shared problem is capable o — doing so, and whether law is helping or quietly destroying that capacity. The key question is whether law promotes governance. Courts and lawyers routinely miss this question. Three well- documented examples show what missing it costs. 12 Law and Governance

In 1959, the New York Stock Exchange ordered its member

irms to cut all private wire connections with Harold Silver, a non- member broker. No charges. No hearing. No explanation. Silver’s business was destroyed. When he sued under the Sherman Act, the Supreme Court held that exchange expulsion decisions are subject to antitrust scrutiny and require notice, an explanation o — charges, and an opportunity to be heard be — ore a member can be excluded.4 The Court was en — orcing a procedural norm rooted in due process and antitrust policy. What the Court did not ask was what the ruling would do to the exchange’s governance architecture. The Securities Acts Amendments o — 1975, which — ollowed Silver directly, required that all exchange disciplinary actions satis — y administrative due pro- cess requirements and that all exchange rule changes receive prior SEC approval be — ore taking e —

ect.5 The exchange, which had gov- erned itsel —

or decades as a sel — -regulating institution with swi — t, in — ormal disciplinary authority, was trans — ormed into a supervised administrative body whose every en — orcement action must survive multi-level procedural review. Paul Mahoney documented what this trans — ormation cost: the exchange’s capacity to discipline members quickly and credibly, the very — eature that sustained market integrity, was systematically eroded by the procedural apparatus that Silver initiated.6 Indeed, the

ederal government directly responded to this incident o — common law causing a governance crisis by passing statutory re — orm that e — -

ectively reversed Silver.7 Chapter 10 traces the — ull sixty-year arc — rom Silver through de- mutualization and the current constitutional crisis: the book’s most detailed case study in what repeated legal intervention does to a gov- ernance institution over time. Error! Use the Home tab to apply Heading 1 to the text that you want to appear here. 13

 In 2019, the Department o ---  Education levied a $4.5 million Clery Act  --- ine against Michigan State University, then the largest in his- tory,  --- ollowing the Larry Nassar abuse scandal. A special investiga- tive report had identi --- ied the core structural de --- ect: athletic medical sta ---

reported to the athletic department rather than to an independ- ent medical authority, which meant complaints about Nassar were evaluated by the very department whose institutional reputation de- pended on suppressing them.8 The — ederal remedy required revised policies, new Title IX coordinators, mandatory training, and retro- spective complaint review. No — ederal mandate required restructur- ing the reporting line between athletic medicine and the athletic de- partment. No mandate required re — orming the Board o — Trustees. Within a year o — the settlement, the interim president appointed to signal a new governance era was himsel —

orced to resign a — ter making statements publicly dismissive o — the victims’ experiences.9 The compliance mandate had addressed the outputs o — governance

ailure. The architecture that produced it was untouched. In ten daycare centers in Hai — a, Israel, a research team intro- duced a — ine — or parents who arrived late to pick up their children. The expected result was less tardiness. The observed result was the opposite: late pickups roughly dou- bled and continued at the elevated rate even a — ter the — ine was re- moved.10 The — ine had converted a social norm, parental guilt about inconveniencing teachers, into a market transaction. Once parents understood that late pickup cost a — ee, they — elt entitled to be late. They were purchasing a service. The moral obligation that had gov- erned their behavior was not supplemented by the external penalty. It was destroyed by it, permanently. These three cases share a structure. In each, a legal rule acted on an institution that was governing a shared problem through its own 14 Law and Governance

internal mechanisms: the exchange governing market integrity through rapid discipline, the university governing institutional con- duct through internal reporting authority, the daycare community governing parental behavior through social obligation. In each case, the legal intervention addressed a visible problem. In each case, it degraded or destroyed the internal governance mechanism in the process. The exchange became a regulated entity. The university adopted policies that satis — ied legal requirements without altering the structural — ailure. The daycare lost a norm that money could not buy back. The pattern is not incidental. It — ollows — rom a structural gap in legal analysis. Legal analysis is trained to see disputes, rights, and parties. It is not trained to see governance, the ongoing institutional architecture through which groups manage shared problems over time. When law acts on a governance institution without seeing it as a governance institution, the governance consequences are invis- ible at the moment o — decision and costly in the aggregate. The im- mediate case gets resolved. The institution gets damaged. Without an ontology o — governance, institutional harm is invis- ible. The present analysis supplies the missing — ramework. It opens the — oundational question: what is governance? It closes with meth- ods that make harms to governance cognizable to law. Error! Use the Home tab to apply Heading 1 to the text that you want to appear here. 15

Chapter 1. The Architecture o

Governance Governance needs a de — inition that is precise enough to do legal work and general enough to travel across institutional settings. The concept must distinguish governance — rom adjacent phenomena — regulation, management, adjudication, spontaneous social order — while identi — ying the structural — eatures that all governance arrange- ments share. What — ollows builds that de — inition in — our steps: it names the problem with the current vocabulary, states the de — ini- tion this book proposes, identi — ies the minimum elements that any governance institution requires, and draws the boundaries that sep- arate governance — rom what it is not.

The word “governance” names everything and there — ore nothing Be — ore answering that question, legal scholarship’s silence on the de — inition problem requires explanation. The di —

iculty is not that scholars have ignored governance. Governance appears throughout the legal literature, in corporate law, administrative law, commons scholarship, organizational theory, and political science. The di —

i- culty is that in each o — these — ields the word names something di — -

erent, and those di —

erences have never been reconciled into a shared legal de — inition. When corporate governance scholars use the term, they typi- cally mean the mechanisms allocating authority between sharehold- ers and boards: voting rights, — iduciary duties, the business judg- ment rule. 11 When administrative law scholars use it, they o — ten mean collaborative and experimentalist approaches to regulation, governance as the alternative to command-and-control rulemak- ing.12 When political scientists use it, the term may encompass any 16 Law and Governance

system o

authoritative rule, up to and including government itsel — .13 When commons scholars use it, the term re — ers to the internal rules and en — orcement mechanisms through which a community man- ages a shared resource. 14 When a corporate lawyer says “govern- ance” at a board meeting, she may mean something more modest: the documents and procedures required to maintain de — ensible de- cision-making and avoid liability. Each o — these usages is coherent within its own — ield. Together, the usages do not add up to a concept. The — ragmentation is not merely a terminological inconven- ience. It has consequences — or legal analysis. A legal scholar analyz- ing corporate governance asks whether shareholder participation in board selection is adequate. A legal scholar analyzing administrative governance asks whether agency rulemaking is su —

iciently collabo- rative. A legal scholar analyzing commons governance asks whether in — ormal norms are sel — -en — orcing. None o — these questions is wrong within its — ield. But the three scholars are analyzing di —

erent ob- jects, using the same word, and producing scholarship that cannot speak to each other’s conclusions. And none o — them is asking the question this book asks: what does law do to governance as such, to the organized system through which a group manages a shared problem, whatever the institutional — orm, whatever the — ield? That question requires a de — inition that travels across — ields. No existing de — inition does. This book’s central claim depends on governance being a dis- tinct legal object with identi — iable elements, speci — ic legal conditions o — existence, and characteristic vulnerabilities to legal degradation. I — governance can mean anything that looks like organized collective action, the claim that legal rules can enable or degrade governance Error! Use the Home tab to apply Heading 1 to the text that you want to appear here. 17

becomes untestable. Without a stable de

inition, legal analysis can- not identi — y which rules a —

ect governance, cannot assess how they a —

ect governance, and cannot determine whether the e —

ect is ben- e — icial or harm — ul. We cannot judge the law at all.

Governance is the organized system by which a group manages a shared problem over time Governance requires no charter, board, statute, or state recognition o — any kind. Governance requires an architecture capable o — sustain- ing collective management o — a shared problem through time. Each element o — the de — inition speci — ies something that distinguishes governance — rom adjacent phenomena that law routinely con — uses with governance, with consequences examined throughout this book. Organized means deliberate institutional structure. Spontaneous social order is not governance, though social order may precede governance and support it once governance exists. A crowd that col- lectively avoids an obstacle has produced order. An institution that has decided how to make collective decisions, how to monitor com- pliance, how to sanction de — ection, and how to revise its rules has produced governance. Because governance rests on deliberate struc- ture rather than on spontaneous coordination, it is subject to design and capable o — analysis in ways that spontaneous order is not. Delib- erate structure can also be deliberately damaged: a legal rule that dis- mantles an institution’s exclusion mechanism or disables its adjust- ment procedure degrades governance in a way that undermines the whole system, not merely one part o — it. A spontaneous order would route around the damage. A governance institution, depending as it does on the integrity o — its architecture, cannot. 18 Law and Governance

 System means a set o ---  connected mechanisms  --- unctioning to- gether, not a collection o ---  separately analyzable rules. Rules alone are not governance. A rule without a monitoring mechanism is an aspiration. Monitoring without a sanction mechanism is observa- tion. Sanctions imposed without recognized decision-making pro- cedures are arbitrary punishment. And decision-making procedures that cannot be revised produce rigidity rather than order. Govern- ance requires all  --- our mechanisms  --- unctioning in relation to one an- other. Treating governance as a collection o ---  separable parts, evalu- ating the decision-making procedure without asking whether the sanction mechanism is intact, or evaluating the sanction without asking whether the monitoring  --- unction is adequate, produces legal analysis that misses the institution's actual vulnerabilities.
 By which a group manages establishes that governance is collec- tive in a speci --- ic sense. The group may be a two-member limited liability company, a  --- ishing community o ---  a hundred households, or a global trading network spanning dozens o ---  countries. What mat- ters is that management o ---  the shared problem runs through mech- anisms the group has established and to which the group is account- able. Regulation, the imposition o ---  rules by an authority external to those it governs, operates di ---

erently: the regulatory authority stands outside the group, prescribes rules — or the group’s conduct, and the group’s members neither established the authority nor can unilaterally revise the requirements. Management in the commer- cial sense operates di —

erently too: a superior directs subordinates within an established hierarchy, and the direction runs downward rather than through collective mechanisms the group itsel — controls. Governance, regulation, and management can all operate simulta- Error! Use the Home tab to apply Heading 1 to the text that you want to appear here. 19

neously on the same institution, and o

ten do. But they are distin- guishable in structure, and law treats them di —

erently — or reasons the distinctions explain. A shared problem gives governance its purpose and de — ines its constituency. A governance system with nothing to solve is a — or- mality, and — ormalities without — unction tend not to survive. More importantly — or legal analysis, the shared problem identi — ies who should have standing to challenge governance decisions, who bears the costs o — governance — ailure, and whose interests a court should weigh when asked to intervene in a governance dispute. Courts that resolve governance disputes without identi — ying the shared prob- lem, treating a network-member expulsion, — or example, as a dis- pute between two parties over a contract, systematically misidenti — y the relevant constituency and there — ore systematically misapply le- gal remedies. Part III o — this book examines that pattern across mul- tiple doctrinal — ields. Over time distinguishes governance — rom a contract and — rom a one-time act o — collective decision-making. A contract allocates rights and obligations between parties to a transaction. A vote re- solves a particular question at a particular moment. Governance cre- ates an ongoing system — or managing a category o — recurring prob- lems, and governance persists as long as the shared problem persists. The temporal dimension generates the requirement — or an adjust- ment mechanism, creates the vulnerability to institutional decay, and explains why legal rules that appear to solve a narrow problem in the present can have e —

ects on governance capacity that become visible only over time. Courts that optimize — or the immediate dis- pute and ignore the institution’s temporal dimension are not merely missing context. They are imposing costs on — uture members o — the institution who had no party in the litigation. 20 Law and Governance

Governance requires

our elements, and the absence o — any one compromises the institution’s capacity to govern This — ramework identi — ies — our elements as the minimum architec- ture o — governance. When any one is absent or has been disabled, the institution may persist through in — ormal means, through histor- ical momentum, or through the unchecked authority o — whoever holds power within it. An institution missing one o — the — our ele- ments lacks the structural capacity to sustain cooperation when the conditions that test cooperation arise: when stakes are high enough to make de — ection attractive, when membership turns over and new members have no stake in prior understandings, or when external pressures challenge established rules. Decision-making. Governance requires a recognized proce- dure — or determining what collective action requires. The key word is recognized: the procedure must be one that members o — the insti- tution regard as authoritative, so that decisions made through the procedure bind the institution even when individual members dis- agree with the outcome. Rules can exist without any recognized process — or making, revising, or authorizing them, and when they do, the institution has no way to respond to members who exploit gaps in existing rules, no way to adapt when the rules prove inade- quate, and no way to resolve disputes that the rules do not clearly cover. In a public company, the decision-making — unction runs through board resolutions and shareholder votes governed by cor- porate statutes and the company’s charter. In the Diamond Dealers Club, the decision-making — unction runs through the club’s internal arbitration system and its council o — senior members. In a — ishing community, decision-making authority may run through seasonal Error! Use the Home tab to apply Heading 1 to the text that you want to appear here. 21

meetings at which those who

ish the waters renegotiate territories and quotas. The institutional — orm varies enormously. Without some recognized procedure, the institution cannot adapt to new cir- cumstances, cannot resolve internal disputes authoritatively, and cannot respond to members who — ind and exploit gaps. Such an in- stitution has rules but not governance. Monitoring. Rules are meaningless without in — ormation about compliance. Governance requires a system to observe member be- havior, detect de — ection — rom institutional norms, and generate the in — ormation on which sanctions depend. Monitoring is a second- order collective action problem: the bene — its o — monitoring, know- ing who is complying and who is not, are shared across all members o — the institution, but the costs — all on whoever does the watching. Because the bene — its are shared while the costs are private, monitor- ing will be systematically undersupplied i — le — t to individual initia- tive. Each member has an incentive to expect that monitoring costs will — all on others. Governance addresses this problem by institu- tionalizing the monitoring — unction: assigning the monitoring role, structuring the monitoring process, and sustaining it through insti- tutional resources rather than individual initiative.15 Courts and regulators who impose external monitoring require- ments on private governance networks sometimes assume that ex- ternal oversight supplements and improves internal monitoring. External monitoring can do that. But external monitoring can also crowd out internal monitoring by signaling to members that their own oversight e —

orts are no longer trusted or necessary, thereby weakening the voluntary compliance on which private governance depends. The Hai — a daycare study illustrates the mechanism at its most vivid: parents who had complied with pickup norms out o — so- cial obligation began treating the — ine as a price — or extended care, 22 Law and Governance

and when the

ine was removed the norm did not return.16 Elinor Ostrom, Roy Gardner, and James Walker documented the same dy- namic in controlled experiments with common-pool resource users: groups that governed themselves through communication and mu- tual monitoring achieved signi — icantly better outcomes than those subjected to external regulation with imper — ect en — orcement, be- cause the external rules displaced the trust and reciprocity that had sustained voluntary cooperation. 17 Whether a particular external monitoring requirement strengthens or weakens governance de- pends on understanding monitoring as a collective good that the in- stitution must actively maintain, not a procedural check that exter- nal oversight can substitute — or at will. Sanctions. When monitoring detects a breach o — institutional norms, governance requires the capacity to impose costs on the vi- olator. Two properties determine whether a sanction mechanism can — unction. First, credibility: the sanction must actually be applied when monitoring detects a violation, not merely threatened and then — orgiven. A governance system in which detected violations are systematically overlooked, or punished only selectively, gives mem- bers an incentive to test the institution’s limits rather than comply with its rules. Second, graduation: the sanction must be calibrated to the severity o — the breach and the circumstances o — the violation. A system that responds to all violations with identical penalties can- not distinguish innocent error — rom strategic de — ection. Innocent error and strategic de — ection are di —

erent problems that require di — -

erent responses, and a governance system that treats them identi- cally will produce the wrong response to both. In private governance institutions, the paradigm sanction is ex- clusion. Private institutions generally lack the state’s monopoly on legitimate physical — orce: they cannot imprison violators or seize Error! Use the Home tab to apply Heading 1 to the text that you want to appear here. 23

property through legal process. What private institutions can do, and what most private governance systems depend upon, is deny the violator access to the bene — its o — membership. The expelled diamond merchant loses access to the trading — loor. The expelled student loses the credential. Exclusion converts membership into a stake that members hold in good standing and — or — eit through misconduct. The credible threat o —

or — eiture is what makes compliance rational

or members who might otherwise de — ect. Excludability is there — ore not merely a — eature o — governance but a structural condition o — it. An institution that cannot meaning — ully exclude de — ectors cannot sustain the sanction mechanism govern- ance requires, which means monitoring and decision-making — unc- tions operate without consequence. Courts and legislatures that weaken the excludability o — private governance institutions, by re- quiring reinstatement o — expelled members, imposing liability — or exclusion decisions, or treating expulsion as a tort, are not adjusting internal rules. They are disabling the sanction mechanism and thereby degrading the institution’s governance capacity as a whole. The governance implications o — judicial intervention in exclusion decisions are examined in detail in Chapters 7 and 10. Adjustment. Governance must be capable o — revising itsel — . Shared problems change as the conditions that generate them change: markets shi — t, technologies create new opportunities — or de-

ection, membership turns over, and members — ind new ways to ex- ploit gaps in existing rules. A governance system that cannot modi — y its own parameters will eventually — ail, not because the original rules were wrong when written but because those rules are no longer ad- equate to the problem the institution exists to solve. An institution that cannot adjust is an institution that can only decay. 24 Law and Governance

 Adjustment has two distinct dimensions. The  --- irst is the capac- ity to revise the rules that govern member conduct: to update spe- ci --- ic requirements, close loopholes, and calibrate sanctions to changed circumstances. The second, and more  --- undamental, is the capacity to revise the governance system itsel --- : to change how deci- sions are made, how monitoring is organized, how sanctions are ap- plied, and who belongs to the institution. An institution with the

irst capacity but not the second can respond to immediate problems but not to the structural — ailures that cause those problems to recur. Ostrom — ound that the ability o — a —

ected parties to modi — y the rules they live under is among the design principles most strongly associ- ated with long-term institutional survival.18 The reason is not mys- terious: a governance system that cannot be changed by those sub- ject to it will eventually lose the legitimacy on which voluntary com- pliance depends. Members who cannot change rules they regard as inadequate will work around those rules instead, and as more mem- bers do so, the system loses both e —

ectiveness and the social support that en — orcement requires. The absence o — any meaning — ul adjust- ment mechanism is there — ore one o — the most reliable predictors o —

governance

ailure, and, as Chapter 12 argues, it is the speci — ic — ailure at the center o — the university governance crisis examined in that chapter. Three levels o — adjustment capacity can be distinguished, each producing di —

erent governance resilience. The — irst level is reactive adjustment: the institution corrects identi — ied — ailures a — ter the — act. The Diamond Dealers Club operated primarily at this level, modi-

ying arbitration procedures and sanctions when speci — ic disputes revealed gaps in existing rules.19 The second level is periodic adjust- ment: the institution schedules regular review cycles that evaluate governance adequacy independent o — any particular — ailure. The Error! Use the Home tab to apply Heading 1 to the text that you want to appear here. 25

NYSE’s post-1975 governance operated at this level, with Section 19(b) o — the Securities Acts Amendments o — 1975 requiring SROs to

ile proposed rule changes with the SEC — or mandatory review be-

ore implementation.20 The third level is adaptive adjustment: the institution maintains continuous monitoring with threshold-trig- gered revision, modi — ying governance parameters automatically when speci — ied conditions are met. Decentralized autonomous or- ganizations provide the clearest contemporary example, with gov- ernance protocols that trigger revision proposals when on-chain metrics cross prede — ined thresholds.21 Higher adjustment levels do not supersede lower ones. An institution with adaptive capacity still requires the ability to correct speci — ic — ailures reactively. The levels are cumulative. An institution that operates only at the reactive level will survive longer than one with no adjustment capacity at all, but it will lag behind evolving problems rather than anticipating them. An institution with periodic review will catch some problems ear- lier, but may be too slow where conditions change between review cycles. The level o — adjustment capacity an institution needs depends on the rate at which the shared problem it addresses changes and the cost o — governance — ailure during the interval between adjustment events.

Output-based

unctionality criteria Governance institutions succeed or — ail by whether they accomplish what they are designed to accomplish. This — ramework proposes

our output-based criteria that distinguish working governance

rom institutional — orm without substance. First, monitoring must generate conduct-governing in — ormation that decision-makers ac- 26 Law and Governance

tually use in their decisions, not in

ormation that documents de — ec- tion but is systematically ignored. Second, sanctioning must be ap- plied at a — requency consistent with the violation rate the institution detects, not selectively imposed or withheld based on political con- siderations internal to the institution. Third, decision-making must govern actual member conduct, not merely express — ormal pre — er- ences that members ignore in practice. Fourth, rules must change in response to documented circumstances in which the existing rules prove inadequate, not persist unchanged through repeated — ailures and shi — ting conditions. An institution satis — ying the — our — unctional elements — deci- sion-making, monitoring, sanctions, and adjustment — satis — ies the structural de — inition o — governance. An institution — ailing any o —

these output-based criteria — where monitoring produces unused in — ormation, sanctions are imposed inconsistently, decisions do not translate to conduct, or rules cannot adapt — is — ailing at governance regardless o — whether the structural elements are — ormally present. This distinction becomes the basis — or evaluating law by its e —

ects on governance, the task Chapter 8 develops through its two-step method. The output-based criteria are what Step 2 o — that method applies: does the governance institution satis — y these — unctionality requirements, and i — not, is law helping or hindering the institu- tion’s capacity to satis — y them?

What governance is not Four distinctions matter — or legal analysis. Misidenti — ying govern- ance leads directly to misapplying legal rules, because the legal

ramework appropriate to regulation is not appropriate to private sel — -governance, the — ramework appropriate to employment and Error! Use the Home tab to apply Heading 1 to the text that you want to appear here. 27

agency is not appropriate to a governing body, and the

ramework appropriate to judicial proceedings is not appropriate to governance decisions. Each o — these errors has doctrinal consequences that Part III o — this book traces across multiple — ields.

Governance is not regulation. Regulation is the imposition o — requirements by an authority external to those it regulates. The regulatory authority stands out- side the group and prescribes rules — or the group’s conduct; the group’s members did not establish the authority, did not design the rules, and cannot unilaterally revise the requirements. Governance operates — rom inside: a group manages its own shared problem through mechanisms the group’s members have established and can modi — y. A state environmental agency imposing emissions stand- ards on a trade association is regulating. The trade association man- aging its members’ compliance obligations through internal rules, monitoring, and sanctions is governing. Regulation and governance can coexist, and the state may regulate the conditions under which private governance operates, but the two are di —

erent in kind, and applying regulatory criteria to governance produces distortions. Le- gal analysis that demands democratic legitimacy, procedural due process, or non-discrimination compliance — rom a private trading network is applying standards calibrated — or state power to an insti- tution that holds no state power. Legal analysis that exempts a pri- vate governance institution — rom any scrutiny on the ground that it is merely sel — -regulatory ignores the public consequences that gov- ernance — ailure can produce. Neither approach is coherent. A — rame- work that distinguishes governance — rom regulation can identi — y which standards apply and why. 28 Law and Governance

Governance is not management. Management is the direction o — an organization’s resources and personnel by those holding hierarchical authority within that or- ganization. A chie — executive manages a corporation. A provost manages an academic program. A general counsel manages the legal department. In each case, superiors direct subordinates within an established hierarchy, and the legal — rameworks governing those re- lationships, employment law, agency law, — iduciary duty to the en- tity, are designed — or that vertical structure. A governing body does something di —

erent: it manages a shared problem on behal — o — a con- stituency that includes but is not limited to the organization’s em- ployees and o —

icers. The university board o — trustees governs the university, setting the institution’s mission, overseeing its — inances, and holding the administration accountable to the institution’s pur- poses. The provost manages the university’s academic operations, directing — aculty and sta —

toward institutional objectives. Con — lat- ing these roles leads courts to apply employment and agency doc- trines to governance decisions, asking whether a board member was acting within the scope o — employment, or whether a governance decision was a managerial prerogative, when the applicable — rame- work is the law o —

iduciary duty and institutional accountability to the governed constituency.

Governance is not adjudication. Adjudication is the resolution o — a speci — ic dispute between identi — ied parties by an authoritative third party. Courts adjudicate. Arbitrators adjudicate. Many governance institutions contain adju- dicative — unctions: internal arbitration systems, grievance proce- dures, discipline committees. Those — unctions resemble adjudica- Error! Use the Home tab to apply Heading 1 to the text that you want to appear here. 29

tion closely enough to generate con

usion. Courts that treat govern- ance institutions as adjudicative bodies tend to evaluate governance decisions by standards appropriate to judicial proceedings: Was the party given notice? Was there an opportunity to be heard? Was the decision-maker impartial? Was the outcome reasoned and ex- plained? These are necessary questions about an adjudication. For a governance decision, they are necessary but not su —

icient, and sometimes they are not even the most important questions. Gov- ernance is ongoing where adjudication is episodic, preventive where adjudication is remedial, prospective where adjudication is retrospective. A governance institution required to satis — y the — ull apparatus o — judicial process be — ore every sanction, exclusion, or rule revision would be administratively unable to govern. The in- stitutional cost o — applying adjudicative procedure to every govern- ance action would consume the resources and attention the govern- ance system depends on. The legal challenge is to identi — y which governance decisions warrant adjudicative scrutiny and which should be evaluated on governance terms, a question Chapter 9 ad- dresses directly.

Governance is not spontaneous social order. Governance — requently grows — rom spontaneous social order and continues to depend on it, which is why this is the hardest dis- tinction to maintain. Ellickson’s study o — Shasta County cattle ranchers showed that close-knit communities develop in — ormal norms resolving disputes e —

iciently without legal intervention, through mutual monitoring, gossip, and graduated social pressure.22 Those norms produce e —

ects resembling governance e —

ects: they allocate entitlements, en — orce expectations, and sustain cooperation across a community o — interdependent actors. Ellickson’s cattle 30 Law and Governance

ranchers rarely used the

ormal legal system even when — ormal legal rules would have supported their claims, because the in — ormal social order resolved their disputes more quickly, more cheaply, and more accurately. What the Shasta County ranchers lack is the institutional struc- ture governance requires. Their social order operates through indi- vidual sel — -help and community social pressure, without any recog- nized collective decision-making procedure — or establishing or modi — ying the norms governing cattle trespass. Individual ranchers observe violations that happen to occur in their vicinity, but no in- stitution has the responsibility or resources to monitor compliance systematically. Violations are met with in — ormal grumbling, recip- rocal sel — -help, and reputation e —

ects that vary with the in — ormal social standing o — the parties. The community is sel — -ordering. It is not sel — -governing. The distinction matters — or legal analysis in two ways. First, courts and regulators that engage with a spontaneous social order as though it were a governance institution, imposing procedural re- quirements, demanding records o — deliberation, treating the com- munity’s in — ormal norms as equivalent to enacted rules, will disrupt the in — ormal social order without creating governance in its place. Second, courts and regulators that treat a genuine governance insti- tution as though it were merely a spontaneous social order, accord- ing it no more protection than the in — ormal understandings o —

neighbors, will

ail to account — or the institution’s legal conditions o — existence and the public consequences o — its — ailure. Drawing the line between social order and governance at the — our minimum ele- ments allows each to receive the legal treatment its structure war- rants. Error! Use the Home tab to apply Heading 1 to the text that you want to appear here. 31

Governance does not require

ormal legal recognition Governance can be highly — ormalized: incorporated, publicly subsi- dized, licensed by the state, and mapped by statute. Governance can also be entirely in — ormal, operating in the shadow o — law or deliber- ately opting out o — legal en — orcement. Legal status and governance architecture are related but distinct. Formal legal status does not guarantee governance, and in — ormal operation does not preclude it. The Diamond Dealers Club illustrates the — irst point. The club has — ormal legal status as an incorporated organization. Its govern- ance operates largely outside the state legal system: members who have disputes submit those disputes to internal arbitration rather than to courts, and using the state court system to resolve what the club regards as an internal matter is itsel — treated as a violation o —

institutional norms, subject to sanction. 23 The club’s governance

unctions because members value access to the trading network and accept the consequences o — exclusion, not because the state provides a backstop to the club’s rules. State recognition o — the club as a legal entity makes it easier — or the club to hold property and enter con- tracts, but the governance that makes the club valuable operates in- dependently o — that recognition. Maine’s lobster — ishing communities illustrate the second point. Acheson’s documentation o — these communities — ound governance operating with no — ormal legal structure at all: no written bylaws, no — ormal tribunal, no articles o — incorporation.24 Territory bound- aries are established and en — orced through graduated sanctions: ver- bal warnings between — ishers escalate to inter — erence with gear, cut- ting o — trap lines, and ultimately exclusion — rom productive — ishing territory. The ostracism mechanism is communicated through so- cial networks spanning the entire Maine coast, so that a — isher ex- cluded — rom one harbor community — inds those sanctions — ollowed 32 Law and Governance

to other harbors. Both the Diamond Dealers Club and the Maine lobster communities solve the same problem, sustaining coopera- tion and deterring de — ection among members who share access to a valuable resource, through the same architectural solution: a credi- ble threat o — exclusion communicated through networks that make the sanction e —

ectively universal. The di —

erence between Manhat- tan diamond exchanges and Maine — ishing communities is institu- tional — ormality. The governance structure is the same. Decentralized autonomous organizations present the most re- cent version o — this problem. Four American states have enacted statutes granting legal personality to DAOs, conditioning that recognition on a minimum threshold o — human governance partic- ipation.25 These statutes reveal something important about the rela- tionship between governance and legal recognition. Be — ore those statutes existed, DAOs already had governance: decision-making through token-holder votes recorded on a blockchain, monitoring through the public and permanent transaction record that block- chains provide, sanctions through automated smart contract execu- tion that imposes costs on accounts that violate protocol rules, and adjustment through governance proposals subject to token-holder vote. The governance existed be — ore legal recognition. The state statutes recognized a governance architecture that was already — unc- tioning and imposed human-governance requirements as condi- tions o — that recognition. The state decided whether and how to ac- commodate governance that already existed. It did not create the governance. Legal scholars who look — or governance by looking — or legal

orm, — or the Delaware charter, the registered nonpro — it, the li- censed exchange, the statutory authorization, will miss governance institutions that have not sought or obtained — ormal recognition. Error! Use the Home tab to apply Heading 1 to the text that you want to appear here. 33

Legal scholars who equate

ormal legal structure with governance will mistake the presence o —

ormal structure — or the presence o —

governance where governance has

ailed, and will miss the absence o — governance behind an elaborate — acade o — legal — orm. A university with a sel — -perpetuating board o — trustees, detailed bylaws, tax ex- emption, and — ederal accreditation may have less — unctional govern- ance than a — ishing community with none o — those things, i — the uni- versity’s — ormal structures have lost the capacity to make meaning — ul collective decisions, en — orce institutional norms, impose costs on vi- olators, or adapt to changed circumstances. The de — inition is analytical, not evaluative. It identi — ies govern- ance by its architecture, not by whether the governance is good or bad. Governance institutions can be captured by dominant — actions and turned against the interests o — those the institutions were cre- ated to serve. Several examples in this book involve precisely that

ailure: governance architecture designed to produce inclusive insti- tutions, captured by organized — actions, producing systematic exclu- sion as a structural output.26 Governance can be coercive. Govern- ance can sustain norms that serve insiders at substantial cost to out- siders. Governance can — ail in ways that harm the members it gov- erns and the public that depends on the institution’s proper — unc- tioning. Identi — ying an institution as a governance institution does not answer whether its governance is well-designed or badly-de- signed, serving its constituency or exploiting it, entitled to legal pro- tection or subject to legal correction. Those questions require a method, developed in Chapter 8, — or evaluating law by what law does to governance. The de — inition in this chapter is the prerequisite to that method: once governance is identi — ied by its — our elements, the evaluative questions become precise. Is the governance system 34 Law and Governance

managing the right shared problem? Does adjustment remain re- sponsive to changing conditions? Are the sanctions proportionate to the violations they address? Is the institution producing bene — its

or members and positive e —

ects — or outsiders, or bene — its — or insid- ers and costs — or everyone else? Those questions are what this book answers. The task o — legal analysis is not to — ind the state’s stamp o — approval. It is to — ind the architecture. Chapter 2: Languages o — Governance Legal scholarship currently lacks a governance vocabulary capable o — doing cross- — ield analytical work. No legal scholar has produced a de — inition o — governance that identi — ies governance wherever it ex- ists, applies consistent minimum elements, and generates a method

or evaluating law by what it does to governance, regardless o — in- stitutional — orm or legal context. This — oundational gap matters ur- gently because law acts on governance institutions in every doctri- nal — ield, yet legal analysis in each — ield uses governance vocabularies designed — or di —

erent institutional problems. The consequence is systematic: courts and legislators degrade governance institutions precisely when they attempt to improve them, because they are an- alyzing governance through — ield-speci — ic concepts that cannot identi — y what governance is as a general institutional object. This chapter surveys the existing vocabularies to establish why the gap exists and why the cross- — ield analytical work this book undertakes cannot proceed without solving it. Legal scholars have produced a vast literature on governance. Corporate law has its governance vocabulary, built around boards,

iduciary duties, and the separation o — ownership — rom control. Ad- ministrative law has its governance vocabulary, built around collab- 36 Law and Governance

oration, participation, and the

ailure o — command-and-control reg- ulation. Political science has several governance vocabularies, built around the state, networks, and international order. Commons scholarship has its governance vocabulary, built around collective action and institutional design. Organizational theory has its gov- ernance vocabulary, built around transaction costs and the bounda- ries o — the — irm. Nonpro — it law has its governance vocabulary, built around charitable missions and — iduciary accountability. Network scholarship has its governance vocabulary, built around reputation, exclusion, and social en — orcement. Each o — these vocabularies is well-developed within its — ield. Each is calibrated to that — ield’s central institutional problem. And because those problems di —

er — rom one another — the manager- shareholder agency problem is not the same as the collective-action problem o — a — ishing community, which is not the same as the legit- imacy problem o — an international regulatory body — a vocabulary calibrated to solve one problem cannot be straight — orwardly applied to the others. When a corporate lawyer reaches — or the concept o —

governance to analyze a

ishing community’s territorial en — orce- ment system, the vocabulary — ails. When a commons scholar reaches

or Ostrom’s design principles to analyze a public company’s board structure, the vocabulary — ails. When an administrative law scholar reaches — or the new governance — ramework to analyze a merchant network’s private arbitration system, the vocabulary — ails. The observation is not, by itsel — , a criticism o — any — ield. Fields develop concepts appropriate to their objects o — study. The criticism is directed at a speci — ic consequence: legal scholarship currently lacks a governance vocabulary capable o — doing cross- — ield analytical work. The consequence — or legal practice is severe. Judicial review o — governance decisions proceeds without a coherent legal concept Error! Use the Home tab to apply Heading 1 to the text that you want to appear here. 37

o

what governance is, what it requires to — unction, or what law does when it acts on governance institutions. The result is that law sys- tematically degrades the very institutions the law intends to im- prove. Two o — the existing vocabularies — Ostrom’s commons govern- ance and Williamson’s transaction cost — ramework — deserve — uller engagement than a survey chapter permits. Both supply — ounda- tional concepts — or the theory this book develops, and both present interpretive di —

iculties speci — ic to the legal analysis o — governance that require care — ul treatment. Chapter 3 provides that treatment. This chapter positions those literatures within the broader land- scape and signals the questions Chapter 3 will answer. 38 Law and Governance

Chapter 2. Governance Babel Seven established literatures have each produced governance vocab- ularies calibrated to their own institutional domains. Surveying them reveals a recurring pattern: each vocabulary captures real — ea- tures o — governance within its — ield but none supplies the cross-cut- ting analytical — ramework that legal analysis requires. The survey proceeds through corporate governance, new governance, political science governance, commons governance, transaction cost eco- nomics, nonpro — it governance, and network governance — in each case identi — ying what the vocabulary accomplishes and where it — alls short.

Corporate governance: sophisticated but entity-bound Corporate governance scholarship is the most developed govern- ance vocabulary in legal scholarship. It has produced canonical de — - initions, institutional economics — oundations, sustained empirical research, and a body o — case law that engages governance questions with real analytical precision. It is also the vocabulary most likely to mislead a lawyer reaching — or a general governance concept, because its sophistication conceals its scope limitation: the entire apparatus is calibrated to the corporation and to the speci — ic principal-agent problem that arises when pro — essional managers control assets on behal — o — dispersed investors. The — oundational problem was identi — ied by Adol — A. Berle, Jr. and Gardiner C. Means: when pro — essional managers direct a cor- poration whose shares are dispersed among many investors, the in- terests o — managers and owners can diverge, and the owners lack the in — ormation and coordination capacity to discipline management directly.27The governance question, in this — raming, is how to align Error! Use the Home tab to apply Heading 1 to the text that you want to appear here. 39

managerial conduct with owner interests. That question organizes every major de — inition that — ollowed. Andrei Shlei — er and Robert W. Vishny de — ine corporate govern- ance as dealing with “the ways in which suppliers o —

inance to cor- porations assure themselves o — getting a return on their invest- ment.”28 The de — inition is tightly calibrated to the investor-return problem. Governance, — or Shlei — er and Vishny, is the set o — mecha- nisms — boards, voting rights, — iduciary duties, legal protections — that prevent managers and controlling shareholders — rom expropri- ating value — rom minority shareholders and creditors. Margaret M. Blair o —

ers a broader de — inition, treating corporate governance as “the whole set o — legal, cultural, and institutional arrangements that determine what publicly traded corporations can do, who controls them, how that control is exercised, and how the risks and returns

rom the activities they undertake are allocated.”29 Blair expanded the scope beyond — inanciers to include all holders o —

irm-speci — ic investments, including employees, but her de — inition still presup- poses publicly traded corporations. The OECD Principles o — Corpo- rate Governance describe governance as involving “a set o — relation- ships between a company’s management, its board, its shareholders and other stakeholders” that “provides the structure through which the objectives o — the company are set, and the means o — attaining those objectives and monitoring per — ormance are determined.” 30 The OECD de — inition is broader than investor protection, but still presupposes a company with a board-management-shareholder triad, and speci — ies no sanction mechanism and no adjustment — unc- tion. The board primacy literature locates governance authority in the board itsel — rather than in shareholders or markets. Stephen M. Bainbridge argues that the board — unctions not as a shareholder 40 Law and Governance

agent but as a centralized decision-making body whose authority is justi — ied by the need — or administrative e —

iciency in complex organ- izations, “a sort o — Platonic guardian” whose independence is the condition o — e —

ective governance. 31 Leo E. Strine, Jr. argues that Delaware law ultimately requires directors to serve stockholder wel-

are as their ultimate end, though they may consider other interests as means. 32 Both positions are detailed accounts o — governance within the corporate — orm. Neither is designed to be applied any- where else. The shareholder-stakeholder debate reveals a deeper problem with the corporate governance vocabulary: it cannot separate the descriptive question o — what governance is — rom the normative question o — what governance is — or. Shareholder primacy holds that governance exists to solve the agency problem between dispersed owners and pro — essional managers.33 Stakeholder theory holds that governance exists to coordinate and balance the interests o — all a — -

ected parties.34 Both sides de — ine governance by re — erence to what it should accomplish. Neither provides a de — inition capable o — iden- ti — ying governance as an institutional object independently o — its purpose. The most detailed internal challenge — Margaret M. Blair and Lynn A. Stout’s team production theory, which argues the board

unctions as a “mediating hierarch” balancing all team members’ claims — remains equally — orm-bound: mediating hierarchs require boards, and boards require corporations.35 Delaware courts de — ine governance operationally through the standards o — review they apply to board conduct. DGCL § 141(a) pro- vides that “the business and a —

airs o — every corporation organized under this chapter shall be managed by or under the direction o — a board o — directors.”36 Chancellor Allen identi — ied the corporation’s conceptual instability — an oscillation between property and social- Error! Use the Home tab to apply Heading 1 to the text that you want to appear here. 41

entity conceptions that Delaware has never resolved.37 Speci

ic de- cisions de — ine governance elements without abstracting them: the shareholder — ranchise as the “ideological underpinning” o — corporate legitimacy in Blasius,38 and the distinction between legal authority and equitable constraint in Schnell.39 The case law is rich. It is also entirely internal to corporate law. Corporate governance vocabulary presupposes a speci — ic entity

orm, a speci — ic principal-agent problem, and a speci — ic legal context. A commons — Ostrom’s irrigation systems, a Maine lobster com- munity — has no board, no shareholders, and no — iduciary duties. The governance problem is collective action among co-equal users. A merchant network has no single entity, no centralized manage- ment, and no hierarchy. A university has shared governance among

aculty, administration, and trustees that does not map onto the board-shareholder dyad. Gerald F. Davis has argued that the corpo- rate governance vocabulary is becoming inadequate even — or its original domain as the publicly traded corporation becomes a less central organizational — orm.40 A vocabulary that struggles with its original object cannot generate a cross- — ield analytical — ramework. Legal analysis that begins and ends with corporate governance con- cepts will not see governance — ailure in institutions that do not look like corporations — which is to say, it will not see most governance

ailure.

New governance: prescriptive program without analytical — oundation The administrative and new governance literature has produced the most extensive discussion o — governance within legal scholarship. It 42 Law and Governance

has also produced the most systematic con

usion between govern- ance as a policy program and governance as an institutional concept. Jody Freeman’s collaborative governance model de — ines govern- ance through — ive — eatures: a problem-solving orientation, direct stakeholder participation beyond notice-and-comment, provisional and revisable solutions, shared responsibility — or — ormulating and implementing standards, and negotiation as the central mecha- nism.41 These are — eatures o — a regulatory design — a prescription

or how the administrative state should operate — not an identi — i- cation o — what governance is as an institutional object. Orly Lobel provides the most thorough mapping o — the govern- ance turn across legal — ields, identi — ying eight — eatures that distin- guish the new governance model — rom the New Deal regulatory model: participation and partnership, collaboration, decentraliza- tion and diversity, integration across policy domains, — lexibility and non-coerciveness, provisionality and dynamic learning, legal plural- ism, and pragmatism and outcome-orientation.42 Lobel’s contribu- tion is important: she demonstrates that something called govern- ance has become a paradigm in legal scholarship. But she does not de — ine governance as an institutional object. For Lobel, governance is “the new approach to regulation” — a collection o — desirable de- sign principles. Her survey cannot identi — y governance in a mer- chant network, a university, or a DAO because it is tethered to the administrative-state context and to a prescriptive re — orm agenda. Michael C. Dor — and Charles F. Sabel’s democratic experimen- talism describes a governance mechanism operating through itera- tion: set provisional — ramework goals, experiment locally, bench- mark per — ormance, revise rules based on learning, repeat.43 Charles F. Sabel and William H. Simon extend the program to the adminis- trative state, identi — ying mandatory reporting, benchmarking, and Error! Use the Home tab to apply Heading 1 to the text that you want to appear here. 43

periodic revision as the core mechanisms.44 Both contributions are care — ully speci — ied institutional designs. They tell us what good gov- ernance looks like in the public-regulatory context. They do not ad- dress the minimum architectural requirements — or governance to exist as a — unctional institution at all. Ian Ayres and John Braithwaite theorize the relationship be- tween state regulation and private sel — -regulation through the en-

orcement pyramid, proposing graduated en — orcement — rom persua- sion and sel — -regulation at the base through escalating civil and criminal penalties to license revocation at the apex.45 The en — orce- ment pyramid is a regulatory technique — a prescription — or how the state should deploy en — orcement tools. It tells us nothing about what governance is as an institutional object. The sharpest internal criticism comes — rom Bradley C. Kark- kainen. He argues that the new governance literature lumps — unda- mentally di —

erent institutional phenomena under a single label, lacks analytical precision such that governance becomes a term o —

approval

or pre — erred regulatory design, presents prescriptive claims as though they were descriptive, and de — ines governance so broadly that nearly any regulatory re — orm quali — ies. 46 Karkkainen calls — or disaggregating the concept into analyzable component mechanisms. His critique directly anticipates this book’s project. Jo- anne Scott and David M. Trubek make a related observation: gov- ernance has become an “umbrella concept” that “risk[s] losing its analytical power i — it re — ers to everything.”47 The new governance literature con — lates two di —

erent things the word “governance” can name. One is governance as institutional architecture: the organized system by which a group manages a shared problem over time. The other is governance as regulatory 44 Law and Governance

technique: a style o

regulatory design characterized by participa- tion, — lexibility, and learning. Freeman and Lobel use the second meaning. This book uses the — irst. That con — lation means the new governance vocabulary cannot identi — y governance in a private merchant guild, recognize governance — ailure in a university, or dis- tinguish a — unctioning governance institution — rom one that has col- lapsed behind a — acade o — policy compliance. Legal analysis built on new governance concepts is designed to evaluate regulatory pro- grams, not to examine institutions.

Political science governance: either too broad or too state-centric Political science has produced the most varied governance vocabu- lary, but that vocabulary occupies one o — two positions that are both problematic — or legal analysis. It is either so capacious that it names any ordering phenomenon at all, or speci — ically calibrated to the problem o — authority exercised beyond or without the state — use — ul

or international relations but unable to address the legal conditions under which institutions exist and — ail. James N. Rosenau’s — oundational de — inition, introduced to ex- plain international order in the absence o — global government, de-

ines governance as “activities backed by shared goals that may or may not derive — rom legal and — ormally prescribed responsibilities and that do not necessarily rely on police powers to overcome de — i- ance and attain compliance.” 48 Rosenau adds that governance “is thus a more encompassing phenomenon than government. It em- braces governmental institutions, but it also subsumes in — ormal, non-governmental mechanisms.” 49 A de — inition that encompasses everything — rom the UN Security Council to a neighborhood watch Error! Use the Home tab to apply Heading 1 to the text that you want to appear here. 45

cannot identi

y what is distinctive about institutions with the spe- ci — ic decision-making, monitoring, sanction, and adjustment archi- tecture this book analyzes. R.A.W. Rhodes catalogued six distinct usages o — governance within political science alone: governance as the minimal state, as corporate governance, as New Public Management, as “good gov- ernance,” as a socio-cybernetic system, and as sel — -organizing net- works.50 Rhodes’s own pre — erred de — inition — “sel — -organizing, in- terorganizational networks characterized by interdependence, re- source exchange, rules o — the game and signi — icant autonomy — rom the state” — describes network-based public administration but presupposes networks o — organizations rather than the individual- member governance this book analyzes. Six contested meanings within a single discipline is itsel — evidence that the term has not achieved de — initional stability. Gerry Stoker — rames governance as a theoretical — ramework through — ive propositions: governance re — ers to actors drawn — rom beyond government; governance identi — ies blurring o — boundaries between public and private sectors; governance identi — ies power de- pendence in collective action; governance is about autonomous sel — - governing networks; and governance recognizes that government steers rather than commands. 51 These propositions characterize a trans — ormation in how the state exercises authority. They do not identi — y what institutional architecture makes governance work at any level o — scale. Jon Pierre and B. Guy Peters de — ine governance as “sustaining co-ordination and coherence among a wide variety o — actors with di —

erent purposes and objectives.”52 This is a macro-political con- cept about state steering capacity. The World Bank’s de — inition — 46 Law and Governance

governance as “the manner in which power is exercised in the man- agement o — a country’s economic and social resources — or develop- ment”53 — is calibrated entirely to country-level state per — ormance. The Worldwide Governance Indicators that operationalize it meas- ure voice and accountability, political stability, government e —

ec- tiveness, regulatory quality, rule o — law, and control o — corruption.54 These indicators tell us nothing about the governance o — a trading network, a pro — essional association, or a university. Gary Marks’ multi-level governance — ramework distinguishes general-purpose, non-overlapping jurisdictions — traditional — ed- eralism — — rom task-speci — ic, overlapping, — lexible jurisdictions.55 Multi-level governance tells us where authority sits within — ederal and quasi- — ederal structures. It does not say what institutional archi- tecture makes governance — unction at any given level. Political science governance vocabulary does not address the le- gal conditions o — institutional existence: what legal rules must be in place — or governance to — orm and persist. It does not address the le- gal consequences o — governance — ailure: what happens when an in- stitution’s monitoring — unction is disabled or its sanction mecha- nism is judicially dismantled. The concept o — a governance void — central to this book’s analysis o — universities and nonpro — its — has no home in any o — the — rameworks surveyed above. A board o — trus- tees that has lost the capacity to en — orce institutional norms against organized — actions within the institution is not a recognizable ana- lytical object in any o — these — rameworks. Legal analysis that draws its governance concepts — rom political science cannot identi — y that

ailure, let alone prescribe a remedy. Error! Use the Home tab to apply Heading 1 to the text that you want to appear here. 47

Commons governance:

oundational but law-silent (developed in Chapter 3) Elinor Ostrom’s commons governance vocabulary comes closest to this book’s de — inition among all the existing literatures. Her eight design principles — or long-enduring common-pool resource institu- tions — clearly de — ined boundaries, congruence between rules and local conditions, collective-choice arrangements, monitoring, grad- uated sanctions, con — lict-resolution mechanisms, minimal recogni- tion o — rights to organize, and nested enterprises — map instruc- tively onto the book’s — our elements. 56 The collective-choice ar- rangement corresponds to decision-making; monitoring corre- sponds to monitoring; graduated sanctions and con — lict-resolution mechanisms correspond to sanctions; and the combination o — col- lective-choice arrangements and nested enterprises corresponds to adjustment. The Institutional Analysis and Development — ramework, devel- oped most — ully in Ostrom’s Understanding Institutional Diversity, identi — ies governance as the product o — nested rule systems at three levels: operational rules, collective-choice rules, and constitutional- choice rules.57 This three-tier structure recognizes that governance includes rules about rules — a — orm o — institutional re — lexivity with direct implications — or the adjustment — unction this book identi — ies. The critical gap is not in the — ramework’s conception o — govern- ance but in its treatment o — law. Ostrom’s design principles treat le- gal recognition — “minimal recognition o — rights to organize” — as Design Principle 7, acknowledging that external authorities can override community sel — -governance. But the — ramework does not theorize how speci — ic legal rules constitute or degrade governance. Law appears as one category o — exogenous variable constraining the action arena, not as a distinct constitutive — orce whose internal logic 48 Law and Governance

shapes which governance architectures can

orm and which cannot. Daniel H. Cole has identi — ied this gap explicitly, arguing that the IAD — ramework needs better integration with — ormal legal analy- sis. 58 Lee Anne Fennell has shown how property law structures commons governance in ways Ostrom’s — ramework underspeci-

ies.59 Chapter 3 develops the Ostrom engagement — ully, examining what the IAD — ramework contributes to the legal analysis o — gov- ernance and what the — ramework requires law to supply that it does not supply itsel — .

Williamson: transactional and dyadic (developed in Chapter 3) Oliver E. Williamson de — ines governance as “the means by which order is accomplished in a relation in which potential con — lict threatens to undo or upset opportunities to realize mutual gains.”60 His — ramework identi — ies three governance structures — markets, hierarchies, and hybrids — di —

erentiated by asset speci — icity, — re- quency o — transaction, and uncertainty.61 The key variable determin- ing e —

icient governance structure is asset speci — icity: the degree to which an investment is specialized to a particular transaction.62 Williamson’s — our levels o — social analysis locate governance at Level 3: below the institutional environment o —

ormal legal rules and above resource allocation and neoclassical price theory.63 Gov- ernance at Level 3 takes law as a given constraint and asks which governance structure minimizes transaction costs within it. The critical gap is the unit o — analysis. Williamson’s — ramework analyzes individual transactions between identi — ied parties. A trad- ing network, a university, or a pro — essional association is not a Error! Use the Home tab to apply Heading 1 to the text that you want to appear here. 49

transaction. It is an institution sustaining complex, multiplex coop- eration among many participants over time. Moreover, governance

or Williamson is a cost to be minimized, not a capacity to be built. His comparative statics ask which structure is e —

icient at a given moment. The book’s — ourth element — adjustment — has no Wil- liamsonian analogue because the — ramework is essentially static: it selects among pre-existing governance structures rather than ana- lyzing how governance — orms, how it is maintained, and how it can be destroyed. Walter W. Powell argued that networks represent a governance — orm that Williamson’s market-hierarchy spectrum cannot accommodate, because networks operate through reciproc- ity and trust rather than price or authority.64 Mark Granovetter ar- gued that Williamson’s — ramework is “undersocialized,” — ailing to account — or the social relations that constitute and constrain institu- tional behavior.65 Chapter 3 develops the Williamson engagement — ully, examin- ing the transaction cost — ramework’s contributions to the legal anal- ysis o — governance and where that — ramework requires supplemen- tation.

Nonpro

it governance: — orm-bound accountability law The nonpro — it governance literature addresses a distinctive ac- countability problem: how to en — orce mission compliance in organ- izations without residual claimants, market discipline, or electoral accountability. The vocabulary is precise and practically important within its domain. It does not generate a cross- — ield governance concept, because its key constructs are arti — acts o — the nonpro — it legal

orm rather than generalizable — eatures o — governance as such. 50 Law and Governance

 Henry B. Hansmann established the  --- oundational  --- ramework by identi --- ying the nondistribution constraint — the prohibition on distributing net earnings to those in control — as the de --- ining struc- tural  --- eature o ---  nonpro --- its.66 Hansmann argued that the constraint serves as a consumer-protection device in markets characterized by severe in --- ormation asymmetry, but simultaneously weakens gov- ernance by eliminating the pro --- it motive that would otherwise dis- cipline management. The constraint creates a governance tradeo ---

: reduced exploitation risk at the cost o — eliminating exit mechanisms (no tradable shares) and voice mechanisms (no shareholder — ran- chise). The leading casebook treatment — rames nonpro — it governance through — iduciary duties — care, loyalty, and the distinctive duty o —

obedience — and the role o

state attorneys general as primary en-

orcers. 67 The duty o — obedience, requiring directors to act in ac- cordance with the charitable mission, is the — iduciary obligation spe- ci — ic to nonpro — it governance and has no corporate analogue.68 Cor- porations are not bound to a — ixed mission; directors have broad strategic discretion. In nonpro — it governance, the mission is the or- ganization’s reason — or existing, and the duty constrains — iduciaries

rom deviating even when deviation would be — inancially advanta- geous. The governance problem that this literature grapples with most persistently is what this book, drawing on prior work, calls the sov- ereign charity problem: elite nonpro — its — major research universi- ties, hospital systems, mega- — oundations — exercise enormous pub- lic power through private governance structures designed — or char- itable organizations rather than — or institutions o — civic conse- quence. Sel — -perpetuating boards select their own successors, — ace no electoral accountability, and are insulated — rom market discipline Error! Use the Home tab to apply Heading 1 to the text that you want to appear here. 51

by large endowments. State attorneys general have en

orcement au- thority but rarely exercise it at the scale the problem requires.69 Eve- lyn Brody’s care — ul work has identi — ied multiple governance gaps in nonpro — it law: doctrinal con — usion between trust law and nonpro — it corporation law, en — orcement gaps created by under-resourced at- torneys general, standing gaps that prevent stakeholders — rom chal- lenging governance — ailures, and systematic disconnection between legal requirements and actual board behavior.70 Nonpro — it governance vocabulary is organized around a speci — ic legal — orm con — ronting a speci — ic accountability gap. Its key con- structs — the nondistribution constraint, the duty o — obedience, at- torney general en — orcement, sel — -perpetuating boards — are arti-

acts o — nonpro — it law. They have no application to corporate gov- ernance, which has residual claimants and market discipline; to commons governance, which has no board, no charter, and no at- torney general; or to network governance, which lacks centralized decision-making and — ormal legal hierarchy. A — ishing community and a major research university can — ail at governance in structurally identical ways — both may lack meaning — ul adjustment mecha- nisms, both may have captured their own sanctioning — unctions — but nonpro — it governance vocabulary can only see the university’s

ailure as a nonpro — it law problem. It cannot identi — y the shared gov- ernance pathology.

Network governance: law is invisible by design The network governance literature provides the most empirically grounded and analytically precise account o — private governance in 52 Law and Governance

action. It also, in its most in

luential — ormulations, explicitly ex- cludes law — rom the de — inition — not by oversight but by theoretical choice. Walter W. Powell argued that networks are a categorically dis- tinct organizational — orm, “neither market nor hierarchy,” charac- terized by reciprocal, trust-based, mutually supportive actions among interdependent but autonomous actors. 71 Joel M. Podolny and Karen L. Page de — ine a network — orm as “any collection o — actors (N ≥ 2) that pursue repeated, enduring exchange relations with one another and, at the same time, lack a legitimate organizational au- thority to arbitrate and resolve disputes.”72 The de — ining — eature — absence o — legitimate organizational authority — builds the exclu- sion o —

ormal legal mechanisms into the concept itsel — . Candace Jones, William S. Hesterly, and Stephen P. Borgatti provide the most comprehensive organizational-theory de — inition o — network governance: “a select, persistent, and structured set o —

autonomous

irms (as well as nonpro — it agencies) engaged in creat- ing products or services based on implicit and open-ended contracts to adapt to environmental contingencies and to coordinate and sa — e- guard exchanges. These contracts are socially — not legally — en-

orced.”73 Jones and colleagues identi — y — our social mechanisms con- stituting network governance: restricted access, macroculture, col- lective sanctions, and reputation. All — our are in — ormal and social. None is legal. The exclusion o — law is not a gap in the theory. It is the theory’s premise. In legal scholarship, Lisa Bernstein provides the most detailed engagement with network governance, de — ining it as “the collective

orce o — reputation-based, non-legal sanctions — lowing — rom a — irm’s position within a network o — interconnected — irms.” 74 Bernstein’s empirical work — on diamond dealers, grain and cotton traders, and Error! Use the Home tab to apply Heading 1 to the text that you want to appear here. 53

Midwestern industrial procurement — demonstrates that network topology does governance work that relational contracts between individual parties cannot explain: in — ormation about de — ection spreads through the network, sanctions are imposed collectively, and network membership itsel — disciplines behavior even in the ab- sence o —

ormal legal en — orcement. Matthew Jennejohn has shown empirically that — irms’ positions in innovation networks predict their contract designs, with network-embedded — irms relying on network-level governance rather than detailed — ormal agreements.75 Keith G. Provan and Patrick Kenis identi — y three structural modes o — network governance — shared governance by all mem- bers, lead-organization governance by one dominant member, and governance by a separate Network Administrative Organization — demonstrating that network governance is not a single institutional

orm but a — amily o —

orms with di —

erent vulnerabilities to legal deg- radation.76 Gerald R. Salancik, noting the — ield’s descriptive richness and theoretical poverty, called — or a theory capable o — explaining how network structure produces governance outcomes rather than merely describing structural patterns.77 That theory has not arrived, in part because the — ield’s — oundational premise — that network governance operates outside legal en — orcement — prevents the — ield

rom asking what law does to network governance, which is the question whose answer would explain when network governance thrives and when it degrades. Network governance vocabulary describes governance through network topology and relational contracting but excludes law by de — initional choice. When a court reviews an expelled merchant’s claim, as in Silver v. New York Stock Exchange,78 the network govern- ance literature provides no analytical tools — or evaluating the inter- action. When antitrust doctrine treats collective network sanctions 54 Law and Governance

as a group boycott, as in Associated Press v. United States,79 the network governance vocabulary cannot ask whether the legal intervention helped or damaged the institution’s governance capacity. Barak D. Richman has documented the erosion o — the diamond network’s pri- vate governance system under pressure — rom market and legal changes, but his account is descriptive and historical rather than providing a legal method — or evaluating what produced the ero- sion.80 The description o — what happened is not a substitute — or a

ramework that explains what law did to governance.

The gap that all seven literatures share Seven developed literatures, each producing a governance vocabu- lary calibrated to its own institutional problem. The calibration is not a de — ect — it is what makes each vocabulary use — ul within its

ield. The de — ect is in what calibration — orecloses: the capacity to an- alyze governance as a general institutional object whose legal con- ditions o — existence and legal vulnerabilities are identi — iable regard- less o — institutional — orm. Corporate governance cannot identi — y governance problems in institutions whose structure does not resemble the board-share- holder dyad. Its analytical categories — the principal-agent problem,

iduciary duty, the business judgment rule — are designed — or the corporation and cannot be extended without distortion to institu- tions organized on di —

erent principles. New governance con — lates governance as institutional architec- ture with governance as regulatory program. Because the vocabu- lary is prescriptive — organized around what good regulatory de- Error! Use the Home tab to apply Heading 1 to the text that you want to appear here. 55

sign looks like — it cannot identi

y governance — ailure in institu- tions that have adopted all the — ormal markers o — good governance while the underlying architecture has collapsed. Political science governance is either so broad that it encom- passes any coordinated activity, or so state-centric that it cannot ad- dress institutions organized and governed outside the state’s — ormal authority structures. It has no concept o — a legal condition o — insti- tutional existence, and no method — or evaluating what law does to a governance institution’s capacity to — unction. Commons governance, in the depth appropriate to a survey chapter, treats law as one exogenous variable among many rather than as a constitutive — orce that determines which governance ar- chitectures can — orm. The IAD — ramework provides a power — ul vo- cabulary — or institutional analysis; what it requires Chapter 3 to sup- ply is an account o — law’s speci — ic role in constituting and degrading the governance conditions the — ramework describes. Transaction cost economics de — ines governance as a cost to be minimized in transaction-level relationships. It cannot accommo- date the institution as distinct — rom the transaction, cannot explain how governance — orms and adjusts over time, and leaves the book’s

ourth element — adjustment — without any theoretical home. Nonpro — it governance is organized around a speci — ic legal — orm con — ronting a speci — ic accountability problem. Its analytical con- structs do not generalize: they illuminate the university’s govern- ance — ailure only as a nonpro — it law problem, which means they can- not identi — y the same — ailure when it appears in a private trading network, a pro — essional association, or a DAO. Network governance excludes law by de — initional choice. The exclusion is theoretically motivated — network governance oper- 56 Law and Governance

ates through social rather than legal mechanisms — but the conse- quence is that the literature cannot ask what law does when it acts on network governance institutions, which is the most practically important governance question that courts and legislators — ace. The pattern across all seven literatures re — lects three shared structural limitations. First, every — ield-speci — ic de — inition presup- poses a particular institutional context — an entity — orm, an organ- izational problem, a regulatory setting — that makes the de — inition inapplicable to institutions organized on di —

erent principles. Sec- ond, no — ield-speci — ic de — inition identi — ies the minimum architec- tural elements required — or governance to exist as a — unctional sys- tem: decision-making, monitoring, sanctions, and adjustment. Third, and most consequentially, no — ield-speci — ic vocabulary treats governance as the distinct legal object on which law acts, the insti- tutional architecture that law can constitute, support, discipline, or destroy. No legal scholar has attempted to supply what all seven litera- tures lack. A thorough search o — legal scholarship — inds no prior work providing a general, cross- — ield de — inition o — governance as a legal concept that travels across corporate, administrative, com- mons, nonpro — it, network, and political contexts with speci — ied minimum elements. Orly Lobel comes closest by surveying the gov- ernance turn across multiple legal — ields, but she catalogs a paradigm shi — t rather than de — ining governance as an institutional object: — or Lobel, governance is the new approach to regulation, not an archi- tectural concept identi — ying what makes collective management o —

shared problems possible. 81 Gillian K. Had

ield and Barry R. Weingast develop a game-theoretic model o — legal order — the con- ditions under which normative systems become distinctively legal through coordination o — decentralized collective punishment — but Error! Use the Home tab to apply Heading 1 to the text that you want to appear here. 57

legal order and governance are adjacent rather than identical con- cepts, and their model addresses en — orcement coordination rather than governance architecture. 82 Gunther Teubner’s re — lexive law

ramework analyzes the nature o — law, not the nature o — governance, and the autopoietic theory underlying it treats law and governance as operationally closed systems that can only perturb rather than constitute each other — a theoretical commitment this book’s ac- count challenges.83 Edward Peter Stringham and Bruce L. Benson demonstrate that private governance creates economic and social order without government but do not construct a general legal de — - inition o — governance as a distinct institutional object.84 The legal pluralism literature — Tamanaha, Merry, and Berman — represents a related tradition that comes close to the present pro- ject but ultimately addresses a di —

erent question.85 Legal pluralism argues that non-state normative orders can constitute law, that the state has no monopoly on legality. This is an ontological claim about what law is. It does not provide a general de — inition o — governance as an institutional object, and it does not build a method — or evalu- ating what law does to governance. Legal pluralism and this book share an interest in non-state ordering, but they proceed — rom di — -

erent questions and toward di —

erent ends. Chapter 3 positions the legal pluralism literature within the intellectual lineage this book ex- tends.

What the survey requires The survey establishes one — inding with clarity: legal scholarship needs a governance vocabulary that does what none o — the existing

ield-speci — ic vocabularies does. That vocabulary must identi — y gov- ernance by its — unctional architecture rather than by its entity — orm. 58 Law and Governance

It must speci

y the minimum elements required — or governance to exist as a — unctional system. It must treat governance as the distinct legal object on which law acts, capable o — being constituted, disci- plined, or destroyed by legal rules. And it must apply consistently across institutional contexts, generating the same analytical ques- tions whether the institution in question is a corporation, a com- mons, a university, a trading network, or a DAO. Chapter 1 supplied that vocabulary. The de — inition — govern- ance is the organized system by which a group manages a shared problem over time, requiring decision-making, monitoring, sanc- tions, and adjustment — was built to — ill exactly the gap the survey reveals. Chapter 3 develops the theoretical — oundations — or the de — i- nition by engaging Ostrom and Williamson at the depth they de- serve, explaining what each — ramework contributes to the legal anal- ysis o — governance and where each requires supplementation by a legal method that neither supplies. Part II builds the original theory on those — oundations. Part III applies it. Chapter 3: Why Legal Theory Still Lacks a Governance Concept Chapter 2 surveyed how di —

erent — ields use the governance vocab- ulary and established that no existing usage provides the cross- — ield legal de — inition this book requires. Chapter 3 examines the intellec- tual traditions that came closest to supplying it — and explains pre- cisely why each stopped short. Consider two ways law can relate to an institution. In the — irst, the institution exists independently and law regulates it — rom the outside: a — irm operates, and securities law governs how the — irm must disclose in — ormation to investors. In the second, law does not merely regulate the institution; law makes the institution what it is. The corporation exists because corporate law created it. The board o — directors has authority because corporate statutes grant it. The

iduciary duty runs to shareholders because courts decided it did. Remove the legal structure and you do not have a corporation with

ewer regulatory burdens. You have no corporation at all. Scholars who study institutions in the second way use the word “constitutive” to describe law’s role: law does not regulate the insti- tution, law constitutes it. That is a power — ul insight, and several im- portant intellectual traditions have built substantial bodies o — work 60 Law and Governance

around it. This chapter examines those traditions, credits what each one achieved, and identi — ies precisely what each one le — t undone. What none o — these traditions supplied is what this book calls the missing ontology o — governance. “Ontology” in this context means a rigorous account o — what something — undamentally is: its basic nature, its minimum requirements, its conditions o — existence. The missing ontology o — governance is a precise legal account o —

what governance is as an institutional object, one that speci

ies its minimum elements, identi — ies the legal conditions under which governance can — orm and persist, and provides a method — or evalu- ating law by what it does to governance. This chapter names that gap explicitly and explains why it remains open. Error! Use the Home tab to apply Heading 1 to the text that you want to appear here. 61

Chapter 3. Governance Ontology Three intellectual traditions have come closest to supplying the missing governance concept, and understanding why each — ell short clari — ies the gap this book — ills. Legal institutionalism demonstrated that law constitutes major institutions but did not develop a general account o — governance itsel — . Commons’s institutional economics identi — ied legal relations as the building blocks o — institutional order but remained — ocused on bilateral transactions. Ostrom’s Institu- tional Analysis and Development — ramework produced the most complete governance architecture in the social sciences but treated law as exogenous to the analysis. Each tradition advanced the pro- ject; none completed it.

Legal institutionalism: law makes institutions, but which ones? Beginning in the 1970s, a body o — historical and political scholarship developed around the claim that law does not merely regulate Amer- ican political institutions. Law created them, shaped them, and con- tinuously remakes them. This body o — work became known as the American Political Development tradition, or APD. It is not primar- ily a law school movement — it grew largely in political science de- partments — but its central claim is one o — the most important — or legal analysis: you cannot understand how American institutions work without understanding what law made them. Morton J. Horwitz, a Harvard legal historian, established the

oundational argument. In The Trans — ormation o — American Law, 1780- 1860, Horwitz demonstrated that American courts did not simply apply pre-existing rules to commercial disputes. They actively re- shaped the rules o — property, contract, and tort in ways that created 62 Law and Governance

the institutional conditions

or industrial capitalism.86 The market was not a natural phenomenon that law then stepped in to regulate. Legal rules on water rights, on corporate liability, on the en — orcea- bility o — commercial contracts made the market’s structure possible in the — irst place. His sequel extended the analysis to the progressive and New Deal periods, showing how legal thought both shaped and was shaped by the rise o — corporate organization and the adminis- trative state.87 Howard Gillman brought the same argument into constitutional theory. In The Constitution Besieged, Gillman showed that the Su- preme Court’s controversial Lochner-era decisions were not simply the imposition o — economic pre — erences on constitutional doctrine. The Court was operating within a coherent constitutional tradition that distinguished legislation serving the general public interest

rom legislation bene — iting particular economic groups, and it used the due process clause to en — orce that distinction. Law was not ex- ternal to the constitutional order; it de — ined the categories through which institutional actors understood what they were permitted to do.88 George I. Lovell showed that statutory ambiguity is sometimes a deliberate institutional design. When legislators dra — t laws with imprecise language, they are sometimes doing so on purpose, dele- gating politically di —

icult decisions to courts. The legal — orm o — the statute determines which institution will resolve the underlying con — lict.89 Keith E. Whittington showed that constitutional mean- ing is built not only through judicial decisions but through the po- litical actions o — presidents, legislators, and agencies that construct institutions and practices in the spaces constitutional text leaves open.90 His companion study showed that judicial review’s authority over the other branches o — government was sustained not despite Error! Use the Home tab to apply Heading 1 to the text that you want to appear here. 63

legislative resistance but through political strategies that

ound ju- dicial de — erence use — ul.91 Karen Orren and Stephen Skowronek developed the — ield’s sharpest theoretical — ramework. At any given moment, they argued, American governance re — lects multiple layers o — institutional ar- rangements built at di —

erent historical periods with di —

erent pur- poses and o — ten inconsistent internal logics. A labor law — rom 1935, a civil rights statute — rom 1964, and a regulatory program — rom 1990 may all be operating simultaneously on the same institution, each embodying a di —

erent theory o — governance, and their interaction produces — riction and contradiction. Orren and Skowronek called this intercurrence. The — ield itsel — they de — ined as the study o — “du- rable shi — ts in governing authority”: changes in who has the author- ity to make binding decisions and how that authority is organized and exercised.92 Their later work showed how the accumulation o —

policy commitments across di

erent historical eras has produced a contemporary administrative state characterized by incoherence and structural rigidity.93 Rogers M. Smith showed that American citizenship law has been constituted by the interaction o — three distinct ideological tra- ditions: a liberal tradition emphasizing individual rights, a republi- can tradition emphasizing civic obligation, and an ascriptive hierar- chical tradition assigning di —

erent rights to di —

erent groups based on race, gender, and other characteristics. The interaction o — these three produced a legal landscape o — structured inequality that neither the liberal nor the republican tradition alone could explain. 94 His “multiple traditions” approach demonstrates that legal analysis o —

institutions must account

or the multiple, o — ten contradictory com- mitments that legal rules simultaneously express. 64 Law and Governance

Law Is Endogenous to Institutions The APD tradition demonstrated — our propositions this book depends on. Law is not external to institutions but shapes what in- stitutions are. Legal change is institutional change. Institutions re-

lect multiple historical periods operating simultaneously, not a sin- gle coherent design. And the meaning o — a legal rule depends on the institutional context in which it operates, so legal analysis must be historical.

Public Governance The APD tradition — ocused on political institutions: courts, leg- islatures, the constitutional order, the administrative state, citizen- ship regimes. These are all institutions created by or in direct rela- tionship with the state. The tradition has not extended its analysis systematically to private governance: trading networks, merchant communities, pro — essional associations, universities, nonpro — its, and the new category o — decentralized digital organizations. The tra- dition has not asked what legal conditions allow private institutions to govern themselves, what legal rules enable or degrade private governance capacity, or how law should be evaluated by its e —

ects on governance in those settings. The tradition has also not supplied a de — inition o — governance with speci — ied minimum elements. Orren and Skowronek de — ine governance as “durable shi — ts in governing authority,” which de- scribes a process o — institutional change rather than speci — ying what governance requires to exist and — unction. No major APD scholar has proposed a de — inition o — governance as an organized system re- quiring decision-making, monitoring, sanctions, and adjustment. None has proposed a method — or evaluating law by what it does to Error! Use the Home tab to apply Heading 1 to the text that you want to appear here. 65

governance as a general concept applicable across corporate, com- mons, network, and nonpro — it governance alike. The gap, stated precisely: the APD tradition established that law shapes what institutions are. It did not supply a legal de — inition o —

governance as a distinct object, minimum elements that de

inition implies, or a method — or evaluating law by its e —

ects on governance.

Commons’s Insight and Its Limits: Legal Relations Structure Institutions but Not Governance Itsel —

Be

ore the APD tradition, there was institutional economics. Its most important — igure — or this book’s purposes is John R. Commons, an economist active in the early twentieth century who built a the- ory o — institutions directly — rom legal — oundations. Commons’s most systematic statement appears in Institutional Economics: Its Place in Political Economy (Macmillan, 1934), with the- oretical — oundations in The Legal Foundations o — Capitalism (Macmil- lan, 1924) and the methodological statement “Institutional Econom- ics,” 21 Am. Econ. Rev. 648 (1931). Commons built his — ramework on the work o — Wesley New- comb Hoh — eld, a legal theorist who in the early twentieth century produced the most precise analysis o — legal relations available in American jurisprudence. Hoh — eld showed that what lawyers call “rights” are actually several di —

erent kinds o — legal relationships: a right in the strict sense (someone owes you a duty), a privilege or liberty (you have no duty yoursel — ), a power (you can change legal relationships), and an immunity (others cannot change your legal position). These distinctions matter because they identi — y exactly what legal protection someone has and against whom. Commons 66 Law and Governance

took Hoh

eld’s typology and used it to analyze economic institu- tions. For Commons, a transaction is not merely an exchange o — goods

or money. A transaction is a legal event involving the trans — er o —

legal rights, the assumption o

legal duties, and the exercise and waiver o — legal powers. Institutional Economics, at 58-65, 87-95. The “working rules” governing transactions, the rules determining who can do what to whom under what circumstances, are legal rules in Hoh — eld’s sense. They are legally en — orceable claims, duties, powers, and liabilities that constitute the structure o — economic relation- ships, not merely social conventions that legal en — orcement happens to back up. Commons built this — ramework through the concept o — the “go- ing concern”: the institution within which transactions take place. A going concern is a legal entity with its own rules, its own internal hierarchy, and its own relationship to the broader legal system. The Legal Foundations o — Capitalism, at 6-8, 138-45. The corporation, the trade union, the public utility, the government agency: each is a go- ing concern whose internal governance is constituted by legal work- ing rules and whose external relationships are constituted by its legal status relative to other going concerns and the state. The connection to this book is direct. What Commons called working rules are what this book calls the legal conditions o — gov- ernance. What he called going concerns are what this book calls governance institutions. Commons showed that legal relations shape what economic institutions are, not merely how they are reg- ulated. This book applies that insight to governance speci — ically: le- gal conditions determine whether governance institutions can

orm, what mechanisms they can employ, and whether they can per- sist over time. Error! Use the Home tab to apply Heading 1 to the text that you want to appear here. 67

Douglass C. North extended institutional economics in a direc- tion relevant here. Institutions, Institutional Change and Economic Per-

ormance (Cambridge University Press, 1990) distinguishes institu- tions — rom organizations: institutions are the rules o — the game, the

ormal rules, in — ormal constraints, and en — orcement characteristics that structure human interaction, while organizations are the play- ers who operate within those rules. 95 The distinction matters be- cause it recognizes that rules themselves require en — orcement. Mon- itoring and sanctions are not automatic — eatures o — institutional li — e but require deliberate institutional supply, which is exactly what this book’s account o — governance architecture requires. Geo —

rey M. Hodgson has been the most sustained critic o — the later institutional economics tradition that departed — rom Com- mons’s legal — oundations. In How Economics Forgot History (Routledge, 2001), Hodgson argues that Williamson and North stripped institutional economics o — its historical and legal speci — icity by treating institutions as e —

iciency-maximizing responses to trans- action costs rather than as historically contingent legal constructs.96 His critique is directly relevant. An approach that assumes surviving institutions are e —

icient cannot explain why legal rules that degrade governance capacity are adopted and maintained. A legal method — or evaluating law by its e —

ects on governance must accommodate the possibility that law systematically degrades governance without sel — - correcting. The lineage — rom Commons through North and Hodgson to this book is not without discontinuity. Each — igure pursued a di —

erent project. But the — oundational commitment links them: legal rela- tions shape what institutional structures are, not merely what con- straints they operate under. This book’s claim that law determines 68 Law and Governance

whether governance can

orm, persist, adapt, and remain accounta- ble is the governance-speci — ic application o — Commons’s insight.

Ostrom’s

ramework: the most complete governance architecture, without a role — or law Elinor Ostrom’s Institutional Analysis and Development — ramework — the IAD — ramework — is the most systematic account o — govern- ance architecture in the social sciences. The IAD — ramework is — oun- dational — or this book’s de — inition o — governance, and the tradition whose treatment o — law most clearly marks the gap this book — ills.

Institutional Analysis and Development (IAD) The IAD — ramework, most — ully stated in Understanding Institu- tional Diversity (Princeton University Press, 2005), analyzes govern- ance by examining what Ostrom calls an “action arena”: a structured social space in which actors interact to produce outcomes. Three categories o — background conditions shape the action arena: the physical nature o — the resource being managed, the characteristics o —

the people involved including their norms and knowledge, and the working rules that actually govern behavior. These background conditions, operating through the action arena, produce outcomes that — eed back to modi — y the conditions over time.97 The — ramework classi — ies rules into seven types, each corre- sponding to a distinct governance — unction: who can participate, what roles participants occupy, what outcomes are permitted, what actions are required or allowed, how individual choices produce col- lective decisions, what in — ormation must be shared, and what costs Error! Use the Home tab to apply Heading 1 to the text that you want to appear here. 69

and bene

its attach to what actions. Understanding Institutional Diver- sity, at 186-215. Together these rule types speci — y what a governance system needs to address i — it is to — unction. The — ramework’s most important — eature — or this book is its three-tier rule structure. Operational rules govern day-to-day be- havior. Collective-choice rules govern how operational rules are made and changed. Constitutional-choice rules govern how the col- lective-choice rules themselves are constituted. Governing the Com- mons, at 50-55. Governance, on this account, is not just a set o — rules. It includes rules about rules: the processes through which the insti- tution can revise its own structure. This is what makes the IAD

ramework a genuine theory o — governance rather than merely a de- scription o — social norms. The collective-choice level corresponds to this book’s decision-making element. Constitutional-choice rules correspond to the adjustment — unction, the capacity to revise the governance system itsel — .

IAD’s eight design principles Ostrom’s eight design principles — or long-enduring commons institutions — clearly de — ined boundaries, congruence between rules and local conditions, collective-choice arrangements, moni- toring, graduated sanctions, con — lict-resolution mechanisms, recog- nition o — rights to organize, and nested enterprises — are the most empirically tested account o — governance architecture available, val- idated across hundreds o —

ield sites spanning multiple continents and resource types.98 Michael Cox, Gwen Arnold, and Sergio Villa- mayor Tomás reviewed the evidence across a large sample o — com- mons institutions and con — irmed that the principles most strongly associated with e —

ective governance, particularly collective choice, 70 Law and Governance

monitoring, and graduated sanctions, show consistent relationships with institutional per — ormance.99 The mapping onto this book’s — our elements is instructive. Col- lective-choice arrangements (Principle 3) correspond to decision- making: those a —

ected by the rules can participate in modi — ying them. Monitoring (Principle 4) corresponds directly to monitoring. Graduated sanctions (Principle 5) and con — lict-resolution mecha- nisms (Principle 6) correspond to the sanctions — unction. Collec- tive-choice arrangements and nested enterprises (Principles 3 and 8) together correspond to adjustment.

Law Is Exogenous to IAD The IAD — ramework’s gap is not in its conception o — governance but in its treatment o — law. Law appears in the — ramework primarily as one subcategory o —

ormal rules within the broader category o —

working rules. Ostrom distinguishes working rules, the rules that actually govern behavior, — rom rules on paper, the — ormal legal rules that may or may not be en — orced in practice. This distinction is cru- cial — or empirical social science. For legal analysis, it creates a speci — ic problem: Ostrom’s — ramework treats law as one input within “rules- in-use” rather than as a constitutive — orce that shapes what govern- ance can be. This is the particular gap this book — ills. Where Ostrom’s — rame- work allows law to — unction as one among many exogenous varia- bles constraining the action arena, legal analysis must ask precisely how speci — ic legal doctrines and legal institutions determine whether the working rules Ostrom describes can — orm and persist. A property law regime that recognizes en — orceable exclusion creates Error! Use the Home tab to apply Heading 1 to the text that you want to appear here. 71

the conditions

or sanctions-based governance. A property law re- gime that prohibits exclusion disables it. A liability regime that treats governance decisions as compensable tort violations trans-

orms the character o — governance authority. Contract law that en-

orces arbitration agreements supplies the mechanism — or con — lict resolution that Ostrom identi — ies as essential. Contract law that voids arbitration agreements on public policy grounds removes it. The IAD — ramework documents these variations across institutions but cannot explain them because law sits outside the — ramework. Design Principle 7, the requirement that institutional arrange- ments receive recognition — rom external governmental authorities, comes closest to addressing law’s role. The principle states that without such recognition, governance institutions can be disrupted or destroyed by legal intervention. Governing the Commons, at 101. But the principle treats legal recognition as a binary condition, pre- sent or absent, rather than as something shaped by speci — ic legal rules in speci — ic ways. It does not theorize the mechanism by which di —

erent property rights regimes determine whether commons governance can — orm. It does not explain what happens when a court converts a governance institution’s exclusion mechanism

rom a categorical rule — you are expelled and the sanction stands — into a compensatory rule — you are expelled but can pay damages to remain — as Silver v. New York Stock Exchange did to the stock exchange’s disciplinary system. It does not analyze how standing rules determine who can challenge governance — ailures in court, which is the mechanism through which governance — ailure either becomes legally actionable or remains legally invisible. Daniel H. Cole has made this point precisely, arguing that the IAD — ramework needs better integration with — ormal legal analysis 72 Law and Governance

because the

ocus on working rules has le — t underdeveloped the anal- ysis o — how — ormal legal rules actually operate to structure govern- ance. Cole, The Varieties o — Comparative Institutional Analysis, 2013 Wis. L. Rev. 383, 400-02. Lee Anne Fennell demonstrated that property law structures commons governance in ways the IAD

ramework underspeci — ies: di —

erent property rights regimes pro- duce di —

erent institutional conditions — or commons governance, and analyzing those conditions requires legal tools the IAD — rame- work does not supply. Fennell, Ostrom’s Law: Property Rights in the Commons, 5 Int’l J. Commons 9, 11-20 (2011). Carol M. Rose sur- veyed the impact o — Ostrom’s work on American legal scholarship and — ound that legal scholars have consistently had to supplement the IAD — ramework with legal analysis that Ostrom hersel — did not provide. Rose, Ostrom and the Lawyers: The Impact o — Elinor Ostrom’s Work on American Legal Scholarship, 5 Int’l J. Commons 28 (2011).

IAD assumes—but does not theorize—law For this book’s purposes, the IAD — ramework requires legal analysis to answer three questions it cannot answer itsel — . The — irst concerns — ormation. When can governance institu- tions — orm? The IAD — ramework takes the existence o — a governance institution as its starting point and asks what design — eatures sustain it. The legal conditions under which governance can begin to exist are prior questions: what property rights does the law recognize, whether the law en — orces the exclusion mechanisms that make sanc- tioning credible, whether the law recognizes the institution’s au- thority to make and en — orce its own rules, whether the law gives standing to those who would challenge governance — ailures. These Error! Use the Home tab to apply Heading 1 to the text that you want to appear here. 73

are speci

ically legal questions whose answers vary across legal re- gimes in ways that explain variation in governance — ormation that the IAD — ramework documents but does not theorize. The second concerns degradation. How does law damage gov- ernance? The IAD — ramework acknowledges that external authori- ties can undermine governance but treats this as a simple binary: legal recognition is either present or absent. It does not analyze how speci — ic legal doctrines interact with speci — ic governance elements. When a court applies antitrust doctrine to a governance institu- tion’s exclusion mechanism, converting it — rom a categorical rule into a compensatory one, it is disabling a speci — ic governance ele- ment, the sanction mechanism, through a speci — ic legal mechanism, with predictable e —

ects on the institution’s capacity to sustain coop- eration. The IAD — ramework provides no tools — or analyzing this because it does not theorize how speci — ic legal doctrines interact with speci — ic governance — unctions. The third concerns evaluation. How should law be evaluated by its e —

ects on governance? The IAD — ramework describes what gov- ernance looks like when it works. This book asks a di —

erent ques- tion: given a legal rule, does it enable or degrade governance? The IAD — ramework does not supply an evaluative method because it is descriptive. This book’s method, identi — y the shared problem, iden- ti — y the governance institution, identi — y the legal conditions under which it exists, assess member bene — its and spillovers, analyze how law enables or degrades it, judge the law, requires the IAD — rame- work’s institutional vocabulary but deploys it toward a normative purpose the IAD — ramework was not designed to serve. 74 Law and Governance

Williamson’s transaction cost

ramework Oliver E. Williamson’s transaction cost — ramework is the most in-

luential account o — governance in economics and organizational theory. Its relationship to this book is complementary but limited. Williamson provides a detailed account o — why governance arrange- ments di —

er in their e —

iciency properties. He cannot address the governance o — groups managing shared problems over time, and he cannot analyze governance as a capacity that must be built and can be destroyed.

E

icient governance, one transaction at a time Williamson’s analysis proceeds — rom two behavioral assump- tions. First, actors have limited in — ormation-processing capacity and cannot write per — ectly complete contracts anticipating all — uture contingencies. Second, actors will pursue sel — -interest opportunisti- cally when they can — not just strategically, but through deception and guile — when the short-term gains outweigh the expected costs o — being caught. Together these create the transaction cost problem: parties to a transaction — ace uncertainty about each other’s behavior, and sa — eguarding investments against opportunistic exploitation is costly. The Economic Institutions o — Capitalism (Free Press, 1985), at 44-52. Governance structures are Williamson’s solution to this prob- lem. A governance structure is “the institutional matrix within which transactions are negotiated and executed.” Id. at 17. Three structures are available: markets governed by price signals and the threat o — exit, hierarchies governed by uni — ied ownership and ad- ministrative control, and hybrids combining market incentives with administrative coordination. Which structure is most e —

icient — or a Error! Use the Home tab to apply Heading 1 to the text that you want to appear here. 75

given transaction depends on how speci

ic the investment is to that particular relationship, how — requently the transaction recurs, and how much uncertainty surrounds it. Comparative Economic Organi- zation, 36 Admin. Sci. Q. 269, 277-83 (1991). Williamson located governance within a — our-level model o — so- cial analysis. At the highest level are deeply embedded social norms and values that change very slowly, over centuries. At the second level are the — ormal legal rules, property rights, contract en — orce- ment mechanisms, and regulatory — rameworks that — orm the insti- tutional environment. Governance structures occupy the third level, below the — ormal legal environment and above the individual trans- actions and resource allocation decisions o — the — ourth level. The New Institutional Economics, 38 J. Econ. Literature 595, 597 — ig.1 (2000). The critical — eature o — this architecture is that law sits at a di —

erent level — rom governance. Law at Level 2 creates the institutional en- vironment within which governance structures at Level 3 are se- lected. Governance takes law as a given constraint and asks which structure minimizes costs within it. This is the precise inverse o — the relationship this book describes: law does not merely constrain gov- ernance options; law shapes what governance institutions are.

Law Shapes Governance Design Three things Williamson provides are directly use — ul to this book. First, his account o — why governance structures di —

er. Not all institutions — ace the same cooperation problems, and understanding why di —

erent governance structures exist in speci — ic contexts re- quires the transaction cost logic Williamson articulates. This book draws on that logic in its analysis o — corporate governance in Chap- ter 11 and contract remedies in Chapter 10. 76 Law and Governance

Second, asset speci --- icity as a variable in governance design. When investments are highly speci --- ic to a particular relationship, parties become vulnerable to hold-up: a --- ter committing to a rela- tionship-speci --- ic investment, a party may  --- ind that its trading part- ner opportunistically renegotiates the terms, knowing that the in- vesting party has no viable alternative. Governance structures pro- tecting against hold-up are not merely e ---

iciency solutions but ena- bling conditions — or cooperation. This insight applies directly to network governance: the more speci — ic members’ investments in a trading network’s reputation, the more governance capacity is needed to protect those investments, and the more damaging it is when a court disables the network’s exclusion mechanism. Third, the concept o — credible commitment. Williamson shows that institutional mechanisms can make promises believable by rais- ing the cost o — breaking them. The credible threat o — exclusion — rom a trading network — unctions exactly this way. Judicial intervention converting the exclusion mechanism — rom a categorical rule into a damages remedy degrades the credibility o — that commitment, which degrades the governance structure’s capacity to sustain coop- eration. Chapter 10 analyzes this mechanism at length.

Static Micro-Level Analysis The unit o — analysis is the transaction between identi — ied parties, not the institution managing shared problems among many mem- bers. A diamond dealers’ club, a — ishing community, or a university is not a transaction. Each is an institution with many members man- aging a shared problem over time. Williamson’s — ramework ana- lyzes governance structures at a single moment in time and asks which minimizes costs — or a given transaction with given character- istics. It cannot ask whether an institution has the — our elements Error! Use the Home tab to apply Heading 1 to the text that you want to appear here. 77

governance requires, whether it is managing the right problem, or whether its governance capacity is being degraded by legal rules that appear to solve a narrower problem. The — ramework is essentially static. It selects among pre-existing governance structures rather than analyzing how governance — orms, how it adapts over time, or how it can be permanently damaged by legal interventions. The adjustment — unction this book identi — ies has no Williamsonian analogue. Governance institutions managing shared problems over time must adapt their rules and their govern- ance mechanisms to changed circumstances. Williamson’s — rame- work cannot address this temporal dimension. By treating govern- ance as a choice among static structures rather than as an institu- tional capacity that must be built, maintained, and adapted, the

ramework — ails to see a critical gap: institutions that satis — y the — or- mal e —

iciency criteria — or a governance structure at time t may lack the capacity to adjust when conditions change at time t+1. The tem- poral dimension is what’s missing. Williamson analyzes governance at the transaction level but cannot address institutions managing shared problems over time — and the institutions this book analyzes are precisely those that must sustain governance across years, dec- ades, or centuries. Law is an external constraint in Williamson’s — ramework, not a

orce that shapes what governance institutions are. Walter W. Pow- ell demonstrated that networks represent a governance — orm that Williamson’s market-hierarchy — ramework cannot accommodate because networks operate through reciprocity and trust rather than price or authority. 100 Mark Granovetter argued that Williamson’s

ramework — ails to account — or the social relations that constitute and constrain institutional behavior, treating actors as isolated eco- nomic agents when they are in — act embedded in social structures 78 Law and Governance

that shape what they can do and what they want.101 Both critiques point to the same structural limitation: the transaction cost — rame- work cannot accommodate the institutional complexity that gov- ernance o — shared problems requires.

The missing ontology Four traditions, each approaching governance — rom a di —

erent an- gle. The APD tradition showed that law shapes what institutions are. Commons showed that legal relations structure economic insti- tutions speci — ically. The IAD — ramework supplied the most complete architectural account o — what governance requires to — unction. Wil- liamson showed why governance arrangements di —

er in e —

iciency and how commitment mechanisms sustain cooperation. What none o — them supplied is a legal de — inition o — governance with speci — ied minimum elements, a set o — legal conditions under which governance can exist and to which law gives or denies access, and a method — or evaluating law by its e —

ects on governance as a distinct legal object. The gap is uni — orm despite the traditions’ di —

erences. The APD tradition established that law shapes institutions but con — ined its analysis to political institutions, never asking what legal conditions allow private governance to — orm and persist. Commons supplied the — oundational insight about legal relations and institutional struc- ture but applied it to the analysis o — capitalism rather than develop- ing a governance-speci — ic de — inition or evaluative method. The IAD

ramework supplied the architectural vocabulary, the elements gov- ernance requires, but treated law as an external constraint rather than a — orce shaping what governance institutions are, leaving the legal analysis o — governance — ormation and degradation undone. Error! Use the Home tab to apply Heading 1 to the text that you want to appear here. 79

Williamson supplied the transaction cost logic explaining why gov- ernance structures di —

er but cannot accommodate institutions man- aging shared problems among many members, cannot analyze gov- ernance dynamically, and treats law as a constraint rather than something that constitutes governance architecture. The convergence on a single gap across — our developed tradi- tions makes two things credible simultaneously. The gap is real: it remains open a — ter decades o — serious work on all — our — ronts. And the book’s de — inition is built on solid — oundations: the constitutive insight — rom APD, Commons’s legal-relational theory, the IAD

ramework’s architectural vocabulary, and Williamson’s transaction cost logic are the materials — rom which the book’s analysis is con- structed. Chapter 4 develops the legal conditions under which govern- ance can exist: the permissions, recognitions, en — orceabilities, and accountability structures that law must supply — or governance to

orm and persist. Part II builds the theory o — how law enables and degrades governance. Part III applies that theory across the doctrinal

ields where the analysis does its most important work. Chapter 4: Legal Conditions — or Governance Law and governance are not the same thing. A governance institu- tion can — orm, — unction, and sustain cooperation without a state’s

ormal endorsement. The Maine lobster communities examined in Chapter 1 had governance long be — ore any statute recognized them. But law shapes whether governance can — orm at all, what tools gov- ernance institutions have available, whether their rules can be en-

orced, who is accountable — or what, and who can challenge — ailures when they occur. This chapter identi — ies the legal conditions that governance institutions require. This analysis identi — ies six legal conditions that recur across governance settings. First, law must permit governance institutions to exist and recognize their authority. Second, law must enable in- stitutions to de — ine membership boundaries and en — orce them. Third, law must make an institution’s internal rules en — orceable be- yond the institution’s own social pressure. Fourth, law must identi — y who holds — iduciary obligations to the institution and give someone standing to en — orce those obligations. Fi — th, law must supply ac- countability structures through which those who exercise govern- ance authority can be held to account. Sixth, satis — ying all — ive prior 82 Law and Governance

conditions does not guarantee governance will succeed. Law sup- plies necessary in — rastructure, not su —

icient cause. Each condition is illustrated here by a documented example showing what the condi- tion enables and a contrasting example showing what its absence costs. Error! Use the Home tab to apply Heading 1 to the text that you want to appear here. 83

Chapter 4. Law as Governance Architecture Six legal conditions recur across governance settings. Each condi- tion is necessary in the sense that its absence predictably compro- mises a governance institution’s capacity to — unction, though no sin- gle condition is su —

icient to guarantee institutional success. The analysis proceeds — rom the most — oundational condition — legal permission to exist — through membership boundaries, internal or- dering, — iduciary structure, and accountability mechanisms, con- cluding with the distinction between legal possibility and institu- tional achievement.

Permission and recognition A governance institution must be permitted to exist. Alabama’s de- mand — or the NAACP’s membership lists in the 1950s shows what the absence o — legal protection — or association means in practice. The NAACP operated across the South as the primary govern- ance institution — or civil rights advocacy: setting strategy, managing litigation, coordinating local chapters, and sustaining a membership base under constant pressure. Alabama’s attorney general, seeking to bar the NAACP — rom operating in the state, demanded the organ- ization’s membership roster. The purpose was transparent. Expo- sure would subject members to economic reprisal, loss o — employ- ment, and physical danger. The Supreme Court held unanimously in NAACP v. Alabama ex rel. Patterson, 357 U.S. 449 (1958), that the demand violated the Fourteenth Amendment. The Court recog- nized that e —

ective advocacy depends on group association and that privacy in group membership is indispensable when a group es- pouses dissident belie — s. Id. at 460, 462. The holding protected the 84 Law and Governance

legal

oundation o

—

governance itsel

—

the right to associate and or- ganize collectively without state inter — erence.102 The contrasting example is Alabama itsel — , — rom 1956 through

1. The NAACP was e

ectively banned — rom the state. The NAACP could not recruit, could not operate chapters, and could not serve the constituency the NAACP existed to govern. When law withholds permission to associate, a governance institution does not merely — unction poorly; the institution ceases to exist. The modern equivalent appears in the legal treatment o — decen- tralized autonomous organizations. A DAO is a blockchain-based organization that governs through token-holder voting, smart con- tracts, and automated en — orcement. Be — ore any state granted DAOs legal recognition, such organizations already possessed governance in the sense Chapter 1 de — ines: decision-making through token votes recorded on a blockchain, monitoring through the permanent pub- lic transaction record, sanctions through automated smart contract execution, and adjustment through governance proposals subject to vote. Without legal entity status, however, participating in that governance created severe legal risk. In Sarcuni v. bZx DAO, 664 F. Supp. 3d 1100 (S.D. Cal. 2023), a — ederal court held that a DAO could be classi — ied as a general partnership, making token holders jointly and severally liable — or the organization’s debts. In CFTC v. Ooki DAO, No. 3:22-cv-05416 (N.D. Cal. 2023), the absence o — legal entity status prevented the organization — rom retaining counsel or appear- ing in court, producing a de — ault judgment. In Samuels v. Lido DAO (N.D. Cal. 2024), institutional investors — aced potential personal li- ability simply — or participating in governance votes.103 Governance participation thus produced personal liability, and once that conse- quence was understood by in — ormed participants, rational actors withdrew — rom governance rather than risk exposure. Error! Use the Home tab to apply Heading 1 to the text that you want to appear here. 85

Wyoming's DAO LLC Supplement, enacted in 2021 at Wyo. Stat. Ann. §§ 17-31-101 to 17-31-116, granted DAO organizations le- gal entity status and limited liability protection. That recognition did not create the governance architecture those organizations al- ready possessed. Legal recognition made participation in govern- ance possible without subjecting participants to unlimited personal liability. Law supplied the condition that allowed the governance to

unction; the governance itsel — was the product o — the institution’s own design. Legal recognition carries limits as well as bene — its. Collective ac- tion by a governance institution can cross — rom legitimate sel — -gov- ernance into illegal combination under antitrust law. The Supreme Court addressed that boundary in Northwest Wholesale Stationers, Inc. v. Paci — ic Stationery & Printing Co., 472 U.S. 284 (1985). A purchasing cooperative o — approximately one hundred o —

ice supply retailers ex- pelled Paci — ic Stationery, which had begun operating as both retailer and wholesaler. The Court rejected per se antitrust condemnation and held that per se treatment applies only when an institution pos- sesses market power or exclusive access to a resource essential to e —

ective competition. Where a cooperative lacks such power, courts apply rule o — reason analysis, asking whether the governance action serves a legitimate business purpose and whether the means employed are reasonably necessary to achieve it. Id. at 296.104 The rule o — reason gives governance institutions room to en — orce mem- bership conditions without automatic antitrust liability, while maintaining judicial oversight when an institution’s market posi- tion trans — orms governance into exclusion — or purely anticompeti- tive ends. 86 Law and Governance

Membership boundaries Governance without the power to exclude is governance without sanctions. Chapter 1 identi — ied exclusion as the paradigm sanction o — private governance institutions: when private institutions cannot imprison violators or seize property, the institution’s primary tool is denying a violator access to the bene — its o — membership. The legal question is whether the institution’s exclusion decisions will stand when challenged. The common law’s de — ault answer is yes. Private associations have inherent power to prescribe membership quali — ications, and members accept those quali — ications when they join. The charter and bylaws — orm a contract between the organization and its mem- bers, and courts review expulsion decisions — or compliance with

our requirements: the association must have acted within its pow- ers, in good — aith, in accordance with its own rules, and consistent with the law. Zechariah Cha — ee, “The Internal A —

airs o — Associa- tions Not — or Pro — it,” 43 Harv. L. Rev. 993 (1930).105 That standard is procedural rather than substantive. Courts will not substitute their judgment — or the association’s evaluation o — whether a member vi- olated the institution’s norms. Blatt v. University o — Southern Cali — ornia, 5 Cal. App. 3d 935 (1970), illustrates the de — ault rule. A night law student who gradu- ated in the top ten percent o — his class was denied admission to an honorary legal society a — ter the chapter retroactively adopted a Law Review participation requirement. The court sustained dismissal, holding that courts will not compel a voluntary honorary society to admit a member where membership is not a practical necessity — or earning a living.106 The court warned that judicial review o — mem- bership selection “would subject to judicial review the membership selection activity and policies o — every voluntary organization.” Error! Use the Home tab to apply Heading 1 to the text that you want to appear here. 87

Governance institutions need

inality in their exclusion decisions, and courts provide that — inality when the institution is voluntary and membership is not essential to pro — essional livelihood. The standard shi — ts when an institution controls access to pro-

essional opportunity. Pinsker v. Paci — ic Coast Society o — Orthodontists produced two Cali — ornia Supreme Court decisions that de — ine the limits o — association autonomy. In Pinsker I, 1 Cal. 3d 160 (1969), the court held that where an association holds quasi-monopolistic con- trol over pro — essional certi — ication, the association’s membership decisions create a justiciable claim and the association bears a — idu- ciary responsibility with respect to membership. In Pinsker II, 12 Cal. 3d 541 (1974), the court imposed two requirements on such institu- tions: substantive rationality, meaning the membership rule must not be arbitrary or contrary to public policy, and procedural — airness, meaning the applicant must receive notice o — the reason — or rejec- tion and a genuine opportunity to respond.107 The court used the phrase “ — air procedure” rather than “due process” to avoid importing constitutional standards designed — or state action into private gov- ernance. Fair procedure requires the institution to apply its own rules consistently. Constitutional due process would subject govern- ance decisions to judicial second-guessing on the merits. The two doctrines impose di —

erent burdens and carry di —

erent governance consequences. Calabresi and Melamed’s — ramework explains why maintaining the property-rule character o — exclusion matters to governance. A property rule protects an entitlement unconditionally: the holder can re — use to yield the entitlement at any price without the holder’s consent. A liability rule permits compelled trans — er at a price set by a court. A governance institution that can exclude members holds its exclusion right as a property-rule entitlement. The expelled 88 Law and Governance

member cannot remain unless the institution consents. When courts convert exclusion into a damages remedy or compel readmis- sion, the entitlement shi — ts to a liability rule. The institution retains the ability to impose a cost, but the institution loses the ability to re — use.108 An institution that must accept a disruptive member at a court- determined price loses the ability to de — ine its own membership. The sanction mechanism degrades, monitoring results stop trigger- ing en — orcement action, and decision-making authority becomes contested by members the institution could not remove. Silver v. New York Stock Exchange, 373 U.S. 341 (1963), shows this mechanism in operation. The NYSE’s in — ormal exclusion system could act swi — tly and without advance notice. The Court converted that system into a process requiring written charges, an opportunity to be heard, and a stated explanation. The exchange retained nominal exclusion au- thority, but the practical cost o — using that authority increased enough to change behavior throughout the institution.109 Chapter 7 examines the — ull consequences o — that shi — t — or exchange govern- ance.

Internal ordering and external en

orceability A governance institution can de — ine its own rules, but whether those rules bind members depends on whether law will support en-

orcement. The contract theory o — association rules provides the primary mechanism. When a member joins an association, the charter and bylaws — orm a contract between the member and the organization. The contract theory gives governance rules legal e —

ect beyond social pressure alone: a member who violates institutional rules and — aces Error! Use the Home tab to apply Heading 1 to the text that you want to appear here. 89

expulsion cannot simply ignore the expulsion and continue operat- ing as though membership were una —

ected. Contract en — orcement requires the institution’s internal proce- dure to satis — y a basic standard. Courts will en — orce association rules i — the association acted within its authority, in good — aith, — ollowing its own published procedures, and with reasonable notice and op- portunity to be heard. Applebaum v. Board o — Directors, 104 Cal. App. 3d 648 (1980). Courts will not independently evaluate whether the substantive decision was correct and will not retry the underlying disciplinary question. The judicial role is to ask whether the institu- tion — ollowed its own process, not to determine whether the insti- tution reached the right outcome. Lisa Bernstein’s empirical work on the National Grain and Feed Association identi — ies a speci — ic threat to this — ramework. When courts apply the Uni — orm Commercial Code’s interpretation provi- sions, speci — ically the code’s direction to consider trade usage, course o — dealing, and course o — per — ormance in resolving commercial dis- putes, they import contextual in — ormal norms into — ormal adjudica- tion. Formal adjudication, however, serves a di —

erent — unction than day-to-day trading relationships. Bernstein — ound that merchant ar- bitrators in the NGFA deliberately apply — ormalistic written trade rules rather than contextual customs, because predictability in the adjudicative setting requires rules that apply uni — ormly regardless o —

the parties’ prior relationship. Lisa Bernstein, “Merchant Law in a Merchant Court: Rethinking the Code’s Search — or Immanent Busi- ness Norms,” 144 U. Pa. L. Rev. 1765 (1996).110 When courts substi- tute the UCC’s contextual approach — or a governance institution’s

ormalistic rules, the courts do not supplement private governance. The courts undermine the predictability that makes — ormal adjudi- cation worth maintaining in the — irst place. 90 Law and Governance

The governance o ---  private companies illustrates both how con- tract law enables internal ordering and how judicial intervention can disrupt legal conditions without warning. Delaware General Corporation Law § 122(18), enacted e ---

ective August 1, 2024, au- thorizes corporations to enter stockholder agreements that restrict, condition, or control corporate governance decisions, expressly notwithstanding § 141(a)’s general mandate that the board manages the corporation. The legislature enacted this provision in direct re- sponse to West Palm Beach Fire — ighters’ Pension Fund v. Moelis & Com- pany, C.A. No. 2023-0309-JTL (Del. Ch. Feb. 23, 2024), in which Vice Chancellor Laster held that a stockholder agreement granting a — ounder pre-approval rights over eighteen categories o — board de- cisions — acially violated § 141(a). The market had relied — or years on such agreements as standard governance instruments — or companies anticipating an initial public o —

ering. The judicial decision invali- dated agreements that private ordering had treated as settled. The Delaware legislature restored the legal condition within — our months, be — ore the appeal was decided.111 Stockholder agreements gave private companies governance structures suited to their particular circumstances, including con- centrated ownership, investor protections negotiated at the time o —

inancing, and — ounder participation in major decisions. A judicial decision that those instruments violated a mandatory statutory pro- vision removed the legal en — orceability o — that governance architec- ture. The legislature’s rapid correction restored the condition that governance required, while leaving — iduciary duty obligations in place to preserve accountability. Error! Use the Home tab to apply Heading 1 to the text that you want to appear here. 91

Fiduciary structure and standing Governance institutions exercise authority over members and re- sources. Legal obligations must attach to that authority, and at least one identi — ied party must have the ability to en — orce those obliga- tions when they are breached. Fiduciary duties are the primary legal mechanism — or attaching obligations to governance authority. Directors o — corporations owe duties o — care and loyalty to the corporation. The duty o — care re- quires directors to in — orm themselves, deliberate, and act with the care an ordinarily prudent person in a like position would exercise. The duty o — care maps onto governance’s decision-making — unction: the institution’s collective choices must rest on genuine, in — ormed deliberation. The duty o — loyalty requires directors to place the or- ganization’s interests above their own personal interests. The duty o — loyalty maps onto governance’s monitoring — unction: con — licts o —

interest must be identi

ied and neutralized be — ore they corrupt the institution’s decisions. Nonpro — its carry a third duty speci — ic to the governance o — charitable organizations, the duty o — obedience, which requires directors to carry out the organization’s mission and ensure that — unds serve the charitable purposes — or which they were raised. In Manhattan Eye, Ear & Throat Hospital v. Spitzer, 186 Misc. 2d 126 (N.Y. Sup. Ct. 1999), the court held that a board violated the duty o — obedience when the board attempted to sell all hospital assets and — undamentally trans — orm the institution’s purpose without ad- equate justi — ication. 112 The duty o — obedience maps onto govern- ance’s adjustment — unction: the institution’s power to revise its own direction is bounded by the purposes the institution was created to serve. Standing determines who can en — orce those duties. In corpora- tions, shareholders can bring derivative suits on the corporation’s 92 Law and Governance

behal

to en — orce directors’ — iduciary obligations. In nonpro — its, no equivalent private en — orcement mechanism exists. The state attor- ney general is the primary en — orcement authority — or charitable or- ganizations and can compel accountings, remove trustees, and im- pose compliance obligations. En — orcement depends, however, on resources, political will, and prosecutorial judgment. Marion R. Fremont-Smith documented this en — orcement gap at length in Gov- erning Nonpro — it Organizations: Federal and State Law and Regulation (Harvard University Press, 2004).113 Individual bene — iciaries o — char- itable trusts generally lack standing under Restatement (Second) o —

Trusts § 391 unless they hold a “special interest,” a standard that is di —

icult to meet.114 When the attorney general declines to act, no other party can compel en — orcement. Three documented cases show what happens when the standing gap combines with governance — ailure. The Bishop Estate in Hawaii administered a $10 billion charita- ble trust created to — und educational opportunities — or Native Ha- waiian children. Trustees steward business deals to political allies, received excessive compensation, and used trust — unds — or personal bene — it, yet no private party had standing to challenge these deci- sions. Only when community members published a widely circu- lated essay and journalists sustained years o — investigative pressure did the attorney general intervene. When asked who held the trus- tees accountable, one trustee answered “nobody.” Samuel P. King and Randall W. Roth, Broken Trust: Greed, Mismanagement and Polit- ical Manipulation at America’s Largest Charitable Trust (University o —

Hawaii Press, 2006).115 Governance existed in

orm, but no account- ability mechanism existed to en — orce it. The Hershey Trust in Pennsylvania controls one o — the wealth- iest charitable institutions in the country. Board members received Error! Use the Home tab to apply Heading 1 to the text that you want to appear here. 93

compensation exceeding ninety-

ive thousand dollars annually, and persistent cross-directorships created con — licts o — interest that went unchallenged — or years. When an alumni association sought to chal- lenge governance decisions in court, the Pennsylvania Supreme Court denied standing in In re Milton Hershey School, 911 A.2d 1258 (Pa. 2006), holding that the alumni lacked a “special interest.” Jona- than Klick and Robert H. Sitko —

documented that political consid- erations constrained the attorney general’s willingness to act against one o — the state’s most prominent institutions. Klick & Sitko —

, “Agency Costs, Charitable Trusts, and Corporate Control: Evidence

rom Hershey’s Kiss-o —

,” 108 Colum. L. Rev. 749 (2008).116 The Getty Trust in Cali — ornia experienced years o — abuse by its chie — executive, Barry Munitz, who used trust — unds — or personal loans, luxury travel, and a seventy-two-thousand-dollar car while receiving $1.2 million annually in compensation. The Cali — ornia at- torney general imposed independent oversight only a — ter the abuse became publicly known through journalism. The governance archi- tecture o — the trust, a board with — iduciary duties and legal recogni- tion, was intact throughout the period o — abuse. The accountability mechanism that would have detected the problem earlier did not ex- ist.117 Evelyn Brody identi — ied the structural problem that underlies all three cases: “charity — iduciaries — requently escape accountability — or their sel — -dealing and neglect or mismanagement” because “ — ew charities have members endowed with voting rights, and state at- torneys general have limited resources to devote to monitoring the nonpro — it sector.” Evelyn Brody, “The Limits o — Charity Fiduciary Law,” 57 Md. L. Rev. 1400, 1401 (1998).118 The absence o — private en — orcement rights is not an incidental — eature o — nonpro — it law. It is the structural condition that allows governance — ailures to persist 94 Law and Governance

long a

ter similar — ailures would have been remedied in institutions where accountability mechanisms are more accessible.

Accountability structures Fiduciary duties identi — y legal obligations. Accountability structures are the mechanisms through which compliance with those obliga- tions is actually checked. A legal duty without an accountability mechanism is a right without a remedy. In corporate governance, the primary accountability mechanism is the shareholder — ranchise. Shareholders vote on directors, and di- rectors who — ail to govern competently or loyally can be removed. Delaware Chancellor William T. Allen described the shareholder

ranchise in Blasius Industries, Inc. v. Atlas Corp., 564 A.2d 651, 659 (Del. Ch. 1988), as the “ideological underpinning” o — Delaware cor- porate governance.119 The — ranchise — unctions as an accountability mechanism when shareholders have the genuine capacity to exercise the — ranchise. Lucian A. Bebchuk documented the structural barriers that make director removal di —

icult in practice: collective action problems among dispersed shareholders, management’s pre — erential access to corporate resources — or proxy campaigns, and staggered board structures that prevent shareholders — rom replacing a major- ity o — directors in any single year. Bebchuk, “The Myth o — the Share- holder Franchise,” 93 Va. L. Rev. 675 (2007).120 The board’s own accountability — or oversight — ailures is gov- erned by the standard established in In re Caremark International Inc. Derivative Litigation, 698 A.2d 959 (Del. Ch. 1996). Chancellor Allen held that directors carry an a —

irmative obligation to ensure that ad- equate in — ormation and reporting systems exist within the corpora- tion. Liability arises — rom “a sustained or systematic — ailure o — the Error! Use the Home tab to apply Heading 1 to the text that you want to appear here. 95

board to exercise oversight.” The court itsel

acknowledged that this standard represents “possibly the most di —

icult theory in corpora- tion law upon which a plainti —

might hope to win a judgment.” Id. at 967.121 The high threshold is deliberate, designed to give boards room to govern without the constant threat o — personal liability — or every compliance — ailure. The practical consequence is that account- ability under Caremark tends to — unction as a prophylactic: boards create monitoring systems primarily to avoid liability rather than because those systems will necessarily detect problems. Marchand v. Barnhill, 212 A.3d 805 (Del. 2019), illustrates what the Caremark standard looks like when it actually applies.122 Blue Bell Creameries su —

ered a listeria outbreak that killed three people. The Delaware Supreme Court held that where — ood sa — ety was “es- sential and mission critical” to a company’s single business line, the complete absence o — board-level — ood sa — ety monitoring systems was su —

icient to support a Caremark liability claim. The holding turned on the total absence o — any monitoring system, not on the adequacy o — an existing one. Boards that maintain no monitoring architecture

or risks central to their business cannot satis — y the oversight duty. In universities and nonpro — its, accountability mechanisms are weaker and the available substitutes are less e —

ective. Regional ac- creditation — unctions as a gatekeeping mechanism — or access to — ed- eral — inancial aid, and accreditors review governance periodically and can sanction institutions that — ail to maintain adequate govern- ance standards. When the University o — Virginia’s Board o — Visitors

orced the summary removal o — President Teresa Sullivan in 2012, the Southern Association o — Colleges and Schools placed UVA on warning — or twelve months and required the adoption o — new gov- ernance policies. Accreditation — unctioned as a governance account- ability mechanism in that instance. 96 Law and Governance

 The record o ---  accreditation in larger governance  --- ailures tells a di ---

erent story. The Accrediting Council — or Independent Colleges and Schools accredited both Corinthian Colleges and ITT Technical Institute despite widespread deceptive practices, dismal graduation rates, and — raudulent job placement statistics. The Department o —

Education withdrew ACICS recognition in 2016,

inding that the agency had “ — ailed broadly in its oversight role.” The — ederal govern- ment subsequently cancelled approximately ten billion dollars in student loans connected to those institutions. Accreditation in those cases provided the appearance o — accountability without the sub- stance. The IRS redesigned Form 990 in 2008, signi — icantly expanding governance disclosure requirements. Nonpro — its must now disclose board composition, compensation, con — lict-o — -interest policies, and executive pay. Research by Harris, Petrovits, and Yetman — ound that donations are positively associated with good governance disclo- sures, which suggests that donor markets respond to transparency. Erica E. Harris, Christine M. Petrovits & Michelle H. Yetman, “The E —

ect o — Nonpro — it Governance on Donations: Evidence — rom the Revised Form 990,” 90 Acct. Rev. 579 (2015).123 Form 990 creates accountability only indirectly, through market pressure rather than legal en — orcement, and the — orm gives no interested party standing to sue — or breach o — governance obligations. The shareholder — ranchise, accreditation, and Form 990 disclo- sure each create a — ormal structure that resembles accountability. Each mechanism works when conditions are — avorable and — ails when they are not. None o — the three replaces the direct accounta- bility o — an interested party with legal standing to en — orce — iduciary duties. Error! Use the Home tab to apply Heading 1 to the text that you want to appear here. 97

Legal conditions do not guarantee institutional success The — ive conditions above are necessary but not su —

icient. Govern- ance institutions can have every legal condition in place and still — ail, sometimes catastrophically. Penn State University had a board o — trustees with plenary legal authority, a president, a campus police — orce with — ull law en — orce- ment powers, — ederal reporting obligations under the Clery Act, and a compliance department. Jerry Sandusky abused children over — i — - teen years. The Freeh Report — ound that Penn State’s most senior leaders “repeatedly concealed critical — acts relating to Sandusky’s criminal conduct” motivated by “the avoidance o — the consequences o — bad publicity,” and that the board had no “regular reporting pro- cedures or committee structures in place to ensure disclosure o — ma- jor risks.” Louis Freeh et al., Report o — the Special Investigative Counsel Regarding the Actions o — The Pennsylvania State University Related to the Child Sexual Abuse Committed by Gerald A. Sandusky, July 12, 2012, at 14, 108.124 Governance — ailed not because the legal conditions were absent, but because the institution’s actual decision-making, moni- toring, and accountability — unctions had been captured by a culture o — de — erence to athletic program reputation. Enron’s board was a model o —

ormal governance compliance. The board comprised — i — teen members with extensive pro — essional experience. The audit committee was chaired by a — ormer dean o —

Stan

ord Business School. The legal architecture o —

iduciary duties, audit requirements, and shareholder accountability was intact throughout the period preceding the company’s collapse. The U.S. Senate Permanent Subcommittee on Investigations — ound six cate- gories o — board — ailure, including knowing tolerance o — high-risk ac- 98 Law and Governance

counting practices and unprecedented approval o

the CFO’s con-

licts o — interest through the LJM partnerships. The Role o — the Board o — Directors in Enron’s Collapse, S. Prt. 107-70, July 8, 2002.125 Senator Lieberman observed at the hearing that the board “didn’t just — iddle while Enron burned. They toasted marshmallows over the — lames.” Legal conditions were present and governance was absent. Lauren B. Edelman identi — ied the systemic dynamic that pro- duces this gap between legal possibility and institutional reality. Or- ganizations respond to legal mandates by creating — ormal structures that signal compliance without changing behavior. EEO o —

ices, anti-discrimination policies, grievance procedures, and compliance training are adopted because law requires them or because their presence reduces liability exposure, not because the structures will change organizational outcomes. Edelman, Working Law: Courts, Corporations, and Symbolic Civil Rights (University o — Chicago Press, 2016).126 Edelman and Talesh extended this analysis speci — ically to legal conditions — or governance, demonstrating that organizations adopt — ormal compliance structures as legal shields without chang- ing underlying behavior — the gap between legal architecture and institutional success is not a — ailure o — the legal conditions but a limit on what law can accomplish alone.127 Kimberly D. Krawiec — ound the same pattern across multiple regulatory domains: compliance programs are adopted primarily as legal shields rather than behav- ioral change mechanisms, and courts treat the existence o — such pro- grams as evidence o — compliance regardless o — whether the pro- grams actually work. Krawiec, “Cosmetic Compliance and the Fail- ure o — Negotiated Governance,” 81 Wash. U. L.Q. 487 (2003).128 The gap between legal possibility and institutional success is not a — ailure o — the legal conditions identi — ied in this chapter. It is a limit Error! Use the Home tab to apply Heading 1 to the text that you want to appear here. 99

on what those conditions can accomplish. Law can make govern- ance possible, but law cannot make governance — unction. The gov- ernance institution must do that work: the institution must build decision-making procedures that actually produce collective choices, monitoring systems that actually detect de — ection, sanction mechanisms that members actually apply, and adjustment processes that actually modi — y the institution when circumstances change. When those — unctions — ail despite adequate legal in — rastructure, the

ailure lies in governance architecture rather than in the legal con- ditions surrounding it. The distinction is central to the method Chapter 8 develops

ully. The method asks whether law is helping or hindering govern- ance institutions in building and maintaining the — our governance elements. Where governance is absent despite legal conditions be- ing in place (Penn State, Enron, and the nonpro — it cases above), legal analysis must ask why governance — ailed and whether di —

erent legal rules would have helped or whether the problem lies beyond what law can address. The method does not claim that good legal condi- tions guarantee good governance. The method does claim that bad legal conditions reliably prevent governance — rom — unctioning.

What law supplies and what it does not The private ordering literature documents governance arising out- side the state and demonstrates that groups create e —

ective ordering mechanisms without — ormal legal in — rastructure. Robert C. Ellick- son, Order Without Law: How Neighbors Settle Disputes (Harvard Uni- versity Press, 1991); Lisa Bernstein, “Opting Out o — the Legal Sys- tem,” 21 J. Legal Stud. 115 (1992); Gillian K. Had — ield & Barry R. Weingast, “What Is Law?,” 4 J. Legal Analysis 471 (2012); Bruce L. 100 Law and Governance

Benson, The Enterprise o

Law (Paci — ic Research Institute, 1990). 129 These accounts are important. They demonstrate that governance is not a creature o — the state and that private institutions can manage shared problems with minimal — ormal legal in — rastructure. Every one o — these accounts, however, underestimates the legal in — rastructure that makes private ordering possible. Ellickson’s ranchers own their land because property law gives them title. Bern- stein’s diamond dealers operate in a building they lease under con- tract law, and their arbitral awards derive binding legal e —

ect — rom the Federal Arbitration Act. Had — ield and Weingast’s classi — ication institutions operate within a legal system they treat as background. Benson’s private governance mechanisms work in part because the state tolerates those mechanisms, recognizes their outputs, and en-

orces the contracts that sustain them. Rather than creating private governance, law supplies the conditions under which private gov- ernance can sustain itsel — . Part II builds the theory o — how law interacts with those condi- tions: how speci — ic legal rules enable or degrade the governance el- ements this book has identi — ied. The conditions established in this chapter are the materials that Part II’s theory will work with.  Part II: The Original Theory: Club Goods, Spillovers, Legal Mechanisms, and the Seven-Step Method

Part I established the vocabulary and legal architecture the book needs. It de — ined governance as a distinct institutional phenomenon, identi — ied its indispensable — unctional elements, and explained the legal conditions under which governance can — orm, persist, and re- main accountable. But de — inition alone is not enough. Once govern- ance is visible as a legal object, the next question is how to under- stand its structure, its social value, and the ways law predictably helps or harms it. 102 Law and Governance

 Part II develops the book's theoretical core. These chapters sup- ply the original  --- ramework that connects governance to the eco- nomic theory o ---  goods, distinguishes private member bene --- its  --- rom broader social spillovers, explains how legal rules act on governance institutions, and converts those insights into a method  --- or legal analysis. I ---  Part I says what governance is, Part II explains why law's e ---

ects on governance escape ordinary doctrinal reasoning. Chapter 5 makes the central theoretical move. It argues that governance is best understood not as a public good or a simple pri- vate good, but as a club good. Governance institutions are excluda- ble because they depend on membership boundaries and the sanc- tion o — expulsion. They are nonrivalrous only up to a point, because participation beyond the institution’s capacity creates congestion and degrades governance quality. And they are sustained through voluntary participation and member contribution rather than uni- versal public provision. This — ramework explains why exclusion is not incidental to governance but o — ten constitutive o — it. Chapter 6 then sharpens the wel — are analysis by separating two things that legal and policy discourse o — ten collapse into one: the bene — its governance institutions provide to their members, and the spillover bene — its their — unctioning creates — or outsiders. That dis- tinction is essential. Without it, analysis either romanticizes exclu- sion by treating all bene — its as club bene — its, or ignores the public cost o — governance — ailure by treating governance as relevant only to members. The chapter shows that good governance o — ten pro- duces value — or nonmembers who neither join nor pay, and that le- gal analysis must there — ore account — or governance e —

ects beyond the institution’s — ormal boundaries. Chapter 7 turns — rom theory to mechanism. I — governance is a socially valuable institutional — orm, law can a —

ect it in patterned Chapter 4: Legal Conditions — or Governance 103

ways. Some legal rules enable governance by protecting member- ship boundaries, en — orcing internal decisions, or reducing the cost o — sustaining institutional order. Other rules degrade governance by disabling sanctioning, removing accountability, converting exclu- sion into — orced access, or substituting — ormal compliance — or — unc- tioning institutional architecture. The chapter identi — ies these mechanisms with precision and shows how legal interventions that appear justi — ied when viewed dispute by dispute can nevertheless erode governance over time. Chapter 8 delivers the methodological payo —

. It presents the seven-step method the rest o — the book applies: identi — y the shared problem, identi — y the governance institution, speci — y the legal con- ditions under which it operates, separate member bene — its — rom spillovers, analyze how law a —

ects the institution, and evaluate the result as enabling, disciplining, or degrading. The method does not displace rights, e —

iciency, or distributional analysis. It adds a varia- ble legal analysis has repeatedly relied on without clearly naming: governance itsel — . Taken together, these — our chapters are the book’s theoretical engine. They explain why governance should be analyzed as a dis- tinct institutional good, why law’s e —

ects on governance are o — ten missed by ordinary doctrinal reasoning, and how those e —

ects can be brought into view through a structured method. Part III then turns — rom theory to application, using this — ramework to analyze commercial networks, stock exchanges, corporations, universities, nonpro — its, and knowledge institutions. Part II is there — ore where the book moves — rom — oundation to originality. It does not simply de — ine governance more care — ully than prior scholarship. It advances a theory o — governance’s struc- 104 Law and Governance

ture, value, and legal vulnerability, then gives that theory opera- tional — orm. The chapters in this Part are the hinge on which the whole project turns. Chapter 5: Governance as a Club Good Lawyers who have taken an economics course will recall the distinc- tion between public goods and private goods. Public goods are those that everyone can use without reducing what is available to others and — rom which no one can be excluded. Private goods are those that one person’s use depletes and — rom which nonpayers can be ex- cluded. These polar cases are use — ul — or thinking about markets and government provision. They are not, by themselves, an adequate ty- pology. Most goods o — institutional consequence — all between the poles. This chapter places governance institutions within a more com- plete typology o — goods, developed originally by Paul Samuelson in the 1950s, extended by James Buchanan in 1965, and re — ined by Vin- cent Ostrom and Elinor Ostrom in the late 1970s.130 The typology organizes goods along two dimensions: whether users can be ex- cluded — rom the good, and whether one person’s use subtracts — rom what is available to others. The intersection o — these two dimensions produces — our categories: private goods, public goods, common- pool resources, and club goods. This chapter argues that most im- portant governance arrangements have the structure o — club goods, and that this structural observation explains — eatures o — governance 106 Law and Governance

that other

rameworks leave obscure. The argument was developed in prior work by this author and is extended here.131 Error! Use the Home tab to apply Heading 1 to the text that you want to appear here. 107

Chapter 5. Club Good Structure The claim that governance institutions share the economic struc- ture o — club goods rests on a typology developed over several decades by economists and institutional theorists. Understanding that typol- ogy requires two preliminary steps: mapping the landscape o — goods along two analytical dimensions, and then examining what Bu- chanan demonstrated about the goods that — all into the club cate- gory. Once the typology is in place, the chapter applies it to govern- ance institutions and draws out the legal consequences.

The

our-cell typology The typology o — goods is a practical analytical tool that identi — ies which institutional arrangements work — or which kinds o — goods. The excludability dimension. Some goods can be supplied to one person while another person is kept out. A private parking lot supplies access to some drivers and denies it to others through gates,

ees, and attendants. Other goods resist exclusion. The light — rom a lighthouse illuminates all ships in range equally, and the lighthouse keeper cannot selectively darken the beam — or ships whose owners did not pay. Excludability is partly a physical property o — the good and partly a — unction o — available technology and institutional de- sign. Broadcast television was non-excludable until encryption technology made scrambling possible. The commons problem in

isheries persists partly because monitoring individual catches at sea is costly. As Elinor Ostrom demonstrated, what counts as excludable depends on institutional arrangements, not only on the physical na- ture o — the thing.132 The rivalry dimension. Some goods are consumed — ully by the person who uses them: one person eating an apple prevents anyone 108 Law and Governance

else

rom eating that apple. Other goods are not consumed in this way: one ship’s use o — the lighthouse beam does not diminish the beam — or other ships. Goods o — the second type are called non-rival- rous, or sometimes low in subtractability. The Ostrom school uses “subtractability” to emphasize that the dimension is a continuum with some goods highly rivalrous, some moderately rivalrous, and some e —

ectively non-rivalrous over wide ranges o — use, rather than a binary category.133 The — our cells. Organizing goods by these two dimensions pro- duces — our types. Private goods are both excludable and rivalrous: a loa — o — bread, a pair o — shoes, a haircut. Public goods are neither ex- cludable nor rivalrous: national de — ense, a weather — orecast, scien- ti — ic knowledge. Common-pool resources are rivalrous but di —

icult to exclude: a — ishery, groundwater, a shared pasture. Club goods are excludable but non-rivalrous, at least up to a point: a gol — course, a cable television network, a private swimming pool. The club good is the cell that matters — or governance. The — our-cell taxonomy was — ormalized by Vincent Ostrom and Elinor Ostrom in 1977, who organized the matrix around subtrac- tability and excludability and added “common-pool resources” as a distinct — ourth type distinct — rom public goods.134 They also renamed “club goods” as “toll goods” to emphasize that the institutional — orm o — provision is secondary to the structural characteristics. Both terms appear in the literature. This book uses “club goods” because the governance analysis concerns the institutional structure o — clubs: their voluntary membership, their sel — - — inancing through member contributions, their governance mechanisms. The term “toll goods” directs attention to pricing structure; “club goods” directs attention to institutional design. The institutional — ocus is what this analysis requires. Error! Use the Home tab to apply Heading 1 to the text that you want to appear here. 109

Buchanan’s club goods James Buchanan introduced the — ormal theory o — club goods in a 1965 paper that opens by identi — ying the gap between Samuelson’s two polar types.135 Samuelson had established the optimality condi- tions — or public goods and had noted that pure public goods and pure private goods represent polar extremes. Buchanan set out to — ill the space between those extremes with a coherent theory. A club, — or Buchanan, is a voluntary membership arrangement

or sharing ownership or consumption o — a good among a group larger than one individual but smaller than the entire population. Three characteristics mark the club-good category. Excludability. The club can deny access to the good by re — using membership or expelling violators. The gate at the tennis court, the wall around the community pool, the credential requirement — or the pro — essional association: all are exclusion mechanisms. Because ex- clusion is possible, the club can charge — or membership and can ex- pel those who violate the club’s rules. Without excludability, there is no club in any meaning — ul sense; the institution becomes a public good with an organizational wrapper that cannot per — orm the gov- ernance — unction exclusion supports. Partial rivalry through congestion. Within the club’s capac- ity, the good is non-rivalrous: one member’s use o — the tennis court on Tuesday does not reduce another member’s ability to use it on Thursday. But as membership grows past some threshold, addi- tional members begin to degrade the experience — or everyone. Court time becomes scarce. The pool becomes crowded. Pro — essional cer- ti — ication loses prestige as standards weaken to accommodate a larger membership. This congestion e —

ect creates an optimal club size. The optimal size is the membership level at which the marginal cost o — adding another member, measured in degraded bene — its — or 110 Law and Governance

existing members, equals the marginal bene

it — rom cost-sharing that a new member brings. Too — ew members imposes excessive per-capita costs. Too many members produces congestion. The op- timal club operates at the point where these pressures balance.136 Voluntary membership and sel — - — inancing. Club members join by choice and bear the costs o — provision through dues, — ees, or contributions. This structure distinguishes clubs — rom government- provided public goods, which are — unded through taxation regard- less o — individual bene — it or pre — erence. Voluntary membership dis- ciplines the club: members who — ind the club’s costs exceed its ben- e — its can exit, joining a competing club or — orming a new one. This exit option, when the market — or club memberships is competitive, creates pressure toward e —

icient provision.137 Buchanan made two — urther observations that matter — or gov- ernance. The — irst is the dual optimization problem: the e —

icient club must simultaneously determine the optimal quantity o — the club good to provide and the optimal membership size. These decisions are made together, not sequentially. The second is the replicability condition: in a competitive market, clubs with suboptimal con — igu- rations — ace discipline through the — ormation o — competing clubs. I —

an existing club is too small, excluded individuals will

orm a new club. I — it is too large, members will de — ect to — orm smaller ones. Competitive discipline in the market — or clubs can produce e —

icient provision without government intervention, provided clubs are replicable and the market — or memberships — unctions well. When a governance institution cannot be replicated because it controls es- sential access, this competitive discipline — ails, and legal analysis must supply the discipline that market pressure cannot.138 Error! Use the Home tab to apply Heading 1 to the text that you want to appear here. 111

How the typology maps to governance institutions The claim that governance institutions have club good structure rests on documented institutional evidence across multiple settings. Establishing the claim requires mapping the de — ining characteristics o — club goods onto the de — ining characteristics o — governance insti- tutions identi — ied in Chapter 1, and being precise about where the map — its cleanly and where it requires quali — ication. Excludability through the sanction o — exclusion. The ex- cludability characteristic o — club goods corresponds directly to what Chapter 1 identi — ied as the paradigm sanction o — private governance institutions. Governance institutions that lack the state’s power o —

imprisonment or asset seizure en

orce their rules through exclusion

rom the bene — its o — membership. The diamond dealer expelled

rom the Diamond Dealers Club loses access to the trading — loor and the trust in — rastructure that makes large-scale oral transactions pos- sible. The licensed pro — essional expelled — rom the bar association loses the credential that permits practice. The merchant network member who cheats and is blacklisted loses access to the trading re- lationships that constitute the member’s livelihood. The structural reason governance institutions rely on exclusion as the primary en — orcement mechanism is that governance institu- tions are club goods. Exclusion works because membership gener- ates non-rivalrous bene — its that members value, and the threat o —

losing those bene

its disciplines behavior. I — governance institutions were public goods, with the bene — its o — governance accruing to eve- ryone regardless o — membership, exclusion would lose its — orce. A trader expelled — rom a club whose bene — its are — reely available to non-members would simply access those bene — its without — ormal 112 Law and Governance

membership. Excludability and the governance sanction o

exclu- sion are the same structural property viewed — rom two angles: one economic, one legal. Non-rivalry up to congestion. The governance mechanisms through which an institution manages its shared problem are not consumed in the way a private good is consumed. One member’s use o — the institution’s dispute resolution system does not reduce an- other member’s ability to use it. One member’s reliance on the in- stitution’s reputation mechanism does not deplete the mechanism’s capacity. One member’s vote in the institution’s collective-choice procedure does not subtract — rom another member’s vote. These governance — unctions are non-rivalrous within the institution’s ca- pacity. Congestion in governance institutions takes a speci — ic — orm. The congestion mechanism in a governance institution is not overuse o —

a physical resource but de

ection — rom the cooperative equilibrium. When too many members de — ect — rom institutional rules, the mon- itoring system is overwhelmed, the sanction mechanism loses cred- ibility, and the institution’s capacity to manage its shared problem degrades. This is Buchanan’s congestion pattern, in which quality degrades as membership grows past the threshold o — e —

ective gov- ernance, rather than Ostrom’s subtractability pattern, in which each unit o — consumption directly subtracts — rom what is available to oth- ers.139 The congestion point determines the optimal size o — a govern- ance institution. A governance institution with too — ew members cannot achieve economies o — scale in governance provision: — ixed costs per member are high, the institution lacks su —

icient diversity o — perspective, and the network e —

ects that increase the value o —

governance bene

its are limited. A governance institution with too Error! Use the Home tab to apply Heading 1 to the text that you want to appear here. 113

many members

aces escalating monitoring costs, more severe — ree- rider problems, and di —

iculty reaching collective decisions across a heterogeneous membership. The optimal governance institution operates at the size where these pressures balance, just as Buchanan’s optimal club operates at the point where cost-sharing bene — its and congestion costs are equated at the margin. Voluntary membership and sel — - — inancing. Governance in- stitutions are sustained by the voluntary participation o — members who bear the costs o — governance through dues, assessments, and the time required — or deliberation and monitoring. This sel — - — inanc- ing structure gives governance institutions their distinctive charac- ter compared to government regulation. A regulatory agency can mandate compliance through threatened sanctions backed by state power. A governance institution must maintain voluntary compli- ance by delivering bene — its worth the cost o — membership and the burden o — adhering to institutional rules. When a governance insti- tution — ails to deliver adequate bene — its, members exit, and the in- stitution either re — orms or collapses.

Diamond Dealers, Stock Exchanges, and Commons: How Club-Good Structure Operates Across Domains Diamond trading networks. Lisa Bernstein’s study o — the New York Diamond Dealers Club documented a governance institution that exempli — ies the club good structure.140 DDC membership pro- vides non-rivalrous access to the trading — loor, the arbitration sys- tem, and the trust in — rastructure that permits billions o — dollars in annual transactions to be conducted on handshake terms. Member- ship is voluntary and sel — - — inanced through dues and arbitration

ees. The exclusion mechanism consists o — expulsion — rom the DDC, 114 Law and Governance

with the expulsion transmitted to every a

iliated bourse worldwide through the World Federation o — Diamond Bourses, and this mech- anism operates as the sole en — orcement sanction. No state apparatus en — orces DDC rules. Because no state en — orcement is available, the credibility o — the exclusion threat — not its actual exercise — is what sustains cooperative behavior among members. The threatened loss o — non-rivalrous membership bene — its converts a collective action problem, in which each trader has an individual incentive to cheat, into a cooperative equilibrium, in which no trader cheats because the expected cost o — exclusion exceeds the expected gain — rom de — ec- tion. Barak Richman’s documentation o — the DDC’s partial erosion con — irms the club good structure by demonstrating the mechanism in reverse: as globalization brought new market entrants who had not grown up in the community and did not — ace — ull exclusion costs, de — ection rates increased, governance quality declined, and — raud costs spread to non-member retailers, bankers, and consumers who had previously bene — ited — rom the DDC’s governance without bear- ing its costs.141 Stock exchanges be — ore Silver. Be — ore the Supreme Court’s decision in Silver v. New York Stock Exchange, 373 U.S. 341 (1963), the NYSE operated as a governance institution with explicit club struc- ture. Seat ownership was an excludable membership right con — er- ring non-rivalrous access to the trading — loor, price discovery mech- anisms, and the exchange’s sel — -regulatory governance system. The exchange governed its members through internal rules, a discipli- nary committee, and the sanction o — expulsion, which meant the loss o — the seat and its associated trading privileges. The Supreme Court in Silver itsel — described the exchange’s historical treatment by courts as that o — a “private club,” noting that the exchange’s limited-entry Error! Use the Home tab to apply Heading 1 to the text that you want to appear here. 115

structure had historically generated broad judicial de

erence in dis- ciplining errant members. Id. at 357. The Securities Acts Amend- ments o — 1975 trans — ormed this club governance structure into reg- ulated governance, requiring procedural sa — eguards — or all discipli- nary actions and prior SEC approval — or all rule changes.142 Paul Ma- honey documented the governance consequences o — this trans — or- mation: the exchange’s capacity to discipline members swi — tly and credibly, the characteristic that had sustained market integrity, was systematically constrained by the procedural apparatus the statute imposed.143 Commons governance institutions. Applying the club good

ramework to commons governance requires a distinction that the existing literature has not drawn clearly. The underlying resource that a commons community governs — a — ishery, a watershed, an alpine meadow — is a common-pool resource: rivalrous and di —

i- cult to exclude. The governance institution that manages the re- source — the — ishing cooperative, the irrigation association, the al- pine commons — is itsel — a club good: excludable through member- ship requirements and non-rivalrous in its governance — unctions up to the congestion threshold. Elinor Ostrom’s Design Principle 1, the requirement that suc- cess — ul commons governance institutions de — ine clearly who is and is not a member, is the excludability condition — or governance as a club good. 144 Ostrom’s irrigation communities in the Philippines and Spain en — orced membership boundaries through monitoring and sanction mechanisms that required participation in mainte- nance work as a condition o — access to irrigation bene — its. Her Swiss alpine communities, some o — which have governed communal pas- tures since the thirteenth century, maintain membership through residence requirements and usage rights tied to membership in the 116 Law and Governance

village community. 145 In each case, the governance institution’s

unctions — collective decision-making, monitoring, graduated sanctions, rule revision — are non-rivalrous within the member- ship. One member’s participation in the seasonal meeting at which quotas are negotiated does not subtract — rom another member’s ca- pacity to participate. The governance institution managing the common-pool resource has club good structure even though the re- source the institution governs does not. The distinction between how the typology classi — ies the govern- ance institution and how it classi — ies the governed resource has di- rect legal implications. When courts or legislators act on commons governance institutions, they must ask whether their intervention a —

ects the resource, which may require di —

erent analysis, or the governance institution, which requires attention to excludability and the sanction mechanism. The two questions receive di —

erent answers — rom the typology.

Edge cases and boundaries Applying the typology honestly requires acknowledging three cate- gories o — governance institutions that do not — it cleanly into the club good model. Mandatory membership institutions. Mandatory bar associ- ations, licensing boards, and regulated sel — -regulatory organizations share club good — eatures — excludable membership and shared gov- ernance — unctions — but they lack the voluntary membership char- acteristic that Buchanan treated as de — initional. A lawyer who must belong to the state bar to practice law has not — reely chosen a club among competing options. The exit option that disciplines clubs in competitive markets does not operate. Error! Use the Home tab to apply Heading 1 to the text that you want to appear here. 117

Even mandatory membership institutions retain club good structure in the sense that excludability through license revocation remains the primary en --- orcement mechanism, governance  --- unc- tions are non-rivalrous, and the institution is sel --- - --- inanced through member dues. What changes is the competitive discipline on the provision decision. Second, the  --- ailure o ---  competitive discipline in mandatory membership institutions explains a pattern that lawyers will recognize: mandatory governance institutions tend toward lower governance quality than voluntary ones because the pressure o ---  member exit is absent. Legal analysis o ---  mandatory membership governance institutions must account  --- or this structural di ---

erence. When a governance institution cannot be disciplined by exit, ac- countability mechanisms become more important, and legal struc- tures that provide accountability in lieu o — competitive pressure de- serve special attention.146 Two accountability mechanisms are compatible with preserving the governance — unction o — exclusion while compensating — or the absence o — exit discipline. First, mandatory internal appeals with in- dependent review allow excluded members to challenge governance decisions within the institution be — ore seeking external judicial in- tervention. An internal appeals body sta —

ed by members without a stake in the original decision, or by outside reviewers with relevant expertise, imposes a check on arbitrary exclusion without convert- ing exclusion — rom a property-rule entitlement to a liability-rule remedy. The institution retains the categorical authority to exclude, but must demonstrate to an internal tribunal that the exclusion — ol- lowed institutional rules and responded to an identi — iable govern- ance violation. The critical design — eature is that the independent reviewer’s authority is limited to procedural compliance and good 118 Law and Governance

aith — not to substituting its judgment — or the institution’s sub- stantive governance decision. This preserves the deterrent — orce o —

exclusion (the expelled member cannot purchase continued access through external litigation) while providing a sa — eguard against the abuse o — exclusion authority that justi — ies heightened scrutiny o —

mandatory-membership institutions in the

irst place. Second, periodic re-authorization or sunset provisions supply temporal exit discipline where spatial exit is unavailable. When a mandatory membership institution’s governing authority must be periodically renewed — whether by legislative reauthorization, member rati — ication vote, or regulatory review — members who cannot leave the institution can nonetheless discipline it by oppos- ing renewal o — its governance powers. Sunset provisions create a re- curring moment at which the institution must justi — y continued au- thority, substituting a temporal check — or the competitive check that voluntary exit provides. This mechanism avoids activating the deg- radation pathway identi — ied in Chapter 7 because it does not inter- vene in any individual governance decision or convert any speci — ic exclusion into a compensated remedy. It operates at the structural level, disciplining the institution’s overall governance quality with- out constraining its operational governance authority between re- view periods. Neither mechanism eliminates the structural vulnerability o —

mandatory membership governance. The absence o

exit creates a persistent accountability gap that no internal procedure — ully closes. Mandatory membership institutions with market power remain the hardest case — or the governance — ramework developed in this book, because the — ramework’s strongest claims — that governance insti- tutions generate positive externalities that law should protect, that exclusion is the paradigm sanction whose categorical character must Error! Use the Home tab to apply Heading 1 to the text that you want to appear here. 119

be preserved — depend on either the availability o

exit or the ade- quacy o — accountability substitutes. Speci — ying these substitutes with su —

icient rigor to match the — ramework’s treatment o — voluntary in- stitutions is an ongoing analytical challenge. Chapter 14 addresses this limitation directly, including the risk that mandatory member- ship institutions could deploy the governance — ramework as a shield

or exclusionary practices that serve institutional interests rather than governance — unctions. Governance institutions whose exclusion has been legally disabled. When courts or regulators disable an institution’s exclu- sion mechanism — converting expulsion — rom a property rule to a liability rule, requiring procedural sa — eguards that make expulsion costly enough to deter its use, or mandating open membership — the governance institution loses the club good structure that sus- tains its governance capacity.147 The institution does not thereby be- come a public good. It becomes a damaged club good: — ormally ex- cludable but practically unable to exercise exclusion credibly. The structure this book identi — ies as governance degradation is the pro- cess by which law converts a governance institution into something that retains governance’s institutional — orm but has lost the exclud- ability that makes governance — unction. Governance institutions with public good characteristics. Some governance institutions generate bene — its so di —

use that ex- cluding non-bene — iciaries is either physically impossible or socially impractical. A pro — essional licensing system that certi — ies physician competence bene — its not only the physicians who are credentialed and the patients who choose among them, but the general public that relies on pro — essional standards. Non-members receive the ben- e — its o — governance quality whether they contribute to governance costs or not. When an institution’s non-member bene — its dominate 120 Law and Governance

its member bene

its, the institution begins to exhibit public good characteristics at the margin. Chapter 6 takes up the analysis o — these spillover bene — its and why they matter — or the legal evaluation o —

governance.

What the classi

ication explains Why exclusion is the paradigm sanction. Under the club good structure, exclusion deprives a member o — non-rivalrous bene — its at zero marginal cost to the institution and the remaining member- ship. This asymmetry gives exclusion unique e —

ectiveness as a de- terrent. A sanction that imposes maximum cost on the violator at zero cost to the en — orcing institution will be pre — erred over sanc- tions that impose costs on both parties. No other en — orcement mechanism available to a private governance institution has this property. Monetary damages require litigation, impose costs on the institution, and may be uncollectable. Suspension imposes adminis- trative costs and requires monitoring the suspended member’s con- tinued exclusion. Expulsion imposes no continuing administrative cost once executed. The club good classi — ication also explains why converting expul- sion — rom a property rule to a liability rule damages governance more severely than the conversion o — most other entitlements. 148 When courts require damages in lieu o — exclusion, they eliminate the cost asymmetry that makes exclusion an e —

ective deterrent. The expelled member can e —

ectively purchase continued access at a price set by the court. The institution must litigate every expulsion deci- sion that a member contests. Monitoring and decision-making

unctions continue, but the sanction mechanism loses the credibility that converts potential de — ectors into compliant members. Error! Use the Home tab to apply Heading 1 to the text that you want to appear here. 121

 Why governance institutions are undersupplied. Because governance  --- unctions are non-rivalrous and excludability limits the institution's ability to charge non-member bene --- iciaries, govern- ance institutions tend to be undersupplied relative to the social op- timum. Members bear governance costs while non-members re- ceive governance bene --- its  --- or  --- ree. Mancur Olson's analysis o ---  group size and collective action provides the complementary mecha- nism.149 As governance institutions grow larger, the  --- ree-rider prob- lem intensi --- ies: each individual member's contribution to govern- ance is a smaller  --- raction o ---  the total, the connection between any member's behavior and governance quality becomes more attenu- ated, and the incentive to shirk monitoring and sanctioning respon- sibilities grows. Governance institutions  --- ace structural pressure to- ward undersupply o ---  governance  --- unctions as membership grows beyond the optimal size. This structural pressure explains why gov- ernance institutions require legal conditions that support their ca- pacity to  --- unction: without legal protection  --- or the club good struc- ture, governance institutions will undersupply governance even be-

ore legal degradation begins. Why the legal conditions identi — ied in Chapter 4 are nec- essary. The club good structure explains why each o — the six legal conditions Chapter 4 identi — ied is necessary — or governance to — unc- tion. Permission and recognition supply the legal basis — or voluntary association that is the — oundation o — club — ormation. Membership boundary protection supplies the legal support — or excludability that is the — oundation o — club governance. External en — orceability makes the institution’s internal rules, which operate as the working rules o — the club, binding beyond social pressure alone. Fiduciary struc- ture attaches legal obligations to the governance authority that club 122 Law and Governance

members delegate to leaders. Standing gives someone the legal ca- pacity to challenge governance — ailures that undermine the club’s ca- pacity to supply governance. The club good — ramework makes clear why each condition is necessary: each condition supports a distinct structural — eature o — governance institutions, and law must sustain all six i — governance is to — unction.

The typology’s limits The club good classi — ication yields three structural consequences — or legal analysis: it identi — ies which conditions support governance

unction, it explains why governance institutions tend toward un- dersupply, and it shows why legal intervention requires calibration rather than prohibition. The classi — ication does not determine whether a governance institution is governing well or badly, whether its exclusion decisions are justi — ied or arbitrary, whether its membership rules are — air or discriminatory, or whether its positive externalities justi — y legal protection o — its governance capacity. Chapter 8 develops the evaluative method that addresses these ques- tions. The typology is an analytical instrument, not a normative ver- dict. Club goods with signi — icant market power require di —

er- ent legal treatment. Buchanan’s wel — are analysis assumes compet- itive markets — or club memberships: ine —

icient clubs lose members to more e —

icient alternatives. When a governance institution ac- quires su —

icient market power that members cannot practically exit, the competitive discipline that justi — ies de — erence to club govern- ance decisions breaks down. This is the antitrust concern that Silver and Northwest Wholesale Stationers address: at some level o — market power, governance decisions can no longer be evaluated simply as Error! Use the Home tab to apply Heading 1 to the text that you want to appear here. 123

club management and require scrutiny

or their e —

ects on competi- tion.150 The club good classi — ication does not resolve the distribu- tive justice challenge. Club goods de — ined by willingness-to-pay systematically exclude those who cannot a —

ord membership. When governance institutions generate substantial bene — its — or the broader public, private provision through a club may be e —

icient in the Pareto sense while being inequitable in the distributive sense. Chapter 14 addresses that challenge directly. The club good — rame- work in this chapter is concerned with the institutional structure o —

governance, not with whether every governance institution serves distributive justice.

Member bene

its and spillovers: setting up Chapter 6 The club good classi — ication sets up the analytical work o — the next chapter. Chapter 6 distinguishes the member bene — its that govern- ance institutions supply to their members — rom the spillover bene-

its that governance institutions generate — or non-members. The distinction matters — or law because it determines who bears the cost o — governance degradation: when governance institutions generate signi — icant spillovers, degradation o — governance capacity harms the public as well as the membership, and the justi — ication — or legal pro- tection o — governance capacity is correspondingly stronger. Making that argument requires a clean separation between member bene — its and spillovers that the club good — ramework sup- plies. Member bene — its are the non-rivalrous governance — unctions the club supplies to its members. Spillovers are the external bene — its that — low to non-members who never pay the costs o — club member- 124 Law and Governance

ship. These are di

erent goods with di —

erent bene — iciaries and di — -

erent wel — are implications. The chapter that — ollows maintains this distinction throughout. Chapter 6: Member Bene — its and Spillovers — rom Governance Chapter 5 established that governance institutions have the struc- ture o — club goods: excludable, non-rivalrous in their governance

unctions up to a congestion threshold, voluntarily joined, and sel — -

inanced through member contributions. The present chapter iden- ti — ies what governance institutions produce, — or whom, and why the distinction matters — or legal design. The prescriptive thesis deserves precise statement be — ore any ar- gument builds toward it: law should protect governance production because governance generates value in two distinct — orms — mem- ber bene — its that — low to members who bear governance costs, and spillover bene — its that — low to non-members who pay nothing. Both outputs depend on the same institutional architecture, and when law damages that architecture, both outputs degrade. Protection o —

governance is not justi

ied by member bene — its alone, but by the joint acknowledgment that governance is produced through club- good mechanisms and consumed partly as a public good — non-ex- cludable bene — its — lowing to bene — iciaries who cannot be charged and did not choose to join. Legal actors encountering governance 126 Law and Governance

invariably encounter it

rom the outside, through the lens o — the ex- cluded or underserved: Harold Silver excluded — rom the NYSE, stu- dents harmed by a university’s governance — ailure, retail investors unable to in — luence proxy outcomes. When law responds by attack- ing excludability — requiring reinstatement o — expelled members, imposing liability — or exclusion decisions, mandating open partici- pation — it attempts to correct the public-good consumption — ailure by dismantling the club-good production mechanism. But produc- tion is prior to distribution. A governance institution that cannot exclude cannot produce governance at all, which means the spillo- vers that made the institution valuable disappear along with the club. Recognizing governance as a dual-structure good, club-good in production and public-good-adjacent in consumption, makes the

ramework’s prescriptive direction precise: law should protect the production mechanism while targeting governance — ailures speci — i- cally, rather than dismantling production in the name o — improving distribution. The two tasks are not equivalent, and con — lating them is the structural error this book traces across six institutional do- mains. Error! Use the Home tab to apply Heading 1 to the text that you want to appear here. 127

Chapter 6. Governance Surplus Governance institutions produce value in two analytically distinct channels. The — irst runs to members: the coordination bene — its, cost reductions, monitoring gains, and dispute resolution that make membership worthwhile. The second runs to outsiders: the spillo- ver e —

ects that — low beyond membership boundaries when govern- ance operates well. Separating these two channels matters because they create di —

erent legal stakes and respond di —

erently to legal in- tervention. The sections that — ollow examine each in turn and then show how the two interact. A necessary quali — ication applies to spillovers speci — ically. Spill- overs — rom governance can be positive or negative. When govern- ance — unctions well, spillovers are o — ten positive: the DDC’s — raud suppression bene — its downstream consumers, exchange surveillance protects retail investors, and community — orest management se- questers carbon — or global bene — it. When governance — ails, the with- drawal o — those bene — its imposes costs on the same outsiders who previously bene — ited, as the 2008 — inancial crisis demonstrated at catastrophic scale. But a distinct category exists that the — ramework must acknowledge — rom the outset: governance that — unctions e —

ec- tively — or members can simultaneously impose negative spillovers on outsiders. A trade association that coordinates member conduct may reduce transaction costs among members while — ixing prices paid by non-member buyers. A pro — essional licensing body that cer- ti — ies practitioner quality may protect consumers — rom incompetent providers while blocking quali — ied competitors and raising costs — or the public. In these cases the governance institution satis — ies the

our-element de — inition o — Chapter 1, produces genuine member bene — its, and operates through a — unctional club-good mechanism 128 Law and Governance

— yet the spillover to outsiders is harm

ul rather than bene — icial. Antitrust law, which Chapter 14 addresses directly, is the primary legal response to governance institutions whose member bene — its come at outsiders’ expense. The present chapter’s analysis o — posi- tive spillovers should not be read to imply that all governance spill- overs are positive. The — ramework’s evaluative method at Step 5 re- quires identi — ying spillovers without presuming their sign.

What governance provides to members Members o — governance institutions receive — our categories o — di- rect bene — it — rom their participation. Each corresponds to a speci — ic market — ailure that governance corrects. Transaction cost reduction. Every transaction between strangers requires e —

ort be — ore, during, and a — ter the exchange it- sel — . Parties must — ind each other, assess each other’s reliability, and negotiate terms be — ore exchange; they must monitor per — ormance during exchange; and they must en — orce obligations and resolve dis- putes when exchange goes wrong. Oliver Williamson identi — ied these costs as the central problem that governance structures exist to solve.151 Governance institutions reduce all three categories — or their members. The New York Diamond Dealers Club provides the most closely studied example. Lisa Bernstein’s study — ound that by the early 1990s the club had approximately two thousand members, o — whom seven hundred to eight hundred used the trading — loor each day. Transac- tions worth hundreds o — thousands o — dollars were concluded on handshake terms, sealed with the Yiddish phrase mazel u’broche, with no written contract. One member’s oral o —

er was binding on both Error! Use the Home tab to apply Heading 1 to the text that you want to appear here. 129

parties under the club’s bylaws.152 The governance mechanism ena- bling this simplicity was the club’s combined arbitration and repu- tation system. The screening requirement — or DDC membership, which required two years in the industry, board approval, and a pro- bationary period, served as a signal that all parties on the trading

loor had been vetted. The bourse’s monitoring o — member behavior through what Bernstein calls the “grapevine” tracked compliance and disseminated in — ormation about who honored obligations dur- ing ongoing trading relationships. When disputes arose, members submitted them to internal arbitration that resolved approximately eighty- — ive percent o — cases during mandatory conciliation, with re- maining cases decided within ten days o — hearing.153 The governance institution reduced the search cost, the negotiation cost, and the en-

orcement cost o — commercial exchange. Members could transact at scale and speed that — ormal legal mechanisms cannot replicate. Robert Cooter and Janet Landa developed the theoretical — rame- work — or this mechanism, arguing against the standard economic assumption that trade is impersonal.154 In a world o — contractual un- certainty, it matters whom one trades with. Governance institutions create and sustain a stock o — institutional trust that substitutes — or

ormal legal en — orcement. This trust stock is itsel — a non-rivalrous good: one member’s reliance on the institution’s reputation moni- toring does not reduce another member’s ability to rely on it, and one member’s use o — the institution’s arbitration system does not di- minish the system’s capacity to serve other members. The govern- ance mechanism is a club good that produces transaction cost reduc- tion as the non-rivalrous membership bene — it. Coordination. Many problems that groups — ace are not con-

licts o — interest but problems o — coordination: multiple equilibria are available, all parties would bene — it — rom settling on the same one, 130 Law and Governance

but without a mechanism

or selection, parties choose di —

erent op- tions and the gains — rom coordination are lost. Governance institu- tions solve coordination problems by establishing the — ocal points around which members converge. Thomas Schelling demonstrated that in coordination games, parties spontaneously converge on solutions that are salient, prom- inent, natural, or conventional, because each party anticipates that other parties will converge on the same option.155 Governance insti- tutions create institutional salience by establishing explicit rules, trade customs, and standard practices that convert the coordination problem into a determinate outcome. All members know the rules, and all members know that all other members know the rules. The institution’s working rules become the — ocal point. Technical standard-setting bodies illustrate the coordination bene — it at scale. The Institute o — Electrical and Electronics Engineers maintains over twelve hundred active standards, including IEEE 802.11, the technical speci — ication underlying Wi-Fi networking. The International Organization — or Standardization coordinates one hundred sixty-one national standards bodies and produces tech- nical speci — ications spanning industrial processes, product sa — ety, environmental management, and in — ormation security. The gov- ernance — unctions o — these institutions, which include setting pro- cedures — or how standards are developed, which proposals are de- bated, how consensus is reached, and how standards are revised, en- able the coordination bene — it that members receive. Members par- ticipate in governance and in return gain access to dra — t standards be — ore publication, voting rights in the development process, and the opportunity to shape speci — ications that their products will need to meet. These are excludable membership bene — its, and they di —

er

rom the public bene — it o — widely adopted standards, which accrues Error! Use the Home tab to apply Heading 1 to the text that you want to appear here. 131

to anyone who implements the standard regardless o

participation in the governance institution. Monitoring and adverse selection. When buyers cannot as- sess the quality o — what they are purchasing, markets — ail: sellers o —

low-quality goods drive out sellers o

high-quality goods because buyers, unable to distinguish them, pay only the average price. George Akerlo — established this dynamic through the analysis o —

used-car markets, and the same logic applies to any market in which quality is observable to sellers but not to buyers at the time o — pur- chase.156 Governance institutions correct this — ailure by monitoring member behavior and certi — ying member quality. The certi — ication — unction is an excludable member bene — it. DDC membership signaled to counterparties that a dealer had been vetted and was subject to ongoing monitoring. Bernstein observed that membership enabled dealers to signal trustworthiness and gave them assurance that all other dealers on the trading — loor had met the same requirements. 157 The credential — unction o — pro — essional association membership operates on the same logic: bar member- ship signals that an attorney has passed a character and — itness re- view and remains subject to pro — essional discipline, allowing clients to transact with attorneys at lower search cost than they could achieve with unregulated practitioners. The monitoring — unction extends beyond initial certi — ication. Governance institutions sustain ongoing reputation tracking, dis- seminating in — ormation about member conduct throughout the membership. This addresses a second in — ormation problem: not merely the initial adverse selection problem o — whether a new coun- terparty can be trusted, but the ongoing moral hazard problem o —

whether an established counterparty will continue to per

orm. Be- 132 Law and Governance

cause the reputation mechanism operates across the

ull member- ship, each member’s knowledge o — every other member’s reputation is maintained and updated at institutional cost rather than individ- ual cost. No single member could e —

iciently maintain reputation in-

ormation spanning all potential counterparties. The institution provides this as a shared, non-rivalrous service. Dispute resolution. When transactions go wrong, governance institutions provide mechanisms — or resolution that are — aster, less costly, and better calibrated to the subject matter than general-pur- pose courts. The National Grain and Feed Association, which Bern- stein documented in a companion study, operates the oldest contin- uously — unctioning industry-based arbitration system in North America, dating to 1901.158 The NGFA’s arbitration committees con- sist o — members with industry expertise and no commercial interest in the case. Arbitrators understand the commercial context o — grain trade disputes in ways that general-jurisdiction courts cannot repli- cate. The system resolves disputes with specialized knowledge and at lower cost than litigation. Comparative resolution timelines con — irm the advantage. Aver- age domestic commercial arbitration cases in the United States re- solve in roughly twelve months. Federal district court civil cases take two to three years on average to reach trial disposition, and complex commercial litigation — requently exceeds that range.159 The cost di —

erential is proportionately similar. These di —

erences are es- pecially pronounced in governance institution dispute resolution, where the specialized subject matter knowledge o — arbitrators — ur- ther reduces the evidentiary burden that parties must satis — y to es- tablish their claims. Error! Use the Home tab to apply Heading 1 to the text that you want to appear here. 133

What governance provides to those who do not pay The member bene — its described above — low to the people who bear the costs o — governance. A separate category o — bene — its — lows to people who never join and never pay. These spillover bene — its are generated by governance quality and are not reducible to the mem- bership mechanism that produces member bene — its. Three domains illustrate the structure. Trading network governance and supply chain integrity. The DDC’s governance mechanism produced bene — its — or members through the trust stock described above. The same mechanism pro- duced bene — its — or people who never set — oot on the trading — loor. Downstream retailers, jewelers, insurers, and ultimately consumers who purchased diamonds bene — ited — rom reduced — raud rates in the supply chain. Because the DDC’s monitoring and sanction mecha- nisms sustained the cooperative norm o — honest dealing, the prod- ucts that moved through the governed supply chain were more re- liably what sellers represented them to be. This quality assurance reached consumers who had no relationship with the DDC and no ability to assess diamond quality directly. Barak Richman’s documentation o — the DDC’s partial erosion con — irms this structure by tracing what happened when governance quality declined. As the trust-based system eroded under pressure

rom globalization and new market entrants who operated outside the governance network, — raud costs spread through the supply chain. Banks reduced credit to diamond dealers, increasing — inanc- ing costs throughout the industry. Retailers who had invested in branded jewelry — aced increased vulnerability to misrepresentation o — synthetic diamonds as natural ones, a — raud risk that DDC gov- ernance had previously suppressed. Vertical integration replaced 134 Law and Governance

trust-based exchange as the dominant transactional

orm, eliminat- ing the cost-e —

icient governance mechanism and substituting a more expensive institutional structure.160 Non-members had been receiving a governance externality they did not pay — or, and when that externality disappeared, the costs — ell on them. Two — eatures distinguish this example. First, the spillover was proportional to governance quality: the spillover existed when gov- ernance — unctioned, degraded when governance weakened, and dis- appeared when governance collapsed. The spillover was not a — ixed endowment but a variable output o — the governance system. Second, the mechanism producing the spillover was the same mechanism that produced the member bene — it: the credible threat o — exclusion sustained the cooperative norm among members, and the coopera- tive norm among members prevented — raud — rom reaching down- stream non-members. These two outputs were jointly produced by the same governance architecture, and weakening the architecture damaged both simultaneously. Financial market governance and investor protection. Stock exchanges and sel — -regulatory organizations govern the be- havior o — broker-dealers and market participants through listing standards, trading surveillance, member discipline, and en — orce- ment actions. The member bene — it o — this governance is access to the exchange’s in — rastructure and the reduced transaction costs that orderly markets provide to member — irms. The spillover bene — it is market integrity: price discovery that produces accurate signals — or all investors, — raud detection that protects all market participants, and systemic stability that protects the broader economy. Paul Mahoney documented this structure in his analysis o — stock exchanges as regulators.161 Exchange sel — -regulation generates mar- ket integrity as a collective good accruing to investors who are not Error! Use the Home tab to apply Heading 1 to the text that you want to appear here. 135

exchange members. Retail investors who purchase publicly traded securities bene — it — rom exchange surveillance systems without con- tributing to their costs. Institutional investors whose assets are managed through broker-dealers bene — it — rom the regulatory — rame- work those dealers operate under. The exchange’s governance pro- duces value — or a constituency vastly larger than its membership. The 2008 — inancial crisis illustrates the magnitude o — the exter- nality through its inverse. A convergence o — governance — ailures in mortgage origination, securitization, credit rating, and risk man- agement produced costs that — ell overwhelmingly on people who had no role in those governance systems. The United States Gov- ernment Accountability O —

ice estimated the output losses and — iscal costs o — the crisis at up to thirteen trillion dollars.162 The households who lost their homes, the workers who lost their jobs, and the pen- sion — unds that saw their assets collapse were not parties to the gov- ernance — ailures that produced the crisis. These parties were recipi- ents o — a negative externality, the inverse o — the positive spillovers that governance, when it — unctions, provides. The scale o — the harm demonstrates the magnitude o — the governance surplus that — unc- tioning — inancial governance sustains. Commons governance and environmental outcomes. The governance institutions that manage common-pool resources pro- duce member bene — its through coordinated resource use and re- duced con — lict over access. The same governance institutions pro- duce spillover bene — its through the environmental outcomes o — sus- tainable management. Ashwini Chhatre and Arun Agrawal studied eighty — orest com- mons across ten countries and — ound that — orests managed by larger groups o — local users with greater rule-making autonomy stored sig- 136 Law and Governance

ni

icantly more carbon than — orests under government manage- ment.163 Carbon sequestration is a global positive externality. The atmospheric bene — its o — carbon stored in a community — orest accrue to populations worldwide who have no relationship with the gov- erning institution and contribute nothing to its costs. The govern- ance institution’s members, the households who participate in rule- making, monitoring, and en — orcement, bear the cost o — maintaining the governance system. The bene — iciaries o — the resulting climate outcome are not con — ined to those members. Nepal’s community — orestry program provides national-scale evidence. Since the 1993 Forest Act enabled the trans — er o — national

orests to community management groups, Nepal has signi — icantly increased its — orest cover. 164 Approximately twenty-two thousand community — orest user groups, comprising millions o — households, now manage millions o — hectares o —

orest under governance sys- tems that include membership boundaries, collective-choice ar- rangements, monitoring, and graduated sanctions, the — our govern- ance elements Chapter 1 identi — ied. The downstream bene — its o — this governance, including reduced soil erosion, improved water quality, watershed protection, and carbon sequestration, — low to popula- tions — ar beyond the governing communities. Valencia’s Tribunal de las Aguas, which has governed water al- location — rom the River Turia since at least the tenth century and was recognized by UNESCO as intangible cultural heritage in 2009, demonstrates that these spillovers are not recent phenomena.165 The governance institution ensures that upstream irrigation does not deprive downstream users o — water and that in drought conditions usage rights are shared equitably throughout the system. The down- Error! Use the Home tab to apply Heading 1 to the text that you want to appear here. 137

stream water quality and availability bene

its o — this governance ex- tend to communities that are not parties to the governance institu- tion and do not participate in its decision-making. The three domains reveal a consistent pattern. Governance spillovers are not incidental byproducts o — an otherwise member-

ocused institution. Governance spillovers can be outputs o — the governance mechanism itsel — , proportional to governance quality when they exist, and dependent on the same institutional architec- ture that produces member bene — its. The DDC’s — raud reduction, the exchange’s market integrity, and the commons’ environmental outcomes can all be — unctions o — how well the governance institu- tion per — orms its core work: making and en — orcing rules, monitor- ing compliance, and sanctioning de — ection. Proportionality, however, operates along each governance di- mension separately, not as a global relationship between an undi — -

erentiated “governance quality” and an undi —

erentiated “spillover.” Each o — the — our governance elements can produce speci — ic spillovers through identi — iable causal mechanisms. Monitoring quality can produce — raud-reduction spillovers: the DDC’s reputation tracking prevented misrepresentation — rom reaching downstream retailers and consumers, and when monitoring eroded, — raud spread through the supply chain in proportion to the monitoring — ailure. Sanction- ing quality can produce deterrence spillovers: the NYSE’s credible threat o — expulsion deterred market manipulation that would have harmed all investors, and when Silver raised the cost o — sanctioning past the point o — absorption, deterrence declined and market integ- rity spillovers declined proportionally. Decision-making quality can produce coordination spillovers: the IEEE’s consensus procedures establish technical standards that enable interoperability across the entire technology sector, and when decision-making is subverted, 138 Law and Governance

as Allied Tube documented, coordination

ails — or members and non- members alike. Adjustment quality can produce adaptive-capacity spillovers: the Swiss alpine commons’ capacity to revise pasture-use rules in response to changing conditions sustained environmental outcomes over centuries, and when adjustment mechanisms rigid- i — y, the environmental spillovers that depend on governance adapt- ability degrade proportionally. Tracing each spillover to its speci — ic governance dimension is essential — or the legal analysis that — ollows, because a legal rule that damages monitoring does not necessarily damage decision-making, and the spillover consequences o — each

orm o — degradation di —

er. Four steps identi — y whether a governance institution generates positive spillovers in a new institutional context. First, identi — y the governance output: which o — the — our governance elements — mon- itoring accuracy, sanction credibility, decision-making quality, or adjustment responsiveness — produces the candidate spillover? Sec- ond, trace the output’s causal pathway beyond the membership boundary: does the governance output generate bene — its that reach non-members, and through what mechanism? The DDC’s monitor- ing accuracy reduced — raud — or downstream retailers through the supply chain; the NYSE’s sanction credibility deterred manipulation that would have harmed all investors through the price-discovery mechanism. Third, apply the inverse test: does the spillover degrade when governance quality declines? I — the candidate bene — it persists even when the governance element that supposedly produces it has deteriorated, the bene — it is not a governance spillover — it is pro- duced by some other mechanism. The DDC example satis — ies this test: when monitoring quality eroded, — raud spread through the sup- ply chain proportionally.166 The — inancial crisis example satis — ies it as well: when exchange governance — ailed to detect and discipline Error! Use the Home tab to apply Heading 1 to the text that you want to appear here. 139

risky trading practices, the market integrity spillovers disappeared and the costs — ell on non-members. Fourth, apply the substitutability test: can the spillover be replicated by a non-governance mecha- nism, such as direct state regulation, market competition, or private contract? I — the spillover is — ully substitutable without governance, then the governance institution’s claim to legal support on spillover grounds is correspondingly weaker, because the spillover does not depend on the governance institution’s continued operation. Where the spillover depends on institutional — eatures that non-gov- ernance mechanisms cannot replicate — contextual monitoring, reputational sanctioning, adaptive norm revision — the governance institution’s spillover contribution is non-substitutable, and legal rules that degrade it impose costs that no alternative institution can o —

set. This — our-step protocol does not claim to predict spillovers in novel institutional settings be — ore governance — ailure reveals them. Spillover identi — ication is principally retrospective: governance sur- plus becomes visible when it is destroyed, as Costanza and col- leagues demonstrated in the analogous context o — ecosystem ser- vices, where the value o — ecological — unctions became quanti — iable only when those — unctions were lost.167 What the protocol does is structure the retrospective analysis so that identi — ied spillovers can be traced to speci — ic governance dimensions, tested — or genuine gov- ernance dependence, and evaluated — or substitutability — the three characteristics that determine whether legal support — or the govern- ance institution is justi — ied on spillover grounds. 140 Law and Governance

The errors o

con — lation Con — lating member bene — its with spillovers produces two analytical errors. The — irst error is overvaluing exclusion. The club good structure o — governance institutions is de — ined by excludability. The club can deny access to the good by denying membership or expel- ling violators. Exclusion is the mechanism that makes club member- ship valuable: the threat o — exclusion disciplines behavior, the en-

orcement o — exclusion removes de — ectors, and the credibility o — ex- clusion sustains cooperation. Chapter 5 explained why exclusion is the paradigm sanction o — private governance institutions and why converting expulsion — rom a categorical rule to a compensated rem- edy degrades governance capacity. I — spillover bene — its are treated as member bene — its, however, the analyst attributes to the club mechanism a value that actually

lows to non-members regardless o — the club’s exclusion decisions. The case — or protecting exclusion as the mechanism producing that value then becomes overinclusive. Spillover bene — its o — governance quality — low to non-members because o — the cooperative norm the exclusion threat sustains among members. Spillover bene — its do not

low to non-members because non-members are excluded. An anal- ysis that credits exclusion with producing the spillover value attrib- utes the non-member bene — it to the wrong mechanism. The correct attribution runs as — ollows: exclusion produces cooperation, coop- eration produces governance quality, and governance quality pro- duces both member bene — its and spillovers as joint outputs. Con — lat- ing member bene — its with spillovers obscures this chain and sup- ports the protection o — exclusion as a direct cause o — social value it does not directly cause. Error! Use the Home tab to apply Heading 1 to the text that you want to appear here. 141

 The second error is undervaluing governance degrada- tion. When a legal rule degrades a governance institution's capacity to  --- unction, by converting its expulsion power  --- rom a categorical rule to a compensated remedy, by imposing procedural require- ments that make discipline too costly to en --- orce, or by mandating open membership that eliminates the screening  --- unction, the direct cost  --- alls on members. Members lose transaction cost savings, coor- dination bene --- its, monitoring services, and dispute resolution ca- pacity.
 I ---  spillovers are ignored and only member bene --- its are counted, the social cost o ---  governance degradation is systematically underes- timated by exactly the amount o ---  spillover value that was being pro- duced. Standard wel --- are economics establishes this point  --- ormally: when a positive externality exists, social bene --- its exceed private ben- e --- its.168 An assessment o ---  degradation costs that counts only private bene --- its misses the harm to non-members who were receiving spill- overs.
 The scale o ---  this underestimation can be substantial. The DDC's membership numbered in the thousands; the downstream supply chain the DDC's governance served extended to millions o ---  retailers and hundreds o ---  millions o ---  consumers.169 Any governance institu- tion whose membership represents a  --- raction o ---  the population its governance serves will generate spillover harm  --- ar exceeding mem- ber harm when governance degrades. A court or legislature that measures the cost o ---  intervention against member bene --- its alone will systematically underweight the  --- ull social consequence o ---  the intervention.
 The correct analytical tool is the joint products  --- rame- work, not pure club theory. James Buchanan's 1965 paper ad- dressed a governance institution in which bene --- its  --- low exclusively

142 Law and Governance

to members: the club excludes non-members, and excludes them

ully. When a governance institution also generates non-excludable spillovers, Buchanan’s — ramework does not cover the — ull analysis. Richard Cornes and Todd Sandler developed the joint products

ramework — or exactly this situation: a good that jointly produces an excludable private characteristic and a non-excludable public char- acteristic. 170 Governance institutions that generate spillovers pro- duce two joint products. The club mechanism, excludability, volun- tary membership, and sel — - — inancing, governs the private character- istic. The externality mechanism, public good provision to non- members, governs the public characteristic. Legal analysis o — gov- ernance institutions requires both — rameworks operating simulta- neously.

Governance surplus Member bene — its plus spillovers equal the total social value o — gov- ernance. This total exceeds the private cost that members bear to produce governance, because members can charge — or the club good they supply to members but cannot charge non-members — or the ex- ternalities those non-members receive — or — ree. The excess o — total social value over member cost is governance surplus. The term is novel, but its components are — irmly established in wel — are economics. Arthur Cecil Pigou demonstrated that goods generating positive externalities, what Pigou called incidental un- charged services, tend to be undersupplied by the market.171 Produc- ers cannot capture the — ull social bene — it o — what they produce, so they may produce less than the socially optimal amount. Applied to governance: members bear all the costs o — governance but capture only part o — the bene — its. Non-members receive spillover bene — its Error! Use the Home tab to apply Heading 1 to the text that you want to appear here. 143

without contributing to costs. Because the

ull social return to gov- ernance may exceed the private return to members, governance in- stitutions may be produced at less than the socially optimal level. This is the standard wel — are economics prediction — or any good with positive externalities, and governance institutions exhibit the same structural characteristics. The undersupply prediction is consistent with available evi- dence, though that evidence does not constitute a direct measure- ment o — governance surplus. The proportion o — the United States work — orce covered by occupational licensing has grown substan- tially over recent decades, but researchers disagree about whether this growth re — lects optimal provision o — the governance — unction or regulatory capture — or the bene — it o — incumbent practitioners.172 Pro — essional associations across sectors report membership plateaus and declines. 173 Commons governance institutions — ace persistent

unding challenges. The structure o — governance institutions, with members bearing costs and non-members receiving bene — its, creates a systematic gap between private and social incentives — or govern- ance provision. The chapter does not claim to measure that gap pre- cisely. It claims only that the gap exists and that its existence has predictable consequences — or how governance institutions will be provided absent countervailing legal support. The cost-o — - — ailure evidence illuminates the magnitude o — gov- ernance surplus without requiring its direct measurement. When a governance institution — ails, the costs include both the loss o — mem- ber bene — its and the loss o — spillovers. I — the spillover component is large, the total cost o —

ailure will substantially exceed what any anal- ysis — ocused on member harm would predict. The — inancial crisis es- timate o — up to thirteen trillion dollars in output losses substantially exceeded what an analysis o — member-institution harm alone would 144 Law and Governance

have produced, because the harm

ell overwhelmingly on non- members. 174 The erosion o — DDC governance spread — raud costs through a supply chain serving — ar more consumers than members. The degradation o — commons governance accelerates resource de- pletion, generating costs borne by populations that never partici- pated in the governance institution. Each case demonstrates a gov- ernance surplus that existed be — ore the — ailure and was destroyed by the — ailure. The surplus went unmeasured precisely because it seemed secure. Its magnitude became visible only when it was de- stroyed. Methodological candor requires acknowledging that govern- ance surplus is identi — ied principally through — ailure costs—the counter — actual analysis o — what is lost when governance breaks down. This approach depends on base-rate estimation problems that are inherent to any in — erence — rom negative to positive states. When governance — ails, multiple causal — actors may contribute to harm, and isolating governance — ailure’s speci — ic contribution re- quires strong assumptions about the counter — actual state o — govern- ance — unctions absent the identi — ied — ailure. Governance institutions operating at di —

erent scales and in di —

erent institutional contexts generate di —

erent spillover magnitudes, and the spillover bene — it per member varies substantially across domains. These variations make any generalized magnitude assessment necessarily approxi- mate. The claim is not that governance surplus can be precisely measured or that its measurement is unambiguous. The claim is that the surplus exists, that it is substantial in demonstrable cases, and that legal analysis must account — or its existence rather than pre- tending that governance costs and bene — its are con — ined to the mem- bership. Error! Use the Home tab to apply Heading 1 to the text that you want to appear here. 145

One quali --- ication requires emphasis. The analysis o ---  governance surplus is not an argument that governance institutions are always bene --- icial, that their exclusion decisions are always justi --- ied, or that legal intervention in governance institutions is always harm --- ul. The analysis is descriptive rather than normative: it identi --- ies what gov- ernance institutions produce, separates the two categories o ---  output, and demonstrates that legal analysis con --- ined to member bene --- its is working with an incomplete accounting. Chapter 14 addresses the governance abuse, distributive justice, and antitrust challenges di- rectly. Whether any particular governance surplus justi --- ies legal protection o ---  a particular governance institution's exclusion mech- anism is a question the method developed in Chapter 8 must answer in context.

What this means

or legal analysis Governance institutions produce two analytically distinct outputs: member bene — its, which — low to the people who bear governance costs through the club mechanism, and spillover bene — its, which

low to non-members who pay nothing. Con — lating these two out- puts produces systematic errors in legal analysis, overvaluing exclu- sion on the one hand and undervaluing degradation on the other. And governance institutions are likely to be undersupplied because members cannot capture the — ull social return on their investment in governance. Chapter 7 develops the implications — or how law acts on gov- ernance in both directions. The undersupply prediction this chapter identi — ies has a speci — ic legal corollary: when a legal rule reduces the private cost o — governance production, it narrows the gap between private and social returns and moves governance supply toward the 146 Law and Governance

social optimum. Chapter 7 identi

ies legal doctrines that do exactly this, describing them as Pigouvian subsidies — named — or the same economist whose externality analysis this chapter applied to gov- ernance surplus. The enabling dimension o — law is easy to miss pre- cisely because the subsidy is structural (judicial de — erence, immunity

rom liability, procedural insulation) rather than — iscal (cash trans-

ers, tax credits). The disabling dimension is equally predictable. A legal rule that weakens a governance institution’s exclusion mecha- nism does not merely reduce member bene — its. The rule also de- grades governance quality, which reduces the spillover bene — its non-members were receiving. The — ull cost o — the rule exceeds what an analysis o — the immediate dispute between the governance insti- tution and the member seeking damages would capture. The parties be — ore the court represent only a — raction o — the parties a —

ected by the court’s decision. This is not a reason — or courts to re — use to de- cide governance disputes. It is a reason — or courts and legislators to understand what is at stake when they act on governance institu- tions: not only the disposition o — a dispute between identi — ied par- ties, but the treatment o — an institutional architecture that produces value — or a constituency — ar larger than either party to the litigation. Chapter 7: How Law Enables and Degrades Governance Chapters 4 through 6 supply the descriptive — oundation: what gov- ernance institutions require, what kind o — good they produce, and what value they generate — or members and non-members alike. This chapter turns — rom description to mechanism. Legal rules act on governance in two directions: enabling and degrading. Some legal rules reduce the cost o — producing governance, make private gov- ernance rules en — orceable, and protect the exclusion mechanism on which governance depends. These rules enable governance. Other legal rules damage governance through identi — iable, recurring mechanisms. These rules degrade governance. A third category o —

legal rules corrects speci

ic governance — ailures without destroying governance capacity more broadly. These rules discipline govern- ance. Enabling law and degrading law di —

er in character. Enabling law subsidizes governance production by reducing the private cost o —

generating a good whose

ull social value producers cannot capture. Degrading law raises the cost o — producing governance past the point that governance institutions can absorb. Discipline, which

alls in neither category, responds to speci — ic identi — ied governance 148 Law and Governance

ailures and is calibrated not to damage governance capacity beyond what correction requires. That last distinction is the hardest to draw. A legal rule that dis- ciplines governance is not the same as a legal rule that degrades gov- ernance, but the di —

erence is not always obvious, and courts and legislators do not always see it. The chapter’s — inal section develops the criteria. The — ramework is applied to speci — ic doctrinal domains in Part III; Chapter 14 addresses those limits most directly, including the claim that governance institutions can use the — ramework as a shield — or discriminatory exclusion. Error! Use the Home tab to apply Heading 1 to the text that you want to appear here. 149

Chapter 7. Law’s Governance E

ects The three categories — enabling, degrading, and disciplining — are analytical, and actual legal rules can shi — t between them depending on context. A rule that enables governance in one institutional set- ting may degrade it in another. The purpose o — the taxonomy is to direct attention to what a legal rule does to governance capacity, measured by its e —

ects on the — our — unctional elements. The sections that — ollow examine enabling mechanisms — irst, then degradation mechanisms, and — inally the criteria — or distinguishing discipline

rom destruction.

How law enables governance Law enables governance through — our mechanisms: de — erence, en-

orceability, property-rule protection, and recognition. Each sup- ports governance quality, not merely governance existence, and each operates through a logic that prior chapters make visible. Judicial de — erence as a governance subsidy. Court de — erence reduces the private cost o — governance production. Every govern- ance institution that exercises authority over members — aces litiga- tion risk: a discipline committee that expels a member — or cheating, an arbitration panel that renders an adverse award, a credentialing body that denies certi — ication. The expected costs o — that litigation, including attorney — ees, management attention, the uncertainty o —

outcome, and the reputational exposure o

contested decisions, — all on the governance institution whether the institution wins or loses. Governance institutions that anticipate heavy litigation invest less in monitoring, moderate their en — orcement, and develop de — ensive documentation that substitutes — or substantive governance. 150 Law and Governance

 De --- erence doctrines correct this by reducing expected litigation costs. The logic is Pigouvian: when a private activity can generate positive externalities, the private producer will undersupply the ac- tivity because the producer cannot charge  --- or the bene --- its that  --- low to non-members. Reducing the producer's private costs moves pro- duction closer to the social optimum. De --- erence doctrines apply this logic to governance, and they do so across  --- our distinct doctrinal settings.175
 The business judgment rule is the corporate governance expres- sion. In its Delaware  --- ormulation, the rule creates a presumption that directors acting on an in --- ormed basis, in good  --- aith, and with- out con --- licts o ---  interest satis --- y their  --- iduciary obligations, and courts will not substitute their judgment  --- or the board's on the merits o ---  a business decision.176 The conventional rationales  --- or this de --- erence, including preventing judicial incompetence in business decisions, reducing managerial risk aversion, and avoiding hindsight bias, ex- plain why de --- erence bene --- its the corporation and its shareholders. They do not explain why de --- erence bene --- its anyone outside the

irm. The governance — ramework supplies the — uller account: corpo- rate governance can generate positive externalities — or employees, creditors, suppliers, and the broader commercial economy, and the business judgment rule reduces the cost o — producing that govern- ance. When the Delaware Supreme Court pierced the rule in Smith v. Van Gorkom, 488 A.2d 858 (Del. 1985), and held directors person- ally liable — or approving a merger a — ter two hours o — deliberation without a — airness opinion, D&O insurance premiums rose sharply, quali — ied candidates grew reluctant to serve on boards, and the Del- aware legislature acted within a year to permit charter-level excul- pation o — directors — rom duty-o — -care damages liability. Del. Code Error! Use the Home tab to apply Heading 1 to the text that you want to appear here. 151

Ann. tit. 8, § 102(b)(7). 177 The governance consequences o

with- drawing de — erence were immediate and measurable. Chapter 11 ex- amines how the — ull spectrum o — Delaware’s review standards, — rom business judgment to entire — airness, maps onto the governance analysis this book develops. The Federal Arbitration Act’s pro-arbitration policy is the pri- vate ordering expression o — the same logic. Section 2 declares arbi- tration agreements “valid, irrevocable, and en — orceable,” and Moses H. Cone Memorial Hospital v. Mercury Construction Corp., 460 U.S. 1, 24 (1983), directed that doubts about arbitrability should be resolved in

avor o — arbitration.178 The FAA eliminates the — ree-rider problem that would otherwise undermine investment in private arbitration in — rastructure: a party who knows courts will not en — orce an arbitral award has diminished incentive to honor the governance system that produced it. Section 9 o — the FAA converts private arbitration decisions into court judgments en — orceable through all mechanisms available to courts, making private governance dispute resolution legally e —

ective while preserving its private character. Chapter 9 ex- amines how these doctrines operate in network governance settings where private arbitration is the primary en — orcement mechanism. The antitrust rule o — reason creates protected space — or cooper- ative governance activity. Per se condemnation o — coordinated com- mercial conduct would prohibit the activities through which gov- ernance institutions — unction: joint standard-setting, collective en-

orcement, shared monitoring, membership criteria, and collective discipline. Northwest Wholesale Stationers, Inc. v. Paci — ic Stationery & Printing Co., 472 U.S. 284, 296 (1985), held that cooperative govern- ance decisions deserve rule o — reason analysis, with per se condem- nation reserved — or institutions that possess market power or con- trol essential resources.179 Rule o — reason analysis asks whether the 152 Law and Governance

governance activity serves legitimate purposes and whether the re- straint is reasonably necessary to achieve them. The institution need not justi — y every governance decision as procompetitive, only those that raise genuine competitive concerns given the institution’s mar- ket position. Common law de — erence to voluntary associations is the oldest expression o — this logic. Courts have consistently re — rained — rom in- ter — ering in associations’ internal a —

airs absent bad — aith, violation o — the association’s own rules, or substantive irrationality.180 Every time a court reviews the merits o — an association’s governance deci- sion, the court imposes costs on the association, exposes the decision to reversal by a tribunal lacking contextual knowledge, and signals to potential rule-violators that sanctions can be contested. De — er- ence removes these costs. The — our doctrines emerged — rom di —

erent doctrinal traditions with di —

erent stated rationales. Each reduces the private cost o — pro- ducing governance, each is calibrated to allow judicial override un- der de — ined conditions rather than granting absolute immunity, and each produces outcomes consistent with positive-externality theory when externalities are present. None was consciously designed as an economic intervention. Each — unctions as an economic intervention nonetheless. En — orceability o — private ordering. De — erence reduces the cost o — governance decisions a — ter those decisions are made. A sec- ond category o — enabling law makes private governance rules bind- ing be — ore any dispute arises. When governance institutions estab- lish rules, those rules bind members through social pressure, repu- tational consequences, and the threat o — exclusion. But when a mem- ber contests a governance decision in court, when an arbitral award requires en — orcement against a non-complying party, or when a Error! Use the Home tab to apply Heading 1 to the text that you want to appear here. 153

governance institution needs to sue a member

or breach o — its rules, law must supply the en — orcement mechanism. The contract theory o — association rules provides the primary instrument. When a member joins an association, the charter and bylaws — orm a contract between the member and the organization. Courts will en — orce association rules when the institution acted within its authority, in good — aith, — ollowing its own published pro- cedures, and with reasonable notice to the member.181 A member who violates institutional rules and — aces expulsion cannot ignore the expulsion and continue trading on the institution’s in — rastruc- ture, because the expulsion has legal e —

ect through the contractual relationship. FAA Section 9 extends this en — orceability to arbitration awards by directing courts to con — irm awards on application by any party. Private governance dispute resolution thereby gains the — orce o — a court judgment. Delaware General Corporation Law § 122(18), which authorizes stockholder agreements restricting corporate governance decisions notwithstanding § 141(a)’s general board-management mandate, il- lustrates en — orceability in corporate governance. The provision was enacted in direct response to West Palm Beach Fire — ighters’ Pension Fund v. Moelis & Company, C.A. No. 2023-0309-JTL (Del. Ch. Feb. 23, 2024), which had invalidated such agreements as violations o —

mandatory law. The legislature’s rapid correction demonstrated the dependence: without en — orceability, private governance instru- ments become unen — orceable aspirations.182 Property-rule protection o — the exclusion mechanism. The third enabling mechanism is legal protection o — expulsion as a cate- gorical right rather than a compensated remedy. Chapter 5 ex- 154 Law and Governance

plained why exclusion is the paradigm sanction o

private govern- ance institutions: exclusion deprives a member o — non-rivalrous bene — its at zero cost to the remaining membership, and the asym- metry between the cost to the expelled member and the cost to the institution gives the exclusion threat its deterrent — orce. That deter- rent — orce depends on the threat being credible, which requires that courts not routinely second-guess expulsion decisions or convert expulsion — rom a categorical right into a compensated remedy. When courts protect expulsion as a property-rule entitlement, they preserve the governance institution’s ability to remove a de — ec- tor without compensating the de — ector — or lost membership. The common law’s traditional rule, under which courts de — er to expul- sion decisions that — ollow the institution’s own procedures and are not arbitrary, gives governance institutions the categorical author- ity that e —

ective deterrence requires. Blatt v. University o — Southern Cali — ornia, 5 Cal. App. 3d 935 (1970), sustained an honorary society’s retroactive eligibility rule on precisely this ground.183 The limits o — this protection are equally important and equally enabling. Pinsker v. Paci — ic Coast Society o — Orthodontists, 12 Cal. 3d 541 (Cal. 1974), and Falcone v. Middlesex County Medical Society, 34 N.J. 582 (1961), establish that when a governance institution controls ac- cess to essential pro — essional resources, its expulsion decisions be- come subject to a substantive rationality requirement. This limit does not undermine the property-rule character o — ordinary expul- sion; it identi — ies the speci — ic condition, monopolistic control over access to livelihood, that justi — ies more intrusive review. By speci — y- ing when categorical protection ends, the law gives governance in- stitutions outside that zone con — idence that their expulsion deci- sions will stand. Error! Use the Home tab to apply Heading 1 to the text that you want to appear here. 155

 Legal recognition and institutional capacity. The  --- ourth en- abling mechanism is recognition o ---  governance institutions as le- gally real entities with capacity to act. A governance institution that cannot contract, cannot sue, cannot hold property, and cannot be legally accountable  --- or its obligations cannot exercise the govern- ance  --- unctions that members and the public depend on.
 Chapter 4 covered the conditions o ---  institutional existence. What the present analysis adds is the distinction between legal per- mission and active legal support. States that have enacted enabling statutes  --- or previously unrecognized governance  --- orms, such as Wyoming's DAO LLC Supplement  --- or blockchain-based govern- ance organizations, the Revised Uni --- orm Unincorporated Non- pro --- it Association Act  --- or associations lacking corporate  --- orm, and the Model Nonpro --- it Corporation Act's  --- lexible governance struc- tures, have done more than permit those institutions to exist. Those statutes have given institutions the legal in --- rastructure to exercise governance authority with predictable e ---

ect. Chapter 13 examines how enabling law — or decentralized governance institutions inter- acts with the governance analysis this — ramework develops. From preservation to construction. The — our enabling mechanisms described above emerged without deliberate coordina- tion: the business judgment rule — rom — iduciary doctrine, FAA en-

orcement — rom — ederal arbitration policy, the rule o — reason — rom antitrust law, and common law de — erence — rom courts’ reluctance to adjudicate associational disputes. Each reduces the private cost o —

producing governance and thereby moves governance supply closer to the social optimum. The Pigouvian logic that justi — ies preserving these existing enabling rules extends — urther: preservation logic sup- ports construction o — new enabling rules where none currently ex- ist. 156 Law and Governance

 The prescriptive implication is this: i ---  governance institutions can generate positive externalities that their members cannot cap- ture, and i ---  reducing the private cost o ---  production corrects the re- sulting undersupply, then the same logic supports constructing new enabling rules in governance domains where positive externalities may be demonstrable but legal enabling mechanisms are absent. The argument is not merely that existing de --- erence doctrines should be maintained. It is that the absence o ---  enabling law in a governance domain where positive externalities may be demonstrable requires investigation o ---  whether externalities are actually present be --- ore concluding a policy  --- ailure is remediable.
 Commons governance institutions illustrate the gap. The Swiss alpine communities that Ostrom documented, the Philippine irriga- tion systems that Chapter 5 describes, and the  --- ishery co-manage- ment regimes that operate across dozens o ---  jurisdictions produce governance bene --- its that extend well beyond their memberships: environmental preservation, resource sustainability, and local coor- dination that neither markets nor governments supply e ---

iciently. Yet these institutions operate without any equivalent o — the business judgment rule to shield their en — orcement decisions — rom judicial second-guessing, without FAA-style en — orcement to give their dis- pute resolution legal e —

ect, without rule-o — -reason protection — or their cooperative restraints, and — requently without the legal recog- nition that would give them standing and capacity to en — orce their own rules. I — the Pigouvian — ramework is correct, commons govern- ance institutions are more undersupplied than corporate or ex- change governance institutions, because they lack every enabling mechanism that the latter enjoy. The governing principle is that leg- islatures and courts should develop enabling law — or commons gov- ernance with the same — unctional logic — cost reduction to correct Error! Use the Home tab to apply Heading 1 to the text that you want to appear here. 157

positive-externality undersupply — that produced the

our existing doctrines, even though commons governance does not map neatly onto the doctrinal categories — rom which those doctrines emerged. The enabling obligation extends beyond preservation o — existing en- abling rules to construction o — enabling rules in governance do- mains where none currently exist.

How law degrades governance Law degrades governance through seven recurring mechanisms that

all into three — unctional clusters organized by which governance el- ement each damages.

Disabling the sanction mechanism Mechanism 1: Converting property rules to liability rules on exclusion. When law converts a governance institution’s expul- sion authority — rom a categorical right to a compensated remedy, the credibility o — the exclusion threat collapses. The collapse occurs not because expulsion is — ormally prohibited, but because the expected cost o — exercising it rises enough to deter en — orcement. Silver v. New York Stock Exchange, 373 U.S. 341 (1963), is the — oun- dational case. The NYSE disconnected Harold Silver’s direct-wire connections without notice or hearing, destroying his brokerage business. The Supreme Court held this violated the Sherman Act and exposed the Exchange to treble damages under the Clayton Act. The holding converted the Exchange’s exclusion authority — rom an exercise o — institutional governance into an actionable antitrust vi- olation, subject to per se condemnation unless the institution could demonstrate procedural compliance. Paul Mahoney documented the consequences: the Exchange’s capacity to discipline members 158 Law and Governance

swi

tly and credibly was systematically constrained by the proce- dural apparatus the decision imposed.184 Nominal exclusion author- ity survived. The practical cost o — using it increased enough to alter en — orcement behavior throughout the institution. Chapter 9 exam- ines the governance consequences o — the Silver line in detail. Expulsion deters de — ection only when the expected cost to a po- tential de — ector o — being expelled exceeds the expected gain — rom de-

ection. When courts subject expulsion decisions to antitrust review with treble-damages exposure, the governance institution must cal- culate the probability o — success — ul legal challenge, the expected damages, and the litigation costs it bears even in de — ense o — a justi-

ied expulsion. When those expected costs are high enough, rational institutions moderate en — orcement rather than bear them. Members observe the pattern. En — orcement becomes unreliable. Unreliable threats lose deterrent — orce, because a member who calculates a meaning — ul probability o — success — ul legal challenge against an ex- pulsion decision will weigh that probability against the governance institution’s sanction, reducing the cooperative discipline that makes the institution — unction. Eric Posner’s analysis o — legal inter- vention in group governance captures the mechanism: even rules that merely duplicate a group’s own norms will undermine sel — -reg- ulation when they introduce legal uncertainty into governance de- cisions that previously had categorical e —

ect.185 Mechanism 2: Procedural requirements that make sanc- tions too costly to exercise. When law imposes notice, hearing, stated-reasons, and judicial-review requirements on governance in- stitution disciplinary decisions, the — ormal expulsion authority re- mains intact but the practical cost o — exercising it rises until the in- stitution e —

ectively stops en — orcing its rules against all but the most egregious violations. Error! Use the Home tab to apply Heading 1 to the text that you want to appear here. 159

 Silver provides the doctrinal expression again. The Court re- quired notice and an opportunity  --- or hearing be --- ore the Exchange could exercise exclusionary authority without antitrust liability. Jus- tice Stewart's dissent identi --- ied the governance cost: "This sel --- -ini- tiating process o ---  regulation can work e ---

ectively only i — the process itsel — is allowed to operate — ree — rom a constant threat o — antitrust penalties.” 373 U.S. at 371. Cali — ornia’s — air procedure doctrine, es- tablished through the Pinsker line o — cases, imposed substantive ra- tionality and procedural — airness requirements on pro — essional asso- ciations exercising quasi-monopolistic control over certi — ication.186 The LMRDA imposes detailed procedural requirements on union disciplinary proceedings. Each o — these requirements has an inde- pendent justi — ication. Each also raises the cost o — governance en-

orcement in ways that governance institutions must absorb rather than pass on. The administrative law literature on rulemaking ossi — ication identi — ies the same mechanism in public governance. Thomas McGarity documented that procedural requirements on agency rulemaking, including notice-and-comment obligations, hard look judicial review, executive oversight, and congressional analytical mandates, caused agencies to shi — t resources — rom substantive gov- ernance to de — ensive documentation and to pre — er in — ormal guid- ance over en — orceable rules.187 The structural parallel between ad- ministrative ossi — ication and private governance proceduralization is imper — ect: agencies — ace constitutional constraints, political ac- countability, and statutory mandates that private associations do not. But the core mechanism is the same. Procedural cost deters gov- ernance output whether the institution is a — ederal agency or a com- modity trade association. 160 Law and Governance

The governance-relevant question is not whether procedural requirements produce any bene --- it. Many do: they reduce the risk o ---

arbitrary en

orcement, provide excluded members a check on insti- tutional overreach, and create a record — or meaning — ul review when review is warranted. The question is whether the procedural cost is proportionate to the governance bene — it. Procedural requirements calibrated to the scale o — the governance decision and the degree o —

monopolistic institutional power discipline governance without de- stroying it. Procedural requirements imposed categorically, without regard to institutional context, produce in private governance the ossi — ication that McGarity documented in administrative govern- ance.

Corrupting the rule system Mechanism 3: Juridi — ication o — contextual norms. When law imports in — ormal, contextual norms into — ormal governance ad- judication, it destroys the predictability on which sustained cooper- ation depends. Bernstein’s NGFA study demonstrates the mechanism directly. NGFA arbitrators deliberately declined to consider trade usage, course o — dealing, and course o — per — ormance, the contextual indicia that UCC §§ 1-303 and 2-208 direct courts to apply, even when evi- dence o — those practices was available. The NGFA’s — ormal written rules deliberately diverged — rom in — ormal trading customs. Bern- stein explained why: governance institutions in repeated-transac- tion environments maintain two distinct normative registers. Rela- tionship-preserving norms govern ongoing trading relationships; they are in — ormal, — lexible, and contextual. End-game norms govern

ormal dispute resolution; they are written, categorical, and predict- Error! Use the Home tab to apply Heading 1 to the text that you want to appear here. 161

able. The registers serve di

erent — unctions and must remain sepa- rate. When courts apply the UCC’s incorporation strategy and im- port relationship-preserving norms into end-game adjudication, they collapse the distinction. In — ormal accommodations become le- gally risky because any accommodation might later be cited as evi- dence o — a binding trade usage. Governance institutions respond by re — using to develop contextual accommodations, reducing the — lexi- bility that sustains long-term cooperative relationships.188 Bernstein’s companion study — ound that geographically co-ex- tensive usages o — trade did not exist even within close-knit commer- cial communities, and that the empirical assumptions underlying the UCC’s incorporation strategy were “highly questionable.” 189 Robert Scott’s analysis o —

ormalism in relational contract extends the argument: contextual interpretation by courts “undermines a parallel goal o — predictable, transparent interpretation o — explicit contract terms” and erodes the reliability on which governance-sus- taining investment depends.190 Cooperation in repeated-transaction environments is sustained when members have reliable expectations about the consequences o — de — ection. Contextual interpretation by external adjudicators makes those expectations unreliable: a member cannot know whether its — ormal obligations mean what they say or what the re- lationship’s in — ormal history suggests. Reliable governance requires reliable rules, and contextual interpretation o —

ormal governance instruments by courts without the institutional knowledge to weigh the relevant context threatens both. Chapter 10 examines how this mechanism operates speci — ically in contract remedies disputes a — -

ecting governed trading networks. 162 Law and Governance

 Mechanism 4: Displacement o ---  private ordering by man- datory rules. When mandatory legal rules displace governance in- stitution rules that were previously privately ordered, institutions lose the adaptive capacity to tailor governance to their particular cir- cumstances.
 The mandatory versus enabling structure o ---  corporate law illus- trates the trade-o ---

. Roberta Romano documented that state compe- tition — or corporate charters produces enabling rules that give cor- porations — lexibility to design governance structures suited to their needs, while a national mandatory regime would eliminate this adaptive capacity.191 Frank Easterbrook and Daniel Fischel made the complementary argument: market — orces ensure e —

icient charter terms, and most corporate law should be enabling.192 The case — or mandatory rules, that asymmetric in — ormation and collective action

ailures prevent e —

icient charter terms — rom emerging, is serious, and some mandatory rules are justi — ied as protections against insider opportunism. The governance-relevant claim is narrower: manda- tory rules reduce governance adaptability, and the optimal domain

or mandatory rules covers the situations where private ordering systematically — ails rather than the governance decisions that insti- tutions have adequate incentives to make well. ERISA preemption demonstrates the degradation mechanism at its most extreme. Section 514(a) displaces all state law that “relates to” an employee bene — it plan, creating a vacuum in which state law is unen — orceable but — ederal law provides only minimal substantive requirements — or sel — - — unded plans. The governance institutions that developed under state law to manage bene — it plan administra- tion, mediate disputes between employers and employees, and en-

orce — iduciary standards have been largely displaced by a — ederal Error! Use the Home tab to apply Heading 1 to the text that you want to appear here. 163

ramework that establishes accountability obligations without providing the institutional mechanisms to en — orce them.193

Removing accountability Mechanism 5: Removing standing without providing sub- stitutes. When law removes private en — orcement rights without re- placing them with alternative accountability mechanisms, govern- ance — ailures go undetected and uncorrected even when — iduciary obligations — ormally remain. Chapter 4 documented this through three nonpro — it cases: the Bishop Estate in Hawaii, the Hershey Trust in Pennsylvania, and the Getty Trust in Cali — ornia. In each case, governance institutions with

ormal legal recognition, — iduciary obligations, and accountability structures su —

ered serious — ailures over extended periods because no private party had standing to challenge governance decisions and the state attorney general lacked the resources, political independ- ence, or institutional capacity to act in time. Evelyn Brody identi — ied the structural cause: “ — ew charities have members endowed with voting rights, and state attorneys general have limited resources to devote to monitoring the nonpro — it sector.”194 The absence o — pri- vate en — orcement rights is not an incidental — eature o — nonpro — it law. It is the structural condition that allows governance — ailures to persist. Chapter 12 examines the governance void in universities and nonpro — its in detail and considers institutional responses that might close the standing gap without compromising governance au- tonomy. Standing gaps arise in other governance contexts as well. The Private Securities Litigation Re — orm Act’s heightened pleading re- quirements raised the cost o — securities — raud litigation enough to 164 Law and Governance

reduce the private en

orcement that had supplemented SEC over- sight.195 Thole v. U.S. Bank N.A., 140 S. Ct. 1615 (2020), held that de-

ined-bene — it pension plan participants lacked Article III standing to challenge plan — iduciaries who had produced substantial investment losses, because the participants’ promised bene — its were technically una —

ected. 196 The Landrum-Gri —

in Act reserves union election challenges to the Secretary o — Labor, — oreclosing direct suits by un- ion members to en — orce governance obligations.197 In each setting, the removal o — private standing concentrates en — orcement authority in a single institutional actor that will predictably under-en — orce. Mechanism 6: Substituting compliance structures — or gov- ernance architecture. When law mandates — ormal compliance programs as evidence o — governance adequacy, organizations create structures that satis — y legal requirements without changing govern- ance behavior. What Kimberly Krawiec has called cosmetic compli- ance satis — ies the legal standard while leaving governance quality unchanged.198 Krawiec documented the pattern across environmental, tort, employment, corporate, securities, and health care contexts: com- pliance programs are adopted primarily as legal shields rather than behavioral change mechanisms, and courts treat the existence o —

those programs as evidence o

compliance regardless o — whether the programs alter outcomes. Lauren Edelman identi — ied the underlying dynamic: organizations respond to legal mandates by creating — or- mal structures that signal compliance, and those structures then shape the legal standards against which compliance is measured, in a cycle that progressively decouples legal accountability — rom gov- ernance quality.199 The corporate governance context illustrates this mechanism with particular — orce. Caremark’s oversight duty nominally requires Error! Use the Home tab to apply Heading 1 to the text that you want to appear here. 165

directors to implement in

ormation and reporting systems. In prac- tice, the near-insurmountable pleading standard meant that almost any compliance program satis — ied the duty, giving boards strong in- centives to create programs that demonstrate compliance rather than programs that detect problems. Roberta Romano’s analysis o —

Sarbanes-Oxley

ound that SOX governance provisions were “ill- conceived,” that independent boards did not reliably improve gov- ernance quality, and that some — irms de-listed or went private to avoid compliance costs that exceeded the governance value the stat- ute was designed to produce.200 The result is predictable under the

ramework: legal mandates that measure governance by its — ormal indicators rather than its substantive — unction produce — ormal indi- cators at the expense o — substantive — unction. Mechanism 7: Mandating open membership that disables screening. When law requires governance institutions to accept members they would otherwise exclude, the institution loses the monitoring and certi — ication — unctions that depend on selective membership. The exclusion mechanism Chapter 5 described serves two pur- poses. The — irst is en — orcement: expulsion punishes de — ectors and deters others. The second, equally important and less o — ten analyzed in legal scholarship, is screening: admission criteria — ilter out poten- tial members whose inclusion would damage governance quality by introducing those with higher de — ection rates, weaker compliance records, or conduct records inconsistent with the institution’s gov- ernance norms. When law requires institutions to admit members regardless o — these criteria, screening — ails and governance quality degrades. Roberts v. United States Jaycees, 468 U.S. 609 (1984), and Board o —

Directors o

Rotary International v. Rotary Club o — Duarte, 481 U.S. 537 166 Law and Governance

(1987), required civic associations to admit women under state pub- lic accommodations laws, balancing associational — reedom against the state’s compelling interest in eradicating discrimination. Neither decision examined whether the contested membership criteria served governance — unctions. 201 Boy Scouts o — America v. Dale, 530 U.S. 640 (2000), held that — orced inclusion o — a member whose pres- ence signi — icantly a —

ected the organization’s ability to express its values violated the First Amendment’s protection o — expressive as- sociation. The Dale — ramework distinguishes organizations de — ined by expressive purposes — rom those that are not, but it does not sup- ply an account o — when membership criteria serve governance — unc- tions that are independent o — expressive purpose.202 No court has explicitly distinguished between membership cri- teria that are constitutive o — a governance institution’s — unction, such as the medical board’s credential requirement, the DDC’s rep- utation screening, or the pro — essional association’s character and — it- ness review, and membership criteria that re — lect social pre — erences without governance signi — icance. The — ormer criteria serve the monitoring and certi — ication — unctions Chapter 6 identi — ied. When law overrides those criteria, governance quality degrades. The latter criteria may warrant override under anti-discrimination or public accommodations analysis without governance consequence. Chap- ter 14 addresses the anti-discrimination challenge directly, including the limits the governance — ramework must acknowledge when as- sociational exclusion crosses — rom governance discipline into dis- crimination against protected classes. Error! Use the Home tab to apply Heading 1 to the text that you want to appear here. 167

When mechanisms interact The seven mechanisms described above are presented individually because each operates through a distinct causal pathway. In practice, legal rules — requently activate more than one mechanism simultane- ously, and the interaction between mechanisms produces govern- ance costs that exceed what any single mechanism would impose alone. Two — orms o — interaction can be distinguished. The — irst is ad- ditive: two mechanisms each impose independent costs on di —

erent governance — unctions, and the total governance degradation is the sum o — the two independent e —

ects. Silver v. New York Stock Exchange illustrates this pattern.203 Mechanism 1 operated by converting the NYSE’s expulsion authority — rom a categorical property-rule enti- tlement to a liability-rule remedy subject to antitrust scrutiny and treble-damages exposure, degrading the sanction — unction’s credi- bility. Mechanism 2 operated independently by imposing proce- dural requirements that made disciplinary action costlier to initiate, degrading the sanction — unction’s — requency. Each mechanism would have imposed governance costs on its own. Operating to- gether, they compounded: not only was the sanction less credible (because subject to judicial override), it was also less — requent (be- cause the procedural cost o — initiating it exceeded the governance bene — it in marginal cases). Mahoney’s empirical observation that NYSE en — orcement activity declined a — ter Silver re — lects the com- bined e —

ect o — both mechanisms, not either one in isolation.204 The second — orm o — interaction is cascading: one mechanism’s activation creates the conditions under which a second mechanism becomes operative or its e —

ects are ampli — ied. Marchand v. Barnhill illustrates this pattern in the corporate governance context.205 The 168 Law and Governance

Delaware Supreme Court identi

ied what appeared to be a monitor- ing — ailure — the Blue Bell Creameries board had implemented no reporting system — or — ood sa — ety compliance. But the monitoring

ailure was not isolated. Because no monitoring system existed, the sanctioning — unction had nothing to act on: no in — ormation about compliance — ailures reached the level at which sanctions could be imposed. Because no sanctions were imposed, the adjustment — unc- tion received no — eedback about the adequacy o — existing rules. And because no adjustment occurred, the decision-making — unction op- erated without the in — ormation that would have prompted correc- tive action. The cascading interaction produced a comprehensive governance collapse — rom what could have begun as a single-ele- ment — ailure: monitoring — ailure cascaded into sanctioning — ailure, which cascaded into adjustment — ailure, which le — t decision-making unin — ormed. A court evaluating the governance cost o — a legal rule must there — ore ask not only which mechanism the rule activates but whether that activation will compound with mechanisms already operating or trigger a cascade through dependent governance — unc- tions. The analytical implication is that legal rules acting on govern- ance institutions already under stress — rom one degradation mecha- nism will impose higher marginal governance costs than the same rule applied to a governance institution operating without prior degradation. Courts and legislatures assessing the governance con- sequences o — a proposed intervention should identi — y the — ull mech- anism pro — ile o — the institution, including mechanisms already acti- vated by prior legal rules, be — ore concluding that the marginal cost o — an additional intervention is acceptable. Error! Use the Home tab to apply Heading 1 to the text that you want to appear here. 169

Discipline versus destruction The seven mechanisms described above are mechanisms o — govern- ance degradation, not arguments against legal intervention in gov- ernance institutions. Law must sometimes intervene to correct spe- ci — ic — ailures: abuse o — members, discriminatory exclusion, anticom- petitive conduct, capture by insiders, systematic neglect o — institu- tional purposes. The governance — ramework demands that such in- terventions be made with awareness o — their governance conse- quences. The analytical challenge is distinguishing intervention that dis- ciplines governance — rom intervention that destroys governance ca- pacity. The distinction is not absolute. The same legal rule can dis- cipline governance in one institutional context and destroy govern- ance capacity in another. The governance — ramework provides — our criteria — or drawing the distinction. Market power and exit availability. The degree o — scrutiny that governance institutions warrant is proportional to the degree o — market power those institutions hold over members who cannot exit. This principle runs through the relevant case law. Northwest Wholesale Stationers calibrated antitrust scrutiny to market power and access to essential resources. Pinsker imposed substantive and procedural requirements on associations with quasi-monopolistic control over pro — essional certi — ication. Common law de — erence ap- plies most strongly to voluntary organizations whose members can readily join alternative institutions. Albert Hirschman’s analysis o — exit, voice, and loyalty provides the theoretical — oundation.206 When exit is readily available, mem- bers dissatis — ied with governance can leave and join alternative in- stitutions. The threat o — exit disciplines governance quality through market competition, and legal intervention is less necessary. When 170 Law and Governance

exit is unavailable or prohibitively costly, when the institution con- trols access to a pro — ession, a market, or a resource — or which no adequate substitute exists, members cannot leave and the market discipline that would otherwise check governance — ailure does not operate. Legal intervention becomes more justi — ied. The calibration this criterion produces is straight — orward: insti- tutions with market power warrant more scrutiny, and institutions without market power warrant less. What the governance — rame- work adds is the recognition that heightened scrutiny still carries governance costs. The question is whether the speci — ic intervention proposed is calibrated to the governance — ailure that market power enables rather than imposed as a categorical override o — governance authority. Mandatory access antitrust. The essential — acilities doctrine and mandatory dealing requirements under antitrust law represent the most aggressive — orm o — the property-rule to liability-rule con- version identi — ied in Mechanism 1. Where standard liability-rule conversion makes expulsion costly to exercise, mandatory access re- quirements a —

irmatively compel the governance institution to ad- mit or deal with parties the institution has chosen to exclude. Aspen Skiing Co. v. Aspen Highlands Skiing Corp., 472 U.S. 585 (1985), permitted a re — usal-to-deal claim where a dominant — irm dis- continued a pro — itable course o — dealing with a competitor — or rea- sons that made no economic sense absent a predatory purpose.207 Applied to governance institutions, this doctrine raises the ques- tion: when does a governance club’s exclusion constitute exclusion- ary monopolization rather than a legitimate governance member- ship decision? The essential — acilities doctrine, at its broadest, would require governance institutions controlling essential in — rastructure Error! Use the Home tab to apply Heading 1 to the text that you want to appear here. 171

to provide access on reasonable terms even to parties they have ex- cluded. Verizon Communications Inc. v. Law O —

ices o — Curtis V. Trinko, 540 U.S. 398 (2004), signi — icantly curtailed mandatory access antitrust by holding that there is no general duty to deal with rivals, that the essential — acilities doctrine remains legally uncertain, and that anti- trust law should not be used to manage ongoing regulatory relation- ships. 208 Trinko implicitly recognized the governance costs that mandatory access imposes: requiring governance institutions to serve parties they have excluded on regulated terms removes the cat- egorical character o — the exclusion right and introduces the same credibility problems identi — ied in Mechanism 1. The decision le — t the governance institution’s exclusion authority more intact than the broader essential — acilities doctrine would have, re — lecting a judicial choice, whether or not consciously — ramed in governance terms, to preserve the property-rule character o — exclusion where a regulatory regime already exists to address access concerns. Where an institution’s exclusion o — a party re — lects genuine an- ticompetitive exclusion rather than legitimate governance disci- pline, antitrust intervention is warranted. Where exclusion re — lects a governance judgment about member quality, conduct history, or compatibility with institutional norms, mandatory access antitrust removes the governance institution’s most e —

ective sanction with- out addressing the problem the sanction was designed to solve. The market power criterion distinguishes the cases: when the govern- ance institution lacks market power and the excluded party has ad- equate alternatives, the exclusion does not threaten competition and antitrust intervention is not justi — ied. When the institution controls essential in — rastructure and exclusion — orecloses e —

ective competi- tion, antitrust scrutiny is appropriate. The governance — ramework 172 Law and Governance

does not oppose this scrutiny; the

ramework requires that manda- tory access remedies, i — imposed, be structured to preserve as much o — the categorical character o — exclusion as the competitive harm re- quires removing. Chapter 14 addresses the equity dimension o —

mandatory access, including access claims grounded in anti-discrim- ination law rather than competition law. Last-period de — ection and structural con — licts. Edward Rock and Michael Wachter identi — ied the conditions under which norm-based governance breaks down: speci — ically, situations in which the repeat-play incentive structures that make sel — -govern- ance e —

ective are no longer operative.209 When a — irm — aces a change o — control, when shareholders — ace a — reeze-out merger, or when an institutional insider has an opportunity to capture governance — or personal bene — it at the expense o — members or the public, the gov- ernance institution — aces a period in which the — uture consequences o — de — ection are small relative to their immediate gains. Sel — -govern- ance through norms — ails because the shadow o — the — uture that nor- mally disciplines behavior has shortened or disappeared. Legal intervention is most justi — ied in these last-period situa- tions and least justi — ied when governance institutions operate in on- going relationships with strong repeat-play structures. Entire — air- ness review in corporate governance, applied to sel — -dealing trans- actions by controlling shareholders, targets exactly the structural con — licts that produce last-period de — ection. The business judgment rule’s withdrawal in those circumstances is discipline, not destruc- tion: it responds to the speci — ic structural — ailure that governance sel — -regulation cannot address without external accountability. Proportionality and comparative institutional compe- tence. Neil Komesar’s comparative institutional analysis provides Error! Use the Home tab to apply Heading 1 to the text that you want to appear here. 173

the organizing principle: the question is not whether the govern- ance institution is per — ect, but whether the intervening institution, whether court, legislature, or regulator, is better positioned to pro- duce the governance outcome sought. 210 Courts lack contextual knowledge, — ace strategic litigation by sophisticated parties, and produce precedents that apply beyond the speci — ic dispute. Legisla- tures are subject to capture by organized interests, produce categor- ical rules that cannot anticipate governance context, and impose compliance costs on all governance institutions in a domain to ad- dress abuses by a — ew. Legal intervention that exceeds the institu- tional competence o — the intervening institution substitutes one

orm o — governance — ailure — or another. Proportionality requires that legal intervention respond to a speci — ic, identi — ied governance — ailure in a manner calibrated to cor- rect that — ailure rather than imposing categorical costs on govern- ance production across the board. When Allied Tube & Conduit Corp. v. Indian Head, Inc., 486 U.S. 492 (1988), imposed antitrust liability on a standard-setting body whose consensus process had been sub- verted by economically interested parties who packed the voting membership, the holding was proportionate: it identi — ied the spe- ci — ic mechanism o — capture, imposed liability on the capturing par- ties, and le — t intact the standard-setting institution’s authority over untainted decisions.211 When Silver imposed antitrust liability on the NYSE — or exclusionary conduct without di —

erentiating exclusion

or governance purposes — rom exclusion — or anticompetitive pur- poses, the holding was disproportionate: a governance cost — ell on all NYSE disciplinary decisions to address a speci — ic — ailure involving a single non-member excluded without notice. 174 Law and Governance

Discipline is targeted. Destruction is categorical. Law that disci- plines governance improves governance quality in the speci --- ic insti- tutional contexts where governance has  --- ailed. Law that degrades governance imposes costs on the governance  --- unction itsel --- , with- out regard to whether any speci --- ic governance  --- ailure has occurred or whether the intervening institution is better positioned to ad- dress it than the governance institution would be.
Calibration, the standard that separates discipline  --- rom destruc- tion, has two independent dimensions that legal analysis must eval- uate separately. Scope calibration asks whether the remedy targets the governance  --- unction that  --- ailed and no other  --- unction. Intensity calibration asks whether the governance cost o ---  the remedy is pro- portionate to the governance bene --- it o ---  correcting the  --- ailure. Scope and intensity calibration are independent conditions: a remedy can satis --- y one while violating the other. Silver  --- ailed on both dimen- sions simultaneously. On scope, the Court imposed antitrust liabil- ity and procedural requirements across all NYSE disciplinary deci- sions, member and non-member alike, when the identi --- ied  --- ailure concerned a single non-member denied access without notice. On intensity, the remedy exposed the Exchange to treble damages under the Clayton Act  --- or exercising a governance  --- unction, sanctioning de --- ectors, that had no causal relationship to the competitive harm the antitrust laws were designed to prevent. Allied Tube, by contrast, satis --- ied both: scope was limited to the speci --- ic standard-setting de- cision that had been corrupted by interest-group packing, and in- tensity was proportionate because liability  --- ell on the parties that had subverted the governance process rather than on the standard- setting institution's legitimate governance  --- unctions. Evaluating calibration along both dimensions prevents the analytical error o ---

treating a remedy as disciplining merely because it is targeted (it may Error! Use the Home tab to apply Heading 1 to the text that you want to appear here. 175

be targeted but disproportionately costly) or merely because it is proportionate in cost (it may impose a proportionate cost on the wrong governance — unction). The discipline-versus-destruction distinction assumes that gov- ernance — ailure is localized: one — unction — ails, and the remedy tar- gets that — unction. But governance — ailure can be systemic. When all

our governance elements have — ailed — when decision-making is captured, monitoring is absent, sanctions are unen — orced, and ad- justment has ceased — a remedy targeting only one element is in- su —

icient, and a remedy addressing all — our appears categorical. The distinction between discipline and destruction breaks down at this boundary unless the — ramework speci — ies a systemic — ailure corol- lary: when the identi — ied governance — ailure is pervasive across all governance — unctions, a comprehensive remedy that addresses all

our elements simultaneously is not categorical in the — ramework’s sense. “Categorical” means imposed without regard to whether — ail- ure is present. A remedy that responds to documented — ailure across every governance dimension is targeted — it is targeted at an insti- tution whose governance has — ailed comprehensively. Marchand v. Barnhill, 212 A.3d 805 (Del. 2019), illustrates the corollary. The Del- aware Supreme Court — ound that Blue Bell Creameries’ board had

ailed to implement any reporting system to monitor — ood sa — ety compliance — the company’s “most central compliance issue.” The

ailure appeared to be a monitoring problem, but in substance the court identi — ied a collapse o — the board’s entire governance architec- ture: no in — ormation systems, no board-level discussion, no agenda items addressing the risk, no management reporting protocols. The remedy — allowing the Caremark claim to proceed past the motion to dismiss stage — addressed what the court recognized as a perva- sive governance — ailure, not a single isolated — unctional de — iciency. 176 Law and Governance

The systemic

ailure corollary preserves the discipline-versus-de- struction distinction by speci — ying that the scope o — a proportionate remedy must match the scope o — the documented — ailure. A remedy that addresses — our governance elements because — our have — ailed satis — ies scope calibration. A remedy that addresses — our governance elements because one has — ailed does not.

What Chapter 8 will do Three categories o — legal rules, organized by their governance e — -

ects, now have precise speci — ications: enabling rules that subsidize governance production, seven degrading mechanisms organized by

unctional cluster, and disciplining rules calibrated to correct spe- ci — ic — ailures without destroying governance capacity more broadly. Chapter 8 converts that taxonomy into a working method. The method asks speci — ic questions o — any legal rule that acts on a gov- ernance institution: what governance — unction does the rule support or damage, through which o — the identi — ied mechanisms does it op- erate, and what cost does it impose on governance capacity relative to the speci — ic governance problem it addresses? The method does not generate automatic answers. It directs legal analysis toward the right questions. Chapter 8: A Method — or Evaluating Law’s Governance E —

ects Rights protection, e —

iciency, distributive justice, legitimacy, anti- discrimination en — orcement, and systemic stability all make legiti- mate claims on legal rules, and those claims — requently con — lict. A rule maximizing rights protection may sacri — ice e —

iciency. A rule disciplining governance abuse may damage governance capacity. A rule en — orcing anti-discrimination may override the membership criteria that governance institutions use to sustain their monitoring and en — orcement — unctions. Weighting these competing objectives against one another, deciding what law is ultimately — or and which values take priority when they collide, is a question no single schol- arly project can answer. Cass Sunstein put the point precisely: legal actors persistently disagree about — oundational principles even while converging on particular outcomes, and that divergence at the level o —

oundations re — lects genuine incommensurability among le- gal values rather than mere — ailure o — analysis.212 This book supplies something more tractable than — oundational answers: a governance optimization — unction, precisely speci — ied, that legal actors can introduce into whatever — ramework they use to weigh competing legal values. Be — ore this — ramework, governance 178 Law and Governance

was a word lawyers used without a stable de

inition, a concern courts expressed without a method — or measuring, and an interest that legal analysis could not reliably trace — rom legal rule to institu- tional e —

ect. Seven chapters have supplied the de — inition, the meas- urement — ramework, and the tracing mechanism. Putting those tools to work is the task o — this chapter. Error! Use the Home tab to apply Heading 1 to the text that you want to appear here. 179

Chapter 8. Governance Evaluation Method The method developed here proceeds in two stages. The — irst makes governance legible as a legal variable — visible and speci — iable in legal argument the way in — ormation asymmetry became legible a — ter Akerlo — ‘s 1970 paper. The second applies a structured sequence o —

analytical steps to any legal rule or decision that a

ects a governance institution. Together, the two stages allow lawyers and courts to trace the path — rom a legal rule to its governance e —

ects and to in- clude those e —

ects in whatever — ramework they use to weigh com- peting legal values.

Making governance legible as a legal variable Akerlo — ‘s 1970 paper on markets — or lemons o —

ers an instructive comparison.213 His contribution was naming and precisely speci — y- ing a phenomenon, in — ormation asymmetry between buyers and sellers, that had always been present in market transactions but had never been made analytically legible, rather than resolving the nor- mative question o — what contract law is — or. Once named and speci-

ied, in — ormation asymmetry could be introduced into doctrinal analysis across contracts, securities, torts, and corporate law. Courts and scholars did not need a uni — ied theory o — contract to ask whether any particular rule exacerbated or corrected an in — ormation asym- metry. Legibility made the variable available — or analysis that was already underway. From that availability — ollowed changes in how lawyers understood what remedies were coherent, what de — ault rules were sensible, and what existing doctrinal categories had been missing. Governance occupies an analogous position. Legal analysis has always recognized that some rules help cooperation and some rules 180 Law and Governance

undermine it. What legal analysis has lacked is a precise account o

governance as an institutional phenomenon: a de

inition stable enough to carry doctrinal argument, a goods-typology classi — ication explaining its production and spillover dynamics, and a taxonomy o — the mechanisms through which law acts on it. Without those an- alytical tools, governance entered legal reasoning as a di —

use intui- tion rather than a speci — ied variable. Di —

use intuitions cannot be weighed rigorously against rights or e —

iciency because the weighing requires both sides o — the scale to be legible. That is why incommensurability makes the contribution o — this book more important, not less. When legal values are incommen- surable, no — ormula correctly weights them. But incommensurabil- ity does not make precise speci — ication o — individual values point- less; it makes precision necessary. Legal actors weighing governance against rights, or governance against e —

iciency, cannot make that weighing more rigorous than their description o — governance al- lows. A di —

use governance intuition will be weighted di —

usely, or discarded entirely, because it cannot be placed on a scale with values that have been re — ined over generations o — doctrinal analysis. A pre- cisely speci — ied governance variable, one with identi — iable structure, identi — iable outputs, and identi — iable legal preconditions, can enter the weighing process with the same analytical standing that rights and e —

iciency analysis have earned by decades o — speci — ication. Three normative premises underlie this — ramework, and stating them explicitly is important because the method’s — orce depends on accepting them. First, governance institutions that satis — y the — our- element — ramework o — Chapter 1 produce value that legal analysis should recognize as a distinct legal interest — not merely as a by- product o — contract, property, or association, but as a structured in- Error! Use the Home tab to apply Heading 1 to the text that you want to appear here. 181

stitutional capacity whose existence is itsel

legally signi — icant. Sec- ond, legal rules that predictably destroy governance capacity impose real costs, costs measurable in the degradation o — cooperation, mon- itoring, sanctioning, and adjustment, even when those rules ad- vance other legitimate objectives such as rights protection, market competition, or anti-discrimination en — orcement. Third, the costs o — governance degradation are cognizable legal harms: they warrant consideration in legal design and adjudication, not as dispositive

actors that trump all other values, but as costs that legal actors must identi — y and weigh rather than ignore. These three premises are modest in scope. The method claims that governance is a real insti- tutional phenomenon, that law a —

ects it, and that the e —

ects o — law on governance should be visible in legal analysis rather than invisi- ble. The method thus makes governance legible without claiming that governance always outweighs rights, that all exclusion is justi-

ied by governance — unction, or that sel — -governing institutions are entitled to legal de — erence regardless o — their conduct.214 Whether to weight governance heavily against other legal val- ues in any particular case, or to treat governance degradation as an acceptable cost o — achieving another legal objective, remains a judg- ment — or legal actors in context. Chapter 14 addresses the limits most likely to justi — y accepting a governance cost, including the anti-dis- crimination challenge, the antitrust inter — ace, and the rights dimen- sion, and examines what the governance — ramework concedes and what it contests in those settings. Here, the task is explaining how the method works. 182 Law and Governance

What the governance evaluation produces This — ramework proposes seven analytical steps through which to evaluate governance e —

ects. Each step poses a speci — ic question, in a speci — ic sequence, and produces an answer with identi — iable content. Taken together, the seven steps yield a governance evaluation: a structured account o — what a legal rule does to governance, through what mechanism, and at what cost to governance capacity. A governance evaluation is one input among several to a legal verdict, not the verdict itsel — . Reaching a legal verdict requires tak- ing the governance evaluation alongside analysis o — the other legal values at stake in the speci — ic decision: rights, e —

iciency, legitimacy, and anti-abuse. Governance analysis supplies one side o — that weigh- ing with a precision previously unavailable. Lawyers and courts in- tegrate this input with the rest o — their analysis. Proper use o — this method requires understanding its scope. When properly used, the method makes governance costs and ben- e — its visible and precisely characterized so that the legal actor decid- ing the question has accurate in — ormation about what the rule does to governance. The method supplies one side o — the weighing legal actors must do; it does not settle the ultimate legal judgment. Even when a legal rule clearly degrades governance capacity, that degra- dation may be justi — ied as the price o — vindicating a right, en — orcing anti-discrimination, correcting anticompetitive conduct, or ad- dressing a governance — ailure more serious than the degradation the rule causes. What the method guarantees is that the governance cost enters the analysis visibly rather than being overlooked. Answers at each step require judgment, not calculation. Legal judgment o — the kind the method requires is in — ormed by text, doc- trine, precedent, and institutional context, the standard materials o —

legal reasoning. What the method does is organize that judgment in Error! Use the Home tab to apply Heading 1 to the text that you want to appear here. 183

a governance-speci

ic sequence, structuring the analysis so that gov- ernance receives sustained attention rather than occasional ac- knowledgment. Governance disputes are polycentric in Lon Fuller’s sense, and that polycentricity explains why structuring the analysis matters.215 Fuller identi — ied polycentric problems as those involving many a — -

ected parties in a — luid state o — a —

airs where a change in one element ripples through all the others. Adjudication handles polycentric problems poorly because party-centered proceedings allow only the parties be — ore the court to present proo — s and arguments, leaving the interests o — all other a —

ected parties without — ormal representation. Governance disputes are polycentric in exactly this way: a court ad- judicating a dispute between a governance institution and an ex- pelled member is resolving a dispute limited to those parties, but the governance question raised by that dispute, whether the rule gov- erning the expulsion enables or degrades governance capacity, a — -

ects every member o — the institution, every bene — iciary o — govern- ance spillovers, and every party who might be excluded in the — uture. None o — these parties has a voice in the party-centered proceeding. Governance analysis compensates — or this structural limitation by expanding what lawyers and legislators consider be — ore a rule is- sues or a statute is enacted. Courts adjudicate disputes between the parties be — ore them and should continue doing so. Lawyers and leg- islators who apply the governance method be — ore rules are made are doing analysis that the court’s own process cannot per — orm — or them. Fuller’s observation about the limits o — adjudication is, in this sense, an argument — or applying the governance method at the leg- islative and regulatory design stages rather than an argument against applying it at all. 184 Law and Governance

 Legal rather than economic analysis is what the method de- mands. Pricing governance outputs, aggregating utilities, and com- puting wel --- are  --- unctions are not what the method requires. Identi-

ying mechanism and direction is. Each step poses the kind o — ques- tion legal analysis already asks: what does this rule do, to what kind o — institution, through what mechanism, with what consequence — or the institution’s capacity to per — orm its — unction? Whether expul- sion is protected as a property right or converted to a compensated remedy, whether — iduciary obligations attach to governance author- ity, whether private en — orcement rights exist to challenge govern- ance — ailures, whether law makes governance institution rules en-

orceable in court or merely aspirational: these are questions lawyers answer through the standard tools o — legal reasoning. The method organizes those questions in a governance-speci — ic sequence.

The seven steps The — ramework proposes seven analytical steps, each building on the prior steps and setting up what — ollows. Selective application

ails: the sequence matters because later steps depend on what earlier steps establish. Step 1 names the object. Steps 2 and 3 describe the institution and its legal environment. Steps 4 and 5 characterize what the institution produces. Step 6 identi — ies how law acts on it. Step 7 evaluates.

Step 1: Identi

y the shared problem Every governance institution exists because a group — aces a shared problem that uncoordinated individual action does not ade- quately address. Naming that problem with precision is the — irst step. Precision means more than a label: it means identi — ying what Error! Use the Home tab to apply Heading 1 to the text that you want to appear here. 185

ailure the institution was organized to address, what evidence shows it has addressed that — ailure, and what governance — unctions are essential to addressing it as opposed to incidental to the institu- tion’s operation. Naming the shared problem constrains all subsequent analysis. A governance institution organized to address — raud in a trading network depends critically on its monitoring and sanction mecha- nisms, because those mechanisms produce the cooperation equilib- rium that prevents — raud. Damage those mechanisms, and govern- ance capacity is diminished at its — unctional core. A governance in- stitution organized to coordinate technical standards depends criti- cally on its decision-making procedures, because those procedures resolve coordination games and select the — ocal points around which members converge. Damage those procedures, and the coordination

unction — ails regardless o — whether the sanction mechanism is in- tact. Legal rules that act on a governance institution produce di —

er- ent governance e —

ects depending on which — unction is essential to the shared problem the institution addresses, and that determina- tion requires naming the shared problem — irst. Framing discipline is part o — Step 1’s work. Two — ramings o — the NYSE setting are available: one treating the shared problem as re- straint o — trade among competing market participants, another treating it as the collective action problem o — maintaining market integrity among interconnected broker-dealers who share in — ra- structure. From the — irst — raming, analysis points toward competi- tive e —

ects and market structure. From the second, analysis points toward monitoring, sanctioning, and discipline — unctions. Only the second — raming identi — ies the governance institution as the object o —

analysis. Choosing between

ramings requires arguing — rom the in- stitution’s actual operation rather than — rom — ormal descriptions or 186 Law and Governance

rom whichever characterization best serves a pre — erred legal out- come.

Step 2: Identi

y the governance institution Chapter 1 de — ined governance as the organized system by which a group manages a shared problem over time, requiring — our ele- ments: decision-making, monitoring, sanctions, and adjustment. Step 2 determines whether a governance institution exists in the rel- evant setting, describes its — our elements with enough speci — icity to analyze legal e —

ects on them, and maps its membership boundaries. Four elements, not two, are required, and the requirement is

unctional rather than — ormal. An organization with decision-mak- ing and monitoring but no e —

ective sanctions and no capacity to adjust its rules is a governance shell, not a governance institution. Governance capacity depends on all — our elements being operative: decision-making authority that members actually — ollow, monitor- ing that produces in — ormation that actually governs conduct, sanc- tions that are actually applied to actual violations, and adjustment procedures that actually revise rules in response to actual changed circumstances. Claims o — governance protection based on institu- tional — orm, meaning — ormal documents that establish all — our ele- ments without any — unctioning mani — estation o — any o — them, do not satis — y Step 2. An institution invoking governance protection must demonstrate that its — our elements are operating, not merely that they appear in its bylaws. Because governance quality is private in — ormation held by insti- tutional leadership, the — unctionality requirement must be tied to observable outputs rather than internal processes, and the eviden- tiary burden must — all on the institution claiming governance pro- Error! Use the Home tab to apply Heading 1 to the text that you want to appear here. 187

tection. Decision-making is

unctioning when institutional deci- sions govern actual member behavior, meaning members can iden- ti — y the institution’s rules and demonstrate that those rules constrain their conduct in ways they would not otherwise choose. Monitoring is — unctioning when it generates conduct-governing in — ormation, meaning in — ormation that has been used, within an identi — iable pe- riod, to initiate at least one sanction proceeding, modi — y at least one institutional rule, or alter at least one member’s behavior through the credible threat o — detection. Sanctioning is — unctioning when sanctions are applied to actual violations at a — requency consistent with the institution’s observable violation rate, not merely when sanction authority exists on paper. Adjustment is — unctioning when rules have changed within an identi — iable period in response to doc- umented changed circumstances, not merely when amendment pro- cedures exist in the bylaws. Each criterion asks — or an output, not a structure: documented arbitration proceedings, not an arbitration clause; recorded rule revisions with identi — ied triggering events, not an amendment provision; sanction decisions with stated bases, not a disciplinary committee roster. An institution that satis — ies all — our output criteria has earned the governance protection the — ramework provides. An institution that can demonstrate structures but not outputs has earned only the presumption that — urther investigation is warranted be — ore protection is granted. In — ormal governance quali — ies. Maine lobster communities gov- erned their — isheries be — ore any statute recognized them. Diamond trading networks operated governance systems through reputa- tional tracking, graduated sanctions, and community arbitration without — ormal legal architecture.216 Whether an in — ormal arrange- ment satis — ies the — our-element test is a — actual question the de — ini- tion makes answerable. Neither the presence o —

ormal structure nor 188 Law and Governance

the absence o

legal recognition determines the answer. What de- termines the answer is whether decision-making, monitoring, sanc- tions, and adjustment are — unctioning. Membership boundaries are part o — the institution’s description, not a separate inquiry. Chapter 5 established that governance insti- tutions have club good structure: membership boundaries de — ine who participates in governance, who bears governance costs, who receives member bene — its, and who is subject to the sanction mech- anism. Boundaries that are unclear or contested signal a structural problem that any legal analysis o — the institution must acknowledge, because legal rules that act on governance produce di —

erent e —

ects depending on whether membership is well-de — ined or porous.

Step 3: Describe the legal conditions under which the governance institution exists Chapter 4 identi — ied six legal conditions governance institutions require: permission and recognition, membership boundaries, ex- ternal en — orceability, — iduciary structure, standing, and accountabil- ity. Step 3 applies that — ramework to the speci — ic institution under analysis, asking not whether the conditions are present in name but whether they are present in e —

ect. Formal recognition without practical en — orceability leaves governance vulnerable at the en — orce- ability condition. Nominal — iduciary obligations without standing to en — orce them leave governance vulnerable at the accountability con- dition. Two kinds o — analytical work make this step essential. First, a diagnostic — unction: describing which conditions are strong, which are weak, and which are absent tells the analyst where governance is structurally vulnerable and there — ore where legal intervention will Error! Use the Home tab to apply Heading 1 to the text that you want to appear here. 189

do the most governance damage. An institution with strong recog- nition and en — orceability but weak standing is highly vulnerable to Mechanism 5, the removal o — standing without providing substi- tutes, while being more resilient to Mechanism 1, which converts expulsion — rom a property rule to a liability rule. Understanding the condition pro — ile allows prediction o — which legal rules pose the greatest governance risk to the speci — ic institution. Second, an etiological — unction: distinguishing between govern- ance — ailure caused by legal degradation and governance — ailure caused by inadequate legal conditions — rom the outset. Both produce

ailing governance, but they call — or di —

erent legal responses. Gov- ernance — ailing because law has degraded a previously — unctional condition calls — or legal restoration o — that condition. Governance

ailing because it was never adequately constituted calls — or legal construction o — the missing condition. Con — lating the two problems produces misdirected legal responses that address the wrong cause. Step 3 also establishes the baseline — or Step 6. When a rule mod- i — ies the en — orceability o — private governance decisions, Step 3’s de- scription o — the pre-rule en — orceability condition is what Step 6’s le- gal e —

ects analysis measures change against. Without that baseline, Step 6 cannot determine whether the rule degrades governance or merely adjusts conditions that were never adequate.

Step 4: Identi

y the member bene — its the governance institution produces Chapter 6 identi — ied — our categories o — member bene — it: transac- tion cost reduction, coordination, monitoring and adverse selection reduction, and dispute resolution. Step 4 identi — ies which o — these the governance institution delivers, through what mechanism, and with what evidence. 190 Law and Governance

 Where empirical evidence exists, the analysis should use it. Bernstein's study o ---  the Diamond Dealers Club quanti --- ied transac- tion cost reduction directly: two thousand members conducted bil- lions o ---  dollars o ---  annual transactions on handshake terms, with dis- putes resolved by DDC arbitration in ten days rather than through years o ---  civil litigation, and the mechanism was a reputation track- ing system that permitted members to extend unsecured credit and accept oral agreements  --- rom counterparties whose conduct history the institution monitored. 217 NYSE sel --- -regulation produced ad- verse selection reduction through surveillance, examination, and the expulsion threat, which together created the assurance that jus- ti --- ied lower transaction costs in trading with known counterparties.
 Where comparable evidence is unavailable, the analyst must identi --- y the mechanism through which member bene --- its are pro- duced and assess whether that mechanism is  --- unctioning. Function- ing monitoring produces in --- ormation that actually governs member behavior. Credible sanctions are sanctions that members actually apply to actual violations, not sanctions that exist in rules but are never used. Predictable governance rules are rules that members can and do rely on to structure their expectations. An institution whose monitoring produces paperwork rather than conduct-governing in-

ormation, whose sanctions are never applied, and whose rules members routinely circumvent is producing symbolic governance rather than actual member bene — its. The distinction between — unc- tioning governance and symbolic governance is observable — rom the

our elements established in Step 2: where those elements are real, member bene — its — ollow; where they are nominal, member bene — its are claimed but not delivered. Step 4 matters — or legal evaluation because the magnitude o — the member bene — its determines the governance stake in any legal rule Error! Use the Home tab to apply Heading 1 to the text that you want to appear here. 191

a

ecting the institution. A rule degrading governance in an institu- tion producing large, well-documented, and causally traceable member bene — its destroys more governance value than the same rule applied to an institution whose member bene — its are speculative or de minimis.

Step 5: Ask whether the governance institution creates spillovers Step 5 distinguishes genuine spillovers—externalities propor- tional to governance quality— — rom coincidental public bene — its. Proportionality is what distinguishes genuine spillovers — rom coincidental public bene — its. Carbon sequestration — rom community

orest governance rises and — alls with governance quality: well-gov- erned — orests store more carbon, and as governance degrades and rules are — louted, sequestration declines proportionally. Market in- tegrity — rom exchange sel — -regulation rises and — alls with en — orce- ment quality: rigorous surveillance and discipline produces reliable prices and — raud deterrence — or all investors, while governance deg- radation reduces market integrity proportionally. Coincidental pub- lic bene — its lack this property; they do not diminish as governance quality diminishes, and con — lating them with genuine spillovers in-

lates the governance analysis. Including spillovers changes the structure o — legal evaluation in a speci — ic and important way. Without spillover analysis, a court evaluating a governance dispute sees legal rule costs borne by the parties be — ore it: the governance institution and the excluded or dis- ciplined member. Spillover analysis reveals that those parties repre- sent only a — raction o — those whose interests the legal rule a —

ects. In the DDC context, the parties were traders; the a —

ected population extended to retailers, jewelers, insurers, and consumers whose ac- cess to reliable gem certi — ication depended on DDC governance 192 Law and Governance

quality. In the NYSE context, the parties were the exchange and Harold Silver; the a —

ected population included every retail investor whose market integrity depended on exchange en — orcement quality. Measuring a rule’s cost against only the parties be — ore the court un- derstates the governance stake in the decision. Spillover magnitude also bears on justi — ications — or legal protec- tion. When spillovers are large and the non-member population substantially exceeds the membership, protecting governance ca- pacity serves a public interest that extends well beyond club mem- bership. That scale o — public interest provides a justi — ication — or legal protection that member-bene — it analysis alone cannot supply.

Step 6: Analyze how law enables, disciplines, or weakens the governance institution Step 6 asks two questions: which enabling mechanisms does the rule supply, and which degradation mechanisms does it activate? Enabling mechanisms reduce the private cost o — producing gov- ernance, bringing governance provision closer to the social opti- mum. Four enabling mechanisms are available: judicial de — erence that reduces governance production costs (the Pigouvian subsidy

unction o — the business judgment rule, the FAA’s arbitration en-

orcement, the antitrust rule o — reason, and common law association de — erence); en — orceability o — private ordering that makes govern- ance rules legally operative beyond social pressure alone; property- rule protection o — the exclusion mechanism that preserves the cate- gorical character o — expulsion as a sanction; and legal recognition that gives the governance institution the capacity to sue, contract, hold property, and be held accountable. Error! Use the Home tab to apply Heading 1 to the text that you want to appear here. 193

 Seven degrading mechanisms  --- all into three  --- unctional clusters. Sanction-disabling mechanisms convert the property right o ---  exclu- sion to a liability rule (Mechanism 1) or impose procedural require- ments that raise sanctioning costs past the point o ---  absorption (Mechanism 2). Rule-corrupting mechanisms import in --- ormal con- textual norms into  --- ormal governance adjudication (Mechanism 3) or displace private ordering with mandatory rules that remove adaptive governance capacity (Mechanism 4). Accountability-re- moving mechanisms strip private en --- orcement rights without providing substitutes (Mechanism 5), substitute compliance struc- tures  --- or governance architecture (Mechanism 6), or mandate open membership that disables the screening  --- unction (Mechanism 7).
 Speci --- ying which mechanism is activated has three analytical consequences that distinguish mechanism identi --- ication  --- rom mere classi --- ication. First, it speci --- ies which governance  --- unction is being damaged. Disabling the sanction mechanism destroys the coopera- tion equilibrium  --- aster and more completely than corrupting rule predictability does, which in turn damages governance more imme- diately than removing accountability mechanisms. Second, it reveals whether the mechanism is calibrated to a speci --- ic governance  --- ailure or imposed categorically. Discipline responds to an identi --- ied  --- ailure through a mechanism directed at that  --- ailure. Degradation activates a mechanism regardless o ---  whether any speci --- ic  --- ailure has occurred or whether the speci --- ic  --- ailure that has occurred justi --- ies the partic- ular mechanism activated. Third, it connects the legal rule to the Part III application chapters, where each mechanism is traced through speci --- ic doctrinal domains with speci --- ic doctrinal examples.
 Mechanism speci --- ication is also what enables the discipline ver- sus destruction analysis. Discipline requires two conditions: a spe- ci --- ic identi --- ied governance  --- ailure to which the legal intervention

194 Law and Governance

responds, and a mechanism o

intervention calibrated to that — ailure. Destruction occurs when an intervention activates a degradation mechanism without satis — ying both conditions: when governance capacity is damaged whether or not any speci — ic — ailure is present, or when the mechanism is broader than the — ailure requires.

Step 7: Evaluate the law’s governance e

ects Step 7 names what the analysis has — ound and positions the — ind- ing — or use in legal decision-making. Four — indings are possible: the legal rule enables governance; disciplines governance by correcting a speci — ic — ailure in a manner calibrated to that — ailure; degrades gov- ernance by activating one o — the seven mechanisms without propor- tionality to any speci — ic governance — ailure; or produces mixed e — -

ects combining enabling or disciplining elements with degrading ones. Naming the governance — inding is distinct — rom reaching a legal verdict. Step 7 also requires identi — ying where the governance — ind- ing intersects other legal values that the larger legal judgment must weigh. When a rule degrades governance but simultaneously pro- tects against discriminatory exclusion, the evaluation should state both — acts: this rule degrades governance through Mechanism 7, at this degree o — magnitude, in service o — this anti-discrimination ob- jective. Step 7 identi — ies the governance costs and bene — its; it leaves to legal actors the judgment whether the anti-discrimination objec- tive justi — ies the governance cost. Step 7 makes both sides o — the question legible so that the legal actor reaching the verdict can weigh them accurately. Step 7 also requires identi — ying the burden o — proo — at each step. The institution claiming governance protection bears the burden at Error! Use the Home tab to apply Heading 1 to the text that you want to appear here. 195

Steps 1-3 o

demonstrating that a speci — ic governance problem ex- ists, that a — unctioning governance institution addresses it, and that legal conditions — or governance are adequate. The party challenging governance bears the burden at Steps 6-7 o — showing that the iden- ti — ied governance e —

ect is degrading rather than disciplining, mean- ing that the identi — ied legal mechanism activates without propor- tional justi — ication tied to a speci — ic documented governance — ailure. This burden allocation prevents both baseless claims o — governance protection and equally baseless attacks on legitimate governance ac- tivity. Step 7’s — inding is one input to a legal judgment. Legal actors use it alongside analysis o — rights, legitimacy, e —

iciency, and anti-abuse concerns to reach verdicts that the governance method itsel — cannot reach. What the method guarantees is that the governance side o —

every such judgment is precisely speci

ied.

Why the method is distinctively legal Three — eatures make the governance method recognizably legal ra- ther than economic or sociological.

Legal categories do the analytic work Property rules and liability rules, — iduciary duties and standing, enabling conditions and mandatory rules, discipline and destruc- tion: these are legal categories, not economic variables or sociologi- cal constructs. Law constitutes, recognizes, and regulates institu- tions through these categories, and the governance method works through them. Because the method’s conclusions are generated through legal analysis, they are contestable through legal argument. 196 Law and Governance

 A governance evaluation  --- inding that a rule converts expulsion

rom a property rule to a liability rule rests on the same kind o — anal- ysis lawyers apply when determining whether a property interest has been taken or merely regulated, whether a duty attaches to a re- lationship, or whether a statutory scheme creates a private right o —

action. Courts and scholars know how to contest and resolve ques- tions o — legal characterization through text, structure, precedent, and purpose. A party arguing that a rule does not actually convert expulsion to a compensated remedy, or that the governance institu- tion’s en — orcement mechanism does not depend on the categorical character o — expulsion, is making a legal argument that engages the method on its own terms. Wel — are economic analysis lacks this — eature. A — inding that a rule reduces aggregate surplus by some estimated magnitude is con- testable only by disputing the estimate, which means providing di — -

erent data, di —

erent models, or di —

erent assumptions about con- sumer pre — erences and market structure. Courts are poorly equipped

or that kind o — contestation; adjudicating competing econometric models is not what courts do. Governance analysis produces — ind- ings amenable to legal argument, which is the — orm o — contestation courts are built to conduct.

Agreement is achievable without

oundational consensus Sunstein’s account o — incompletely theorized agreements iden- ti — ies the characteristic structure o — success — ul legal reasoning: legal actors agree on particular outcomes and speci — ic explanations with- out requiring agreement on — oundational principles.218 Governance analysis operates at exactly this level. Legal actors can agree that a speci — ic rule activates the property-to-liability-rule conversion Error! Use the Home tab to apply Heading 1 to the text that you want to appear here. 197

mechanism and degrades governance capacity through that mecha- nism’s causal chain, without agreeing on whether governance should be weighted heavily against rights as a general matter, whether the governance — ramework is the correct theoretical ac- count o — governance, or what law is ultimately — or. Agreement at the level o — mechanism identi — ication and causal description is achievable through legal analysis because it requires only one — oundational commitment: that governance institutions have club good structure, produce member bene — its and spillovers, and are acted upon by legal rules in the ways Chapter 7 identi — ies. A legal actor who accepts those empirical claims can apply the method without committing to any comprehensive theory o — legal value. Sunstein’s insight applies with precision: where legal systems suc- ceed in producing stable doctrine, they do so by generating agree- ment at levels o — speci — icity that do not require resolution o — the

oundational disagreements that persist between them.

Portability without reductionism Legal scholarship has generated power — ul specialized — rame- works — or evaluating law within doctrinal domains. Corporate law has the business judgment rule and the entire — airness standard. An- titrust has the rule o — reason. Association law has the — air procedure doctrine. Compliance law has Caremark’s oversight duty. Each

ramework evaluates governance decisions in domain-speci — ic terms, using concepts and precedents that have developed in that domain. None provides a cross-domain — ramework — or evaluating any legal rule by its governance e —

ects. Governance analysis provides that cross-domain — ramework without displacing the specialized — rameworks. Applied to a corpo- 198 Law and Governance

rate governance dispute, a diamond trading network dispute, a com- mons governance dispute, and a university governance dispute, the method asks the same seven questions. Answers will di —

er because the shared problems, governance institutions, legal conditions, member bene — its, spillovers, and degradation mechanisms di —

er across these settings. Portability means the questions travel intact across doctrinal — ields; it does not mean the answers converge on a single conclusion. Specialized — rameworks in each — ield remain available to supply content to the method’s steps: the business judg- ment rule — unctions as the enabling mechanism analyzed in Step 6; the rule o — reason creates the antitrust space — or cooperative govern- ance that Step 6 maps; the — air procedure doctrine is the procedural requirement whose governance cost Step 6 measures through Mechanism 2. Governance analysis gives these — rameworks a com- mon analytical home without collapsing them into one another.

Where the method does its best work The method’s in — ormational demands are not uni — orm across its seven steps, and this asymmetry determines where the method is most and least e —

ective as a practical tool. Steps 1 through 3 (identi — y the shared problem, evaluate the governance institution’s structure, and assess whether governance is — unctioning) and Step 6 (identi — y the legal mechanism through which a rule acts on governance) re- quire the kind o — institutional analysis that courts and legislators routinely per — orm: characterizing an institution’s structure, exam- ining its rules and their application, and tracing the causal pathway through which a legal rule a —

ects institutional behavior. These steps produce — indings through the legal categories the method is built on, and courts can apply them with the same con — idence they bring to Error! Use the Home tab to apply Heading 1 to the text that you want to appear here. 199

characterizing

iduciary relationships, property interests, or statu- tory schemes. Steps 4 and 5, which require evaluating member bene — its and spillovers, impose higher in — ormational demands. Governance quality is private in — ormation held by the institution and its mem- bers. Spillover bene — its are di —

use, o — ten unpriced, and attributable to governance only through causal in — erence that requires domain- speci — ic knowledge. In litigation limited to the parties be — ore the court, no party has adequate incentives to produce spillover evi- dence: the governance institution bene — its — rom overstating its gov- ernance value, the challenging party bene — its — rom understating it, and the court lacks independent access to the institutional knowledge required to evaluate competing claims. The adversarial process, which produces reliable — indings when both parties have access to the relevant evidence and incentives to present it accu- rately, is poorly suited to questions where the relevant evidence is dispersed among non-parties and the relevant causal in — erences re- quire expertise the litigation process does not supply. These in — ormational constraints mean that the method per-

orms di —

erently in di —

erent institutional settings. In legislative and regulatory design, where the in — ormational investment can be made ex ante through hearings, commissioned studies, notice-and-com- ment proceedings, and expert consultation, the method’s demands at Steps 4 and 5 can be met with reasonable rigor. A legislature con- sidering whether to extend rule-o — -reason protection to a new cat- egory o — governance institutions can commission an empirical as- sessment o — the spillovers those institutions generate and design the enabling statute to re — lect the — indings. The method’s contribution in legislative design is both structural (organizing the governance 200 Law and Governance

inquiry through the right questions) and quantitative (directing the in — ormational investment to produce — indings at each step). In litigation, the method’s contribution is primarily structural. Courts applying the method should expect to produce precise — ind- ings at Steps 1 through 3 and Step 6, where legal characterization does the analytical work, and approximate — indings at Steps 4 and 5, where the in — ormational demands exceed what adversarial proceed- ings reliably produce. This asymmetry is a — eature o — the method’s design, not a weakness. The method’s comparative advantage lies in identi — ying settings where legal analysis can proceed with con — i- dence and settings where in — ormational constraints require appro- priate epistemic humility. A method that claimed equal precision everywhere would lack operational purchase; a method that speci-

ies where its — indings are strongest — and where they should be treated as structural rather than quantitative — gives courts and leg- islators reason to trust it in the settings where it per — orms best, and reason to apply appropriate caution in settings where in — ormational demands cannot be — ully met.

Governance analysis in the presence o

rights, legitimacy, and anti-abuse Governance analysis does not operate in isolation. Legal systems simultaneously pursue rights protection, democratic legitimacy, and the prevention o — institutional abuse — values that can con — lict with governance preservation. The — ramework must speci — y how gov- ernance analysis interacts with each o — these competing commit- ments, because a method that cannot accommodate rights, legiti- macy, and anti-abuse concerns cannot — unction as a general tool o —

legal evaluation. Error! Use the Home tab to apply Heading 1 to the text that you want to appear here. 201

Rights analysis When a legal rule that protects an individual right simultane- ously degrades governance capacity, the governance — ramework does not instruct legal actors to sacri — ice the right. Rights claims have independent standing in legal analysis, and the governance method does not contest that standing. What the method changes is the quality o — the balance that legal actors draw. Roberts v. United States Jaycees shows how mandatory mem- bership overrides governance screening.219 Civic associations exer- cised screening — unctions, selecting and vetting members based on character, conduct, and compatibility with institutional norms, that serve the monitoring and membership-quality purposes Chapter 6 identi — ied. Mandatory open membership activates Mechanism 7, disabling the screening — unction and reducing the institution’s ca- pacity to maintain governance norms that sustain member cooper- ation. That governance cost is real: the institution becomes less able to exclude those whose inclusion would damage the cooperation equilibrium producing member bene — its and spillovers. The state’s compelling interest in eradicating discrimination against women is a claim o — the — irst constitutional order, and the governance analysis does not diminish it. Governance analysis sup- plies what was previously missing — rom the constitutional balance: identi — ication o — the speci — ic governance — unction that mandatory membership overrides (screening), the mechanism through which the override damages governance (Mechanism 7), and the dimen- sion o — governance quality a —

ected (monitoring capacity and coop- eration equilibrium maintenance). Courts conducting Roberts-style balancing have generally proceeded without this speci — ication, leav- 202 Law and Governance

ing the governance side o

the constitutional scale empty or approx- imate. Making it precise changes the quality o — the balancing with- out altering its structure or its constitutional logic. Rights analysis is already structured as a balancing inquiry in most contexts where governance costs arise: compelling interest analysis, proportionality review, heightened scrutiny. Governance analysis supplies one side o — those balancing inquiries with greater precision, enabling conclusions to be reached through analysis ra- ther than assumption.

Legitimacy concerns Procedural requirements on governance institution disciplinary decisions implicate two genuine legal values simultaneously. Proce- dural protection — or members subject to governance authority, spe- ci — ically the opportunity to be heard be — ore being expelled or disci- plined, serves rule-o — -law values that are independent o — the gov- ernance analysis and anterior to it. Raising the cost o — exercising sanctions by imposing procedural burdens degrades the cooperation equilibrium that governance sustains, because members who antic- ipate that en — orcement will be costly begin to discount the sanction threat. Both consequences are real. Calibration is where governance analysis adds value to legiti- macy analysis. Procedural protection — or excluded non-members targets a genuine governance — ailure: parties signi — icantly a —

ected by governance decisions without any — ormal relationship to the gov- ernance institution deserve process, and governance institutions ex- ercising authority over such parties without providing it have — ailed in a way that legal intervention should correct. Procedural protec- tion extended categorically to all member discipline decisions im- Error! Use the Home tab to apply Heading 1 to the text that you want to appear here. 203

poses governance costs without a corresponding governance bene-

it, because the speci — ic — ailure the protection is designed to prevent, arbitrary exclusion o — members who had no opportunity to respond, has not been established to have occurred in the member discipline context. Governance analysis produces the — inding: this rule im- poses governance costs through Mechanism 2, at this magnitude, in response to this identi — ied — ailure (or without any identi — ied — ailure). Whether the costs are justi — ied in light o — legitimacy bene — its is the normative judgment that legal actors must make with that — inding in hand.

The anti-abuse concern Governance language can be used to insulate abusive conduct

rom legal scrutiny. Private institutions have invoked associational autonomy to shield discriminatory exclusion, anticompetitive mar- ket division, and insider capture — rom precisely the legal interven- tions that should correct them. That the governance — ramework provides grounds — or resisting some legal interventions creates a risk that it will be misused to resist interventions that are legitimate. This concern is the most structurally serious challenge the — rame- work — aces, because the risk is internal rather than external: it arises

rom within the — ramework’s own logic. Steps 2 and 6 address the concern structurally rather than through external limiting principles. Step 2 requires that a govern- ance institution actually exist in the relevant sense, which means demonstrating that its — our elements are — unctioning: that monitor- ing produces conduct-governing in — ormation, that sanctions are ac- tually applied, that decision-making governs actual behavior, and that adjustment tracks actual changed circumstances. An institution 204 Law and Governance

that nominally monitors but never generates conduct-governing in-

ormation, nominally sanctions but never en — orces, and nominally adjusts rules but does so only to entrench insiders — ails Step 2. When the governance protection is claimed by an institution that Step 2 shows to be a governance shell, the claim — ails at its — oundation. Step 6’s discipline versus destruction analysis addresses the con- cern at the remedial level. Discriminatory exclusion is either a gov- ernance — ailure that the discipline criterion covers, in which case le- gal intervention is warranted by the — ramework’s own terms, or a non-governance abuse that the institution is attempting to shield under governance language, in which case Step 2 should already have revealed that the institution lacks genuine governance — unc- tions in the relevant respect. The — ramework’s response to anti- abuse is embedded in the method’s structure because Steps 2 and 6 together require demonstrating both that governance exists and that any challenged legal intervention responds to an identi — ied govern- ance — ailure rather than to legitimate governance activity. Chapter 14 develops this analysis through cases in which courts have correctly overridden governance institution claims on anti- discrimination, antitrust, and rights grounds. Governance analysis endorses those outcomes as discipline responses to identi — ied gov- ernance — ailures, and it shows how the method reaches those char- acterizations through its structured analysis rather than assuming them. Error! Use the Home tab to apply Heading 1 to the text that you want to appear here. 205

The method applied: Silver v. New York Stock Exchange Silver v. New York Stock Exchange, 373 U.S. 341 (1963), appears throughout this book because it illustrates simultaneously what gov- ernance institutions do, how law acts on them, and what happens when legal analysis addresses only the dispute between the parties be — ore the court and misses the governance institution that the dis- pute implicates. Applying the — ull method makes the illustration precise.

Step 1: The shared problem NYSE members shared trading in — rastructure and depended on the reliability o — market prices and the creditworthiness o — trading counterparties. Market manipulation, — raud, and breach o — trading obligations were the shared problem: le — t unchecked, such conduct could propagate through the interconnected trading system and de- stroy the market integrity that made exchange membership valuable in the — irst place. No individual market participant could e —

iciently monitor all its counterparties; the coordination di —

iculties and in-

ormation costs o — individual monitoring would have been prohib- itive. Collective governance in — rastructure was the solution to a col- lective action problem.

Step 2: The governance institution NYSE governance had all — our required elements in — unctioning

orm. Decision-making occurred through listing standards, trading rules, and member conduct requirements adopted by the exchange’s governing body. Monitoring occurred through market surveillance and member examination programs that generated actual conduct- 206 Law and Governance

governing in

ormation. Sanctions operated through the exchange’s disciplinary apparatus, with expulsion as the ultimate sanction re- moving a member — rom all trading — loor access and bene — its. Adjust- ment occurred through periodic revision o — trading and conduct rules in response to changing market conditions. Each element was

unctioning, not merely — ormal.

Step 3: Legal conditions Be — ore Silver, NYSE governance rested on strong legal condi- tions across all six dimensions. Legal recognition was unambiguous: the Securities Exchange Act registered the exchange as a sel — -regu- latory organization with delegated regulatory authority. Member- ship boundaries were en — orced as property-rule entitlements: expul- sion was categorical, not convertible to a damages remedy at the ex- pelled member’s option. Exchange rules were externally en — orceable as contractual terms o — membership through the contract theory o —

association rules. Fiduciary and quasi-

iduciary obligations attached to exchange authority through the Exchange Act’s regulatory — rame- work. Implied antitrust immunity protected sel — -regulatory activi- ties — rom per se condemnation. Those conditions collectively gave the exchange’s governance mechanisms their legal — orce. Silver degraded three o — the six conditions simultaneously. Ex- posing expulsion decisions to antitrust liability reduced the prop- erty-rule character o — expulsion by conditioning categorical author- ity on procedural compliance. Imposing procedural requirements raised the practical cost o — exercising that authority. Narrowing the implied immunity reduced the legal protection — or sel — -regulatory activities generally. Step 3’s — unction here is establishing the base- line: initially strong legal conditions that the decision under analysis degraded. Error! Use the Home tab to apply Heading 1 to the text that you want to appear here. 207

Step 4: Member bene

its NYSE governance produced member bene — its across three o —

Chapter 6’s

our categories. Transaction costs among member — irms were reduced through shared market in — rastructure, standardized trading rules, and a reputation mechanism permitting — irms to con- duct large transactions on the basis o — known counterparty conduct histories that the exchange tracked. Adverse selection was reduced through surveillance and examination: a member transacting with another member had assurance that the counterparty had been vet- ted — or compliance with exchange requirements and remained sub- ject to ongoing monitoring. Dispute resolution was available through the exchange’s arbitration system at lower cost and with greater specialized expertise than civil litigation. Coordination was also present in the exchange’s role setting standardized protocols en- abling e —

icient price discovery.

Step 5: Spillovers Mahoney documented that exchange sel — -regulation produced market integrity — or all investors, not only member broker-deal- ers.220 Price discovery e —

iciency, the accuracy o — equity prices as sig- nals o — underlying — irm value, bene — ited all economic actors who re- lied on equity market valuations: pension — unds, insurers, creditors, and ultimately ordinary savers. Fraud detection and en — orcement o —

trading rules protected retail investors who never became exchange members. The Supreme Court acknowledged this in Silver itsel — , rec- ognizing that exchange sel — -regulation served the broader public purpose o — maintaining market integrity. Proportionality holds: ex- change governance that — unctions less e —

ectively produces less mar- ket integrity, less accurate prices, and less — raud deterrence — or non- members, in proportion to the governance quality decline. 208 Law and Governance

Step 6: Legal e

ects Silver activated two degradation mechanisms simultaneously, both within the sanction-disabling cluster. Mechanism 1 operated by converting the NYSE’s expulsion au- thority — rom a categorical property-right entitlement to a liability- rule remedy subject to antitrust scrutiny and treble damages expo- sure. Be — ore Silver, disciplinary authority had the categorical charac- ter o — a property rule: the exchange could expel a member — or con- duct rule violations without exposing itsel — to antitrust liability — or having done so. A — ter Silver, exercising that authority in any case ar- guably involving competitors created antitrust exposure. Expected liability costs, multiplied by the probability o — success — ul antitrust challenge, raised the private cost o — disciplinary decision-making throughout the institution. Formal authority survived; the practical cost o — using it increased enough to alter en — orcement behavior across the institution’s entire disciplinary apparatus. Mechanism 2 operated through procedural requirements im- posed as preconditions — or antitrust immunity. Notice and an op- portunity to be heard be — ore exclusionary decisions added adminis- trative cost to every disciplinary decision, required the exchange to maintain records adequate to demonstrate procedural compliance, and created litigation exposure whenever members contested whether adequate process had been provided. Both mechanisms targeted the exchange’s governance capacity at its sanction — unction, the element most essential to maintaining the cooperation equilibrium that addressed the shared problem o —

raud and market manipulation. The governance — ailure Silver addressed was genuine and speci — - ically located. Harold Silver was a non-member broker. His direct- wire connections were terminated without notice and without an Error! Use the Home tab to apply Heading 1 to the text that you want to appear here. 209

opportunity to be heard. Silver had no membership relationship with the exchange and no — ormal mechanism — or contesting the de- cision. An institution exercising authority over parties outside its membership, a —

ecting those parties signi — icantly, without provid- ing any process at all, has — ailed in a way that legal intervention should correct. The Court’s remedy, however, extended — ar beyond the identi-

ied — ailure. Broad antitrust liability and categorical procedural re- quirements across all NYSE disciplinary activities, member disci- pline included, addressed a — ailure concerning non-member access to exchange in — rastructure. Member discipline is a di —

erent govern- ance — unction. Silver was not a member; member discipline was not the mechanism through which he was harmed. Governance costs

ell on the member discipline — unctions that had no causal relation- ship to the speci — ic — ailure the case addressed.

Step 7: Evaluation Silver yields a mixed governance evaluation, predominantly de- grading. Legitimate discipline is present. Governance institutions exer- cising authority over parties outside their membership, a —

ecting those parties signi — icantly without providing any process, have

ailed in a way that warrants correction. A targeted procedural re- quirement limited to non-member access decisions would have ad- dressed that — ailure without activating degradation mechanisms on unrelated member discipline — unctions. That targeted intervention would constitute discipline as Chapter 7’s analysis de — ines it: re- sponding to a speci — ic governance — ailure through a mechanism cal- ibrated to correct it. 210 Law and Governance

 Predominant degradation  --- ollows  --- rom the remedy's  --- ailure on both dimensions o ---  calibration Chapter 7 identi --- ies. On scope, anti- trust liability and procedural requirements were imposed across all exchange governance decisions, member and non-member alike, degrading the member discipline governance that bore no causal re- lationship to the identi --- ied  --- ailure. Mechanisms 1 and 2 are both ap- plied beyond the governance  --- unction that  --- ailed: they damage gov- ernance capacity whether or not the speci --- ic  --- ailure involving non- member access is present. On intensity, the remedy exposed the Ex- change to treble damages under the Clayton Act  --- or exercising a governance  --- unction — disciplining members — that served market integrity rather than anticompetitive exclusion, imposing a govern- ance cost disproportionate to the governance bene --- it o ---  correcting a procedural  --- ailure in non-member access decisions. Mahoney's em- pirical analysis documents the institutional e ---

ect: constrained en-

orcement behavior across the exchange’s disciplinary apparatus, not only in non-member access decisions. The gap between identi-

ied — ailure and categorical remedy, on both the scope and intensity dimensions, is what makes the decision predominantly degrading rather than purely disciplining. Governance analysis does not conclude that Silver was wrongly decided as a matter o — law. Antitrust scrutiny o — exchange govern- ance, procedural protection — or parties a —

ected by exchange deci- sions, and judicial oversight o — sel — -regulatory organizations all have legitimate grounding in competition policy and administrative law. What governance analysis concludes is that the governance cost was not measured, the — ailure was not precisely identi — ied as one about non-member access rather than member discipline generally, and the remedy was accordingly broader than the — ailure required. A Error! Use the Home tab to apply Heading 1 to the text that you want to appear here. 211

court with this analysis available would have been equipped to ad- dress the non-member access problem through targeted process re- quirements while preserving the categorical character o — member discipline. Supplying that analysis so that calibration is possible is the method’s contribution: not reversing Silver, but preventing its

ailure o — calibration — rom recurring.

The

ramework in Part III Part III applies the governance method to six doctrinal domains. Each chapter asks the same seven questions, and the answers di —

er because the institutional settings di —

er. Network governance (Chapter 9) presents governance institu- tions whose in — ormal character courts have consistently misread as the absence o — governance. Exclusion — rom commercial networks has governance — unctions, including maintaining the cooperation equilibrium, certi — ying member quality, and deterring de — ection, that courts have analyzed as anticompetitive conduct rather than governance activity. Governance analysis reveals both the govern- ance institution courts are missing and the spillovers that make the governance costs o — misreading it signi — icant. Contract remedies (Chapter 10) presents what this book calls the wrong plainti —

problem: courts protecting the individual promisee while — ailing to account — or the harm to the governance institution that made the promise possible. Breach o — a governance-embedded commitment injures not only the individual promisee but the insti- tutional — ramework through which similar commitments could be made and relied on. Governance analysis identi — ies the institution that is the real plainti —

and examines what contract doctrine does to it. 212 Law and Governance

 Corporate governance (Chapter 11) presents the domain where governance analysis is most developed in existing doctrine, includ- ing the business judgment rule, entire  --- airness review, Caremark, and Blasius, but least explicitly  --- ramed in governance terms. Gov- ernance analysis supplies the explicit governance  --- raming, which changes how the doctrine's enabling and degrading  --- unctions are understood and which doctrinal developments appear as govern- ance improvements versus governance erosions.
 Universities and nonpro --- its (Chapter 12) present governance in- stitutions with the largest spillovers relative to their capacity to sel --- -

inance. Educational and philanthropic institutions produce public goods whose bene — its extend — ar beyond their membership, which means governance voids in these institutions impose costs on ben- e — iciary populations who have no — ormal voice in governance. The standing gap that Mechanism 5 identi — ies is most consequential here, and governance analysis reveals why the nonpro — it accounta- bility problem is a structural rather than incidental — eature o — non- pro — it law. Knowledge institutions (Chapter 13) turn the governance anal- ysis toward the legal academy itsel — . Law reviews and other scholar- ship-producing bodies have governance institutions with — our ele- ments, namely editorial decision-making, peer review, citation practices as social monitoring, and methodological adjustment, whose quality determines the reliability o — the knowledge that legal scholarship and legal practice depend on. Governance analysis o —

knowledge institutions asks what legal rules enable or degrade epis- temic governance, which is the governance o — the mechanisms that produce reliable legal knowledge. Error! Use the Home tab to apply Heading 1 to the text that you want to appear here. 213

 Limits, Gaps, and Research Horizons (Chapter 14) develops the cases in which governance analysis correctly supports legal inter- ventions that impose governance costs: anti-discrimination man- dates, antitrust en --- orcement against anticompetitive exclusion, and due process requirements in governance institutions exercising co- ercive authority. Those interventions are the discipline side o ---

Chapter 7’s discipline versus destruction analysis. Chapter 14 shows how governance analysis reaches them rather than treating them as external exceptions to a pro-governance rule. Di —

erent shared problems, di —

erent governance institutions, di —

erent legal conditions, di —

erent member bene — its, di —

erent spill- overs, and di —

erent legal mechanisms produce di —

erent governance evaluations across these six domains. An accurate analytical method should produce variation rather than uni — ormity, and the govern- ance method does. What it does not do is predetermine outcomes. It determines what questions must be answered to evaluate any legal rule by what that rule does to governance, and it supplies a vocabu- lary precise enough to make those answers carry doctrinal weight.

Evidentiary standards The seven-step method speci — ies the analytical questions at each step but has not yet speci — ied what evidence satis — ies each step. A court or regulator applying the method needs to know: what counts as ad- equate evidence that a governance institution satis — ies Step 2’s — our- element test? What counts as adequate evidence o — member bene — its at Step 4? What counts as adequate evidence o — spillovers at Step 5? The — ramework proposes the — ollowing burden structure. A governance institution seeking the — ramework’s protective implica- tions bears the initial burden o — demonstrating, by a preponderance 214 Law and Governance

o

documentary evidence, that its — our governance elements are

unctional within the meaning o — Step 2. Functionality must be es- tablished through the output-based criteria Chapter 1 speci — ies: monitoring that generates conduct-governing in — ormation actually used in decisions, sanctioning applied at a — requency consistent with the observable violation rate, decision-making that governs actual member conduct, and rules that change in response to documented inadequacies. Formal structures that satis — y none o — these output cri- teria do not satis — y Step 2, regardless o — their organizational com- plexity. Once the governance institution has established — unctionality at Step 2, the party challenging the governance institution’s exclusion or other governance decision bears the burden o — demonstrating, by a preponderance o — the evidence, that the challenged decision does not serve any o — the governance — unctions identi — ied at Step 4. This allocation re — lects the institutional competence point: the govern- ance institution possesses contextual knowledge about the relation- ship between its membership criteria and its governance quality that external challengers typically lack, and requiring the governance in- stitution to establish its bona — ides be — ore shi — ting the burden en- sures that the — ramework’s protections extend only to institutions with genuine governance — unction. The court should not require the governance institution to quanti — y spillover bene — its with economic precision at Step 5. Qual- itative evidence o — a causal mechanism linking governance quality to non-member bene — it, supported by at least one documented in- stance o — spillover degradation when governance quality declined in a comparable institution, su —

ices — or Step 5’s analytical purpose. Re- quiring quantitative precision would impose an evidentiary stand- Error! Use the Home tab to apply Heading 1 to the text that you want to appear here. 215

ard that the underlying economics does not support: spillover ben- e — its are by de — inition not priced in markets, and demanding mar- ket-equivalent measurements would systematically undervalue pre- cisely the bene — its the — ramework exists to identi — y.

Remedial protocol When a court identi — ies a speci — ic governance — ailure under Steps 2 and 6, the remedial question is what the court should do about it. The — ramework proposes that courts consider remedies in the — ol- lowing sequence, ordered by ascending governance cost: First, a declaratory judgment identi — ying the governance — ailure and stating the legal expectation. A declaration that a governance institution has — ailed at a speci — ic element — that its monitoring sys- tem does not generate conduct-governing in — ormation, or that its adjustment mechanism has not responded to documented inadequa- cies — creates a public record o — the — ailure without imposing direct governance costs. Declaratory relie — preserves the institution’s au- thority to correct the — ailure through its own governance processes while putting the institution on notice that — urther — ailure may war- rant escalation.221 Second, a mandatory disclosure order requiring the governance institution to document its corrective actions within a speci — ied pe- riod. Disclosure orders impose modest governance costs — the cost o — documenting what the institution is doing to address the identi-

ied — ailure — while generating in — ormation that the court, regula- tors, and stakeholders can use to evaluate whether correction has occurred. Third, a targeted injunction limited to the governance — unction that — ailed. Where the — ailure is speci — ic — a monitoring system that 216 Law and Governance

ignores a category o

misconduct, a sanctioning mechanism that has not been applied in documented cases o — violation — the injunction addresses that — ailure without extending to governance — unctions that are operating adequately. Scope calibration, as Chapter 7 devel- ops it, requires that the injunctive remedy target the — ailed — unction and no other. Fourth, appointment o — a special governance monitor — or a de-

ined term. The governance monitor remedy parallels the corporate compliance monitor structure used in de — erred prosecution agree- ments, adapted to the governance context.222 Governance monitors impose the highest governance costs o — any remedy in the sequence because they introduce an external actor into the governance pro- cess itsel — , potentially displacing the internal governance mecha- nisms the — ramework exists to protect. Monitor appointment is there — ore appropriate only where the — irst three remedial tiers have

ailed to produce correction, or where the governance — ailure is so pervasive across all — our elements that no internal correction mech- anism is plausibly available. The court should impose the least governance-costly remedy reasonably likely to correct the identi — ied — ailure. I — the initial rem- edy — ails to correct the — ailure within a stated period, the court may escalate to the next remedial tier upon renewed motion supported by evidence that the initial remedy was insu —

icient. This escalating structure re — lects the — ramework’s core commitment: governance degradation is a real cost, and remedies that impose unnecessary governance costs are themselves a — orm o — governance harm even when they respond to genuine governance — ailures. Error! Use the Home tab to apply Heading 1 to the text that you want to appear here. 217

Evaluating Part III applications Part III applies this method to six doctrinal domains. Three criteria determine whether an application succeeds. First, the application must identi — y a shared problem and governance institution that the reader would not otherwise recognize as governance — revealing institutional structure that existing doctrine obscures. Second, the seven-step method must produce a governance evaluation that di — -

ers — rom the evaluation existing doctrine provides, identi — ying costs or bene — its that standard analysis misses. Third, the governance evaluation must generate prescriptive implications speci — ic enough to guide doctrinal re — orm or judicial reasoning, rather than restating at a higher level o — generality what existing analysis already pro- vides. Where an application satis — ies all three criteria, the — rame- work has demonstrated portability with analytical value. Where an application satis — ies — ewer than three, the limits o — the — ramework’s reach become visible, and those limits are themselves in — ormative

or the research agenda Chapter 15 develops.  Part III: Applications and Evaluation

Part II developed the book’s theory. Part III puts that theory under pressure. The chapters that — ollow do not merely illustrate the

ramework with convenient examples. They test whether the same method can explain institutional domains that di —

er sharply in structure, doctrine, and stakes: private commercial networks, stock exchanges, corporate boards, universities, nonpro — its, and knowledge-producing institutions. I — the — ramework is genuinely portable, it should reveal something important in each setting with- out collapsing their di —

erences into a single abstract pattern. That is the task o — this Part. Each chapter asks the same basic questions: what shared problem is the institution managing, what governance structure addresses it, what legal conditions sustain it, and what law does to that structure when courts, legislatures, or reg- ulators intervene. But the answers di —

er by domain. Some chapters 220 Law and Governance

show law degrading governance by weakening sanctioning author- ity. Others reveal law — ailing through accountability voids. A third group addresses misrecognition o — institutional — unction. Chapter 9 begins with the domain in which private governance is easiest to see: commercial networks and private dispute resolu- tion. The New York Diamond Dealers Club and related merchant institutions show governance operating in a relatively pure — orm, with membership boundaries, credible sanctions, and internal dis- pute resolution mechanisms that sustain cooperation where ordi- nary legal en — orcement would be too slow or costly. The chapter uses this setting to demonstrate the book’s central claim that gov- ernance institutions can generate substantial spillover value — or out- siders while depending — or their success on exclusionary powers that legal analysis o — ten treats with suspicion or ignores altogether. Chapter 10 then — ollows the historical arc o — exchange govern- ance through the New York Stock Exchange. The chapter traces how Silver and the legal developments that — ollowed trans — ormed a sel — -governing institution into a regulated entity whose disciplinary authority became progressively less autonomous. This is the book’s

ullest case study in governance degradation across time. It shows how a legal intervention can begin as a response to a legitimate abuse yet still produce broader institutional costs when the remedy is not calibrated to the governance — unction at stake. Chapter 11 turns to corporate governance, where the relevant institution is more — amiliar and the doctrine more developed. Here the — ramework yields a di —

erent insight. The problem is not simply that law inter — ered too much with governance, but that re — ormers repeatedly misunderstood governance’s structure. By treating cor- porate governance as i — it were a public good rather than a club good, Chapter 8: A Method — or Evaluating Law’s Governance E —

ects 221

shareholder-rights re

orms expanded participation in ways that de- graded governance quality, encouraged proxy advisor concentra- tion, and produced the pattern o —

ormal re — orm without — unctional success. The chapter also shows that Delaware doctrine, especially its review spectrum, o — ten tracks governance calibration more closely than its vocabulary suggests. Chapter 12 shi — ts — rom market institutions to universities and nonpro — its, where the central problem is not excessive inter — erence but an accountability void. Sel — -perpetuating boards, hollow — iduci- ary duties, and the absence o — stakeholder standing create govern- ance structures that can exercise enormous power while resisting meaning — ul correction. The resulting — ailures are not incidental; they are outputs o — the legal architecture. This chapter demonstrates that compliance- — ocused interventions o — ten leave those structures intact, which is why they repeatedly — ail to produce genuine re — orm. Chapter 13 applies the — ramework re — lexively to knowledge in- stitutions themselves. Law reviews and related scholarship-produc- ing institutions do not just comment on law; they certi — y claims that courts, agencies, and scholars rely on. The governance question is there — ore unavoidable: what screening institution produced this knowledge, and what con — idence does that process justi — y? The chapter argues that student-edited law reviews are structurally mis- matched to the epistemic demands o — much empirical legal scholar- ship, producing predictable problems o — adverse selection and meth- odological opacity. Chapter 14 then turns the — ramework on itsel — . A theory that travels across domains must identi — y its limits as well as its reach. That chapter addresses the danger that governance language can le- gitimate private power, the coercive potential o — private govern- ance, the antitrust inter — ace, the relationship between governance 222 Law and Governance

autonomy and anti-discrimination law, the problem o

doctrinal de- terminacy, and the historical relationship between governance ex- clusion and systematic discrimination. These are not objections to be brushed aside. They are the places where the — ramework is tested most seriously and where — uture work must extend it. Chapter 15 concludes by gathering the method’s implications — or legal analysis and identi — ying the research agenda that must — ollow. This conclusion previews the application chapters, then addresses the book’s own relationship to law review certi — ication and knowledge production. Chapter 8 speci — ied three criteria — or evaluating whether an ap- plication succeeds: the application must reveal governance structure that existing doctrine obscures, generate a governance evaluation that di —

ers — rom standard analysis, and produce prescriptive impli- cations speci — ic enough to guide re — orm. Each o — the six application chapters satis — ies these criteria, but each chapter most prominently demonstrates a di —

erent one. Chapters 9 and 10 most prominently satis — y the — irst criterion: they reveal governance institutions — the DDC’s commercial network and the NYSE’s sel — -regulatory appa- ratus — that courts have consistently misidenti — ied or missed en- tirely. Chapter 11 most prominently satis — ies the second: it shows that the governance — ramework generates predictions about share- holder-rights re — orms that diverge — rom standard corporate govern- ance analysis and that the available evidence supports the govern- ance prediction. Chapter 12 most prominently satis — ies the third: it produces speci — ic structural re — orm proposals — standing re — orm, duty o — obedience re — orm, operational test re — orm — that — ollow di- rectly — rom the governance diagnosis. Chapters 13 and 14 serve a di — -

erent but complementary — unction: they test the — ramework’s limits Chapter 8: A Method — or Evaluating Law’s Governance E —

ects 223

by applying it re

lexively to knowledge institutions and then sub- jecting the entire analysis to the strongest available objections. Across these chapters, the seven-step method — rom Chapter 8 remains constant, but its application is not mechanical. The method is a disciplined inquiry, not a checklist that generates the same an- swer wherever it is used. What changes — rom chapter to chapter is the institution under examination, the legal mechanism acting on it, and the — orm o —

ailure or success at issue. That variation is the point. A use — ul — ramework should travel while remaining sensitive to institutional di —

erence. Part III there — ore serves two purposes at once. It evaluates par- ticular bodies o — law, and it demonstrates the — ramework’s practical value by showing that governance becomes visible as a legal variable in settings where existing doctrine o — ten notices only rights, incen- tives, or dispute-centered litigation. The claim o — this Part is not that governance analysis replaces those modes o — reasoning. It is that without governance analysis, each o — them remains incomplete. Chapter 9: Network Governance and the Missing Institution Members o — the New York Diamond Dealers Club almost never sue each other. In a business where a single handshake commits a mer- chant to paying six — igures — or rough stones (no written contract, no

ormal title trans — er, no recourse to warranty law), the near-zero de-

ault rate is not the product o — judicial en — orcement. It is the product o — governance architecture: mandatory arbitration, a reputation network that tracks compliance histories across thousands o — trad- ing relationships, and a credible threat o — expulsion — rom the only market where this trade is conducted. A member who dishonors a handshake deal can be expelled — rom the bourse within days. Expul- sion does not cost one relationship; it costs every — uture purchase, every — uture sale, every — uture line o — credit throughout an entire industry. Lisa Bernstein documented the result a — ter years o — empir- ical study: disputes within the DDC are resolved overwhelmingly through internal procedures, not courts, and de — ault rates remain close to zero across decades o — trading.223 Members avoid litigation not because legal remedies are unavailable but because litigation is an in — erior substitute — or a governance system that resolves disputes

aster, at lower cost, and with better-calibrated outcomes than any 226 Law and Governance

court. Chapter 1 introduced the DDC as the book’s

irst illustration

o

—

governance without

—

ormal legal recognition; Chapter 5 placed

its exclusion mechanism within the club-good typology; and Chap-

ter 6 traced the spillover bene

—

its its governance produces

—

or parties

who never join the bourse. This chapter applies the complete seven-

step method to what those earlier treatments only anticipated.

That governance architecture is what courts have almost never

seen clearly, because it presents to them (when it presents at all) not

as a governance institution but as one side o

—

a dispute-centered

controversy. When a purchasing cooperative expels a member and

the expelled member challenges the expulsion as an antitrust viola-

tion, the court sees an organization and a

—

ormer participant. When

a securities sel

—

-regulator like FINRA moves to expel a broker-

dealer, the court sees a constitutional administrative law question

about private delegated authority. When a pro

—

essional society re-

moves a member

—

or violating its rules, the court sees a dispute

about whether the rules were

—

airly applied. What no court sees, in

any o

—

these settings, is what every remaining member o

—

the gov-

ernance institution observes: that the institution’s authority to ex-

clude, the speci

—

ic mechanism that sustains cooperation among all

the others, is being converted

—

rom a categorical entitlement into a

procedural contingency whose outcome depends on strangers with

no stake in the governance institution’s continued

—

unction.

This chapter identi

—

ies that structural gap, traces its mani

—

esta-

tions across three doctrinal settings, and shows what governance

analysis would have asked that dispute-centered analysis did not.

The object o

—

analysis is the governance institution itsel

—

not the dispute-centered relationship between the institution and the ex- pelled member, but the shared en — orcement in — rastructure on which every other member depends and which courts, through no malice Governance Evaluation Method 227

and by the logic o

their own analytical — ramework, render invisible at the moment o — decision. 228 Law and Governance

Chapter 9. Network Governance Commercial trading networks supply the book’s — oundational appli- cation because they isolate the governance phenomenon in its pur- est — orm. When merchants build systems o — rules, monitoring, sanc- tions, and dispute resolution that — unction as alternatives to court- centered en — orcement, the resulting institution is governance in the sense this book de — ines. Courts, however, routinely analyze disputes arising within these networks without recognizing the governance institution that made cooperation possible — treating each case as a bilateral contract problem or an antitrust restraint rather than an institutional question.

How trading networks produce compliance without courts Standard economic theory predicts that unsecured credit between strangers will be routinely exploited. When the only consequence o — de — ault is a judgment payable to the seller, a rational buyer will skip payment whenever the savings exceed the probability- weighted cost o — the judgment. Four independent bodies o — research document that prediction — ailing, comprehensively and across cen- turies. Bernstein’s DDC study — ound that New York and Antwerp dia- mond merchants conduct tens o — thousands o — dollars in transac- tions annually on — orty- — ive-day oral credit, with no written con- tract and a near-zero de — ault rate over decades.224 Avner Grei — doc- umented the same structure among eleventh-century Maghribi traders, who sent goods on consignment across the Mediterranean to agents they had never met, relying on a trading coalition that co- ordinated collective re — usal to deal against cheaters.225 Janet Landa Network Governance 229

ound a parallel institution among Hokkien merchants in the Ma- laysian rubber markets o — the 1960s, sustained by kinship ties and commercial ethics operating without — ormal legal in — rastructure.226 Paul Milgrom, Douglass North, and Barry Weingast reconstructed a centralized reputation system at the medieval Champagne Fairs: a record-keeper who tracked compliance histories across thousands o — traders, enabling punishment by merchants who had never dealt with the cheater personally.227 Elinor Ostrom documented the same structure in common-pool resource communities managing — isher- ies, irrigation systems, and grazing lands — or centuries without re- course to courts.228 Bernstein’s study o — the cotton trade added an- other: trade associations maintain arbitration systems backed by ex- pulsion authority, producing compliance rates that contract law alone cannot explain.229 Industries, centuries, and continents separate these examples. What they share is not a cultural disposition toward honesty. Oliver Williamson made the essential point in 1993: commercial actors ex- tend credit not because they believe in a counterparty’s moral char- acter but because the institutional structure makes de — ault more costly than compliance.230 Bernstein sharpened this in her 2015 arti- cle introducing network governance as the analytical concept that replaces trust in the legal literature. Network governance, as Bern- stein de — ines it, is the collective — orce o — reputation-based, non-legal sanctions — lowing — rom a — irm’s position within a network o — inter- connected — irms.231 Networks generate the conditions under which cooperation is rational. Trust (the — elt sense o — security in a transac- tion) is an output o — the governance mechanisms that produce those conditions, not an independent input. 230 Law and Governance

 Network structure, not cultural homogeneity, produces govern- ance capacity. Grei ---  had modeled the Maghribi traders as an ethni- cally bounded coalition: a centralized institution that collected in-

ormation about misconduct, disseminated — indings to members, and coordinated multilateral punishment. 232 Bernstein’s 2019 rea- nalysis drew on network topology research to propose a di —

erent mechanism. The Maghribi operated within a small-world network: dense local clusters o — merchants connected by sparse long-distance ties through bridge — igures she called merchants’ representatives.233 These bridge nodes aggregated reputational in — ormation and al- lowed news o — cheating to propagate rapidly across the Mediterra- nean without requiring a central clearinghouse. Bernstein’s reanal- ysis preserves Grei — ‘s conclusion that reputation-based governance worked among the Maghribi while redirecting the causal account

rom ethnic exclusivity to network structure. Governance capacity

ollows — rom the density, speed, and credibility o — in — ormation — lows within the network. What all these networks share, once the trust — rame is removed, is a single en — orcement mechanism: exclusion. A diamond dealer who de — aults loses access to the entire bourse, including every — uture purchase, every — uture sale, and every — uture line o — credit. A Ma- ghribi merchant who cheats — aces collective re — usal to deal — rom every Maghribi trader across the Mediterranean. A cotton merchant who re — uses to comply with a trade association arbitration award

aces expulsion and loss o — the bonding system. One-on-one reputa- tion is too weak to explain the observed compliance rates in large networks, where any particular pair o — traders transact in — requently. I — the only consequence o — cheating were losing one relationship, the cheater would — ace a low cost. Expulsion changes the calculus entirely. A dealer deciding whether to skip a six- — igure payment Network Governance 231

weighs the savings against the loss o

every — uture business relation- ship in the industry, a loss worth multiples o — any single transaction over a career. Barak Richman has documented the vulnerability o — this gov- ernance structure in detail. The DDC’s governance did not erode through legal intervention. It eroded because exogenous market

orces diluted the cost o — exclusion: De Beers shi — ted — rom buyer o —

last resort to aggressive competitor, squeezing midstream margins; Indian diamond merchants entered the market without the inter- generational — amily-business structure that reputation-based coop- eration requires; rough diamond prices rose while internet retail compressed polished prices; and members’ children chose other ca- reers, reducing the intergenerational commitment that sustained re- ciprocal norms.234 DDC membership declined — rom roughly 2,000 to — ewer than 1,200. Governance deteriorated as members could in- creasingly trade outside the DDC’s governance perimeter (an Israeli channel, an Indian network, an internet plat — orm), reducing the cost o — DDC exclusion to the point where the threat lost its — orce. Richman’s account illuminates something essential to the gov- ernance analysis: the DDC’s cooperation equilibrium depended on exclusion remaining costly. When exogenous — orces made exclusion cheaper, compliance declined in proportion. Legal intervention that

urther reduces the cost o — exclusion produces the same result through a di —

erent channel. A market already under pressure — rom Indian entrants and margin compression becomes more — ragile, not less, when a court order converts the expulsion o — a cheater into a procedural dispute lasting years. 232 Law and Governance

Why courts miss the institution The governance institution is invisible to dispute-centered analysis because dispute-centered analysis asks only about parties. Every le- gal proceeding between a governance institution and an expelled or disciplined member — rames itsel — as a two-party dispute: the institu- tion on one side, the expelled member on the other. Legal analysis inquires into what the institution did to the member, what the member is owed, and whether the institution’s procedures were ad- equate. No one in the proceeding represents the remaining mem- bers. No one represents the downstream non-members whose ac- cess to reliable goods or services depends on the network’s govern- ance quality. No one asks what the governance institution’s contin- ued membership community loses when expulsion becomes subject to judicial revision. Lon Fuller identi — ied the structural source o — this — ailure in his account o — polycentric problems.235 Fuller observed that some dis- putes involve such dense networks o — interdependency that a judi- cial resolution a —

ecting any one relationship reverberates through all the others, reaching parties who are not be — ore the court and whose interests cannot be — ully represented in a dispute-centered proceeding. Network governance disputes are polycentric in pre- cisely this sense. Every remaining member’s cooperation depends on the governance institution’s en — orcement credibility. Every downstream participant who relies on the network’s quality- maintenance — unctions has a stake in what happens when the insti- tution disciplines a member. A court adjudicating the relationship between the institution and the expelled member resolves the — rac- tion o — the dispute that has a party to represent it. The governance institution itsel — , the shared en — orcement in — rastructure on which Network Governance 233

every other member’s cooperation depends, has no standing and no voice. Chapter 8 applied the governance method to Silver v. New York Stock Exchange to show how a court can correctly identi — y a speci — ic governance — ailure while simultaneously imposing governance costs

ar exceeding what the identi — ied — ailure required. 236 Chapter 10 traces the sixty-year doctrinal arc that — ollowed. Here the relevant point is what Silver illustrates about the dispute-centered — rame: the Court saw Harold Silver’s injury, exclusion without notice or hear- ing, and — ashioned a remedy calibrated to Silver’s injury rather than to the NYSE’s governance — unction. No part o — the majority opinion asked what the NYSE’s other members would lose as governance production became more costly. Justice Stewart’s dissent identi — ied the problem exactly: “The purpose o — the sel — -regulation provisions o — the Securities Exchange Act was to delegate governmental power to working institutions which would undertake, at their own initi- ative, to en — orce compliance with ethical as well as legal standards in a complex and changing industry. This sel — -initiating process o —

regulation can work e

ectively only i — the process itsel — is allowed to operate — ree — rom a constant threat o — antitrust penalties.” 237 Stewart’s dissent articulates, without the vocabulary this book has developed, the mechanism Chapters 5 through 7 analyze — ormally: governance production costs determine governance output, and le- gal rules that raise those costs reduce governance quality, imposing costs on members who have no voice in the proceedings that impose them. 234 Law and Governance

The governance institution as a club good Buchanan’s theory o — club goods provides the — ramework — or under- standing what the network’s exclusion authority produces and why judicial intervention degrades it. 238 Club goods are nonrivalrous among members, meaning one member’s reliance on the shared en-

orcement system does not diminish another member’s ability to rely on it, and excludable — rom rule-breakers. Excludability is not incidental to the club good. Excludability is the — eature that separates a club good — rom a public good and prevents — ree-riding — rom de- stroying the good’s provision. Chapters 5 and 6 developed this analysis in — ull. What Chapter 5 established applies here with precision: the credible threat o — ex- clusion in a collectively governed trading network maps exactly onto Buchanan’s de — inition. When one dealer relies on the knowledge that cheaters — ace expulsion — rom the bourse, that reli- ance does not reduce any other dealer’s ability to rely on the same knowledge. Each en — orcement action strengthens every member’s con — idence in the threat, because each action con — irms the network will — ollow through. Expelled members lose the bene — it. A dealer re- moved — rom the bourse cannot invoke the expulsion threat to make counterparties extend credit, because the expelled dealer is no longer part o — the community that en — orces the norm. Remove excludability by allowing expelled members to remain inside the governance system through judicial order, procedural de- lay, or compensated reinstatement, and the club good degrades into a public good. Olson’s account o —

ree-riding then applies: when the bene — it o — the governance system is available to members who vio- late the rules, the incentive to comply diminishes.239 Bohnet, Frey, and Huck demonstrated experimentally that medium-level external en — orcement crowds out cooperative behavior more e —

ectively than Network Governance 235

either no en

orcement or total en — orcement: when participants can- not tell whether cooperation or external authority is sustaining the relationship, voluntary compliance erodes.240 Partial judicial inter- vention, including reviewing expulsions one at a time, staying en-

orcement during appeal, and imposing procedural requirements case by case, produces precisely the medium-level en — orcement that the experimental literature identi — ies as most destructive to cooper- ative norms. Ali and Miller’s — ormal analysis adds a dimension that Chapter 7 introduced and that bears directly on what courts destroy when they override expulsion decisions.241 Ali and Miller proved that perma- nent ostracism is sel — -de — eating in networks with strategic commu- nication, because once a member knows a cheater will be perma- nently expelled, the member loses — orward-looking incentives to re- port cheating accurately. Temporary ostracism with the possibility o —

orgiveness solves this problem. But Ali and Miller’s result de- pends on the network retaining unilateral control over calibration: who gets expelled, — or how long, and under what conditions read- mission becomes possible. When a court orders reinstatement, it does not merely override one expulsion decision. It removes the network’s calibration authority: the ability to distinguish deliberate

rom inadvertent breach, to modulate sanctions by severity, and to condition readmission on demonstrated re — orm. A court applying general principles o —

airness sees two — acially similar cases o — non- per — ormance. The network sees members with di —

erent histories, di —

erent stakes, and di —

erent possibilities — or redemption. Calibra- tion is what Ali and Miller showed makes the exclusion system e — -

ective, and calibration requires exactly the contextual institutional knowledge that courts do not have and that the governance institu- tion does. 236 Law and Governance

Three doctrines that miss the governance institution Courts encounter the missing-institution problem across at least three doctrinal contexts: contract law’s incorporation o — trade usage, antitrust law’s treatment o — cooperative exclusion, and association law’s review o — membership decisions. Each involves a distinct legal mechanism. Each produces the same structural error: the court an- alyzes what the expelled or disciplined party lost without analyzing what the governance institution and its remaining members lost.

Contract law: trade usage and the severance o

norms

rom en — orcement UCC Section 1-303(c) permits courts to supplement written con- tract terms with trade usages, meaning customs so widely — ollowed in an industry that parties are assumed to have incorporated them.242 When a trade usage is actually a governance norm, a standard en-

orced through collective expulsion rather than individual lawsuits, treating the norm as an implied contract term severs it — rom the en-

orcement mechanism that gave it meaning. Bernstein’s empirical — indings illuminate the gap between how norms — unction within governance systems and what happens when courts receive them.243 In the diamond industry, customs governing payment terms, inspection procedures, and dispute resolution oper- ate within a sel — -contained governance system backed by mandatory arbitration and the expulsion threat, not as implied terms en — orcea- ble in civil litigation. In the cotton industry, Bernstein — ound that the norms governing actual behavior within the trading network di —

er materially — rom the norms courts would recognize as trade us- ages. Her distinction between relationship-preserving norms and Network Governance 237

endgame norms captures the

unctional di —

erence precisely. Dia- mond dealers — ollow relationship-preserving norms in ongoing trading: pay promptly, resolve disputes through the bourse, accept the arbitrator’s judgment. Those norms — unction because dealers ex- pect to keep trading and because the expulsion threat makes com- pliance rational. When a court picks up those norms and treats them as implied contract terms, it applies them as endgame norms: the relationship is already over, the parties are in litigation, and the court uses the norms to calculate damages in a proceeding the gov- ernance system was designed to prevent. A court asked to identi — y the trade usage governing diamond payment terms will state a rule precise enough to generate a dam- ages calculation: payment is due within — orty- — ive days o — delivery. The actual norm is more contextual: payment is expected promptly; the community tolerates short delays — or good reasons but not — or bad ones; and the community, not a — ixed rule, determines what counts as a good reason. Judicial restatement will be accurate in sur-

ace content and wrong in operative — unction, because it strips the norm o — the — lexibility and collective en — orcement that made it e — -

ective. Barrow-Shaver Resources Co. v. Carrizo Oil & Gas, Inc. illus- trated the problem concretely: the Texas Supreme Court permitted a jury to decide whether oil and gas industry custom incorporated a reasonableness standard into an express consent clause, with — ive separate amici — iling brie — s asserting competing versions o — industry norms.244 What had been a contextual community judgment became a question — or twelve strangers to resolve on the basis o — competing expert testimony. Juridi — ication does not distort the norm inci- dentally. It is the distortion: collective en — orcement is replaced by judicial en — orcement, and the two mechanisms are not substitutes. 238 Law and Governance

 The Federal Arbitration Act's presumption o ---  arbitrability works in the opposite direction and illustrates the enabling mecha- nism by contrast. By en --- orcing arbitration clauses and severely lim- iting judicial review o ---  arbitral awards, the FAA reduces the cost o ---

maintaining private dispute resolution systems.245 Without it, par- ties could agree to arbitrate and then litigate when the arbitral out- come proved un — avorable, — ree-riding on the arbitration system’s in- vestment in in — rastructure, expertise, and procedural — airness while re — using to be bound by its results. Arbitration systems are govern- ance club goods: the FAA’s presumption o — arbitrability is a Pigou- vian subsidy that reduces the production costs o — private dispute res- olution by eliminating the — ree-rider problem that would otherwise undermine investment in it.246 Courts en — orcing network arbitra- tion agreements and declining to review arbitral outcomes except on narrow grounds are per — orming the governance-enabling — unc- tion that the book’s enabling mechanisms describe: reducing the pri- vate cost o — governance production to bring supply closer to the so- cial optimum.

Antitrust: exclusion as restraint rather than governance Antitrust law treats network exclusion as a potential restraint o —

trade. When a cooperative or trade association expels a member, the expelled party may claim a concerted re — usal to deal under Section 1 o — the Sherman Act. Northwest Wholesale Stationers, Inc. v. Paci — ic Sta- tionery & Printing Co., 472 U.S. 284 (1985), established the current doctrinal — ramework: purchasing cooperative expulsion is evaluated under the rule o — reason rather than the per se rule, with the analysis organized around whether the cooperative possesses market power Network Governance 239

and whether the expelled member lost access to something essential

or e —

ective competition.247 Both the doctrinal — ramework and the underlying — acts reward examination. Northwest Wholesale was a cooperative buying asso- ciation o — independent retail stationery stores in the Paci — ic North- west, pooling purchasing volume to obtain wholesale discounts un- available to individual retailers. Paci — ic Stationery, one member, ac- quired a wholesale operation while remaining a member, in e —

ect competing with the cooperative while bene — iting — rom cooperative prices. Northwest expelled Paci — ic — or — ailing to disclose the owner- ship change in violation o — the cooperative’s bylaws. Paci — ic sued un- der the Sherman Act. The — actual record was sparse: Paci — ic lost ap- proximately $10,000 in annual patronage rebates and access to warehouse — acilities.248 The Court applied the rule o — reason, — ound no market power and no essential — acility, and remanded — or — urther proceedings. What the Court analyzed: whether Paci — ic’s competitive posi- tion was damaged, and i — so, whether the damage was justi — ied under the rule o — reason. What the Court did not analyze: what every re- maining member o — the cooperative lost when its expulsion author- ity became subject to judicial review. The Court acknowledged that “wholesale purchasing cooperatives must establish and en — orce rea- sonable rules in order to — unction e —

ectively,” but treated this only as a reason to apply rule-o — -reason rather than per se analysis, not as a reason to ask whether the en — orcement mechanism itsel — was the governance asset on which every other member depended.249 Proce- dural — ailures by the cooperative may well have been present; the Court observed that “no explanation — or the expulsion was advanced at the time” and that “Paci — ic was given neither notice, a hearing, nor any other opportunity to challenge the decision.”250 Governance 240 Law and Governance

costs imposed on the remaining members were equally present and entirely invisible. The governance method applied to Northwest Wholesale would have organized the analysis di —

erently. The shared problem is re- ducing procurement costs through collective purchasing by inde- pendent retailers who individually lack the volume to obtain whole- sale prices. The governance institution is the cooperative’s member- ship structure, dues obligations, rule- — ollowing requirements, and expulsion mechanism. The legal conditions include the rule o — rea- son’s antitrust space — or cooperative governance, — unctioning as the Pigouvian subsidy that GCG identi — ies, enabling cooperative exclu- sion to exist without per se liability and reducing governance pro- duction costs at the membership boundary.251 Member bene — its are access to prices available only through pooled purchasing, shared warehouse in — rastructure, and the mutual assurance that all mem- bers are bearing the cooperative’s costs rather than — ree-riding. Spillovers extend to retail competition in the Paci — ic Northwest, which bene — its — rom the cooperative’s ability to keep independent retailers viable against chain stores. Legal e —

ects: the Court’s ruling imposed judicial review on the cooperative’s expulsion authority, activating Mechanism 2 — rom Chapter 7, procedural cost increase, without calibrating the intervention to any speci — ic identi — ied gov- ernance — ailure. Paci — ic’s dual wholesale-retail operation was a gen- uine con — lict o — interest that the cooperative’s governance could le- gitimately address. The Court’s — ailure to ask what the remaining members lost and its sparse — actual record le — t the governance costs unexamined and there — ore unmeasured. Evaluated under Step 7: the rule-o — -reason — ramework is correct as a legal structure, because the rule o — reason creates precisely the protected space that governance needs; the Court’s analysis was incomplete because it did not assess Network Governance 241

the governance costs o

subjecting expulsion authority to case-by- case judicial review. Klor’s, Inc. v. Broadway-Hale Stores, Inc., 359 U.S. 207 (1959), illus- trates the distinction the governance — ramework draws between ex- clusion — or governance purposes and exclusion — or competitive pur- poses.252 Broadway-Hale, a department store chain, used its buying power to induce ten national appliance manu — acturers to re — use to sell to Klor’s, a neighboring retailer, or to sell only at discriminatory prices. No governance — unction was at stake: Broadway-Hale was not a cooperative with shared governance obligations; the manu — ac- turers were third parties with no governance relationship to Klor’s; and the purpose o — the exclusion was competitive suppression rather than norm en — orcement. Klor’s was correctly decided as a per se vi- olation. The distinction between Klor’s and Northwest Wholesale is not between per se and rule-o — -reason analysis, though it produces that doctrinal di —

erence. It is between exclusion that serves no gov- ernance — unction, speci — ically withdrawing — rom one competitor what is available to its rivals purely — or competitive advantage, and exclusion that maintains the shared en — orcement in — rastructure on which an entire membership community depends. The governance

ramework makes that distinction analytically tractable where anti- trust doctrine, organized around market power and competitive in- jury, leaves it approximate.

Association law:

air procedure and the displacement o — calibration Courts reviewing membership decisions in voluntary associations, pro — essional societies, and similar organizations apply a standard drawn — rom the common law o — voluntary associations: the decision 242 Law and Governance

must comply with the organization’s own rules, must not be arbi- trary or capricious, and must a —

ord the expelled member procedural

airness: notice, an opportunity to be heard, and punishment pro- portionate to the o —

ense.253 Each requirement is reasonable as a pro- tection against speci — ic governance — ailures. Collectively, they pro- duce a judicial review — ramework that measures governance by whether the expelled member was treated — airly, without asking whether the governance institution’s remaining members were pro- tected. MFS Securities Corp. v. SEC, 380 F.3d 611 (2d Cir. 2004), illustrates the temporal dimension o — governance cost. 254 On February 25, 1998, the NYSE expelled MFS Securities — rom membership on the same day two o — its — loor brokers were arrested — or illegal stock — lip- ping. MFS contested the expulsion through internal NYSE proce- dures, — ederal antitrust litigation, SEC review, and ultimately a pe- tition to the Second Circuit. Final resolution came in August 2004, more than six years a — ter the initial expulsion. Throughout those six years, every other NYSE member observed that an expelled member could contest and delay, pursue multiple procedural tracks simulta- neously, and remain a market participant while challenging its gov- ernance institution’s authority. The Second Circuit ultimately de- nied MFS’s petition on procedural grounds without reaching the merits. No court in those six years asked what the NYSE’s other members lost during the period o — governance uncertainty; every member whose con — idence in the exchange’s disciplinary authority was diminished by the spectacle o — a — irm arrested — or — raud remain- ing in the market while its expulsion wound through successive proceedings. Alpine Securities Corp. v. Financial Industry Regulatory Authority, 121 F.4th 1314 (D.C. Cir. 2024), cert. denied, No. 24-904 (U.S. June Network Governance 243

2, 2025), updates the same story at constitutional scale.255 FINRA in- itiated expedited proceedings against Alpine a — ter — inding it had vi- olated a preexisting cease-and-desist order more than 35,000 times, with documented customer misappropriation totaling $54.5 million and a FINRA hearing panel characterizing the violations as “among the most egregious character — ound in the securities industry.” FINRA sought immediate expulsion; under its rules, expulsion or- ders in expedited proceedings take e —

ect be — ore SEC review. The D.C. Circuit reversed in part, holding that FINRA, as a private en- tity, could not impose expulsion, characterized as e —

ectively irre- versible and a corporate death penalty, without prior governmental superintendence, on private nondelegation grounds. FINRA re- sponded by — iling proposed rule changes to codi — y the SEC-review requirement.256 The constitutional analysis is sound as — ar as it goes: the private nondelegation concern the court identi — ies is genuine, and the SEC review requirement addresses a structural problem with unreview- able private authority. What the opinion does not contain is any analysis o — what Alpine’s continued operation during the pendency o — review means — or FINRA’s governance authority over its other members. Every other FINRA member observes that a — irm with Alpine’s documented record can continue operating while its expul- sion winds through SEC review and — ederal court proceedings. That observation is itsel — governance in — ormation: it tells every other member something about the credibility o — the expulsion threat that sustains their own compliance. The constitutional question whether FINRA’s expedited-expulsion procedure satis — ies due process is dis- tinct — rom the governance question how the answer to that consti- tutional question a —

ects FINRA’s capacity to govern the behavior o —

244 Law and Governance

every other member. The court answered the constitutional ques- tion. Nobody asked the governance question. Higgins v. American Society o — Clinical Pathologists, 51 N.J. 191, 238 A.2d 665 (1968), extended judicial intervention to the reinstatement o — an expelled member over the association’s governance judg- ment.257 The New Jersey Supreme Court ordered the ASCP to rein- state a medical technologist it had expelled, distinguishing between compelling initial admission to a voluntary association (which courts generally will not do) and compelling reinstatement o — a wrong — ully expelled member. The court’s articulation o — why rein- statement rather than damages is the appropriate remedy when a member is wrong — ully expelled captures something the governance analysis con — irms: the member’s interest is relational and positional, and it can only be restored through membership, not monetized through a damages award. What the court did not examine is the inverse problem. When a court orders reinstatement over the asso- ciation’s governance judgment, it does not merely restore one mem- ber’s status. It overrides the association’s assessment o — whether that member’s continued presence is consistent with the governance standards the association en — orces, and it does so with in — ormation in — erior to the association’s. Ali and Miller’s result applies: the asso- ciation’s calibration authority over readmission, precisely what makes graduated sanction systems e —

ective, has been removed by judicial order in the individual case. In none o — these proceedings, MFS Securities, Alpine Securities, or Higgins, did any court ask what the governance institution’s re- maining members lost. The expelled or disciplined party was visible and individually sympathetic; the governance costs were di —

use and belonged to parties who were not in the room. Fuller’s polycentric problem is not a — ailure o — judicial analysis in any particular case. It Network Governance 245

is a structural

eature o — adjudication that the governance method is designed to address.

Plat

orm deactivation and the same mechanism Online plat — orm markets replicate the governance structure o — the trading network in a new institutional setting. A buyer on Amazon Marketplace has never met the seller and has no realistic ability to assess the seller’s reliability be — ore transacting. Deactivation, mean- ing the permanent removal o — a seller, — unctions as the plat — orm’s expulsion mechanism because it is the plat — orm’s credible deactiva- tion threat that sustains buyer con — idence and makes the market possible. Tadelis demonstrated the parallel explicitly: plat — orm — eed- back systems per — orm the same in — ormation-dissemination — unction as the centralized record-keeper Milgrom, North, and Weingast documented at the Champagne Fairs.258 Klonick developed the gov- ernance analysis — or content moderation: digital plat — orms are pri- vate governance systems with their own rules, procedures, and en-

orcement mechanisms, exercising authority over their members’ access to essential digital in — rastructure.259 Keller’s analysis o — the EU’s Digital Services Act identi — ies the governance cost that procedural requirements impose on plat — orm en — orcement with precision.260 DSA Articles 17, 20, and 21 require plat — orms to provide reasons — or moderation decisions, o —

er inter- nal complaint mechanisms, and submit to external dispute resolu- tion. The — unctional parallel to Silver v. NYSE is exact: Silver required the NYSE to provide notice and hearing be — ore excluding a non- member — rom market in — rastructure; the DSA requires plat — orms to provide reasons and appeals be — ore excluding a user — rom digital in-

rastructure. Keller observed that plat — orms responding rationally to 246 Law and Governance

the DSA’s procedural obligations will en

orce less: — or a plat — orm with billions o — moderation decisions annually, the cost o — compli- ance with appeal and explanation requirements makes narrowing the en — orcement scope economically attractive. A plat — orm that en-

orces only violations clear enough to withstand appeal produces less governance, not better governance. Douek has generalized this point: the procedural rights — raming addresses individual — airness in particular deactivation decisions while missing the systemic govern- ance question o — how en — orcement policies a —

ect the entire popula- tion o — plat — orm participants. 261 The governance-cost mechanism that Silver imposed on NYSE en — orcement is operating at scale in the DSA context, be — ore it has been empirically measured. Plat — orm governance di —

ers — rom collective network govern- ance in one important structural respect: plat — orm deactivation is a unilateral commercial decision, not a collective membership deci- sion. A diamond bourse expels a cheater through a collective process in which the bourse’s members participate, and the bourse’s govern- ance incentives align with quality en — orcement because the bourse’s members collectively bear the costs o — bad en — orcement and bene — it

rom good en — orcement. A plat — orm operator deactivates a seller through an internal corporate decision, and the operator’s govern- ance incentives are commercial: deactivation serves buyer con — i- dence and, at some — requency, imposes costs on incorrectly deac- tivated sellers who generate revenue. Where the plat — orm’s govern- ance incentives align with quality en — orcement, the governance analysis applies in the same way it applies to the trading network. Where they diverge, speci — ically where the plat — orm has pro — it mo- tives that con — lict with governance quality, a separate analysis is re- quired, one that accounts — or the plat — orm operator’s dual role as Network Governance 247

governor and pro

it-maximizer and — or the structural di —

erence be- tween collective membership governance and unilateral corporate governance. That analysis belongs to a di —

erent project. What this chapter establishes is the baseline: the Mechanism 2 e —

ect that pro- cedural requirements impose on governance production costs oper- ates identically when applied to plat — orm deactivation systems, and the governance — ramework predicts the DSA’s consequences be — ore they have been — ully documented. The DSA’s procedural — ramework activates the governance cost mechanism identi — ied in Chapter 7 with particular precision. The Act’s requirements—providing reasons — or moderation decisions, establishing internal complaint procedures, and submitting to ex- ternal dispute resolution—directly parallel the procedural obliga- tions that Silver v. NYSE imposed on exchange governance. But where Silver operated at the scale o — one market’s membership deci- sions, the DSA operates at the scale o — billions o — moderation deter- minations annually. Mechanism 2, the procedural cost-increase e — -

ect, predictably intensi — ies as the volume o — decisions subject to procedural requirements grows. A plat — orm operator responding rationally to these requirements will narrow the en — orcement scope—not because the plat — orm becomes more libertarian, but be- cause the cost o — de — ending en — orcement decisions becomes ration- ally insurmountable. The governance degradation pattern that Sil- ver imposed on a single exchange’s credibility now extends to plat-

orm moderation systems across all member states simultaneously. The — ramework predicts these consequences be — ore they have been

ully measured: procedural requirements intended to enhance — air- ness in individual deactivation decisions will produce systematic 248 Law and Governance

underen

orcement that degrades the governance quality — or the en- tire population o — plat — orm participants. This is the same mecha- nism, operating at unprecedented scale.

Governance costs that dispute-centered analysis cannot count The contrast between conventional doctrinal analysis and the seven-step governance evaluation is sharpest in Northwest Wholesale Stationers. Set them side by side. Conventional rule-o — -reason analysis asked: did the cooperative possess market power? Had Paci — ic Stationery lost access to some- thing essential — or e —

ective competition? The Court — ound no mar- ket power and no essential — acility. Paci — ic had lost approximately $10,000 in annual patronage rebates and access to warehouse — acili- ties. Rule o — reason satis — ied; remand on procedural grounds. The analysis concluded when the analysis o — Paci — ic’s injury concluded. The seven-step governance analysis asks seven additional ques- tions, all o — them about parties who were not in the room. Step 1: the shared problem is reducing procurement costs — or independent retailers individually too small to obtain wholesale prices. Step 2: the governance institution is the cooperative’s membership structure, its dues obligations, its rule- — ollowing requirements, and its expul- sion mechanism, speci — ically the mechanism that makes — ree-riding irrational by converting membership into a stake that members hold conditional on compliance. Step 3: the legal conditions enabling that institution include the antitrust rule o — reason itsel — , which — unc- tions as the Pigouvian subsidy giving cooperative governance pro- tected space to exist without per se liability. Steps 4–5: member ben- Network Governance 249

e

its are the below-wholesale pricing available only through collec- tive purchasing; spillovers extend to Paci — ic Northwest retail mar- kets, which bene — it — rom independent retailers remaining viable against chain competitors. Step 6: the Court’s holding subjected the cooperative’s expulsion authority to case-by-case judicial review without calibrating that review to any speci — ic identi — ied govern- ance — ailure, thereby activating the procedural cost-increase mecha- nism identi — ied in Chapter 7. Step 7 requires evaluating calibration along both dimensions Chapter 7 disaggregates. On scope, the Court’s holding extended judicial review to the cooperative’s expul- sion authority generally, when the identi — ied problem was one member’s undisclosed con — lict o — interest — the remedy reached be- yond the governance — unction that — ailed (membership screening — or con — licts) to the entire sanctioning — unction. On intensity, the Court’s sparse — actual record le — t the governance cost unmeasured: Paci — ic’s $10,000 in lost rebates was quanti — ied, but the degraded credibility o — every subsequent expulsion threat was not. The rule- o — -reason — ramework is doctrinally correct as a structure; the Court’s application — ailed both calibration dimensions because it assessed Paci — ic’s competitive injury without measuring the governance costs imposed on every remaining member. The divergence is not about who wins. Paci — ic Stationery may well have deserved to lose on the merits. The divergence is about what legal analysis counts as a cost. Conventional analysis counted Paci — ic’s $10,000 in lost rebates. The governance analysis also counts the degraded credibility o — every subsequent expulsion threat the co- operative can make, the increased cost o — en — orcing membership rules against every — uture member who knows the outcome is judi- cially contestable, and the diminished cooperative surplus available to the hundreds o — remaining members whose compliance depends 250 Law and Governance

on the en

orcement mechanism remaining credible. These costs be- long to parties who were not be — ore the court and cannot be counted without a — ramework that identi — ies them as costs in the — irst place. Seven steps applied to the DDC, the Northwest Wholesale co- operative, and FINRA’s disciplinary system produce governance evaluations that can be stated precisely, even though no court in any o — these proceedings per — ormed them. In each setting, the shared problem shares a common structure: reducing the cost o — exchange among parties who cannot individu- ally monitor all their counterparties, by building collective en — orce- ment in — rastructure that makes de — ection irrational. Each govern- ance institution has all — our required elements in — unctional — orm, measurable through the output-based criteria Chapter 8 speci — ies. Monitoring is — unctioning when it generates conduct-governing in-

ormation: the DDC’s reputation network produces exactly this — Bernstein documented that compliance histories circulate through the bourse with su —

icient speed and accuracy that members extend unsecured credit to counterparties they have never met, relying on the network’s in — ormational output rather than individual due dili- gence. Sanctioning is — unctioning when sanctions are applied to ac- tual violations at a — requency consistent with the institution’s ob- servable violation rate: the DDC’s arbitration system resolves the overwhelming majority o — disputes internally, and expulsion deci- sions are executed within days o — a — inding. Decision-making is

unctioning when institutional decisions govern actual member be- havior: DDC trading rules — the — orty- — ive-day oral credit terms, the handshake deal conventions, the mandatory arbitration require- ment — govern real transactions worth billions annually. Adjust- ment is — unctioning when rules change in response to documented Network Governance 251

changed circumstances: this is the dimension on which DDC gov- ernance — altered as exogenous market pressures intensi — ied, and Richman’s documentation o — the institution’s erosion is, in the

ramework’s terms, a record o — adjustment — ailure preceding the degradation o — the other three elements. FINRA’s regulatory appa- ratus satis — ies all — our criteria through more — ormalized mechanisms; the Northwest Wholesale cooperative satis — ies them through less

ormalized ones. The legal conditions enabling these institutions are primarily the antitrust rule o — reason, which creates protected space

or cooperative governance; the FAA’s arbitration policy, which en-

orces the private dispute resolution mechanisms that networks use; and the common law o — voluntary associations, which a —

ords pre- sumptive de — erence to membership decisions. Member bene — its are identi — iable and in some settings quanti — ied: the DDC’s two thou- sand members conducted billions o — dollars o — transactions annually on handshake terms, with disputes resolved in days rather than years o — litigation, through a reputation system that permitted unsecured credit at scales — ormal markets would not support. Spillovers extend to downstream market participants through speci — ic governance di- mensions: monitoring produces supply chain reliability (down- stream retailers and consumers rely on the DDC’s quality-veri — ica- tion outputs); sanctioning produces market integrity (the credible expulsion threat deters — raud that would otherwise impose costs on every participant in the distribution chain); and decision-making produces coordination bene — its (standardized trading conventions reduce transaction costs — or parties who interact with DDC mem- bers but are not themselves members). The DDC’s governance ero- sion under market pressure illustrates the dimensional speci — icity o —

spillover degradation: as adjustment

ailed — irst, the institution lost 252 Law and Governance

the capacity to respond to changing market conditions, which de- graded monitoring outputs (reputational in — ormation became less reliable as the membership shrank and trading moved outside the bourse), which in turn reduced the supply chain reliability on which downstream participants depended. The legal rules courts applied in these settings activated degra- dation mechanisms — rom Chapter 7. UCC Section 1-303’s trade us- age doctrine applied to governance norms activates Mechanism 3: it juridi — ies contextual norms by importing the substance o — in — ormal governance rules into — ormal adjudication while stripping them o —

the collective en

orcement that made them operative.262 Judicial re- view o — expulsion decisions activates Mechanisms 1 and 2: Mecha- nism 1 because the possibility o — judicial reinstatement converts ex- pulsion — rom a property-rule entitlement, under which only the net- work can revoke membership, to a liability-rule remedy in which a court determines whether membership should be restored at what amounts to a judicially set price; Mechanism 2 because procedural requirements imposed on expulsion decisions raise the cost o — exer- cising the sanction past the point where the institution can absorb the cost, deterring en — orcement and degrading the cooperation equilibrium. The DSA’s plat — orm governance requirements activate Mechanism 2 on a scale proportional to the volume o — decisions plat — orms must explain and de — end. None o — these governance costs appeared in any judicial opinion in any o — these cases. Courts resolving disputes between governance institutions and expelled members proceeded without any — rame- work — or identi — ying or measuring those costs. The governance method does not claim that every judicial intervention imposing these costs is wrongly decided. It claims that decisions made without measuring those costs cannot accurately evaluate what the legal rule Network Governance 253

does to governance, and that the cost o

that analytical gap is borne by every member o — the governance institution who was not present when the court resolved the dispute-centered controversy. The in — ormational demands o — the governance evaluation di —

er between the litigation settings in which these disputes arise and the legislative or regulatory settings in which governance policy is de- signed. In litigation, where the method’s in — ormational constraints are most acute, Steps 1 through 3 and Step 6 produce precise — ind- ings: the shared problem, the governance institution’s structure, whether governance is — unctioning under the output-based criteria, and the mechanism through which the legal rule acts on governance are all determinable through the institutional analysis courts rou- tinely per — orm. Steps 4 and 5, which require evaluating member bene — its and spillovers, are more di —

icult: no party in an expulsion dispute has incentives to produce spillover evidence with the rigor the method ideally demands. In Northwest Wholesale, the Court’s sparse — actual record le — t not only Paci — ic’s competitive injury poorly developed but the cooperative’s governance value entirely unex- amined. The method’s structural contribution in litigation — or- ganizing the governance inquiry so that courts ask the right ques- tions about institutional e —

ects rather than only dispute-level e —

ects — is available even when quantitative precision at Steps 4 and 5 is not. In legislative settings, by contrast, the method’s — ull quantitative potential is available: a legislature considering whether to extend rule-o — -reason protection to a new category o — cooperative govern- ance can commission the empirical assessment o — spillovers that no litigant will produce. Asking the governance questions does not predetermine the an- swers. Be — ore imposing procedural requirements on an expulsion decision, a court can ask whether those requirements are calibrated 254 Law and Governance

to an identi

ied governance — ailure or imposed categorically. Be — ore awarding expectation damages to a promisee — or breach within a governed trading network, a court can ask whether the party-cen- tered remedy adequately captures the network-level harm or whether the remedial design requires recognizing the governance institution as an injured party. Be — ore reviewing a cooperative’s ex- pulsion under the antitrust rule o — reason, a court can ask not only whether the expelled member lost something competitively but whether the expulsion served the governance — unction on which every remaining member depends. The strongest version o — the objection to this chapter’s analysis does not dispute the existence o — governance costs. It argues that the governance — ramework systematically privileges institutional insid- ers by converting every judicial protection o — excluded individuals into a “governance cost,” thereby making it analytically impossible

or any legal intervention to be justi — ied. I — procedural requirements are a cost, and judicial review is a cost, and reinstatement is a cost, then the — ramework treats every constraint on institutional power as a harm — which is precisely what power — ul institutions claim and precisely what courts imposing those constraints are designed to re- sist. The objection deserves its — ull weight. The response is that the governance — ramework does not treat every governance cost as dis- positive; it treats every governance cost as countable. The current analytical de — ault, which counts party-centered injury to the ex- cluded party and counts zero governance costs to remaining mem- bers, is not a neutral baseline. It is a systematic undercount o — one category o — real costs. A — ramework that makes those costs visible does not automatically — avor the institution; it enables a more com- plete accounting in which both the excluded party’s injury and the remaining membership’s governance loss appear on the same Network Governance 255

ledger. Courts that see both costs may still conclude that procedural protection, judicial review, or reinstatement is justi — ied — particu- larly in mandatory membership institutions or institutions with market power, where the — ramework’s own criteria call — or height- ened scrutiny. What the — ramework prevents is reaching that con- clusion without knowing what it costs. Chapter 10 turns to the setting where that — ailure has the longest documented history and the largest available empirical record: the sixty years o — exchange governance law that began when the Su- preme Court in 1963 resolved Harold Silver’s dispute-centered case without asking what the NYSE’s governance institution would cost to repair. Chapter 10: Exchange Governance and the Silver Arc Chapter 8 analyzed the 1963 — acts o — Silver v. New York Stock Exchange as the governance method’s worked example, identi — ying the NYSE’s unilateral wire-termination decision as a — ailure to provide the procedural mechanism that Silver needed to contest the expul- sion, and identi — ying the Supreme Court’s remedy as a property-to- liability-rule conversion that imposed governance costs that a party- centered analysis could not measure.263 The analysis stopped in 1963. This chapter begins in 1964 and does not stop until 2025. The sixty-year arc — rom Silver through the 1975 Amendments, through the crystallization o — absolute SRO immunity, through de- mutualization and the regulatory-commercial bi — urcation, and into the current constitutional crisis is the most — ully documented case study available o — law repeatedly acting on a speci — ic governance in- stitution. Exchange governance has attracted sustained scholarly at- tention across sixty years, generating detailed doctrinal histories, empirical accounts o — governance behavior, and theoretical — rame- works — or understanding what securities exchanges actually do.264 None o — that scholarship has read the arc through the governance

ramework this book develops. The governance reading produces a 258 Law and Governance

narrative that is both more coherent and more troubling than the standard account. The standard account reads the immunity arc as a story about implied antitrust repeal, an ongoing judicial e —

ort to reconcile the Sherman Act’s hostility to private restraints on competition with the Exchange Act’s authorization o — private regulatory cartel activ- ity. That — raming is not wrong, but it is incomplete. Read through the governance — ramework, the arc is a story about law repeatedly recalibrating its relationship to a speci — ic governance institution: de- grading that institution’s capacity in 1963, partially restoring it through alternative accountability mechanisms in 1975, completing the restoration through absolute immunity doctrine in the 1990s and 2000s, and then con — ronting the structural change demutualiza- tion introduced, without adequate analytical tools, when it trans-

ormed governance institutions into pro — it-seeking commercial ac- tors. The governance — ramework explains why each major doctrinal development was the legal system’s response to observed govern- ance consequences, whether or not the courts used that vocabulary. Exchange Governance 259

Chapter 10. Exchange Governance The Supreme Court’s 1963 decision in Silver v. New York Stock Ex- change recognized that exchange sel — -regulation served — unctions that antitrust law alone could not replicate, but the Court declined to speci — y how — ar that recognition should extend. The six decades that — ollowed produced a sequence o — legislative, regulatory, and ju- dicial developments that progressively trans — ormed the exchange’s governance architecture — each responding to a speci — ic problem, none guided by a systematic account o — what exchange governance produces and what legal intervention costs. Tracing that arc through the governance — ramework reveals a pattern o — institutional consequences that the standard doctrinal narrative obscures.

What Silver le

t unresolved The Supreme Court’s 1963 decision in Silver was deliberately nar- row.265 The Court held that the NYSE had exceeded the legitimate scope o — its sel — -regulatory authority by acting without providing Silver notice and an opportunity to be heard. The holding identi — ied a speci — ic governance — ailure (the absence o — a procedural mecha- nism — or non-member access decisions) and imposed a speci — ic rem- edy (antitrust liability as the accountability check when adequate procedures were absent). What the holding did not provide was a

ramework — or governance conduct that was procedurally adequate. Justice Stewart’s dissent captured the problem precisely: the major- ity had decided that the Exchange could not act as it had acted, but had not decided what the Exchange could do instead.266 Every exchange, every trading network, and every pro — essional association that exercised governance authority over market partic- 260 Law and Governance

ipants

aced the same question a — ter Silver: what procedural require- ments now apply to our disciplinary decisions, and what antitrust liability do we — ace i — we discipline a member or deny access to a non-member? The Court had established that the Exchange Act did not create blanket antitrust immunity, that some accountability mechanism was required, and that the courts could provide that ac- countability when the Exchange — ailed to do so internally. It had not established what procedures would be adequate, what standards would govern substantive review o — exchange governance decisions, or whether Congress or the SEC, rather than the antitrust courts, was the proper institution — or providing ongoing governance over- sight. The governance consequences o — this doctrinal vacuum were immediate and operating through exactly the mechanisms Chapter 7 identi — ies. Mechanism 2 (increased procedural costs reducing gov- ernance production) operated through the litigation risk Silver cre- ated. An exchange con — ronting a disciplinary decision had to assess whether its procedures were adequate to de — eat antitrust liability, with no authoritative guidance on what adequacy meant. Mecha- nism 1 (property-rule to liability-rule conversion) operated struc- turally: the categorical character o — the exchange’s expulsion author- ity had been replaced by a regime in which the antitrust courts could review whether any given expulsion decision was justi — ied and pro- vide a remedy i — they — ound it was not. The exchange retained gov- ernance authority in — orm but exercised it against a background o —

potential judicial override. The accommodation threshold implicit in the Court’s holding warrants precise speci — ication because it determines the boundary between governance discipline and governance degradation in the exchange context. The Court required “notice and an opportunity Exchange Governance 261

to be heard” — a procedural minimum that the NYSE could have satis — ied without converting its exclusion authority — rom a property rule to a liability rule.267 An internal hearing at which the a —

ected party received reasons — or the adverse decision, conducted under procedures subject to SEC review under the Exchange Act’s existing supervisory — ramework, would have addressed the Court’s concern about unchecked authority over non-members while preserving the categorical character o — the exchange’s disciplinary power over members. That accommodation threshold — internal process with stated reasons, subject to regulatory review, but without exposure to antitrust treble-damages liability — is precisely what the 1975 Amendments later codi — ied. Section 19(d) required SROs to provide

air procedures in disciplinary proceedings and subjected those pro- ceedings to SEC review on the merits.268 The Amendments recog- nized that the accommodation the Court sought in 1963 could be provided through regulatory architecture rather than through anti- trust liability. The twelve-year gap between Silver and the 1975 Amendments is the period during which exchange governance op- erated under the wrong accountability mechanism: antitrust liability imposed governance costs disproportionate to the procedural — ail- ure the Court identi — ied, because the accommodation threshold could have been satis — ied by a — ar less costly institutional design. Commissioner Loomis stated the point explicitly in November 1975: the Senate Committee intended SEC oversight to be “more

ormal and pervasive,” substituting structured regulatory accounta- bility — or the blunt instrument o — antitrust litigation.269 Paul Mahoney’s historical study o — exchange governance identi-

ies the Silver period as a doctrinal turning point — rom the private contract model o — exchange regulation toward what Douglas Mi- 262 Law and Governance

chael would later call “audited sel

-regulation”: governance author- ity exercised privately under governmental supervision rather than autonomously under private contract rights.270 Mahoney’s account supports the governance — ramework’s diagnosis: the private con- tract model, in which exchange governance rested on property rights and reputational sanctions without governmental oversight, could not survive judicial imposition o — antitrust liability. The ques- tion was what would replace it.

The 1975 Amendments as governance restoration Congress answered that question twelve years a — ter Silver with the Securities Acts Amendments o — 1975.271 The Amendments substan- tially restructured the Exchange Act’s — ramework — or SRO oversight in ways that existing scholarship has analyzed primarily as regula- tory expansion. The governance reading identi — ies what the Amendments accomplished in governance terms: they substituted a di —

erent accountability mechanism — or the antitrust check that Sil- ver had imposed, restoring governance capacity by providing the in- stitutional — ramework that Silver had — ound missing. Governance restoration centered on procedural codi — ication that resolved the uncertainty Silver had created. Section 19(b), as substantially amended, required SROs to — ile proposed rule changes with the SEC and obtain Commission approval be — ore implementa- tion.272 Section 19(d), newly enacted, required SROs to — ile notice o —

disciplinary sanctions and subjected those sanctions to SEC re- view.273 Section 19(e) authorized the SEC to abrogate, add to, or de- lete SRO rules it — ound inconsistent with the requirements o — the Exchange Governance 263

Exchange Act.274 The 1975 Amendments trans

ormed the SEC’s re- lationship to SRO governance — rom occasional supervision to struc- tured, mandatory, pre-en — orcement review. Be — ore the Amendments, SEC oversight o — exchange govern- ance was intermittent and largely reactive. The Commission had au- thority under § 19(b) o — the original 1934 Act to require rule changes, but the Exchange could implement rules without advance Commis- sion review. The Silver Court had noted that the SEC lacked author- ity to review the NYSE’s speci — ic wire-termination decision, which was why the antitrust courts had to step in. 275 The 1975 Amend- ments closed precisely that gap. Post-1975, exchange rule changes, including disciplinary procedures, required Commission approval be — ore taking e —

ect. Post-1975, disciplinary sanctions were subject to SEC review. The accountability mechanism that Silver had sought through antitrust courts was now built into the regulatory structure itsel — . The doctrinal consequence emerged in the Supreme Court’s de- cision in Gordon v. New York Stock Exchange, Inc., decided ten weeks a — ter the Amendments’ statutory enactment and twenty-two days a — ter the Amendments’ signed date.276 The timing is crucial — or un- derstanding what Gordon did and did not hold. Gordon had been lit- igated and argued under the pre-amendment — ramework: the case was argued in March 1975, be — ore the Amendments were enacted in June. The Court’s June 26, 1975 decision was there — ore rendered in a transitional moment — the Amendments were now statutory law, but the case had been brie — ed, argued, and was being decided on the basis o — pre-amendment legal doctrine. The Court’s antitrust im- munity holding rested explicitly on pre-amendment § 19(b)’s “active and ongoing” SEC supervisory authority over commission rates, not 264 Law and Governance

on the new Amendments themselves.277 The NYSE’s

ixed commis- sion schedule had been actively supervised, reviewed, and eventually required to be competitive by SEC order; that active supervision sat- is — ied the Silver requirement — or an accountability mechanism other than antitrust liability, and there — ore immunized the rates — rom an- titrust challenge.278 Gordon proves the Amendments were not themselves the source o — immunity because active SEC supervision predated them. Immunity — ollowed — rom the pre-amendment — ramework’s struc- tured oversight. The Amendments, which had been signed into law just be — ore Gordon was decided, provided the statutory — oundation that would make such active supervision a mandatory institutional

eature going — orward, but the Court’s holding rested on authority that had existed be — ore the Amendments’ passage. The Amendments were already in — orce when the decision issued, and the Court ob- served their signi — icance. But Gordon’s holding established the prin- ciple that the pre-amendment — ramework o — active SEC oversight was su —

icient to immunize exchange governance decisions — rom an- titrust attack, a principle the Amendments would make easier to sat- is — y going — orward, not because the Amendments created new im- munity but because they made comprehensive SEC oversight o —

SRO governance a mandatory institutional

eature rather than a contingent regulatory practice. The 1975 Amendments achieved governance restoration through substitution o — accountability mechanisms. The standard account, endorsed across the SRO immunity literature — rom Ma- honey through Edwards, characterizes the Amendments as expand- ing regulatory oversight, with courts subsequently using the ex- panded — ramework as a basis — or broader immunity protection. 279 Exchange Governance 265

That characterization is accurate but analytically incomplete. Ex- panded SEC oversight and broader immunity protection are, in gov- ernance terms, a single mechanism operating in two phases. Phase one: the Amendments created the institutional conditions — or ade- quate governance accountability: mandatory rule — iling, mandatory pre-approval, mandatory disciplinary review. Phase two: once those conditions were satis — ied, antitrust courts could no longer provide the marginal governance improvement Silver had sought through li- ability exposure, because the SEC oversight — ramework already pro- vided the accountability mechanism Silver required. The Amend- ments did not grant immunity. They created the conditions under which existing immunity doctrine applied. The Amendments achieved governance restoration through substitution o — accountability mechanisms, the enabling pattern Chapter 7 identi — ies: law restores governance capacity by providing an alternative institutional structure that addresses an identi — ied

ailure. Silver had imposed antitrust liability as the accountability mechanism — or exchange governance decisions that lacked adequate procedural sa — eguards. The 1975 Amendments provided compre- hensive procedural sa — eguards through the SEC oversight — rame- work. Once those sa — eguards were in place, the justi — ication — or an- titrust liability as an additional governance check was substantially undermined: the governance — ailure Silver identi — ied had been rem- edied through a di —

erent institutional design. Evaluated through the disaggregated calibration — ramework de- veloped in Chapter 7, the 1975 Amendments achieved scope calibra- tion — or some governance — unctions — the § 19(b) rule- — iling re- quirement targeted the speci — ic governance activity (rulemaking) where regulatory accountability was needed — but created intensity problems o — their own through the procedural ossi — ication that 266 Law and Governance

comprehensive pre-approval requirements impose on governance production. The scope dimension was calibrated because the Amendments addressed the speci — ic — ailure Silver had identi — ied (in- adequate procedures — or consequential governance decisions) rather than imposing accountability mechanisms across all exchange con- duct. The intensity dimension was less certain: procedural ossi — ica- tion — rom mandatory pre-approval requirements can itsel — degrade governance by slowing the institution’s ability to respond to emerg- ing regulatory problems. But the Amendments’ intensity problem was minor compared to Silver’s intensity — ailure. Silver had imposed disproportionately high governance costs — treble-damages expo- sure and antitrust liability — — or a narrow remedial purpose (requir- ing notice and hearing on membership decisions). The 1975 Amendments imposed signi — icant but proportionate costs by re- quiring pre-approval procedures that directly addressed the identi-

ied — ailure. Commissioner Philip Loomis stated the accountability-substi- tution rationale in November 1975, just months a — ter the Amend- ments’ enactment. Addressing a joint con — erence sponsored by the NASD, the Boston Stock Exchange, and the SEC, Loomis reported the Senate Committee’s view that the Commission’s oversight should now be “more — ormal and pervasive” and that SROs were ex- ercising “delegated governmental powers” whose exercise required more accountable procedures.280 Loomis was describing governance architecture, not antitrust doctrine, but the two were connected as the governance — ramework predicts: comprehensive SEC oversight substituted — or private antitrust litigation as the accountability check

or exchange governance, preserving governance capacity while ad- dressing the speci — ic — ailure the Court had identi — ied. Exchange Governance 267

The crystallization o

absolute immunity The shi — t — rom contingent to absolute SRO immunity took two dec- ades a — ter the 1975 Amendments, but the governance logic was es- tablished early. I — exchange governance decisions made under com- prehensive SEC oversight were exempt — rom antitrust liability be- cause the oversight — ramework provided the necessary accountabil- ity, the same logic extended to securities — raud and other private- right-o — -action claims. An SRO per — orming regulatory — unctions under SEC supervision did not merely escape antitrust liability; it deserved the — unctional immunity that any governmental entity ex- ercising delegated regulatory authority would receive. Austin Municipal Securities, Inc. v. National Association o — Securities Dealers, Inc. crystallized the — irst phase o — absolute immunity in 1985.281 The Fi — th Circuit held that the NASD was “entitled to abso- lute immunity — or its role in disciplining its members and associates” when per — orming regulatory — unctions. The court’s reasoning com- bined the Silver — ramework with the 1975 Amendments’ institu- tional structure: Congress had delegated en — orcement authority to the NASD; that delegation carried governmental immunity; a pri- vate damages action against the NASD — or its regulatory conduct would undermine the congressional design as surely as an antitrust suit would undermine exchange governance authority under Silver. Austin Municipal made the doctrinal step — rom contingent-immun- ity-when-oversight-is-present to absolute-immunity- — or-regula- tory- — unctions, and the Fi — th Circuit’s reasoning held even be — ore the Second Circuit extended it. The Second Circuit’s extension came in 2001 with D’Alessio v. New York Stock Exchange, Inc.282 D’Alessio articulated the doctrine’s modern — orm with a — ormulation that has survived — or more than 268 Law and Governance

two decades: the NYSE “stands in the shoes o

the SEC in interpret- ing the securities laws — or its members and in monitoring compli- ance with those laws” and there — ore deserves “the same immunity enjoyed by the SEC when it is per — orming — unctions delegated to it under the SEC’s broad oversight authority.” 283 The “stands in the shoes” doctrine did two things simultaneously. It grounded SRO im- munity in the delegated governmental authority — ramework that Austin Municipal had introduced. And it extended immunity to all regulatory oversight — unctions, not merely disciplinary proceedings, by treating the SRO’s regulatory role as comprehensive rather than

unction-speci — ic. The governance analysis o — the crystallization period identi — ies the period’s doctrinal contribution as the Pigouvian subsidy e —

ect that Chapter 7 describes — or enabling law. Absolute immunity — rom civil suits reduced the expected private cost o — governance decisions. Every expulsion, every disciplinary sanction, every regulatory de- termination an SRO made without — acing civil liability was a gov- ernance decision made at lower cost than it would have been under a regime o — contingent immunity. Lower governance production costs move governance supply toward the social optimum when governance generates positive externalities, as exchange govern- ance does through each o — its governance dimensions separately: monitoring produces market surveillance outputs that sustain price integrity — or all market participants, not only exchange members; sanctioning produces deterrence against manipulation and — raud that bene — its every investor who relies on market prices; decision- making produces trading rules and listing standards that create the coordination in — rastructure on which non-member market partici- pants depend; and adjustment produces the capacity to respond to new trading technologies, new market structures, and new — orms o —

Exchange Governance 269

misconduct that would otherwise degrade market quality

or all par- ticipants. These spillovers are not — ungible, and legal rules that de- grade one governance dimension may leave others intact — a point the demutualization analysis below illustrates, where commercial incentives compromised the sanctioning — unction while leaving the decision-making and monitoring — unctions partially operational.284 Absolute immunity — or regulatory — unctions is, in governance terms, a Pigouvian subsidy: it reduces private governance produc- tion costs to correct — or the positive externalities that would other- wise cause systematic underproduction. Courts reached this struc- tural position without the vocabulary; the governance — ramework describes what the doctrine was doing. The Second Circuit completed the doctrinal architecture in 2007 with In re NYSE Specialists Securities Litigation.285 The case in- volved allegations that NYSE specialists had misused their privi- leged market position to — ront-run customer orders, conduct that, i —

true, would have constituted securities

raud. The court held that no

raud exception applied to SRO absolute immunity: the NYSE’s spe- cialists were per — orming a regulatory — unction in their capacity as market-makers operating under NYSE oversight, and absolute im- munity extended to “potentially abusive” governance decisions when made in the regulatory capacity. The court’s explanation re- pays care — ul attention: “I — an SRO’s exercise o — a governmental power delegated to it deserves absolute immunity, the SRO’s non- exercise o — that power also entitles it to immunity.”286 NYSE Specialists pushes the Pigouvian subsidy logic to its limit. The subsidy rationale explains why governance decisions made un- der appropriate oversight should not — ace civil liability: liability ex- posure at the margin deters governance production by raising costs, and deterrence is socially costly when governance generates positive 270 Law and Governance

externalities. But the subsidy logic also has a boundary condition: it applies when the governance institution is actually per — orming gov- ernance — unctions, and when the oversight — ramework provides a genuine substitute — or private litigation as an accountability mecha- nism. I — the governance institution’s regulatory decisions are sys- tematically unchecked by the oversight — ramework, the Pigouvian subsidy becomes a subsidy — or governance — ailure rather than gov- ernance production. NYSE Specialists acknowledged that the Mado —

supervision

ailures, where the SEC’s own oversight o — FINRA-ad- jacent regulatory — unctions had produced catastrophic results, com- plicated this story, but the court treated accountability — ailures as ar- guments — or improving SEC oversight rather than restoring private litigation. The governance — ramework identi — ies this as a legitimate but contestable judgment: the SEC oversight — ramework under § 19 provides genuine pre-en — orcement review o — rule-making but less e —

ective post-en — orcement review o — individual disciplinary deci- sions, and the substitution o — SEC oversight — or private litigation is imper — ect in exactly the respects that matter most — or individual members who have been wrongly sanctioned.

Demutualization and the regulatory-commercial distinction The absolute immunity crystallization o — the 1990s and 2000s rested on a governance assumption the Birdthistle, Henderson, and Dom- balagian analysis o — the “Fi — th Branch” identi — ied clearly: SROs were governance institutions per — orming regulatory — unctions under governmental delegation, and their immunity tracked the govern- mental character o — those — unctions.287 Demutualization challenged Exchange Governance 271

that assumption at its structural

oundations. When the NYSE con- verted — rom a member-owned non-pro — it association to a publicly traded — or-pro — it corporation in 2006, when NASDAQ transitioned to public company status, when exchanges became shareholder- driven commercial entities competing — or listing — ees and order

low, the governance institution the immunity doctrine had been built around ceased to exist in its original — orm. The governance problem demutualization created is structural rather than merely behavioral. A member-owned exchange’s gov- ernance incentives align with governance quality because the mem- bers who bear the costs o — bad governance and bene — it — rom good governance are the same parties who control the governance insti- tution. Demutualization severs that alignment. A publicly traded ex- change operator has shareholders whose interests are returns on eq- uity, not governance quality. Those interests overlap when govern- ance quality attracts listings and order — low that generate revenue, the alignment Mahoney described as the original basis — or private exchange governance.288 They diverge when governance quality im- poses compliance costs that reduce margins, when en — orcement de- cisions risk alienating large-volume traders who generate signi — i- cant revenue, or when the exchange’s competitive position in the listings market depends on regulatory leniency that would disad- vantage a member but bene — it the exchange’s market share. A — ter demutualization, the exchange per — orms governance — unctions while simultaneously operating as a pro — it-seeking actor in the mar- kets it governs. Courts responded to this structural change by developing the regulatory-commercial distinction that the governance — ramework predicts must emerge — rom any principled attempt to apply immun- ity doctrine to a trans — ormed institutional structure. In re Facebook, 272 Law and Governance

Inc., IPO Securities and Derivative Litigation drew the distinction’s in- itial boundary: NASDAQ’s decision not to halt trading during the Facebook IPO’s technical mal — unction was a regulatory activity en- titled to absolute immunity, while the design, testing, and promo- tion o — the trading so — tware that mal — unctioned were commercial activities not protected by immunity. 289 The court’s analysis was

unctional: regulatory activities received immunity because they were exercises o — delegated governmental authority; commercial ac- tivities received no immunity because they were exercises o — the ex- change’s authority as a market operator competing — or commercial advantage. The Second Circuit extended and re — ined the distinction in City o — Providence, Rhode Island v. BATS Global Markets, Inc.290 The court held that exchanges were not entitled to absolute immunity — or their provision o — proprietary data — eeds, co-location services, and com- plex order types to high- — requency trading — irms. These products and services did not constitute regulatory activities in the relevant sense: they were commercial o —

erings through which exchanges ex- tracted revenue — rom market participants who purchased access ad- vantages unavailable to ordinary investors. The conduct served the exchange’s commercial interests as a product vendor, not its regula- tory interests as a market overseer. Immunity applied “only to con- duct that constitutes a delegated quasi-governmental prosecutorial, regulatory, or disciplinary — unction,” and selling co-location services to hedge — unds was not such a — unction regardless o — the exchange’s regulatory status in other respects. The governance analysis o — the regulatory-commercial distinc- tion — inds the doctrine structurally correct and practically unwork- able — or connected reasons. Structurally correct: the distinction asks Exchange Governance 273

precisely the right governance question: is this institution per

orm- ing governance — unctions when it does this? Governance analysis in Step 2 o — the seven-step method requires identi — ying the governance institution and con — irming that the challenged conduct — alls within its governance — unction. BATS and Facebook apply this question im- plicitly. An exchange providing co-location services is not govern- ing anything; it is selling proximity. An exchange deciding not to halt trading during a mal — unction is governing market integrity, which is the shared problem the exchange exists to manage. The courts’ — unctional analysis tracks the governance method’s analytical logic even without the governance vocabulary. Practically unworkable: applying a — unctional test case-by-case to an institution that per — orms regulatory and commercial — unctions simultaneously, o — ten through overlapping personnel and institu- tional structures, produces governance uncertainty that is itsel — a governance cost. Standard Investment Chartered, Inc. v. NASD con — irmed that the party asserting immunity bears the burden o —

demonstrating that the challenged conduct constitutes delegated regulatory activity.291 A — ter BATS and Facebook, that burden requires exchange counsel to predict, — or every category o — exchange conduct, whether a court will characterize it as regulatory or commercial, and that prediction requires litigating structural questions about ex- change governance architecture in every individual case. The gov- ernance — ramework predicts exactly the resource allocation the Na — day critique documents: resources that could support substan- tive governance are diverted to de — ensive legal positioning that at- tempts to place conduct on the right side o — a line courts draw un- predictably.292 Na — day’s 2010 account o — the “doctrinal bait-and-switch” in Standard Investment Chartered diagnosed a related problem earlier.293 274 Law and Governance

The Second Circuit in Standard Investment Chartered had held that the NASD’s — ailure to disclose material in — ormation to its members about the merger with NYSE Regulation (the transaction that cre- ated FINRA) was not subject to member challenge because the members lacked standing to challenge SRO rule changes under § 19(b). The Na — day critique is that the court’s standing analysis was correct under the pre-1975 doctrine but was incoherent a — ter the 1975 Amendments, which had created § 19(b) as a mechanism — or member participation in rulemaking precisely to provide accounta- bility — or governance decisions a —

ecting members. The governance

ramework supports and extends the Na — day critique: the 1975 Amendments had built member participation into the accountabil- ity mechanism that justi — ied absolute immunity; cutting members out o — that participation under the standing doctrine created by Standard Investment Chartered removed a critical component o — the SEC-oversight substitute without restoring the antitrust accounta- bility it had replaced.

The constitutional crisis and the governance

ramework’s diagnosis The demutualization-era complications were structural but man- ageable, and courts could continue developing the regulatory-com- mercial distinction through case-by-case adjudication, imper — ectly and at governance cost, but within the existing immunity — rame- work. The challenge that emerged in Alpine Securities Corp. v. FINRA in 2024 is categorically di —

erent in that it attacks the constitutional

oundations o — delegated sel — -regulatory authority itsel — .294 Alpine Securities, a registered broker-dealer, challenged FINRA’s authority to conduct an expedited disciplinary proceeding Exchange Governance 275

that threatened Alpine with e

ective market exclusion be — ore judi- cial review was available. Alpine’s constitutional theory combined two lines o — the Supreme Court’s recent structural constitution ju- risprudence: the Appointments Clause argument that FINRA’s hearing o —

icers were “O —

icers o — the United States” who must be appointed through constitutionally prescribed procedures, and the nondelegation argument that Congress had delegated legislative power to FINRA without an intelligible principle con — ining the del- egation’s scope.295 The D.C. Circuit rejected Alpine’s arguments and upheld FINRA’s authority in a decision that cert was denied in June 2025.296 But the constitutional arguments survived the decision in important respects: the D.C. Circuit’s analysis devoted substantial analysis to why FINRA’s hearing o —

icers were not O —

icers o — the United States, and the nondelegation argument was not de — initively resolved but rather — ound premature given the absence o — a clear Su- preme Court mandate to apply the major questions doctrine to SRO delegations. Benjamin Edwards’s account o — the constitutional threat docu- ments the doctrinal trajectory that made these arguments credible rather than — rivolous.297 The combination o — Lucia v. SEC on the Ap- pointments Clause,298 West Virginia v. EPA on the major questions doctrine,299 and the academic campaign against the administrative state’s structural — oundations had created the intellectual and doctri- nal conditions in which a serious constitutional challenge to SRO authority was not merely available but was the natural next doctri- nal move. The governance — ramework identi — ies this constitutional pressure as a Phase 5 development in the sixty-year arc, a challenge that would not have reached doctrinal credibility without the insti- tutional trans — ormation demutualization had completed. 276 Law and Governance

The governance stakes o ---  Alpine's constitutional arguments, had they prevailed, would have been severe and symmetrical with Silver's 1963 governance degradation. Silver imposed governance costs by exposing expulsion decisions to antitrust liability when procedural sa --- eguards were absent. A success --- ul nondelegation or Appoint- ments Clause challenge to SRO authority would have imposed gov- ernance costs o ---  comparable magnitude: it would have required re- structuring FINRA's hearing o ---

icer appointment process, created uncertainty about which past disciplinary decisions remained valid, and subjected the structural basis o — SRO governance to ongoing constitutional litigation. The MFS Securities timeline documents the governance cost o — regulatory uncertainty in this domain. The 6.5-year interval between the initiation o — disciplinary proceedings and — inal judicial resolution produced documented consequences — or market participants who could not predict whether FINRA’s disci- plinary authority was constitutionally sound.300 FINRA’s post-Alpine response, the — iling o — proposed rule changes in June 2025 that would require SEC approval be — ore ex- pulsion orders take e —

ect, illustrates the accountability-substitution pattern the governance — ramework identi — ies as the recurring adap- tive mechanism in this arc. 301 The 1975 Amendments had substi- tuted SEC oversight — or antitrust accountability as the governance check — or exchange rule-making. The post-Alpine rule changes pro- pose substituting SEC approval — or FINRA-autonomous expulsion authority as the governance check — or the most consequential disci- plinary decisions. In governance terms, the response is internally coherent: reducing the constitutional vulnerability o — SRO author- ity by building in the SEC oversight that makes the delegation ac- countable enough to survive constitutional scrutiny. Whether that substitution succeeds will depend on whether courts — ind that SEC Exchange Governance 277

pre-approval o

expulsion orders adequately addresses the Appoint- ments Clause and nondelegation concerns that Alpine raised.

What the sixty-year arc shows about law and governance The sixty-year arc, read through the governance method, produces a coherent account o — law repeatedly acting on exchange governance institutions through identi — iable mechanisms and in response to ob- served governance consequences. Silver degraded exchange governance capacity through Mecha- nisms 1 and 2 as Chapter 8 documented. Evaluated through the dis- aggregated calibration — ramework, Silver — ailed on both dimensions: on scope, antitrust liability reached across all NYSE disciplinary de- cisions when the identi — ied — ailure concerned a single non-member denied access without process; on intensity, the treble-damages ex- posure imposed governance costs wildly disproportionate to the governance bene — it o — requiring notice and hearing. The 1975 Amendments partially restored governance capacity by substituting SEC oversight — or antitrust liability as the accountability mecha- nism. Evaluated on the same calibration dimensions, the Amend- ments achieved scope calibration — or some governance — unctions — the § 19(b) rule- — iling requirement targeted the speci — ic governance activity (rulemaking) where regulatory accountability was needed — but created intensity problems o — their own through the proce- dural ossi — ication that comprehensive pre-approval requirements impose on governance production. Gordon con — irmed that pre- amendment active SEC supervision had already been providing the 278 Law and Governance

accountability mechanism Silver required, and the Amendments in- stitutionalized and expanded that supervision, making it compre- hensive rather than contingent. The absolute immunity crystallization o — the 1990s and 2000s completed the Pigouvian subsidy — unction that the 1975 governance restoration had begun. Evaluated on calibration dimensions, abso- lute immunity achieved a — orm o — scope calibration — it applied spe- ci — ically to regulatory- — unction governance decisions rather than to all exchange conduct — but arguably — ailed intensity calibration in the opposite direction — rom Silver: where Silver imposed dispropor- tionately high governance costs, absolute immunity may impose disproportionately low accountability costs, shielding governance

ailures that SEC oversight does not e —

ectively catch. The account- ability de — icit that absolute immunity creates (the removal o — private litigation as a check on governance abuse) is partially addressed by SEC oversight under § 19, and the adequacy o — that partial address is the genuinely di —

icult governance question that courts have not di- rectly con — ronted. NYSE Specialists acknowledged the question and de — erred it; the governance — ramework suggests the de — erral is un- justi — ied because the adequacy o — the substitution varies systemati- cally with the type o — governance decision at issue, with individual disciplinary actions receiving substantially less e —

ective SEC over- sight than rule-making. The sixty-year arc thus demonstrates that the scope and intensity dimensions track real doctrinal variation across time: the law has oscillated between interventions that are scope-uncalibrated and intensity-excessive (Silver), scope-calibrated but intensity-uncertain (1975 Amendments), and scope-calibrated but potentially intensity-de — icient (absolute immunity). Demutualization introduced a governance problem the absolute immunity doctrine was not designed to handle: structural con — licts Exchange Governance 279

between commercial interests and governance

unctions within a single institutional actor. The regulatory-commercial distinction the courts developed in Facebook and BATS applies the governance

ramework’s core analytical question (is this institution actually per-

orming governance — unctions?) but applies it case-by-case in cir- cumstances where ex ante institutional design would be more e —

ec- tive. The governance — ramework’s prediction that case-by-case

unctional analysis would impose governance uncertainty costs through de — ensive legal positioning is con — irmed by the post-BATS pattern o — litigation over the regulatory-commercial line. The constitutional challenge in Alpine represents the latest phase o — the same structural dynamic. Each time the law imposes a new accountability mechanism on exchange governance (antitrust liabil- ity in 1963, SEC oversight expansion in 1975, absolute immunity’s boundaries drawn a — ter demutualization), the governance institu- tion adapts and the legal system recalibrates. The post-Alpine FINRA rule changes propose the next calibration: SEC pre-approval o — ex- pulsion orders as the accountability mechanism that resolves the constitutional vulnerability the demutualization-era SRO model created. Whether that calibration restores governance capacity at acceptable accountability cost is the governance question the next decade o — exchange regulation will answer. What the governance — ramework adds to existing SRO scholar- ship is the common vocabulary that explains each doctrinal devel- opment as law acting on a governance institution, enabling or de- grading governance capacity through identi — iable mechanisms, and responding to observed governance consequences whether or not the responding institution used that vocabulary. The implied anti- trust repeal — raming accounts — or Silver and partially — or Gordon. The 280 Law and Governance

Fi

th Branch account explains demutualization’s structural compli- cations.302 Edwards’s constitutional crisis account narrates the Alpine challenge. None o — these accounts, individually or together, explains why the 1975 Amendments produced broader immunity (govern- ance restoration through alternative accountability mechanism sub- stitution), why absolute immunity is enabling rather than merely protective (Pigouvian subsidy — or governance production), or why the regulatory-commercial distinction is structurally correct but practically costly (it applies the right governance question in the wrong institutional context, individual cases rather than ex ante governance design). The exchange governance arc also provides the book’s clearest empirical test o — the governance method’s predictive capacity. The method predicts that legal rules imposing new accountability re- quirements on governance institutions will generate governance costs through Mechanisms 1 and 2; that alternative accountability structures can restore governance capacity when they adequately substitute — or the removed mechanism; that governance production below the social optimum generates pressure — or enabling legal re- sponses; and that structural trans — ormation o — the governance insti- tution creates governance problems that existing doctrine, cali- brated to the pre-trans — ormation institution, cannot adequately handle. Silver, the 1975 Amendments, the immunity crystallization, and demutualization have, in sequence, con — irmed each o — these pre- dictions over sixty years o — traceable doctrinal development. The exchange governance setting also illuminates the govern- ance method’s di —

erential e —

ectiveness in litigation and legislative contexts. The SEC’s regulatory in — rastructure provides precisely the in — ormational investment that Steps 4 and 5 demand: the Commis- sion has access to trading data, market surveillance outputs, and Exchange Governance 281

SRO compliance records that no individual litigant could produce. The 1975 Amendments’ governance restoration succeeded in part because it placed the governance evaluation in a regulatory setting where spillover evidence — market integrity data, price discovery quality, investor con — idence metrics — is systematically collected. Courts adjudicating individual immunity claims under BATS or Fa- cebook lack comparable access to this data, which is why the regula- tory-commercial distinction produces the case-by-case uncertainty the governance — ramework predicts. The exchange governance arc is the book’s strongest illustration o — the principle that the govern- ance method’s — ull quantitative potential is available in regulatory and legislative design settings, while its contribution in litigation is primarily structural. The strongest counterargument this chapter must con — ront is the institutional willingness objection. The governance — rame- work’s pre — erence — or targeted remedies presupposes a governance institution willing to implement them. Silver suggests that govern- ance institutions exercising power over non-members lack that willingness: the NYSE excluded Harold Silver without notice or hearing, and there is no evidence the Exchange would have provided process voluntarily. Categorical procedural requirements were nec- essary because the institution would not correct the — ailure on its own. The objection is strongest in the non-member access context that Silver itsel — presented. Where the governance institution has no ongoing governance relationship with the excluded party — Silver was not an NYSE member — the institution’s sel — -correction incen- tives are weakest, because the institution bears no governance cost

rom the non-member’s exclusion. The — ramework’s pre — erence — or targeted remedies applies with — ull — orce to member discipline, 282 Law and Governance

where the institution has ongoing governance incentives to cali- brate sanctions correctly, and with diminished — orce to non-mem- ber access, where those incentives are absent or attenuated. The proper reading o — Silver, under this analysis, is that the Court cor- rectly identi — ied a governance — ailure (exclusion o — a non-member without process) and correctly imposed a remedy targeted to that

ailure (mandatory process — or non-member access decisions), but that the remedy’s scope extended beyond what the — ailure warranted by reaching member discipline as well, where the willingness objec- tion does not apply and the governance costs o — procedural require- ments are highest. Chapter 11 applies the same governance analysis to business as- sociations, where the governance institution is smaller, the legal

ramework more developed, and the doctrinal mistakes correspond- ingly more visible. Chapter 11: Corporate Governance and the Governance o — Governance Thirty years o — shareholder rights re — orm have produced results the re — orm movement did not predict and cannot explain. Say-on-pay became law in 2011 on the premise that shareholders voting on ex- ecutive compensation would restrain excess. Median S&P 500 CEO compensation stood at $7.5 million that year. By 2024, it reached $17 million, a 127% increase over the re — orm period.303 Say-on-pay

ailure rates never exceeded — our percent in any year — rom 2011 through 2024, meaning shareholders approved essentially every package presented to them — or a vote.304 A mechanism designed to discipline compensation through shareholder voice instead rati — ied compensation growth that dwar — ed pre-re — orm levels. The explana- tion requires a di —

erent theoretical — rame. The club-good — ramework, developed in Chapter 5, supplies that — rame. Corporate governance is a club good: excludable, non- rivalrous up to congestion, and dependent — or its quality on main- taining appropriate membership constraints. The shareholder rights movement committed a category error. It treated a club good as a public good, proceeded on the assumption that more access pro- duces better governance, and — or three decades designed re — orms 284 Law and Governance

around maximizing shareholder participation without regard to whether that participation would improve governance quality or congest it. The result, documented across — our domains, is govern- ance in — orm without governance in — unction. This chapter makes a normative claim, not a descriptive one. Delaware courts do not per — orm governance analysis in the terms this book develops, and nothing here asserts that they do. What this chapter claims is that the club-good analysis predicts the shape o —

Delaware’s review spectrum, and that where Delaware doctrine di- verges — rom that prediction, the divergence identi — ies a re — orm tar- get. Corporate Governance 285

Chapter 11. Corporate Governance Corporate governance scholarship has produced more sustained an- alytical attention than any other governance domain this book ex- amines, yet the — ield’s central debates have largely proceeded with- out the — ramework Chapter 5 develops. Re — raming corporate gov- ernance as a club good — excludable, subject to congestion, and pro- ducing both member bene — its and spillovers — reorients several long-running disputes and identi — ies a recurring analytical error in re — orm proposals that treat governance outputs as simple commod- ities rather than institutional products.

The shareholder rights movement and its category error The shareholder rights movement encompasses a generation o — stat- utory, regulatory, and judicial developments sharing a common ar- chitecture: identi — y a domain o — corporate decision-making previ- ously reserved to the board, and introduce a mechanism — or share- holder participation. Congress enacted say-on-pay through the Dodd-Frank Act o — 2010, requiring public companies to hold advi- sory votes on executive compensation at least once every three years. The SEC’s 2003 mandatory voting rule required registered in- vestment advisers to implement proxy voting policies and to vote client shares in clients’ best interests. Delaware’s Court o — Chancery expanded Caremark — rom Chancellor Allen’s nearly-impossible claim into an active litigation category addressing mission-critical compliance — ailures, and subsequent decisions extended oversight duties to — inancial reporting, aircra — t sa — ety, and human capital man- agement. 286 Law and Governance

Each intervention  --- ollowed the same logic: re --- ormers identi --- ied a domain o ---  corporate decision-making as inadequately governed, and prescribed more participants as the correction. This is the logic o ---  a public good. National de --- ense is the canonical example: non- excludable, nonrivalrous, and subject to chronic undersupply be- cause no private actor can capture su ---

icient bene — it to justi — y the — ull cost o — provision. The standard corrective is mandatory contribu- tion. The shareholder rights movement treated governance identi- cally, as a resource whose quality increases as participation expands. Mandatory voting, compulsory say-on-pay, and expanded oversight duties — unction as the taxation mechanism, — orcing contributions

rom participants who would otherwise — ree-ride. That analogy is backwards. The next section explains why, and uses the explanation to make precise predictions about what three decades o — governance re — orm would produce.

Corporate governance as a club good James Buchanan’s 1965 — ramework identi — ies a category o — goods that — its neither the private good nor the public good model.305 Club goods are excludable (the club can deny access to nonmembers), nonrivalrous up to a threshold o — congestion (multiple members do not diminish each other’s enjoyment below that threshold), and val- uable precisely because membership constraints maintain quality. The canonical example is a swimming pool: members can be ex- cluded through a locked gate, multiple swimmers do not diminish each other’s enjoyment within a reasonable membership range, but crowding beyond optimal size degrades the experience — or everyone. The optimal club size balances the bene — its o — additional members against the congestion costs they impose. Crucially, quality depends Corporate Governance 287

on the club’s ability to manage its own membership. A club that can- not exclude cannot prevent congestion. Corporate governance has all three — eatures. Board deliberation time and director attention are — inite: board governance is rival at the margin, which is to say that adding governance participants be- yond a threshold produces congestion rather than enrichment. Mul- tiple shareholders exercising governance rights simultaneously do not diminish each other’s access to the ballot within a reasonable range. And governance quality depends on the selection and com- position o — the board, which means that expanding access beyond the capacity o — the resulting participants degrades what governance produces. Delaware section 141(a) is the — oundation o — the excludability mechanism: “The business and a —

airs o — every corporation orga- nized under this chapter shall be managed by or under the direction o — a board o — directors.” This provision does not merely authorize board management. It establishes the board as the governance insti- tution o — the corporation and excludes shareholders — rom day-to- day governance. Shareholders elect directors, approve — undamental transactions, and vote on charter amendments, but they do not manage. Judicial de — erence to board decisions, the business judgment rule, is the institutional protection that makes this excludability credible. Without de — erence, every board decision is subject to post- hoc judicial second-guessing, and the board’s management authority becomes hostage to whatever any disappointed shareholder — inds objectionable. Stephen Bainbridge characterizes the rule as an “ab- stention doctrine”: courts simply re — use to review the merits o —

board decisions made in good

aith by disinterested directors with adequate in — ormation.306 Abstention is the mechanism. By declining 288 Law and Governance

to review board decisions on the merits, courts preserve the board’s e —

ective authority to make them. Corporate governance also generates positive externalities be- yond the membership. Employees, creditors, communities, and markets depend on governance quality through channels that do not run through board membership, and those channels trace to speci — ic governance dimensions. Monitoring produces the disclosure out- puts — audited — inancials, material event reports, risk — actor updates — on which stock price discovery depends; the spillover runs — rom monitoring quality to capital allocation e —

iciency across the entire market. Sanctioning produces the deterrence that prevents manage- rial sel — -dealing — rom destroying — irm value on which creditors, em- ployees, and counterparties depend; the spillover runs — rom sanc- tioning credibility to employment stability and supply chain relia- bility. Decision-making produces the strategic coordination that de- termines how — irms deploy capital, hire, and invest; the spillover runs — rom decision-making quality to the broader economic productivity that depends on well-governed — irms allocating re- sources e —

iciently. Adjustment produces the capacity to respond to changed circumstances — new technologies, new competitive threats, regulatory shi — ts — whose bene — its extend to every stake- holder a —

ected by the — irm’s operations. These spillovers are not in- terchangeable: a governance re — orm that degrades monitoring while leaving decision-making intact produces di —

erent externality losses than one that degrades decision-making while leaving monitoring intact. The corporate governance setting also presents distinctive in — ormational advantages — or the governance method: SEC-man- dated disclosure, proxy — ilings, and compensation committee re- ports provide a richer empirical — oundation — or Steps 4 and 5 o — the Corporate Governance 289

governance evaluation than is available in any other governance do- main. Where the method operates in a legislative or regulatory de- sign context — evaluating proposed governance re — orms such as say-on-pay or mandatory voting rules — the quantitative data in-

rastructure corporate law has developed makes the — ull analytical potential o — Steps 4 and 5 available.307 Congestion in a governance club does not resemble congestion in a swimming pool. Governance congestion is a quality phenome- non. When the shareholder base expands to include participants who lack the incentive to govern attentively, governance quality de- clines even i — the — ormal voting procedures continue undisturbed. A board that must justi — y every business decision to an audience o —

passive

und managers — ollowing proxy advisor recommendations, activist hedge — unds seeking short-term extraction, and retail share- holders submitting proposals with three percent support does not govern the same way as a board operating within the protected in- stitutional space that board authority and judicial de — erence create.

The empirical record o

ailure The shareholder rights movement’s empirical record con — irms the governance congestion prediction across — our domains. Say-on-pay. Academic analysis o — say-on-pay’s e —

ects — inds no evidence o — reduced compensation growth. Fisch, Palia, and Solo- mon conclude that say-on-pay votes primarily re — lect shareholder dissatis — action with — irm per — ormance rather than compensation levels, and that the provision has — ailed to constrain CEO pay. 308 Subsequent analysis — inds that say-on-pay compressed compensa- tion toward ISS-approved templates without reducing market me- dian levels, producing standardization rather than restraint. The 290 Law and Governance

structural explanation is straight

orward. Boards with near-certain approval in hand — ace no marginal incentive to constrain packages. Shareholders, con — ronting a binary vote on a package already nego- tiated and presented, have no mechanism to propose alternative structures. Say-on-pay reduced the board’s excludability — rom one governance domain, the setting o — executive compensation, and moved compensation governance — rom a small, expert, in — or- mation-rich group to a large, di —

use, in — ormation-poor group — il- tered through ISS and Glass Lewis criteria. Quality degraded. The proxy advisor duopoly. Institutional Shareholder Ser- vices and Glass Lewis together controlled over ninety percent o — the proxy advisory market by 2024.309 The mandatory institutional vot- ing regime created this concentration. Unable to evaluate each port-

olio company’s proposals independently, institutional investors sought third-party voting recommendations to satis — y their — iduci- ary obligation to vote. Robo-voting, the automatic implementation o — proxy advisor recommendations without independent review, grew — rom seven percent o — ISS customers in 2007 to twenty-three percent by 2021.310 ISS and Glass Lewis are private — irms, answerable to no regulator, subject to no disclosure requirements comparable to those imposed on the public companies they evaluate, and oper- ating under acknowledged con — licts o — interest in markets — or con- sulting services they sell to the companies they rate. The mandatory voting rule was intended to ensure that institutional investors gov- ern port — olio companies in clients’ interests. Instead, it delegated e — -

ective governance authority over public companies to two unac- countable private — irms. The ISS-Glass Lewis duopoly is not a mar- ket — ailure. It is an emergent club structure that — ormed spontane- Corporate Governance 291

ously to manage the congestion the mandatory voting rule pro- duced, capturing authority that belongs to a governance club it did not build. Passive indexing and the output-based — unctionality test. The output-based — unctionality criteria Chapter 8 speci — ies provide a precise diagnostic — or the governance degradation these re — orms produce. Monitoring is — unctioning when it generates conduct-gov- erning in — ormation used in sanctions, rules, or behavior; proxy ad- visor recommendations based on standardized checklists do not generate — irm-speci — ic monitoring outputs that govern the conduct o — the — irms they evaluate. Sanctioning is — unctioning when sanc- tions are applied at a — requency consistent with the observable vio- lation rate; say-on-pay’s sub- — our-percent — ailure rate against 127% compensation growth is prima — acie evidence that the sanctioning mechanism is non- — unctional. Adjustment is — unctioning when rules change in response to documented changed circumstances; the mandatory voting regime has not been re — ormed despite two dec- ades o — documented governance degradation, suggesting the adjust- ment mechanism has itsel —

ailed. By the output-based criteria, shareholder-rights governance is governance in — orm that — ails — unc- tionality at three o —

our dimensions. Index — unds grew — rom approximately three percent o — equity market value in 2000 to over twenty percent by 2020. BlackRock, Vanguard, and State Street together hold between twenty and twenty- — ive percent o — every S&P 500 company. The economic logic o — index investing destroys governance incentive: a — und tracking the S&P 500 cannot sell an underper — orming company without abandoning its index mandate, and because the — und holds every company in the index, the bene — it o — improving governance at any single — irm accrues to the — und only in proportion to that — irm’s 292 Law and Governance

weight in the index. By 2024, the major passive

und complexes had expanded their stewardship teams — rom the skeleton crews o — the prior decade, but the ratio o — pro — essionals to monitored companies remained acute: BlackRock employed approximately sixty- — ive to seventy stewardship pro — essionals, Vanguard approximately sixty, and State Street Global Advisors approximately twelve, together re- sponsible — or thousands o — port — olio companies each.311 Even with expanded sta —

, no team o — that size can deliberate meaning — ully about governance at thousands o —

irms. What they can do is apply standardized checklists and vote in coordinated blocs, which is what they do. The mandatory voting rule — orced participants whose insti- tutional design produces governance indi —

erence to exercise gov- ernance authority. The result is governance in — orm without gov- ernance in — unction. Caremark expansion. Chancellor Allen’s original 1996 — ormu- lation characterized Caremark claims as “possibly the most di —

icult theory in corporation law upon which a plainti —

might hope to win a judgment.” 312 The subsequent decades trans — ormed Caremark

rom a near-impossible claim into an active litigation category. Marchand v. Barnhill in 2019 established that boards o — companies with mission-critical regulatory obligations bear heightened moni- toring obligations — or those speci — ic risks. 313 Later decisions ex- tended the doctrine — urther. In re McDonald’s Corporation Share- holder Derivative Litigation subjected the board’s oversight o — cor- porate culture and human capital management to Caremark liabil- ity, a domain where — irm-speci — ic expertise is most valuable and where standardized judicial criteria are least equipped to substitute

or it.314 Scholars who have studied Caremark’s e —

ect on actual cor- porate compliance — ind no evidence that the expanded doctrine has reduced misconduct. What it has produced is increased compliance Corporate Governance 293

spending, more time devoted to compliance committee meetings, and board attention directed toward litigation-avoidance rather than value creation, the precise pattern Chapter 7 identi — ies as Mechanism 6: compliance substituting — or governance rather than instantiating it. The divergence between standard corporate governance analy- sis and the governance — ramework is not incidental. It is structural. Standard corporate governance analysis since Berle and Means has treated the central problem as one o — agency costs — the divergence between management’s interests and shareholders’ interests — and has accordingly optimized — or shareholder participation as the cor- rective mechanism.315 The governance — ramework identi — ies a di — -

erent central problem: the quality o — the governance output itsel — , measured by whether the — our elements — unction according to the output-based criteria. These two optimization targets produce con-

licting prescriptions in at least three domains where the di —

erence is observable. On say-on-pay, the shareholder-participation model predicted that mandatory voting would constrain excessive com- pensation by giving shareholders a voice; the governance — rame- work predicted that opening a small, expert, in — ormation-rich gov- ernance process to a large, di —

use, in — ormation-poor voting popu- lation would degrade governance quality through the congestion mechanism, and the empirical record con — irms the governance pre- diction. On Caremark expansion, the shareholder-participation model predicted that expanding judicial oversight o — compliance would improve board monitoring; the governance — ramework pre- dicted that expanded oversight would produce compliance substitu- tion under Mechanism 6, and the compliance-spending data con-

irm — ormal investment without measurable misconduct reduction. 294 Law and Governance

On proxy access and mandatory institutional voting, the share- holder-participation model predicted that — orcing institutional in- vestors to vote would produce accountability; the governance

ramework predicted that — orced participation by parties whose in- stitutional design produces governance indi —

erence would produce voting-without-deliberation, and the rise o — the ISS-Glass Lewis proxy advisory duopoly con — irms exactly that pattern. In each do- main, the governance — ramework generates a di —

erent prediction

rom standard analysis, and the available evidence supports the gov- ernance prediction. The analytical di —

erence is that standard corpo- rate governance analysis asks how to maximize shareholder partici- pation, while governance analysis asks how to maximize govern- ance output quality — and the two objectives diverge precisely where participation degrades quality.

Delaware’s review spectrum as a governance- calibration instrument The chapter’s normative contribution is reading the Delaware re- view spectrum as a governance-calibration instrument designed to maintain governance quality through doctrinal de — erence calibrated to the degree o — governance — ailure present. A legal system that took governance quality seriously would provide minimal review where governance is — unctioning, intensive review where governance has structurally — ailed. Delaware has done precisely that, without the governance vocabulary and without any — ormal commitment to club-good theory. The three-level spectrum maps governance qual- ity onto doctrine in a way that makes governance accountability proportional to governance — ailure. Corporate Governance 295

 The business judgment rule applies where no structural govern- ance  --- ailure is present. 316 Governance analysis predicts de --- erence here. Where governance is  --- unctioning — the deliberative process has engaged participants with appropriate capacity — the govern- ance output is presumed valid. The business judgment rule  --- unc- tions as a Pigouvian subsidy  --- or governance provision, reducing the litigation costs that Chapter 7 identi --- ies as Mechanism 1 and 2. By declining to review the merits o ---  board decisions made in good  --- aith by disinterested directors with adequate in --- ormation, courts reduce the governance production costs that would otherwise deter gov- ernance supply. This subsidy addresses the undersupply o ---  govern- ance quality that results when boards cannot capture spillover ben- e --- its: a board whose governance bene --- its employees, creditors, mar- kets, and communities cannot capture those bene --- its in the  --- irm's valuation, so private governance producers invest less than would be socially optimal. The business judgment rule moves governance supply toward the social optimum by reducing the cost o ---  making governance decisions. It is an enabling doctrine, not an insulating one. Without it, every board decision would invite litigation costs su ---

icient to deter attentive governance. With it, boards can govern with con — idence that routine business decisions will not trigger re- view. The strongest objection to this characterization is that de — erence subsidizes governance — ailure equally with governance success. The business judgment rule protected Blue Bell Creameries’ board — which had implemented no monitoring system — or — ood sa — ety, the company’s most central compliance obligation — just as it protects a diligent board with robust compliance in — rastructure. In Pigou- vian terms, de — erence is an untargeted quantity subsidy: it increases 296 Law and Governance

governance production across the board without selecting

or gov- ernance quality. The objection is structurally sound: the BJR in iso- lation does not distinguish competent — rom incompetent govern- ance. But the BJR does not operate in isolation. The complete Pigou- vian mechanism is de — erence-plus-accountability: the business judg- ment rule increases governance quantity by reducing production costs; Caremark, entire — airness, and Revlon duties control govern- ance quality by withdrawing de — erence where governance has struc- turally — ailed. The combined system produces both increased supply and quality assurance. Marchand’s holding — that Caremark requires mission-critical monitoring systems, and that a board without any such system — or — eits de — erence — is the quality-control complement that makes the quantity subsidy targeted at the system level even though it is untargeted at the individual-rule level. Neither compo- nent works alone. De — erence without accountability produces the untargeted subsidy the objection identi — ies. Accountability without de — erence produces the governance-cost escalation that Mecha- nisms 1 and 2 describe. The doctrinal structure that Delaware has developed is the combined mechanism that the Pigouvian analysis requires. Enhanced scrutiny under Unocal Corp. v. Mesa Petroleum Co. and Revlon, Inc. v. MacAndrews & Forbes Holdings, Inc. applies where structural con — licts create what Chapter 7 identi — ies as the last-period de — ection problem. 317 A board adopting de — ensive measures that may entrench management, or a board selling the company on terms that may — avor management over shareholders,

aces distorted incentives: the usual governance disciplines are weakened relative to the immediate gains available through govern- ance sel — -dealing. Enhanced scrutiny addresses this by withdrawing the presumptive de — erence the business judgment rule provides and Corporate Governance 297

requiring instead that the board’s decision

all within a range o — rea- sonableness given the structural con — lict. The doctrine does not im- pose entire — airness’s requirement that the controlling party prove both — air dealing and — air price. It asks a narrower question: given the con — lict present, did the board’s decision-making process oper- ate within the bounds o — institutional competence? This calibration re — lects the governance analysis. The board’s incentive to govern is compromised, not eliminated; the institutional capacity to deliber- ate remains present; the corrective is to intensi — y scrutiny o — the de- liberative process while preserving space — or the board’s ultimately in — ormed judgment. Entire — airness review applies where governance has been cap- tured by a controlling stockholder whose interests have displaced governance — unction as the decision-making — ramework.318 Where the party controlling the governance institution is the same party whose conduct the institution is supposed to check, internal correc- tion mechanisms are unavailable. The club good has been appropri- ated by a member who controls it entirely. Judicial review at the price level ( — air price) and process level ( — air dealing) substitutes — or governance the institution can no longer produce. The entire — air- ness standard presumes that the controlling shareholder cannot es- tablish business judgment de — erence and requires instead that the controller prove both — air dealing and — air price. This is the govern- ance system’s most demanding scrutiny, applied precisely where governance’s internal accountability has completely — ailed. These three doctrinal tiers map governance quality calibration onto existing Delaware law in a way that makes precise predictions. The business judgment rule provides a Pigouvian subsidy that moves governance supply toward the social optimum by reducing governance production costs, but only in settings where governance 298 Law and Governance

is not structurally compromised. Enhanced scrutiny targets the last- period de — ection problem without requiring proo — o — un — air price, matching the scope o — intervention to the scope o — the structural

ailure. Entire — airness applies when governance is captured and in- ternal accountability is impossible, requiring the most searching re- view because the governance institution itsel — has — ailed. This is governance-responsive law.

Where Delaware doctrine diverges

rom governance predictions Two divergences identi — y re — orm opportunities. Caremark overextension. Allen’s original — ormulation admitted no judicial second-guessing o — the board’s choice among reasonable compliance systems. Courts could — ind liability only where the board had implemented no oversight system at all. Marchand imposed a more demanding standard: boards o — companies with mission-crit- ical regulatory obligations must create board-level compliance mon- itoring speci — ically attuned to those obligations. This contraction o —

excludability is de

ensible, because it en — orces the membership cri- terion that the business judgment rule requires: a board that builds compliance in — rastructure — or peripheral risks while neglecting mis- sion-critical ones has not satis — ied the quality threshold — or de — er- ence. The McDonald’s extension crosses — rom criteria en — orcement into congestion introduction. Human capital management is pre- cisely the domain where — irm-speci — ic expertise is most valuable and where judicial oversight using standardized criteria is least produc- tive. Admitting judicial review into that domain does not en — orce Corporate Governance 299

governance quality criteria. It imposes compliance-optimized gov- ernance on — irms where governance should be optimized — or value creation. Shareholder activism as Mechanism 7. Proxy access, mandatory in- stitutional voting, and say-on-pay opened corporate governance to participation by parties whose governance capacity did not justi — y the access provided. Chapter 7 identi — ies mandating open member- ship that disables screening as Mechanism 7: the governance degra- dation mechanism that eliminates the quality control on which club good value depends.319 Each o — these three re — orms added a category o — low-capacity participants to a governance process the participants could not meaning — ully engage. The ISS-Glass Lewis duopoly — illed the resulting gap, capturing authority the re — orms created without the governance accountability that makes club governance legiti- mate.

State competition on the excludability dimension The Delaware-Texas competition in 2025 tests the club-good anal- ysis directly. States are competing not on shareholder rights but on the reliability and scope o — board excludability — rom judicial and shareholder override. Delaware’s Senate Bill 21, enacted March 25, 2025, clari — ied the conditions under which the business judgment rule applies to con- trolling stockholder transactions and restored predictability to the rati — ication doctrine — ollowing the doctrinal disruption that Tornetta v. Musk produced.320 The Delaware Supreme Court’s decision in Tor- netta initially appeared to reject established doctrine by ordering re- scission o — a controlling shareholder’s compensation package. On December 19, 2025, the Court unanimously reversed, holding that 300 Law and Governance

rescission was an improper remedy

or a rati — iable con — lict transac- tion and that the compensation arrangement satis — ied the business judgment rule’s conditions once rati — ied by disinterested sharehold- ers.321 More signi — icantly, the Court a —

irmed its earlier — inding o — li- ability — or the compensation-setting process while reversing the re- scission remedy, and reduced the attorney — ees award — rom the trial court’s $345 million to approximately $54 million on a quantum me- ruit basis. This divergence between liability — indings (a —

irmed) and remedy (reversed) underscored that Delaware law distinguishes be- tween governance — ailure in the decision-making process and the appropriate judicial response to that — ailure — an important distinc- tion — or calibrating remedies to actual governance harm. During the interval between the trial court’s January 2024 order and the Supreme Court’s December 2025 reversal, Tesla reincorpo- rated to Texas, and during the 2025 proxy season, 64.3% o — U.S. companies with pending reincorporation proposals proposed leav- ing Delaware.322 Texas Senate Bill 29, enacted in response, codi — ied the business judgment rule by statute, o —

ering a legislative guaran- tee o — board de — erence that cannot be modi — ied by a single chancel- lor’s opinion.323 The club-good — ramework resolves the race-to-the-top versus race-to-the-bottom debate by identi — ying the variable being com- peted over. States that maintain optimal governance club structure — excludability su —

icient to prevent congestion, subject to mem- bership criteria that maintain quality — will attract — irms that pro- duce governance externalities non-shareholders depend on. The competition is not simply between pro-management and pro-share- holder rules. It is a competition between regimes that calibrate gov- ernance structure correctly and those that distort it in either direc- tion. Delaware’s SB 21 clari — ies that it competes on the dimension o —

Corporate Governance 301

governance club integrity: the statute codi

ies that rati — ication by disinterested shareholders — unctions as a business judgment rule sa — e harbor — or controlling stockholder transactions, and that courts should respect that rati — ication as evidence o — governance quality ra- ther than override it through expansive remedial doctrines. This is state competition on governance structure, not on shareholder rights. The private equity comparison con — irms the analysis. Private equity governance concentrates decision-making authority in in- vestors with governance capacity, aligns incentive compensation with long-term value, and imposes monitoring through active board engagement. Net annualized returns — or buyout — unds ex- ceeded public equity returns by roughly — ive percentage points over the two decades — ollowing 2000, with operational improvements, averaging 370 basis points o — EBITDA margin expansion, as the documented mechanism.324 That advantage has declined as the pri- vate market matured and leverage advantages eroded, but its histor- ical existence provides a natural experiment: the governance struc- ture with the least shareholder participation produced the highest documented returns during the period o — comparison. Founders structuring the most sophisticated new companies have drawn the same lesson, adopting dual-class share structures that preserve gov- ernance club integrity by restricting voting authority to investors whose economic interests align with long-term value creation.325

The club-good standard

or evaluating governance re — orms The governance — ramework implies a speci — ic evaluative standard. Courts, regulators, and legislatures should ask whether a proposed 302 Law and Governance

re

orm en — orces governance quality criteria, removing low-quality governance — rom the protected institutional space, or introduces congestion, admitting governance participants who degrade quality without satis — ying the membership criteria the excludability protec- tion was designed to maintain. Applied to say-on-pay, this standard recommends eliminating the mandatory advisory vote regime, or restructuring it to require direct shareholder engagement with the compensation committee rather than passive rati — ication o — a binary choice. The current struc- ture does not trans — er compensation governance — rom boards to shareholders. It trans — ers compensation governance — rom boards to proxy advisors, which apply standardized criteria across thousands o — companies without — irm-speci — ic adaptation. Applied to the mandatory voting rule, the standard recommends an exemption or sa — e harbor — rom — iduciary liability — or non-voting by index — und managers whose economic model renders — irm-spe- ci — ic governance engagement irrational. Requiring participants to vote when they lack the incentive and capacity to vote thought — ully does not improve governance. It generates demand — or ISS and Glass Lewis recommendations and the robo-voting concentration those recommendations produce. Applied to Caremark, the standard supports Marchand’s mis- sion-critical calibration applied consistently. Oversight duty should be demanding — or governance — ailures that could destroy the insti- tution, moderate — or — ailures that could signi — icantly harm it, and minimal — or peripheral risks that compliance departments can man- age without board-level engagement. The McDonald’s extension in- verts this calibration: it imposes board-level governance engage- ment on the domain where — irm-speci — ic expertise matters most and judicial criteria are least equipped to substitute — or it. Corporate Governance 303

Applied to the Delaware review spectrum, the standard recom- mends preserving its governance-calibration logic while correcting the Caremark overextension. Enhanced scrutiny should turn on whether the governance institution's decision-making process was structurally compromised by the relevant con --- lict, not on whether the outcome seems reasonable in retrospect. Entire  --- airness review should be available where governance has been captured by a con- troller in a speci --- ic transaction, not triggered by controlling stock- holder identity divorced  --- rom evidence o ---  governance  --- ailure.

Three challenges to the governance analysis Three challenges to the club-good analysis o — corporate governance merit direct engagement. Accountability and optimal excludability holds that governance quality requires accountability, and accountability requires the pos- sibility o — shareholder override. A governance club that can exclude all shareholder participation can also exclude accountability — or sel — - dealing, entrenchment, and value extraction. The club-good — rame- work does not endorse unrestricted excludability. It endorses ex- cludability su —

icient to prevent governance congestion while sub- ject to membership criteria that maintain governance quality. The business judgment rule does not protect bad — aith, sel — -dealing, or decisions made without adequate in — ormation. Entire — airness re- view applies to controlling stockholder transactions involving ac- tual con — licts. These doctrines en — orce membership criteria. They exclude — rom the protected governance space decisions that do not satis — y the quality threshold. Accountability and appropriate exclud- ability are not in tension; they are di —

erent aspects o — the same in- stitutional design problem. 304 Law and Governance

Entrenchment and governance quality criteria holds that Dela- ware's board-centric governance structure enables managerial en- trenchment at shareholders' expense, and that stronger shareholder rights are the solution. Lucian Bebchuk's empirical work on pay without per --- ormance, entrenched boards, and perpetual dual-class structures consistently  --- inds that concentrated board authority ena- bles management to extract value without accountability.326 That di- agnosis is accurate. But the proposed solution has thirty years o ---  ev- idence against it. CEO compensation has risen  --- aster since say-on- pay than be --- ore it. Proxy advisors are extracting rents  --- rom the man- datory voting regime in ways that bene --- it their institutional clients rather than ultimate bene --- icial shareholders. The club-good  --- rame- work's response to entrenchment is governance quality criteria, not participation expansion. Entire  --- airness review o ---  controlling stock- holder sel --- -dealing, meaning --- ully applied, is the appropriate mech- anism. What the movement has pursued instead, participation ex- pansion that produces congestion, has not addressed the entrench- ment Bebchuk identi --- ies and has created new institutional problems alongside it.
The activist per --- ormance literature holds that Bebchuk, Brav, and Jiang's study o ---  approximately two thousand hedge  --- und activist in- terventions  --- inds operating per --- ormance improvements persisting

or — ive years post-intervention, challenging the claim that ex- panded shareholder participation produces congestion rather than improvement. 327 Several responses are available. First, deHaan, Larcker, and McClure’s reanalysis using value-weighted rather than equal-weighted returns — inds no signi — icant long-term per — ormance improvement — or the typical shareholder; the Bebchuk results are concentrated in the smallest twenty percent o — targeted — irms meas- Corporate Governance 305

ured in equal-weighted terms. Second, even accepting those

ind- ings, governance improvement at targeted — irms tells us nothing about systemic e —

ects on non-targeted — irms whose boards preemp- tively alter strategy to avoid becoming targets. Third, activist hedge

unds, unlike index — unds, hold concentrated positions and have concentrated economic incentives to govern the — irms they target. I — activist governance improves outcomes, it may be precisely be- cause activists are not the di —

use, poorly-incentivized participants whose mass produces governance congestion. An activist — und with a — ive percent position and a — ocused governance thesis is closer to the governance club member the excludability mechanism was de- signed — or than to the index — und manager with — i — teen pro — essionals monitoring thirteen thousand companies. Chapter 12 applies the same governance analysis to institutions that lack even the market discipline that keeps corporate governance

rom complete capture: universities and nonpro — it institutions, where the governance void is structural and the accountability mechanisms are absent by design. Chapter 12: Universities, Nonpro — its, and the Governance Void Chapter 11 identi — ied the shareholder rights movement’s category error: treating corporate governance as a public good and congest- ing the club good with participants who lack the incentive or capac- ity to govern attentively. That — ailure operated within a legal — rame- work that at least provides shareholders derivative standing, an ac- tive judiciary, and competitive state incorporation law. This chapter examines governance — ailure under — ar weaker legal accountability structures. The governance void has a speci — ic legal structure. It is not the absence o — governance activity. Universities have bylaws, mission statements, compliance o —

ices, — aculty senates, student handbooks, and board committees. The governance void is the absence o — ac- countability. Governance activity proceeds without any mechanism by which a —

ected parties can challenge decisions, en — orce stated commitments, or compel response to identi — ied — ailures. The result is organized — actions occupying governance structures that no ex- ternal constituency can contest, producing institutional outputs in- consistent with the institution’s stated values. 308 Law and Governance

 American universities demonstrate the governance void in its most developed  --- orm. They are institutions that combine public

unction with private governance design: they educate — uture civic leaders, produce knowledge that in — orms public policy, certi — y cre- dentials that serve labor markets, and receive more than $200 billion annually in — ederal research grants, tax exemptions, and subsidized student loans. They do all o — this while governing themselves as pri- vate corporations with no meaning — ul external oversight. Their boards are sel — -perpetuating. Their — iduciary duties are judicially un- en — orceable. The parties most directly a —

ected by governance deci- sions, students and — aculty, have no standing to challenge those de- cisions in court. The combination o — public — unction and private im- punity is not an accident. It is a product o — legal architecture, and the consequences are predictable — rom the architecture. The 2023-2025 period on American campuses provided one o —

the most visible demonstrations o

this architecture’s consequences. University presidents equivocated when asked by Congress whether calling — or the genocide o — Jews violated their institutions’ codes o —

conduct. Federal agencies launched Title VI investigations into sixty institutions. Columbia University reached a nine- — igure settlement; most other institutions extracted commitments in the tens o — mil- lions. Not one settlement required governance re — orm.328 The com- pliance approach treated systematic institutional exclusion as a pol- icy de — icit curable by better policies. The structural conditions pro- ducing the exclusion remained unchanged. Legal rules governing nonpro — it institutional design have cre- ated accountability voids that predictably produce governance — ail- ure. The appropriate response is structural re — orm targeting the void rather than compliance intervention targeting its outputs. University Governance 309

Chapter 12. University Governance Universities and major nonpro — its occupy a distinctive position in the governance landscape. They per — orm — unctions o — public signi — - icance — educating students, producing research, providing chari- table services — while operating under governance structures that insulate decision-makers — rom the accountability mechanisms that discipline comparable institutions in the public and — or-pro — it sec- tors. The analysis begins with universities, where the accountability void is most visible, and then extends to the broader nonpro — it sec- tor, where similar structural — eatures produce similar governance

ailures.

The sovereign charity: public

unction, private governance Elite private universities occupy a distinctive institutional position. They are not government agencies, so administrative law and dem- ocratic accountability do not reach them. They are not market ac- tors, so shareholder discipline does not reach them. They are non- pro — it corporations, so the mechanisms that discipline — or-pro — it governance, shareholder derivative suits, market exit, and competi- tive displacement, are absent or attenuated. Yet they wield power that is in every substantive sense public: they allocate economic op- portunity through credentials, produce and certi — y knowledge through research and publication, set pro — essional standards through law and medical schools that supply the pro — essions, and receive subsidies that dwar — those o — most regulated industries.329 The term “sovereign charity” captures this institutional posi- tion.330 A sovereign charity is an institution that wields public in — lu- ence, receives public subsidy, and per — orms public — unctions while 310 Law and Governance

governing itsel

as a private corporation answerable to no external constituency. Sovereign charities are not new. Colonial-era univer- sity charters created sel — -perpetuating boards precisely because the

ounding donors wanted institutional independence — rom temporal and ecclesiastical authority. Harvard’s Corporation has governed by internal appointment since 1650. Yale’s board is structured as a hy- brid: ten sel — -perpetuating Successor Trustees, six Alumni Fellows elected by degree-holders, and three ex o —

icio members, but the Successor Trustee component controls succession and dominates board composition across generations. What is new is the scale o —

the public subsidy, the range o

the public — unction, and the accu- mulation o — legal shields that protect this private governance — rom external accountability. The governance analysis begins with the shared problem. Uni- versities are designed to manage a cluster o — related problems: the production and transmission o — knowledge, the development o — hu- man capital, the certi — ication o — pro — essional competence, and the maintenance o — institutional conditions — or — ree inquiry. These problems are real, recurring, and interconnected. The governance institution charged with managing them is the board o — trustees, ex- ercising authority delegated to administration and — aculty in various domains. The legal conditions under which that governance exists are the subject o — this section.

Three

eatures o — the accountability void State nonpro — it law creates three — eatures o — university governance that together constitute the accountability void. Each — eature is a le- gal rule. Each produces a speci — ic governance — ailure. Together they University Governance 311

create institutional conditions under which systematic exclusion o

dis

avored groups is not merely possible but structurally likely. The sel — -perpetuating board. State nonpro — it corporation law gives boards the authority to select their own members. The board determines its own composition. No stakeholder constituency, in- cluding students, — aculty, alumni, donors, or the communities in which the institution operates, has a binding vote on board mem- bership. The consequence is that boards are not accountable to the constituencies whose interests they nominally serve. Reputational pressure can in — luence board behavior, but no — ormal mechanism translates that pressure into compelled action. A board can absorb and ignore external criticism inde — initely, because the people with authority to respond are the same people with authority to do noth- ing.331 The governance analysis identi — ies this as a — ailure o — the adjust- ment mechanism. Governance requires not only decision-making and monitoring but adaptation: the capacity to correct errors, re- spond to changed conditions, and remove participants who have captured governance — or private purposes. Sel — -perpetuating boards disable the adjustment mechanism, the — ourth governance element de — ined in Chapter 1 as the capacity to revise institutional architec- ture when it — ails, by making board composition immune — rom ex- ternal correction. When a board — ails, there is no legal process by which a —

ected parties can compel its replacement. The hollow duty o — obedience. Nonpro — it law nominally re- quires trustees to act within the institution’s stated mission, the duty o — obedience, which distinguishes nonpro — it governance — rom the duty o — loyalty and duty o — care — amiliar — rom business organiza- tions.332 But when trustees write the mission statement themselves, 312 Law and Governance

and when the statement is vague enough to accommodate any con- ceivable set o — decisions, the constraint is — ormal rather than opera- tive. Harvard’s mission, “to educate — uture leaders” and “strive to- ward a more just, — air, and promising world,” does not describe a standard capable o — judicial en — orcement. Courts de — er to trustees’ own interpretations, and state attorneys general rarely challenge even — lagrant mission inconsistency. The governance analysis identi — ies this as a monitoring — ailure. Monitoring requires a standard against which per — ormance can be assessed. A duty o — obedience to a vague mission provides no such standard. The Robertson v. Princeton University litigation demon- strated the obverse: when a mission commitment is speci — ic enough to be justiciable, courts can evaluate whether institutional conduct is consistent with the — iduciary obligation. The Woodrow Wilson School commitment to placing graduates in — ederal government ser- vice was speci — ic, and a six-year litigation, though settled, demon- strated that speci — icity enables en — orcement.333 The problem is not the duty’s existence but the vagueness that prevents its operation. The absence o — stakeholder standing. In a — or-pro — it corpo- ration, shareholders can bring derivative suits to discipline the board — or breaches o —

iduciary duty. In a nonpro — it university, no comparable mechanism exists. Students, — aculty, alumni, and donors generally lack standing to challenge governance decisions. The state attorney general is the nominal en — orcer, but attorneys general are chronically under-resourced and politically disinclined to challenge wealthy, prestigious institutions. The general rule that only the at- torney general has standing to en — orce charitable obligations re-

lected an era when charities were small, local, and managed simple trusts. Applied to institutions with billion-dollar endowments, tens University Governance 313

o

thousands o — students, and governance decisions a —

ecting mil- lions o — lives, the rule produces structural impunity: power exercised without meaning — ul accountability.334 Limited exceptions do not — ill the void. The Cali — ornia Supreme Court’s decision in Turner v. Victoria held that a director who loses her board seat a — ter — iling suit retains standing to pursue the chari- table trust breach claim, and the court’s reasoning acknowledged the legislature’s authority to expand en — orcement standing beyond di- rectors.335 Some states permit donor en — orcement o — restricted gi — t conditions. But none o — these mechanisms reaches the constituen- cies most directly a —

ected by governance — ailures: students subjected to systematic exclusion, — aculty silenced by captured governance structures, or communities that depend on universities — or eco- nomic and civic — unctions.

Exclusive inclusion as a governance output The three — eatures o — the governance void combine to produce a speci — ic and predictable governance — ailure. This chapter names it “exclusive inclusion”: the governance state in which an institution claims to protect all members while structurally ensuring that some are unprotected, because the accountability void allows organized

actions to capture governance structures and deploy them against dis — avored groups.336 Exclusive inclusion is not primarily an attitudinal phenomenon. It does not require that board members, administrators, or — aculty harbor hostility toward the groups they exclude. It requires only that the governance architecture creates an opportunity — or organized

actions to capture delegated authority, that the captured authority can be deployed against particular identity groups, and that no 314 Law and Governance

mechanism exists by which a

ected parties can challenge the de- ployment. Given those structural conditions, exclusive inclusion

ollows as a predictable governance outcome. The mechanism has two steps. First, the absence o — external ac- countability (no stakeholder standing), internal accountability (un- en — orceable — iduciary duties), and electoral accountability (sel — -per- petuating boards) creates conditions — or capture. Capture occurs when a — action gains control over decision-making in a particular domain, a hiring committee, a student conduct o —

ice, a depart- mental programming budget, without e —

ective oversight or con- straint. Second, once a — action captures governance authority in a domain, it can implement policies that systematically exclude iden- tity groups the — action dis — avors while claiming to advance the in- stitution’s — ormal commitments. The disentanglement matrix clari — ies which institutional acts warrant governance scrutiny.337 The matrix sorts contested institu- tional actions along two axes: who is acting (an individual exercising personal judgment, or an institution deploying organizational au- thority and resources), and what is targeted (a perspective, or an identity). Individual expression o — a perspective, a pro — essor who de- clines to collaborate with a colleague over a disagreement about pol- icy, is protected. Institutional deployment o — authority against an identity group, a department that uses o —

icial channels to advance a campaign excluding scholars and students based on national origin or religious a —

iliation, is not. The matrix exposes the standard de-

ensive move in governance capture cases: actors who have de- ployed institutional authority against an identity group retreat to individual expression claims, converting an institutional act in the bottom-right quadrant into a personal speech act in the top-le — t quadrant. The analysis cuts through that retreat by — orcing the University Governance 315

threshold question: did the institution take this act through institu- tional channels using institutional authority and resources? The pattern documented across American universities applies across identity categories. At multiple University o — Cali — ornia cam- puses, — aculty governance bodies used delegated authority to ad- vance coordinated campaigns that produced systematic exclusion and intimidation o — Jewish students.338 At — our public universities recognized as leaders in computer science diversity e —

orts, depart- mental governance structures de — ined diversity in ways that system- atically excluded Women o — Color through race-avoidant pro- cesses.339 A predominantly White liberal arts college with explicit DEI commitments applied a — acially neutral behavioral policy that produced systematic racial inequality through di —

erential en — orce- ment against Black students.340 Faculty hiring at many institutions operates through in — ormal culture and committee discretion that screens out conservative and libertarian candidates in ways con- sistent with ideological bias. 341 Students with disabilities — ace sys- tematic exclusion not through individual hostility but through ac- commodation systems that convert legal entitlements into contin- gent administrative privileges.342 The cross-identity pattern con — irms that the mechanism is structural, not cultural. I — the problem were cultural hostility toward any particular group, the same governance — ailures would not a —

ect Jewish students, Women o — Color, Black students, conservative — ac- ulty, and students with disabilities simultaneously. The governance architecture does not discriminate among the groups it — ails to pro- tect. It simply — ails to provide accountability, and organized — actions

ill the resulting void. 316 Law and Governance

Federal coercion as a case study in compliance substitution The Trump administration’s 2025 campaign against universities tested whether — ederal coercion can produce governance re — orm. The results demonstrate that it cannot. The administration — roze billions in research — unding, opened Title VI investigations against sixty institutions, and imposed conditions on restored — unding ranging — rom governance changes to de — initional mandates.343 Co- lumbia agreed to approximately $200 million in settlements. Brown committed $50 million over ten years. Universities capitulated within weeks. But what did capitulation produce? Settlement agreements required universities to adopt speci — ic de — initions, appoint student liaisons, conduct campus climate sur- veys, and submit demographic data.344 Notice what the settlements did not require: governance re — orm. No settlement mandated changes to board composition or selection procedures. None created stakeholder standing — or students or — aculty to challenge depart- ment-level exclusion. None strengthened the duty o — obedience or required mission statements speci — ic enough to constrain — uture board discretion. The settlements treated exclusive inclusion as a compliance de — icit curable by policy adoption. Harvard’s judicial victory illustrates the same point — rom the op- posite direction. In September 2025, — ederal judge Burroughs issued a consolidated 84-page opinion in two consolidated cases (AAUP– Harvard Faculty Chapter v. DOJ and Harvard v. HHS, the same judge who had presided over Students — or Fair Admissions v. Harvard) — inding that the — unding — reeze violated the Administrative Procedure Act and the First Amendment, concluding that the administration had “used antisemitism as a smokescreen — or a targeted, ideologically- University Governance 317

motivated assault on this country’s premier universities.”345 The rul- ing was correct on administrative law. But Harvard’s legal victory le — t the governance void that produced the original crisis entirely intact. Harvard won in court; the governance architecture remains unchanged on campus. University counsel grasped the asymmetry. Columbia could ac- cept a settlement worth approximately $200 million, adopt new de — - initions, hire liaisons, and conduct surveys — all reversible once

ederal attention shi — ted. Restructuring board governance or creat- ing en — orceable mission standards would have been permanent. Re- versibility has value. Governance re — orm eliminates that value. Compliance substituted — or governance rather than producing it — Mechanism 6 operating at institutional scale. The compliance structure (de — initions, liaisons, surveys, reports) satis — ies the exter- nal requirement without changing the internal governance archi- tecture that produced the — ailure. Lauren Edelman explains why: or- ganizations internalize legal requirements through symbolic struc- tures that courts de — er to as evidence o — legal compliance, without the behavioral compliance the law was designed to produce.346

Beyond universities: the nonpro

it sector Universities are the most visible sovereign charities, but the gov- ernance analysis applies across the nonpro — it sector. Hospitals gov- erned by sel — -perpetuating boards, — oundations with vague mission statements, major nonpro — it organizations with public — unctions and private governance, and associations o — pro — essionals with reg- ulatory authority all share the structural conditions that make the governance void possible. 318 Law and Governance

The nonpro --- it hospital sector illustrates both the structural problem and the existing legal response. Nonpro --- it hospitals receive approximately $37.4 billion in annual tax exemptions. In exchange, Revenue Ruling 69-545 conditions hospital tax exemption on pro- moting "the health o ---  a broad class o ---  individuals in the community served," and the A ---

ordable Care Act’s section 501(r) requirements mandate triennial community health needs assessments, community input processes, and implementation strategies.347 The operational test — or hospitals gives the duty o — obedience substantive content: the mission commitment to community bene — it is speci — ic enough to be evaluated, and the regulatory — ramework creates external ac- countability mechanisms that supplement the state attorney gen- eral’s standing. No comparable — ramework exists — or nonpro — it universities. Universities receive comparable public subsidies through tax ex- emptions, deductible donations, and — ederal grants, but — ace no gov- ernance accountability requirement that speci — ies and en — orces the content o — their obligations to their communities. The hospital sec- tor demonstrates that speci — icity and accountability are achievable through existing legal tools. The absence o — those tools in university governance re — lects a legal architecture that has not been updated to re — lect the scale, scope, and public signi — icance o — these institutions. The structural analysis also applies to private — oundations. Large private — oundations control billions in assets dedicated nominally to charitable purposes, govern themselves through boards that are in most cases donor-controlled or sel — -perpetuating, and deploy those assets with minimal legal accountability to the communities whose wel — are the — oundations claim to advance. The duty o — obedience is nominally present. The en — orcement mechanism is the same tooth- less attorney general standing rule that governs universities. The University Governance 319

governance void at major private

oundations is less visible than the void at universities because — oundations do not have enrolled stu- dents to document exclusion and — ile complaints. But the structural conditions are identical.

Applying the governance method The seven-step governance method generates a speci — ic analysis — or university governance. First, the shared problem: universities manage the production and transmission o — knowledge, the development o — human capital, the certi — ication o — pro — essional competence, and the maintenance o — institutional conditions — or — ree inquiry. These are genuine, re- curring, and socially important problems. The institution exists to address them. Second, the governance institution: the board o — trustees, which delegates authority to administration, — aculty, and various subunits. The board is the institution’s ultimate decision-making authority. Applying the output-based — unctionality criteria Chapter 8 speci — ies, the governance analysis reveals a systematic — ailure across all — our

unctional dimensions. Decision-making is — unctioning only to the extent that board decisions govern actual institutional behavior — and at many universities, — aculty governance bodies and administra- tive subunits exercise e —

ective authority that the board neither monitors nor overrides, making board decision-making nominal rather than — unctional. This means the board cannot ensure that de- cisions (once made) actually shape what the institution does. Moni- toring is — unctioning only when it generates conduct-governing in-

ormation used in sanctions or rule changes — and the hollow duty 320 Law and Governance

o

obedience provides no standard against which monitoring out- puts can be assessed. Without a speci — ic, en — orceable mission stand- ard, monitoring cannot distinguish success — rom — ailure; it cannot generate the conduct-governing in — ormation on which sanctions depend. Sanctioning is — unctioning only when sanctions are applied at a — requency consistent with the observable violation rate — and the absence o — stakeholder standing means governance violations go unsanctioned because no party has legal capacity to trigger en — orce- ment. Departments systematically exclude identity groups, but no student or — aculty member can bring a claim; the attorney general lacks resources to investigate hundreds o — institutions. Adjustment is — unctioning only when rules change in response to documented changed circumstances — and sel — -perpetuating boards, as the anal- ysis above establishes, have disabled the adjustment mechanism by making board composition immune — rom external correction. By these criteria, university governance — ails at all — our — unctional di- mensions, not merely at the structural level the accountability void describes. The governance system is not broken in one place; it is non- — unctional as a system. Universities are also, — or students and in many cases — or — aculty, mandatory membership institutions in the sense Chapter 5 speci-

ies. A student enrolled in a pro — essional program has no practical exit: leaving means — or — eiting years o — investment, credential pro- gress, and pro — essional relationships, with no assurance that another institution will accept the trans — er. The accountability mechanisms Chapter 5 identi — ies — mandatory internal appeals with independ- ent review, and periodic re-authorization — apply here with partic- ular — orce. Internal appeals with independent review would give stu- dents subjected to governance — ailures a check on arbitrary exclusion University Governance 321

without converting the university’s governance authority into a li- ability-rule remedy. Periodic review o — departmental governance authority — requiring departments to justi — y continued delegated authority at regular intervals — would supply the temporal exit dis- cipline that the absence o — practical exit removes. Neither mecha- nism would require courts to adjudicate the substance o — academic governance decisions; both would require the governance institu- tion to demonstrate procedural compliance and good — aith to a re- viewer with independence — rom the challenged decision. Third, the legal conditions under which that governance exists: state nonpro — it corporation law, which authorizes sel — -perpetuating boards; the duty o — obedience, which is nominally binding but struc- turally unen — orceable due to mission vagueness; the attorney-gen- eral-only standing rule, which eliminates stakeholder en — orcement; and the tax-exempt status, which creates — ederal and state subsidy relationships without corresponding accountability requirements. Fourth, the member bene — its: degree programs and credentials, research resources, — aculty employment, and institutional in — ra- structure — or knowledge production. Fi — th, the spillovers, traced to speci — ic governance dimensions. Here the accountability void becomes critical, because universities’ spillovers are extraordinarily broad and the constituencies bene — it- ing — rom them are poorly positioned to participate in or oversee the institution’s governance. Monitoring o —

aculty research quality produces the knowledge reliability on which non-members who cite, apply, or build upon university research depend — the spillover runs — rom monitoring quality to the integrity o — public discourse and policy — ormation. Students will cite research that the university

ailed to vet; policymakers will act on studies that never satis — ied basic reliability standards. Decision-making about curriculum and 322 Law and Governance

admissions produces the human capital development that labor markets and civic institutions consume — the spillover runs — rom decision-making quality to economic productivity and civic compe- tence among graduates and the communities they enter. Sanction- ing o — academic misconduct produces the certi — ication reliability on which employers, licensing bodies, and patients depend — the spill- over runs — rom sanctioning credibility to the trustworthiness o —

every credential the institution issues. A degree

rom a university that has systematically excluded entire identity categories — rom — air treatment has lost its certi — ication reliability: employers cannot dis- tinguish which graduates were developed through legitimate pro- cesses and which advanced through captured governance that de- ployed institutional power against dis — avored groups. Adjustment capacity produces the university’s responsiveness to new knowledge, new societal needs, and new threats to inquiry — the spillover runs — rom adjustment quality to the long-term relevance and reliability o — the university’s contributions to public li — e. When governance is captured and adjustment mechanisms are disabled, universities — ail to respond to documented governance — ailures until external pressure — orces capitulation. The in — ormational demands o —

the governance evaluation are most severe in this nonpro

it setting, where spillover bene — iciaries are the broadest, least organized, and hardest to represent in any adversarial proceeding. No party in any litigation involving university governance has incentives to pro- duce the spillover evidence Steps 4 and 5 require: universities over- state their governance contributions, challengers understate them, and the di —

use public that bene — its — rom university governance has no institutional presence. The governance method’s contribution to university re — orm is there — ore primarily a legislative and regulatory University Governance 323

design tool — in

orming accreditation re — orm, state attorney gen- eral authority, board composition requirements, and operational test conditions — rather than a litigation tool. The structural re-

orms this chapter proposes re — lect that in — ormational reality: they target governance architecture through legislation and regulation rather than case-by-case judicial evaluation o — governance quality. Sixth, how law a —

ects governance: state nonpro — it law has de- graded university governance through all three — eatures o — the ac- countability void. Sel — -perpetuating boards (Mechanism 5, remov- ing standing) disable the adjustment mechanism. Hollow — iduciary duties (Mechanism 6, compliance substitution) allow symbolic gov- ernance to satis — y legal requirements without behavioral compli- ance. The standing bar (Mechanism 5 again) eliminates the stake- holder en — orcement that would make monitoring e —

ective. Federal compliance intervention has applied Mechanism 6 at the regulatory level, adding symbolic compliance requirements that universities satis — y without changing governance architecture. Seventh, evaluation: law has systematically degraded the gov- ernance institution that manages American higher education. The degradation is not incidental. It — ollows — rom legal rules that con- centrate authority in sel — -perpetuating institutions while eliminat- ing the accountability mechanisms that could discipline that author- ity.

Toward structural re

orm Three procedural re — orms address the accountability void. They do not require universities to adopt particular values, admit particular students, or reach particular conclusions about contested political questions. They require only that universities govern themselves 324 Law and Governance

through structures that make institutional capture harder and stake- holder recourse available. Standing re — orm. Legislatures could expand standing by permit- ting stakeholders with sustained institutional ties, currently en- rolled students and currently employed — aculty, to bring derivative actions to en — orce existing — iduciary duties against nonpro — it educa- tional institution boards. The model is the corporate derivative ac- tion, adapted — or the nonpro — it context. Corporate derivative suits have produced primarily deterrent e —

ects rather than litigation

loods; knowing that shareholders can en — orce duties, boards exer- cise greater care. Procedural sa — eguards, a demand requirement, heightened pleading standards, a bond requirement, and — ee-shi — t- ing — or — rivolous claims, would prevent harassment litigation while permitting meritorious en — orcement. Standing re — orm addresses the accountability void by giving the constituencies most directly a — -

ected by governance — ailures a mechanism to challenge those — ail- ures. This directly counters the standing elimination that is Mecha- nism 5, and it enables the monitoring — unction that Chapter 8 spec- i — ies as essential to governance: external parties can now generate conduct-governing in — ormation through litigation threat, which disciplines board and administrative behavior toward compliance. A potential tension warrants direct resolution. The club-good

ramework developed in Chapter 5 treats excludability as a struc- tural condition o — governance quality, and standing re — orm appears to weaken excludability by admitting external parties into the gov- ernance accountability process. The resolution is that en — orcement standing is analytically distinct — rom governance participation. The corporate derivative action model already embodies this distinction: shareholders in a public corporation can bring derivative suits chal- University Governance 325

lenging the board’s breach o

iduciary duty without thereby acquir- ing authority to manage the corporation or participate in its gov- ernance decisions. 348 Aronson v. Lewis established the demand re- quirement precisely to maintain this separation: shareholders must demonstrate that the board’s own governance process cannot ad- dress the alleged — ailure be — ore equity permits the shareholder to act on the corporation’s behal — . 349 Extending derivative en — orcement standing to stakeholders with sustained institutional ties — enrolled students, employed — aculty — replicates this separation in the non- pro — it context. En — orcement standing permits accountability — or governance outputs without altering membership boundaries, deci- sion-making authority, or exclusion criteria. The club-good — rame- work is preserved because the standing re — orm adds an external check on governance quality, not an internal participant in govern- ance production. An institution subject to derivative en — orcement by its stakeholders retains — ull authority over its membership crite- ria, its decision-making procedures, its monitoring apparatus, and its sanctioning mechanisms. What it loses is the immunity — rom ac- countability that the standing void currently provides. Duty o — obedience re — orm. State legislatures could require mission statements o — tax-exempt educational institutions to include spe- ci — ic, measurable commitments to equal treatment across identity categories. Such speci — icity would give courts and attorneys general a standard against which institutional conduct can be evaluated. An institution that does not wish to make that commitment could say so explicitly, but could not claim the commitment while maintain- ing governance structures that produce systematic exclusion. The Robertson litigation demonstrated the principle: when a commit- ment is speci — ic, courts can evaluate whether conduct is consistent with the obligation. This re — orm addresses the monitoring — ailure 326 Law and Governance

that the hollow duty o

obedience produces, by giving monitoring a concrete standard against which it can operate. With a speci — ic mis- sion standard, universities cannot de — end systematic exclusion by invoking mission vagueness. The duty o — obedience becomes — unc- tional rather than symbolic. Operational test re — orm. The IRS could reinterpret the operational test to condition educational institution tax exemption on demon- strated governance structures that prevent systematic identity-based exclusion, parallel to the community bene — it standard that already governs nonpro — it hospitals. The hospital community bene — it stand- ard was adopted through IRS rulemaking rather than congressional legislation, demonstrating the administrative authority — or this ap- proach.350 A re — ormed operational test would condition public sub- sidy on governance architecture, not on any particular viewpoint. It would require only that the governance structure make exclusive in- clusion structurally unlikely. This re — orm addresses the adjustment

ailure that sel — -perpetuating boards produce, by making board composition and governance structure part o — the tax-exemption evaluation. Universities seeking continued tax-exempt status would need to demonstrate that they have implemented accountability mechanisms — either standing re — orm or equivalent mechanisms providing stakeholder recourse — that make exclusive inclusion costly rather than consequence- — ree. The three re — orms are mutually rein — orcing. Standing without en — orceable standards leaves en — orcement without a clear legal basis. Standards without standing leave en — orcement dependent on the chronically under-resourced attorney general. External accountabil- ity through the operational test strengthens both but cannot substi- University Governance 327

tute

or the internal mechanisms the — irst two re — orms create. To- gether they create overlapping accountability that makes exclusive inclusion structurally unlikely even i — any single mechanism — ails. The connection to output-based — unctionality is direct and crit- ical. Standing re — orm enables sanctioning by creating a mechanism

or applying sanctions at a — requency consistent with documented violations. Duty o — obedience re — orm enables monitoring by provid- ing a speci — ic standard against which monitoring can measure per-

ormance. Operational test re — orm addresses the adjustment mech- anism by linking governance structure to institutional authoriza- tion (tax-exempt status), creating consequences — or — ailure to main- tain governance quality. The three re — orms together target the — our

unctional dimensions at which university governance currently

ails: standing re — orm (sanctioning), duty o — obedience re — orm (monitoring), operational test re — orm (adjustment), and indirectly all three enhance decision-making by making the board aware that decisions will be subject to review under clearly speci — ied standards. Chapter 13 applies the governance — ramework to a narrower in- stitutional type whose outputs directly shape legal reasoning: knowledge-producing institutions, speci — ically the law reviews and research journals whose scholarship this book cites and courts rely on. The governance question is whether the institutions designed to produce reliable knowledge are themselves reliably governed. Chapter 13: Knowledge Institutions and the Governance o — Scholarship Every knowledge-producing institution governs an epistemic — unc- tion: it determines what claims are reliable enough to certi — y and what claims are not. Medical journals make that determination — or clinical evidence. Patent o —

ices make it — or novelty. Financial rating agencies make it — or credit risk. The governance quality o — each in- stitution determines the epistemic weight that downstream users, courts, regulators, investors, and physicians, can justi — iably place on certi — ied outputs. When governance — ails, the — ailure does not stay inside the institution. It propagates outward through every system that relies on the certi — ied claims. Legal scholarship is the epistemic good this chapter examines, produced and certi — ied primarily through the American law review system. Courts rely on that certi — ication. The Supreme Court has treated law review scholarship as authoritative on how law operates in practice, — rom the Brandeis brie — through the Warren Court’s so- cial science citations to the Roberts Court’s empirical studies on cap- ital punishment’s deterrent e —

ect and campaign — inance’s economic impact. When a proposition o — law rests partly on a law review ci- tation, the governance quality o — the institution that certi — ied the 330 Law and Governance

scholarship is a legal question, not merely an academic one. Student- edited law reviews exhibit governance — ailures that the book’s ana- lytical — ramework explains and its seven-step method can diagnose and address. The governance void in legal knowledge production operates by the same mechanism as the governance — ailures exam- ined in preceding chapters, produces predictable outputs, and is sus- ceptible to the same structural analysis: what shared problem does the institution address, what accountability mechanisms exist, and what has law done to those mechanisms? Those questions explain why law review scholarship is systematically less reliable than its in- stitutional prestige suggests, and why that gap matters — or everyone who relies on scholarship to understand how law works. Knowledge Governance 331

Chapter 13. Knowledge Governance Knowledge governance presents the book’s most re — lexive applica- tion. The institutions that produce and certi — y legal knowledge — law reviews, editorial boards, peer review processes — are them- selves governance institutions subject to the same analytical — rame- work this book develops. Applying the club-good analysis to legal scholarship’s own production system reveals structural governance

ailures with downstream consequences — or the legal system that re- lies on that scholarship.

The knowledge institution as a governance problem Knowledge institutions produce epistemic goods: claims about how the world works, methods — or evaluating those claims, and creden- tials certi — ying that particular claims have survived adequate scru- tiny. Legal scholarship is an epistemic good. It tells courts, agencies, scholars, and legislators how law operates in practice, what e —

ects legal rules produce, and what changes to legal rules would improve those e —

ects. The legal system uses these outputs. Knowledge institutions need governance because, like commer- cial networks and universities, they manage a shared problem — the production o — reliable claims — under conditions where quality is di —

icult to observe directly. The shared problem that knowledge governance addresses is the production o — reliable claims under con- ditions where the quality o — those claims is di —

icult to observe di- rectly. Governance is what makes a claim more than an assertion. Governance structures, peer review, citation standards, replication norms, data transparency requirements, determine whether knowledge claims carry the institutional warrant that makes them usable by parties who cannot independently evaluate them. 332 Law and Governance

 The club-good analysis applies directly: knowledge governance is excludable through editorial selection and nonrivalrous up to con- gestion. Knowledge governance is excludable: not every person who writes about law can publish in a top law review, and the credentials attached to publication in those venues signal that some screening process occurred. Knowledge governance is nonrivalrous up to con- gestion: one scholar's publication in a journal does not prevent an- other's, but i ---  the screening process breaks down and the publication signal no longer carries in --- ormation about quality, the credential loses its value  --- or everyone. And knowledge governance quality de- pends on the selection and composition o ---  the screening institution: a peer-reviewed journal with expert reviewers in the relevant  --- ield screens di ---

erently than a student-edited journal without methodo- logical expertise, and the outputs carry correspondingly di —

erent epistemic weight. Knowledge governance can produce bene — its that are wide- ranging and di —

use. Courts that cite care — ully screened scholarship are less likely to embed — actual errors into precedent. Legislators and agencies that rely on high-quality empirical legal research are more likely to produce e —

ective regulation. Law students trained on methodologically sound scholarship develop more accurate mental models o — how law operates. The parties adversely a —

ected by bad precedent, ine —

ective regulation, or miseducated lawyers are not participants in the editorial governance o — the journals whose out- puts they depend on. They cannot compensate journals — or invest- ing in quality screening. The club good may be undersupplied. Knowledge Governance 333

How law reviews are structured and why that matters

or governance The American law review is an unusual institution by the standards o — academic publishing. Nearly every other scholarly discipline re- views manuscripts through expert editorial boards whose members have substantive expertise in the relevant — ield. Submitted manu- scripts are evaluated by peer reviewers with the methodological knowledge to assess whether the methods support the claims. Re- jection rates at leading journals are high, the process is slow, and the bottleneck is expert judgment. Legal scholarship operates di —

erently. The large majority o —

American law reviews are edited by students who are selected by academic per — ormance and writing ability, not by methodological expertise in the — ields the scholarship addresses. A law review edit- ing a quantitative empirical article on the economic e —

ects o — a reg- ulatory regime typically cannot veri — y whether the statistical meth- ods support the causal claims. A law review editing a qualitative em- pirical article on how compliance o —

icers interpret ambiguous reg- ulatory requirements cannot evaluate whether the interview meth- odology distinguishes pattern claims — rom mechanism claims. The mismatch is structural, not personal: the screening institu- tion’s capacities do not — it the epistemic demands o — what it screens. Student editors can evaluate argument structure, citation accuracy, prose quality, and the logic o — doctrinal claims. They cannot evaluate methodological soundness in empirical work that draws on social science methods they have not been trained in. The governance in- stitution is mismatched to the epistemic challenge. Recent meta-research con — irms the predicted transparency de — - icit across top law journals. A 2024 study examining 300 quantita- tive empirical articles — rom highly ranked journals across 2018–2020 334 Law and Governance

ound data-availability statements in 19% o — articles, with script availability at 6% and preregistration at only 3%. A comprehensive 2025 study examining every empirical article in the top twenty law journals — rom 2010 to 2022 — ound that only 15% had readily accessi- ble datasets. The equivalent — igure in economics and political science is 99%. Even journals with mandatory data availability policies showed poor en — orcement: o — 48 articles published under such poli- cies, only 10 actually made data available. The in — rastructure exists elsewhere. Law reviews have not adopted it. A causal complication challenges the governance analysis. Mat- thews and Rantanen’s 2025 examination o — data availability in the Journal o — Empirical Legal Studies, a peer-reviewed law journal,

ound that only six percent o — articles made data available, a — igure comparable to the worst student-edited journals and worse than the overall law review average. That — inding does not re — ute the gov- ernance analysis; it re — ines it. I — peer review were su —

icient to solve the transparency problem, JELS would look like economics. It does not. The deeper cause is disciplinary culture: legal academic norms do not treat data availability as a condition o — publication credibility, and that norm persists across editorial structure. Student editing compounds the problem by eliminating any screening in — rastruc- ture that could reward methodological transparency. Expert peer re- viewers could, in principle, require transparency as a condition o —

acceptance, even i

the discipline’s prevailing culture does not de- mand it. Student editors cannot per — orm that — unction because they lack the methodological expertise to assess whether the claimed methods were actually per — ormed or whether the underlying data would support the claims made. The two problems compound: a disciplinary culture that does not value transparency, governed by a Knowledge Governance 335

screening institution that cannot en

orce it when authors choose to assert rather than demonstrate. These — igures concern quantitative work, where mature trans- parency standards exist. For qualitative and mixed methods empiri- cal research, which lacks law-speci — ic reporting standards alto- gether, the institutional baseline is weaker still. Authors make claims about how legal actors interpret rules, why institutions be- have as they do, and what mechanisms produce observed outcomes. Yet the evidence rarely appears in reviewable — orm. Selection crite- ria go undisclosed. Analytic procedures remain implicit. Negative cases disappear.

The adverse selection problem and its governance explanation When editors cannot evaluate methodological quality, they evaluate what they can observe: prose quality, topic salience, credentials, and

it with the journal’s perceived identity. Under those conditions, methodological investment con — ers no competitive advantage in law review placement. An article with veri — ied claims based on transparent methods looks identical to a student editor as an article with well-cra — ted prose dressing up methodologically weak conclu- sions. Chapter 6 identi — ied adverse selection as the predictable out- come when a club good’s screening institution cannot distinguish quality — rom the appearance o — quality. Law review governance ex- hibits the same mechanism. George Akerlo — ‘s analysis o — markets — or lemons identi — ies the predicted equilibrium. When quality is di —

i- cult to observe, high-quality producers migrate to markets that can 336 Law and Governance

recognize their investment. Low-quality producers remain in mar- kets where screening cannot detect weakness. Applied to legal schol- arship: high-quality qualitative empirical researchers increasingly route their best work to peer-reviewed social science venues that can identi — y and reward methodological rigor. Law reviews receive a pool increasingly composed o — work that could not survive rigor- ous methodological scrutiny. The pool skews toward methodologi- cal weakness over time, because the incentive to invest in rigor is systematically attenuated at the submission-screening stage. The governance — ramework identi — ies this outcome as adverse selection, the predictable consequence o — weak screening institu- tions. The problem is not that law review editors are careless or that legal scholars are methodologically unsophisticated. The problem is structural: the governance institution lacks the screening technol- ogy that would reward methodological investment, and without that technology, adverse selection is the predicted outcome regard- less o — individual intentions. The downstream consequences — or legal reasoning are direct. When courts cite law review scholarship — or empirical claims about law in action, they are citing outputs produced by a governance in- stitution that cannot distinguish rigorous in — erence — rom sophisti- cated assertion. The Macaulay model o — legal scholarship, qualita- tive, empirical, and revealing the gap between law on the books and law in action, trans — ormed contract theory precisely because it drew on interview data systematically gathered — rom a de — ined population o — businesspeople and lawyers. That work would have satis — ied any reasonable disclosure standard. Scholarship that looks like the Ma- caulay model but lacks the underlying evidentiary discipline pro- duces misleading claims that courts cite with the same con — idence they extend to the genuine article. Knowledge Governance 337

 The governance analysis adds a dimension that conventional critiques o ---  law review editing miss. Critics o ---  student editing typi- cally  --- ocus on student editors' inability to evaluate methodological strength, their tendency to  --- avor  --- ormal complexity over substantive insight, or their bias toward established scholars and prestigious in- stitutions. These are real problems. But they are not the governance problem this chapter identi --- ies. The governance problem is epis- temic: the screening institution cannot evaluate whether empirical claims are warranted, and the resulting adverse selection degrades the pool o ---  scholarship on which courts and scholars depend. This is a collective action problem, not a problem o ---  individual editor quality, and it requires institutional solutions.

Legal constraints compound the governance

ailure A distinctive — eature o — legal scholarship creates a second govern- ance problem that compounds the — irst. Much o — the most important in — ormation about how law operates in practice is legally protected

rom disclosure. Attorney-client privilege shields con — idential com- munications between lawyers and clients. Judicial sealing and pro- tective orders preserve settlement con — identiality. Contractual con-

identiality agreements and trade secret doctrine bind researchers to nondisclosure. Institutional review board restrictions impose con-

identiality requirements on research involving human subjects. Privilege encourages candid legal advice; sealing protects settle- ment — inality; IRB restrictions protect research subjects — rom disclo- sure. Each serves a legitimate governance purpose. But they create a structural consequence — or empirical research: the transparency that veri — ication requires con — licts with the con — identiality that law re- quires. A scholar who studies how settlement operates must rely on 338 Law and Governance

con

idential sources that disclosure requirements would prohibit exposing. A scholar who studies how inside counsel advises on reg- ulatory risk cannot disclose the privileged communications that document that advice. The governance consequence is a di —

erent kind o — selection problem than adverse selection. Where adverse selection skews the pool toward methodologically weak work, the legal constraint prob- lem skews the pool toward easily observable phenomena. Scholars can study appellate decisions, published regulations, and public company — ilings without legal barriers. They — ace systematic barriers studying settlement dynamics, compliance behavior, and pro — es- sional practice advice, precisely the phenomena that matter most to law in action. The result is a body o — empirical legal scholarship sys- tematically biased toward what the governance structure can accom- modate rather than toward what the legal system needs to under- stand.

Applying the governance method to law reviews The seven-step governance method produces a speci — ic analysis o —

legal scholarship governance. First, the shared problem: the legal system produces and uses knowledge claims about how law operates in practice. Courts, agen- cies, legislators, and scholars need reliable methods to distinguish well-warranted claims — rom sophisticated assertions. The accuracy o — legal reasoning about empirical matters depends on the quality o —

the screening processes that certi

y knowledge claims as warranted. Second, the governance institution: the law review editorial board, which selects manuscripts — or publication and thereby con-

ers the institutional signal that legal scholarship carries. For most Knowledge Governance 339

law reviews, this institution is composed o

second- and third-year law students. The output-based — unctionality criteria Chapter 8 speci — ies produce a stark diagnostic. Monitoring is — unctioning when it generates conduct-governing in — ormation — but law re- view editorial boards do not peer-review empirical methodology, so they generate no in — ormation about whether the empirical claims in submitted manuscripts are warranted by the evidence. Decision- making is — unctioning when institutional decisions govern actual member behavior — but editorial acceptance decisions do not gov- ern authors’ methodological practices, because the screening process cannot distinguish rigorous — rom non-rigorous work. Sanctioning is — unctioning when sanctions are applied to actual violations at a

requency consistent with the observable violation rate — but law reviews have no mechanism — or post-publication correction, retrac- tion, or sanction — or methodological — ailure; the low disclosure rates documented above suggest a violation rate — ar exceeding any sanc- tioning — requency. Adjustment is — unctioning when rules change in response to documented changed circumstances — and the law re- view system has not meaning — ully re — ormed its screening processes despite decades o — documented critique. Law review governance

ails all — our — unctional tests — or empirical scholarship. The — rame- work’s protective implications — that legal rules enabling govern- ance are justi — ied because governance generates positive externali- ties — do not apply to an institution that — ails Step 2 entirely. This is actually a strong result — or the chapter’s argument: the governance

ramework correctly identi — ies law reviews as governance-de — icient, which means the — ramework’s prescriptive — orce points toward structural re — orm rather than toward preservation o — the existing editorial architecture. 340 Law and Governance

 Third, the legal conditions under which that governance exists: law reviews are governed by the law schools that house them as stu- dent organizations, and are subject to the law school's institutional constraints but not to any external accreditation  --- or methodological standards. No licensing body, accrediting organization, or regula- tory  --- ramework requires law reviews to implement disclosure standards  --- or empirical claims. Law reviews are also subject to the legal constraints discussed above: the doctrines that create opacity in the subject matter o ---  scholarship limit the transparency that ver- i --- ication requires.
 Fourth, the member bene --- its: publication credit, citation counts, reputational bene --- its, and the career advancement attached to place- ment in highly ranked journals.
 Fi --- th, the spillovers, traced to governance dimensions: i ---  moni- toring were  --- unctioning, it would produce the epistemic reliability on which every downstream user depends — courts citing scholar- ship  --- or empirical claims, agencies incorporating scholarly evidence into regulatory records, and students building mental models o ---  how law operates. I ---  sanctioning were  --- unctioning, it would produce the deterrence against methodological carelessness on which the credi- bility o ---  legal scholarship as a genre depends. I ---  decision-making were  --- unctioning, it would produce the selection output that distin- guishes warranted claims  --- rom unwarranted ones, the signal that makes law review placement in --- ormative. Because all three govern- ance dimensions have  --- ailed, the spillovers are degraded at each di- mension simultaneously: downstream users receive unreliable sig- nals about empirical claims,  --- ace no institutional assurance that methodological  --- ailures will be caught, and cannot distinguish gen- uine  --- rom spurious scholarship by publication venue alone. The in-

ormational demands o — the governance evaluation are di —

erent Knowledge Governance 341

rom other governance domains: the subject matter o — legal scholar- ship governance is itsel — epistemic, which means Steps 4 and 5 can draw on meta-research about transparency rates, replication out- comes, and citation patterns that is available without the adversarial limitations that constrain other governance domains. The low dis- closure rates and accessible-dataset rates documented above are the kind o — quantitative spillover evidence that Steps 4 and 5 demand — and they are available because the governance — ailure in knowledge institutions is itsel — empirically documented by the kind o — research the governance — ramework calls — or. Sixth, how law a —

ects the governance institution: existing law creates the legal constraints on transparency that compound the governance — ailure without providing any compensating require- ment that law reviews implement adequate disclosure standards. There is no legal requirement — or data availability, methodological disclosure, or screening protocols — or empirical claims. The govern- ance institution is le — t to develop voluntary practices in the absence o — any external accountability mechanism, and the adverse selection dynamic predicts that voluntary practices will be insu —

icient. Seventh, evaluation: the governance — ramework identi — ies the law review editorial structure as a club good operating without ad- equate membership criteria — or the epistemic — unction it per — orms. The screening institution cannot evaluate the claims it certi — ies. The institutional signal it con — ers there — ore carries less in — ormation than it appears to carry. Users who rely on that signal, particularly courts and agencies that cannot independently evaluate methodology, make decisions based on epistemic credentials that do not reliably re — lect epistemic quality. 342 Law and Governance

The QELS governance

ramework as structural repair The governance — ramework — or qualitative empirical legal scholar- ship proposed in the author’s companion work, Qualitative Empirical Legal Scholarship, o —

ers three linked institutional tools that — unction as governance — or the knowledge club. These tools address the gov- ernance — ailures that the seven-step analysis has identi — ied. The — irst tool is a claim typology. Pattern claims assert that a phenomenon is — requent, typical, or distributed in a particular way across a de — ined population. Mechanism claims assert that a causal process produces an observed outcome. Interpretive claims assert that a text, practice, or institution carries a particular meaning within a social context. Each claim type has distinct in — erential risks and requires distinct evidentiary warrant. A pattern claim requires a de — ined population, a disclosed selection procedure, and a search

or negative cases. A mechanism claim requires, in addition, evi- dence o — the causal process rather than mere correlation. An inter- pretive claim requires documentation o — the interpretive commu- nity whose conventions give the meaning its authority. The typology provides editorial in — rastructure — or claim-evi- dence — it evaluation. A student editor who cannot evaluate statistical methods can ask: is this a pattern claim? Has the author de — ined the population? Disclosed the selection procedure? Reported negative cases? These are answerable questions that the typology makes leg- ible without requiring methodological expertise. The second tool is the standardized methodological abstract, a structured disclosure document requiring authors to provide seven key elements: claim classi — ication, population de — inition, selection procedure, discon — irmation criteria, negative case analysis, valida- tion measures, and legal constraints. The standardized abstract — unc- Knowledge Governance 343

tions as the excludability mechanism

or the knowledge club. Au- thors who conducted rigorous research can complete it. Authors who did not cannot — abricate the underlying materials the abstract would require them to disclose. The asymmetry is what makes the tool work: disclosure requirements are easier to satis — y — or authors who did the work, and harder to satis — y — or authors who dressed up thin methodology in sophisticated prose. The third tool is the best available evidence doctrine, which pro- vides a narrow accommodation — or research constrained by legal barriers to transparency. When law itsel — prevents the — ull disclosure that veri — ication requires, a researcher should not be excluded — rom publishing empirical scholarship about legally important but con — i- dential phenomena. The doctrine permits constrained research when the legal barriers are precisely identi — ied by speci — ic legal cita- tion, the claims are appropriately narrowed to match what the con- strained evidence can support, and compensatory rigor is provided through additional validation measures. The doctrine addresses the topic-selection bias that strict transparency requirements would produce, without eliminating the credibility discipline that disclo- sure requirements provide. The governance analysis shows why voluntary adoption is in- su —

icient. Voluntary standards work when producers can capture the reputational bene — its o — compliance. In the legal scholarship con- text, the bene — its o — methodological compliance — low primarily to users who cannot independently evaluate methodology, not to au- thors who invest in rigor. The adverse selection dynamic predicts that voluntary rigor generates modest reputational bene — its while imposing real costs, and the predicted equilibrium is underinvest- ment in methodological quality. Institutional in — rastructure that changes the screening technology, by making rigor visible to editors 344 Law and Governance

who otherwise cannot evaluate it, changes the equilibrium by mak- ing methodological investment identi — iable and rewardable.

Why governance

ailures in knowledge institutions matter to law The connection between scholarship governance and legal out- comes is not speculative. Several documented patterns show the downstream consequences o — weak epistemic governance. Courts have embedded empirically contested claims into consti- tutional doctrine because the law review scholarship those courts cited did not survive methodological scrutiny. The empirical record on whether capital punishment deters crime has been cited in both directions by justices reasoning about the constitutionality o — the death penalty, yet the quality o — the underlying empirical work has varied enormously and was largely invisible to the judicial opinions that cited it.351 The empirical record on the e —

ects o — gun regula- tions, campaign — inance rules, and racial disparities in criminal sen- tencing has generated legal doctrine that turns in part on what courts believe the evidence shows. 352 That belie — is structured by what scholarship says, and what scholarship says is structured by what governance institutions certi — y. Regulatory agencies — ace the same dependence. The administra- tive law standard — or arbitrary and capricious review 353 requires agencies to engage with signi — icant empirical evidence in the record. When advocacy organizations submit law review scholarship as ev- idence in rulemaking proceedings, the quality o — that scholarship a — -

ects the quality o — agency reasoning. An agency that relies on meth- Knowledge Governance 345

odologically weak scholarship in support o

a regulatory determina- tion may — ace arbitrary and capricious challenges precisely because the underlying scholarship could not survive veri — ication. These downstream consequences illustrate a critical dimension o — how governance — ailures propagate. When courts and agencies cite scholarship — or empirical claims about law in action — whether capital punishment deters criminal conduct, whether particular gun regulations reduce homicide, whether campaign — inance restrictions a —

ect electoral competition, whether sentencing disparity tracks protected characteristics — those citations embed institutional cer- ti — ication into legal doctrine. The governance institution that certi-

ied the scholarship (the student-edited law review) did not possess the methodological expertise to veri — y the claims. Courts, which also lack methodological expertise in social science, rely on the institu- tional signal the law review con — ers. The result is doctrine that turns on empirical propositions that no participant in the doctrinal devel- opment chain was positioned to veri — y rigorously. This is govern- ance — ailure cascading into legal — ailure: the output o — a knowledge institution whose screening cannot evaluate what it certi — ies be- comes the basis — or rules a —

ecting the behavior o — millions o — people. The doctrinal implication o — the governance analysis is that legal actors who rely on scholarship — or empirical claims should ask the governance question: what screening process produced this claim, and what does that process tell me about the reliability o — the war- rant? A law review article by a credentialed scholar citing other law review articles is not sel — -certi — ying. The governance institution that certi — ied each link in that chain may have lacked the capacity to evaluate the methodology. Courts have implicitly recognized this problem in di —

erent ways. The Daubert standard — or expert testimony requires — ederal 346 Law and Governance

courts to evaluate whether scienti

ic evidence is based on su —

icient

acts or data, is the product o — reliable principles and methods, and applies those methods reliably to the — acts o — the case. The Daubert inquiry is a governance question: did the knowledge-production process satis — y the quality criteria that make the output reliable? The same question, applied to academic scholarship rather than — orensic expert testimony, is what the governance — ramework — or legal schol- arship asks.

The sel

-re — erential problem and the governance de — ense The chapter’s strongest test is re — lexive: does the governance — rame- work apply to the scholarship institutions it analyzes? This book re- lies heavily on law review scholarship: Bernstein’s empirical work on the diamond trade, Mahoney’s historical analysis o — exchange governance, Ostrom’s institutional — ramework, Akerlo — ‘s adverse selection model, and dozens o — other sources certi — ied and published by the very editorial process this chapter argues is de — icient. I — the governance — ramework correctly diagnoses law review screening as incapable o — evaluating methodological rigor, the question arises: should readers trust the scholarship this book cites, and should law review readers and courts trust this book’s assertions about govern- ance? The honest answer requires distinguishing what the governance

ramework claims — rom what it does not claim. The governance

ramework does not validate prior sources or grant them a retroac- tive governance seal o — approval. The — ramework provides a method

or evaluating the reliability o — the institutions that certi — ied them. Knowledge Governance 347

For each source this book relies on, a reader could ask: what screen- ing process produced this claim, and does that process tell me any- thing about its reliability? The evaluation must be source-by-source, not institution-wide. For empirical claims, the transparency indicators this chapter identi — ies — disclosed populations, reported selection procedures, accessible data, identi — ied analytic procedures — provide a partial signal even without — ormal methodological review. Scholarship produced by the Macaulay method (disclosed sampling, transparent selection, negative case reporting) carries more epistemic weight than scholarship that makes empirical claims without methodology. Scholarship published in a venue with mandatory data availability and — unctional en — orcement o — that mandate carries more epistemic weight than scholarship in a venue with voluntary norms and no en — orcement structure. The reader can evaluate each claim on these governance dimensions without accepting the chapter’s claim that law reviews as a class are governance-de — icient. The governance analysis applies to this book’s own claims with equal — orce. This manuscript makes arguments about governance that rest partly on empirical premises: the claim that law review dis- closure rates are low, that the NYSE’s post-Silver governance capac- ity declined, that universities without independent monitoring o —

their Title IX processes

ailed to catch abuse. These empirical claims rest on published sources, administrative records, empirical re- search, and historical analysis that a skeptical reader should evaluate on the governance criteria the book sets — orth. The — ramework does not provide a de — ense against that scrutiny. It provides a method — or conducting it. At bottom, the governance analysis is not a claim about the in- trinsic virtue o — this book or its citations. It is a structural analysis o —

348 Law and Governance

institutions. The law review system has governance

ailures docu- mented by this chapter’s citations, and those — ailures mean that law review publication does not carry the guarantee o — methodological rigor that it appears to carry. That recognition does not condemn every law review article or validate every peer-reviewed article in another discipline. It identi — ies a systematic institutional weakness. Readers who understand that weakness can adjust their con — idence calibration accordingly when they encounter a law review citation to empirical claims. The governance — ramework’s structural de — ense against the re-

lexive objection is this: the governance analysis is presented as a structural diagnosis, not as a personal claim about this book’s intel- lectual virtues. The — ramework applies to law reviews as one insti- tutional domain among several, including the peer-reviewed social science journals where this book’s governance analysis also relies on sources. The critique is structural: it identi — ies how the law review’s institutional architecture produces predictable governance — ailures, a —

ecting which kinds o — claims are certi — iable and which must be taken on — aith. The appropriate response is not to reject all scholar- ship produced by imper — ect institutions but to evaluate the speci — ic institutions that certi — ied each piece o — work by the governance cri- teria this chapter and earlier chapters have speci — ied. The law re- view’s governance — ailures are real; acknowledging them is the be- ginning o — institutional re — orm, not a reason to dismiss the entire body o — legal scholarship the institution has produced.

Portability o

the governance analysis The book’s — ramework turns on itsel — here, and turns productively. Each chapter has applied the governance method to an institutional Knowledge Governance 349

domain: commercial networks, contract remedies, corporate gov- ernance, universities and nonpro — its. This chapter applies the method to the institution that produces legal scholarship, including the scholarship on which this book relies. The re — lexive application o — the governance method to scholarship-producing institutions is the — ramework’s — ullest demonstration o — its generality: i — the method can identi — y governance — ailures in any institutional do- main, it applies equally to the domain that produces the scholarship on which the — ramework relies. I — governance analysis identi — ies the conditions under which in- stitutions can be reliably trusted to manage shared problems, then governance analysis o — scholarship-producing institutions identi — ies the conditions under which their outputs can be reliably trusted. Scholarship produced under institutional conditions with adequate screening, transparent methods, and accountability — or claim-evi- dence — it carries a di —

erent epistemic warrant than scholarship pro- duced under conditions where those governance — eatures are absent. The method does not yield a simple conclusion that student-ed- ited law reviews should be abolished. The American law review has distinctive — eatures, including the training it provides students, the breadth o — doctrinal coverage it enables, and the responsiveness to legal development that annual editorial cycles — acilitate, that alter- native institutional — orms may not replicate. The governance anal- ysis asks a narrower question: what structural changes would im- prove the epistemic reliability o — scholarship this institution certi-

ies? The standardized methodological abstract and the audit rubric are governance tools — or a speci — ic institutional context, not a blue- print — or general peer review. They are designed to — unction within the law review’s existing editorial structure while improving its claim-screening capacity. 350 Law and Governance

Chapter 14 maps the  --- ramework's limits, gaps, and research ho- rizons across six dimensions, ranging  --- rom the risk that governance language masks private power to the relationship between govern- ance exclusion and the history o ---  systematic discrimination. Each limit marks a place where the  --- ramework's current speci --- ication reaches its edges and where serious scholarship can extend the anal- ysis.

Chapter 14: Limits, Gaps, and Research Horizons A — ramework that applies across institutional domains reveals di — -

erent problems in each. The governance — ramework developed in this book, precisely because it engages antitrust law, civil rights law, constitutional theory, political philosophy, and legal methodology, encounters phenomena at each crossing that it does not — ully re- solve. Naming those limits honestly is part o — the — ramework’s con- tribution. A — ramework that cannot identi — y its edges cannot guide research into what lies beyond them. Six limits, gaps, and open questions structure this chapter. The

irst — our are substantive: the risk that governance language masks private power, the problem o — coercion within governance institu- tions, the intersection o — governance bene — its with antitrust con- cerns, and the relationship between governance autonomy and anti- discrimination and rights claims. The — i — th is methodological: the

ramework’s current level o — generality leaves determinacy gaps that speci — ic doctrinal applications will need to — ill. The sixth is norma- tive and historical: the relationship between the — ramework’s ana- lytical treatment o — exclusion and the history o — systematic discrim- ination that exclusionary arrangements have produced. 352 Law and Governance

 Each o ---  these limits is also a research horizon. The  --- ramework's current speci --- ication does not resolve them, and that is not a de --- i- ciency. It is the ordinary condition o ---  a  --- ramework at the beginning o ---  its development. The analysis that  --- ollows re --- ines the  --- ramework by speci --- ying what it does and does not claim, and in doing so iden- ti --- ies where serious scholarship can extend the analysis.
 A  --- oundational scope question should be addressed be --- ore the substantive limits: why does this book analyze governance produced by groups rather than governance produced by markets, algorithms, plat --- orms, or state bureaucracies? The limitation is structural rather than arbitrary. The  --- ramework's analytical tools — excludability, shared use, congestion dynamics, the member-bene --- it/spillover dis- tinction — depend on the club-good structure that group govern- ance institutions exhibit. Market governance lacks the excludability on which the club-good  --- ramework's analytical power depends: prices are public goods, and market discipline operates through exit rather than through the internal sanctioning mechanisms the

ramework analyzes.354 Algorithmic governance lacks membership in the relevant sense: users o — an algorithm do not jointly produce governance outputs through collective decision-making, monitor- ing, and sanctioning, and the — our-element test o — Chapter 1 does not apply to automated systems that adjust parameters without member participation.355 State governance lacks voluntary exit, the de — ining — eature that gives club-good analysis its disciplinary — orce — citizens cannot withdraw — rom state authority in the way that members o — a voluntary governance institution can leave, which means the market competition among governance institutions that Tiebout identi — ied as the mechanism driving club-good e —

iciency does not operate.356 Extending the — ramework to institutions that Knowledge Governance 353

lack club-good structure would dilute its precision without increas- ing its explanatory reach. The — ramework’s analytical power is a

unction o — its scope limitation, not a concession in spite o — it. Fu- ture work may adapt the — ramework — or hybrid institutions — plat-

orms that combine club-good — eatures with public-good — eatures, or decentralized organizations that combine algorithmic govern- ance with member-driven decision-making — but that adaptation requires its own theoretical development and should not be accom- plished by stretching a — ramework beyond the institutional struc- tures it was built to analyze. 354 Law and Governance

Chapter 14. Framework Limits Every analytical — ramework has limits, and stating them precisely is part o — the — ramework’s value. Six challenges deserve direct engage- ment, each identi — ying a boundary condition where the governance analysis either requires quali — ication or encounters a competing le- gal value that may properly override governance considerations. Addressing these limits strengthens rather than undermines the

ramework, because a method that claims to resolve every con — lict is less use — ul than one that speci — ies where its analytical leverage is greatest and where other considerations must enter the analysis.

Governance Language and the Risk o

Capture The most direct challenge to this book’s analysis is not that it is wrong but that it is dangerous. Governance language is not merely descriptive. It is normatively loaded. Calling an arrangement a “gov- ernance institution” imports a legitimacy claim: governance is how groups manage shared problems, and institutions that manage shared problems deserve the legal protections that enable them to do so. The risk is that private actors will invoke governance lan- guage to shield arrangements that are, beneath the legitimating ter- minology, exercises o — private power with no genuine governance

unction. The concern is legitimate. The history o — private ordering in- cludes examples in which governance claims concealed anticompet- itive conduct, discriminatory practices, and simple rent-seeking by organized — actions. Trade associations described as governance in- stitutions — or their industries have — unctioned as price- — ixing car- tels.357 Pro — essional licensing bodies described as governance insti- tutions — or their pro — essions have — unctioned as barriers to entry Framework Limits 355

protecting incumbents. 358 The language o

governance, including this book’s language o — governance, is available to any private actor seeking legal cover. The most pressing version involves strategic deployment. Rob- erts v. United States Jaycees illustrates it precisely:359 i — governance cost analysis becomes a required element o — civil rights adjudication, dis- criminatory institutions gain a procedural tool — or delaying compli- ance. Every institution — acing a civil rights challenge would invoke the governance — ramework to demand that courts evaluate the gov- ernance costs o — the remedy be — ore ordering compliance, converting the governance evaluation into a litigation weapon against en — orce- ment. The — ramework must state its boundary clearly: governance analysis does not create an a —

irmative de — ense in civil rights litiga- tion. Courts applying anti-discrimination law should not be re- quired to conduct a — ull governance evaluation be — ore ordering com- pliance with Title VII, Title VI, or constitutional equal protection requirements. The — ramework’s appropriate role in the civil rights context is as a legislative and regulatory design tool — in — orming how remedies can be structured to achieve civil rights objectives while minimizing unnecessary governance destruction — not as a litigation-stage de — ense that governance institutions invoke to resist en — orcement. Legal actors at the remedial stage, a — ter a violation has been established, can use governance analysis to ask whether spe- ci — ic remedies are calibrated to correct the identi — ied violation with- out destroying institutions per — orming valuable — unctions. But that question comes a — ter the determination that a violation occurred, not be — ore compliance is required. Governance laundering is how capture operates in practice. An entity can adopt — ormal governance trappings — a stated mission, a membership structure, bylaws, a dispute resolution procedure — 356 Law and Governance

and then invoke those trappings to claim the property-rule protec- tion the — ramework provides — or genuine governance exclusion. The

ramework’s answer to governance laundering is the output-based

unctionality criteria Chapter 8 speci — ies and the evidentiary bur- den-shi — ting those criteria establish. An institution claiming gov- ernance protection bears the burden o — demonstrating that its gov- ernance elements are — unctional in the output-based sense: that its monitoring generates conduct-governing in — ormation actually used in sanctions, rules, or behavior; that its sanctioning is applied at a

requency consistent with its observable violation rate; that its deci- sion-making governs actual member conduct; and that its rules have changed in response to documented changed circumstances. These output-based criteria are harder to — abricate than structural criteria, because outputs require actual institutional activity — real arbitra- tion proceedings with discoverable records, real sanction decisions with documented grounds, real rule changes with identi — iable trig- gering conditions — that sophisticated entities cannot generate cos- metically without engaging in the substantive governance the crite- ria require. The criteria are imper — ect: a su —

iciently determined en- tity can generate cosmetic outputs alongside cosmetic structures. But the evidentiary bar is substantially higher than the structural test alone, and the burden o — production — alls on the institution claiming protection rather than on the party challenging it. The — ramework’s actual claim di —

ers — rom what this concern as- sumes. The — ramework de — ines governance operationally, as the or- ganized system by which a group manages a shared problem over time, and asks whether any given arrangement satis — ies that de — ini- tion. Legitimacy is a separate question. The — ramework then asks whether the legal protections available to the arrangement are cali- brated to the governance — unction actually per — ormed. Framework Limits 357

The critical question is what work the governance analysis does. The seven-step method begins by identi --- ying the shared problem and asking whether the institution addresses it. Chapter 9's analysis o ---  diamond dealer networks identi --- ies the shared problem as con- tract en --- orcement reliability in a context where  --- ormal legal mech- anisms are too slow and costly to support high-volume oral transac- tions. The governance institution's sanctioning mechanism, credi- ble threat o ---  worldwide ostracism, addresses that problem by mak- ing de --- ection costly at scale. A trade association that calls itsel ---  a gov- ernance institution but whose primary activity is  --- ixing prices at above-market levels  --- ails the  --- irst step: it does not address a shared governance problem. It exploits one. The governance  --- ramework identi --- ies that  --- ailure rather than hiding it.
More precisely: governance arrangements can pursue member bene --- its while also generating spillovers, or they can pursue mem- ber bene --- its through coordination that harms outsiders. The dia- mond bourse generates supply chain integrity that  --- lows to buyers, retailers, and consumers outside the bourse.360 A price- --- ixing cartel extracts surplus  --- rom outside buyers without generating o ---

setting externalities. The — ramework distinguishes these cases by asking whether the institution’s outputs, at Steps 4 and 5, include genuine bene — its to non-members or only the capture o — bene — its at non- members’ expense. The response does not eliminate the concern. An institution can pursue genuine governance purposes while also suppressing com- petition or discriminating against outsiders. A pro — essional bar as- sociation can set genuine competency standards while also using those standards to exclude quali — ied practitioners — rom minority groups. The governance analysis identi — ies the governance — unction; it does not certi — y that every governance institution’s practices are 358 Law and Governance

benign. The next three sections address the speci

ic legal regimes that constrain governance institutions even when their governance

unction is real.

Coercion Within Governance: A Structural Limit Governance institutions can be coercive. The ostracism mechanism that makes the diamond bourse e —

ective also gives bourse leaders power to harm individual members through procedures that may be biased, corrupt, or captured by — actions with private interests in a particular outcome. The private governance arrangements this book analyzes as alternatives to court-centered dispute resolution impose sanctions, restrict economic activity, and make determina- tions o —

act and liability without the due process protections that public law en — orcement provides. The concern that animated Herbert Wechsler’s neutral princi- ples argument about constitutional adjudication has a counterpart in private law: governance by private institutions operating beyond judicial scrutiny is governance without the rule o — law. Mandatory arbitration arrangements that require employees to waive class ar- bitration rights, private property owner associations that impose as- sessments without meaning — ul democratic accountability, and pro-

essional organizations that expel members without due process are all invoking governance language to shield exercises o — private power that courts should not hesitate to police. The concern is well- — ounded and is not — ully answered by point- ing to the governance — unction such arrangements per — orm. The re- sponse has two components. First, the governance — ramework already builds discipline into its analysis. Chapter 7’s account o — governance-enabling law and Framework Limits 359

governance-degrading law includes a category o

law that properly disciplines governance rather than destroying it. Judicial review o —

private arbitration awards

or — undamental procedural un — airness does not destroy governance; it en — orces the membership criteria on which governance quality depends. A governance institution whose sanctions process is captured by — actions pursuing private interests has — ailed the — ourth element o — the governance de — inition: adjust- ment. Governance that cannot sel — -correct when its processes are corrupted is not per — orming the governance — unction. Judicial inter- vention to restore procedural integrity to a captured governance process is governance discipline, not governance destruction. The deeper point is that the coercion concern identi — ies a real limit on the — ramework’s prescriptive ambitions. The governance

ramework does not counsel de — erence to private governance insti- tutions regardless o — what they do to members. It counsels asking whether legal intervention improves governance or degrades it. In- tervention that restores due process to a captured arbitration mech- anism improves governance. Intervention that destroys the sanc- tioning mechanism altogether destroys the governance — unction along with the abuse. The — ramework’s prescriptive claim is about calibration: law should target the governance — ailure speci — ically ra- ther than using the — ailure as grounds to eliminate the governance institution entirely. The diamond bourse case illustrates the calibration. The bourse’s sanctioning mechanism is subject to manipulation: a — ac- tion that controls the bourse’s arbitration processes can use ostra- cism — or private advantage rather than trade integrity en — orcement. Legal oversight that reviews bourse arbitration awards — or basic procedural adequacy, without requiring bourse arbitration to repli- 360 Law and Governance

cate

ull civil procedure, calibrates judicial intervention to the gov- ernance — ailure without eliminating the governance — unction. The

ramework predicts that this is the right approach. It predicts, equally, that an antitrust challenge that simply declares the bourse’s exclusionary practices unlaw — ul destroys governance while correct- ing an identi — ied abuse.

The Antitrust Inter

ace The antitrust inter — ace is the most doctrinally speci — ic challenge the

ramework addresses. Governance institutions that manage shared problems through exclusion and sanctioning create exactly the con- ditions that antitrust law is designed to police. An agreement among competitors to en — orce shared rules, however described, is a hori- zontal agreement subject to per se condemnation or rule-o — -reason analysis under Sherman Act section 1. The governance language this book deploys does not change the economic analysis. I — the arrange- ment restricts competition, the antitrust — ramework applies. The challenge has two versions. The strong version holds that governance bene — its can never justi — y antitrust concern: the per se rule covers most genuine governance coordination because most governance coordination among competitors involves horizontal price-setting, market allocation, or output restriction. The weak version holds that governance bene — its do not automatically justi — y antitrust immunity: the institutions this book discusses should be analyzed under the rule o — reason, and some will survive and some will — ail that analysis. The strong version is descriptively incorrect about how anti- trust law actually operates. The Supreme Court has consistently rec- ognized that the Sherman Act’s condemnation o — every combination Framework Limits 361

in restraint o

trade does not reach every governance arrangement, because some arrangements are ancillary to productive cooperation rather than substitutes — or competition. The rule o — reason is pre- cisely the doctrinal space in which governance bene — its enter the an- titrust analysis. NCAA v. Board o — Regents recognized that college athletics requires horizontal agreement on rules that restrict mem- ber conduct, and while the Court — ound the speci — ic output re- striction at issue unjusti — ied, it did not condemn the governance

ramework. Broadcast Music, Inc. v. CBS recognized that blanket licensing arrangements enabling rights management would be con- demned as naked price- — ixing without recognizing that they pro- duce what the Court characterized as “to some extent, a di —

erent product” that competition without coordination could not have supplied. The weak version is correct. Governance institutions that per-

orm genuine governance — unctions should receive rule-o — -reason treatment, not per se condemnation, because per se condemnation does not account — or the governance bene — its the institution pro- duces. But receiving rule-o — -reason treatment does not mean receiv- ing automatic immunity. The — ramework requires the — ull method at Step 6: asking how antitrust law a —

ects the governance institu- tion, and whether applying antitrust liability at this margin destroys the governance — unction, disciplines a speci — ic governance — ailure, or eliminates an arrangement whose governance bene — its do not justi — y the competitive harm. The Silver v. New York Stock Exchange analysis in Chapter 9 illustrates this directly. The Court held that antitrust liability at- tached to the NYSE’s re — usal to maintain wire connections with a broker it had decided to exclude, because the exchange had not — ol- 362 Law and Governance

lowed its own procedural rules. The holding did not destroy ex- change governance. It en — orced the membership criteria, procedural

airness, on which governance quality depends. The 1975 Amend- ments then substituted SEC oversight — or antitrust liability as the accountability mechanism, which Chapter 10 analyzes as govern- ance restoration. Antitrust doctrine does not contradict the — rame- work; it is explained by it. One — rontier deserves identi — ication even though this book does not resolve it. Decentralized autonomous organizations, gig-econ- omy plat — orms, and other hybrid institutions that combine algorith- mic coordination with member-driven governance raise antitrust questions that existing doctrine does not squarely address. Market concentration measured by conventional indices (the Her — indahl- Hirschman Index) may coexist with genuine governance decentral- ization measured by governance-speci — ic metrics such as the Naka- moto coe —

icient (the minimum number o — entities that must collude to control a protocol) or Gini coe —

icients applied to token-voting distributions. A companion paper develops a dual-metric en — orce- ment — ramework — or these institutions, proposing that antitrust analysis o — decentralized governance requires measuring both mar- ket power and governance power and that the two can diverge in ways that produce distinct competitive harms. 361 Extending this book’s club-good — ramework to institutions that blend algorithmic and member-driven governance is a task — or — uture work, but the antitrust inter — ace described in this section supplies the analytical

oundation: the question remains whether a given institution’s co- ordination serves a genuine governance — unction or masks compet- itive harm, with the additional complication that decentralized gov- ernance structures can exhibit both simultaneously. Framework Limits 363

Rights, Anti-Discrimination Law, and Governance Analysis Anti-discrimination law and constitutional rights create constraints on private governance that the — ramework must respect and cannot override. Title VII o — the Civil Rights Act prohibits employment discrimination on the basis o — protected characteristics. Title VI prohibits discrimination in — ederally — unded programs. Anti-dis- crimination law applies to private governance institutions that qual- i — y as employers, program recipients, or public accommodations, without regard to the governance — unctions those institutions per-

orm. Constitutional and common-law rights impose independent constraints on governance institutions. Individual rights, including

ree expression, due process, equal protection, and the various asso- ciational rights recognized under constitutional and common law, create claims that governance institutions cannot simply override. The — ramework’s insistence that governance institutions need the authority to exclude and sanction members does not settle whether any particular exclusion or sanction violates a right. The governance

unction is relevant but not determinative. The answer is direct: the — ramework treats governance as one input alongside rights, e —

iciency, and legitimacy. Chapter 2 de — ines what governance is. Chapter 4 explains what legal conditions enable governance to exist. Chapter 7 explains how law can enable or de- grade governance. The — ramework identi — ies governance as a legal object and develops a method — or analyzing law’s e —

ect on govern- ance. It is structured to work within, not against, anti-discrimina- tion law, rights claims, and constitutional requirements. The practical question is what the — ramework adds to rights analysis, not whether it replaces rights analysis. Anti-discrimination 364 Law and Governance

law asks whether a protected characteristic is the basis

or an adverse action. Rights analysis asks whether a constitutional interest was in-

ringed and whether adequate justi — ication exists. The governance

ramework asks, in addition, what the adverse action does to the governance institution whose authority was exercised. These three questions are complementary, not competing. A governance institution’s exclusion o — a member on the basis o — a protected characteristic violates anti-discrimination law regardless o — what governance — unction the institution per — orms. What the governance analysis adds is the recognition that legal remedies — or anti-discrimination violations should be calibrated, where possible, to avoid destroying governance institutions that per — orm valuable

unctions while achieving the anti-discrimination objective. A tar- geted remedy that compels reinstatement o — the wrong — ully ex- cluded member is di —

erent — rom a systemic remedy that eliminates the governance institution’s authority to exclude any member under any circumstances. The — irst addresses the anti-discrimination vio- lation. The second destroys governance while correcting the viola- tion. The Chapter 12 exclusive inclusion analysis provides a use — ul il- lustration. The three structural re — orms proposed there, stakeholder standing, mission speci — icity, and operational test re — orm, do not override institutional autonomy. They create accountability mech- anisms that enable institutions to be held to their own stated com- mitments without dismantling the governance structures through which those commitments are meant to be pursued. The re — orms are procedural. They do not substitute judicial or regulatory judg- ment — or institutional judgment. They create the conditions under which the institution’s own governance can — unction as its stated values require. Framework Limits 365

The Determinacy Gap Among the limits worth noting is the challenge o — indeterminacy. A

ramework that produces di —

erent prescriptions depending on how the shared problem is characterized, which institution is identi — ied as the governance institution, and how member bene — its and spillo- vers are measured cannot provide — ully determinate legal guidance. Lawyers and judges need rules that produce predictable outcomes across the range o — cases the legal system must resolve. A — ramework that produces di —

erent answers depending on — raming does not im- prove on the common law’s traditional approach o — reasoning — rom principle to case; it adds complexity without adding determinacy. The concern re — lects a genuine — eature o — legal scholarship that proposes — rameworks — or resolving contested cases. Frameworks that depend on the characterization o — inputs are susceptible to stra- tegic manipulation: parties will characterize their arrangements as governance institutions, their exclusionary practices as sanctioning mechanisms, and their private interests as shared problem manage- ment in whatever way the — ramework would support their pre — erred outcome. I — the — ramework cannot constrain those characteriza- tions, it cannot guide legal decision-making. Two responses are available. The — irst is that indeterminacy is comparative. The alternative to the governance — ramework is not a — ully determinate doctrinal analysis that produces single right answers. The alternative is doc- trinal analysis without governance vocabulary, which produces di — -

erent outcomes depending on which doctrinal category the case is assigned to, with no principled method — or choosing among catego- ries. Whether the diamond bourse’s exclusionary practices are ana- lyzed as an antitrust violation, an exercise o — property rights, a pri- 366 Law and Governance

vate contractual arrangement, or an expression o

associational au- tonomy produces very di —

erent legal outcomes, and current doc- trine provides no principled basis — or choosing among those char- acterizations. The governance — ramework does not eliminate char- acterization problems, but it directs attention to a consistent set o —

questions: what is the shared problem, what institution addresses it, what does the legal intervention do to that institution. These ques- tions do not guarantee determinate answers, but they direct charac- terization toward the institutionally relevant issues. The second is that the — ramework’s indeterminacy is concen- trated at the margins, not at the core. The cases where governance analysis produces determinate answers are a substantial portion o —

the domain the

ramework addresses: the diamond bourse’s sanc- tioning mechanism is clearly governance; Kodak’s re — usal to supply parts to competitors is clearly not; the NYSE’s en — orcement o — its own procedural rules is clearly governance-enabling; per se anti- trust liability — or horizontal governance coordination is clearly gov- ernance-degrading. The hard cases exist, but they are hard cases — or every — ramework. The governance — ramework is not more indeter- minate than the alternatives; it is more explicit about what it is ask- ing and why.

Exclusion, History, and Analytical Responsibility The sixth limit addresses the normative orientation o — the theory. By treating exclusion as an institutional — eature that governance quality depends on, the — ramework provides analytical support — or institutional arrangements that have historically been used to ex- clude women, racial minorities, and members o — marginalized groups — rom economic and pro — essional li — e. The diamond bourse Framework Limits 367

was not always an open trading institution. Bar associations, real es- tate boards, and medical licensing bodies operated exclusionary gov- ernance arrangements — or much o — American history in ways that rein — orced racial, gender, and class strati — ication. Treating excluda- bility as a constitutive governance — eature imports a historical legacy that scholars who care about justice should scrutinize, not launder through value-neutral institutional analysis. The concern carries the most political and historical — orce o — the six, and it is not answered by noting that the — ramework distin- guishes legitimate governance exclusion — rom anti-discrimination violations. The concern is not that the — ramework authorizes cur- rent anti-discrimination violations. The concern is that the — rame- work’s analytical sympathy — or exclusion provides cover — or ar- rangements whose exclusionary histories should make them pre- sumptively suspect, and that treating governance exclusion as insti- tutionally — unctional normalizes a practice whose costs have — allen systematically on particular groups. The response requires honesty about what the — ramework does and does not claim. The — ramework treats exclusion as a mechanism requiring jus- ti — ication. It identi — ies excludability as a structural — eature o — club goods, and recognizes that the quality o — certain governance institu- tions depends on the institution’s ability to screen membership. These are descriptive structural claims that in — orm, but do not de- termine, normative conclusions. The analysis o — why diamond bourse governance quality depends on the ostracism mechanism is a claim about causal structure, not a claim that the diamond bourse’s past or present membership practices are just. 368 Law and Governance

 But the response cannot stop there. Structural claims about club goods have normative uses. The observation that governance qual- ity depends on excludability is available to parties who want to maintain exclusionary arrangements and to parties who want to re-

orm them while preserving governance — unction. The — ramework’s prescriptive ambitions, its claim that law should be evaluated by what it does to governance, are not neutral between those uses. A prescriptive — ramework that counsels against legal intervention that degrades governance provides stronger support — or governance- quality arguments against re — orms than — or re — orm arguments against exclusionary arrangements. The honest response acknowledges this asymmetry while iden- ti — ying its limits. First, the — ramework explicitly does not recom- mend legal de — erence to governance institutions as a general matter. It recommends calibrated legal intervention: legal rules should tar- get speci — ic governance — ailures rather than destroying governance institutions while correcting identi — ied problems. The recommen- dation applies to legal intervention on any side o — governance dis- putes, including intervention that addresses exclusionary govern- ance practices. Second, the — ramework’s analysis o — governance deg- radation mechanisms is use — ul — or re — ormers. Understanding how law can disable governance institutions by eliminating their sanc- tioning capacity or mandating open membership is analytically use-

ul — or identi — ying which governance re — orms will achieve their goals and which will produce the compliance-substitution pattern that de — eated thirty years o — shareholder rights re — orms and the 2025 university settlements. Third, and most — undamentally: the — rame- work does not conclude that every governance institution deserves the legal protections that enable it to per — orm its — unctions. It con- cludes that legal analysis should ask what a legal intervention does Framework Limits 369

to governance be

ore applying it. That question is available to law- yers who seek to discipline exclusive governance arrangements, not only to lawyers who seek to protect them.

Two Additional Framework Constraints Two additional limits deserve statement be — ore the chapter con- cludes, both identi — ied by the — ormal analysis in Appendix C. The complementarity requirement. The — ramework’s pro- posals are jointly necessary and cannot be sa — ely adopted in isolation. Property-rule protection o — the exclusion mechanism (the — irst pro- posal) without targeted remedies — or identi — ied governance — ailures (the — ourth proposal) creates blanket immunity: a governance insti- tution with property-rule protection and no mechanism — or chal- lenging speci — ic governance abuses has unchecked authority to ex- clude without accountability. Targeted remedies without property- rule protection leave governance institutions exposed to the Mech- anism 1 conversion that Silver illustrates. The enabling obligation without the governance laundering sa — eguards invites entities to adopt governance trappings — or litigation advantage. The comple- mentarity is not merely a theoretical concern; it is a practical con- straint on legislative and regulatory adoption o — the — ramework. Partial adoption, adopting one element without the others, pro- duces worse outcomes than the status quo because it provides the bene — its o — the — ramework (governance protection, enabling mech- anisms) without the constraints (accountability, targeted remedies, burden-shi — ting). Legislatures and courts implementing the — rame- work must adopt its components as a system, not as a menu — rom which individual items can be selected — or convenience. 370 Law and Governance

The in --- ormational asymmetry and the disclosure comple- ment. Governance quality is private in --- ormation held by institu- tional leadership and not currently subject to any legal disclosure re- quirement. The  --- ramework prescribes legal protection  --- or institu- tions that generate governance surplus, but governance surplus can- not be evaluated without in --- ormation about governance quality, and no existing legal mechanism  --- orces its production. This in --- orma- tional asymmetry creates a structural problem  --- or the  --- ramework's operationalization: the institution claiming governance protection possesses the in --- ormation needed to evaluate its claim, while the party challenging governance or the court evaluating it does not. The  --- ramework requires a disclosure complement: a requirement that institutions claiming governance protection produce evidence o ---

unctional governance through standardized reporting o — the out- put-based criteria. Such a requirement would parallel the SEC’s dis- closure regime — or corporate governance, which makes governance quality partially observable through mandatory reporting o — board composition, committee structure, executive compensation, and au- dit outcomes. For non-corporate governance institutions, a compa- rable disclosure requirement — reporting arbitration caseloads, sanction — requencies, rule-change histories, and monitoring outputs — would create the in — ormational in — rastructure that the — rame- work’s governance evaluation requires at Steps 2, 4, and 5. Without such a requirement, the — ramework’s evaluative method depends on governance in — ormation that no party has adequate incentives to produce accurately, and the governance evaluation remains struc- turally incomplete. Framework Limits 371

The Research Agenda Chapter 15 draws together the book’s analytical threads and states the evaluative payo —

. The — ramework developed across these chap- ters is a vocabulary — or asking institutional questions that legal anal- ysis currently cannot consistently ask. Those questions are: What is the shared problem? What governance institution addresses it? What does the legal rule under analysis do to that institution? Is the e —

ect enabling, disciplining, or degrading? And is the law, in light o — those e —

ects, good, bad, or mixed? The limits, gaps, and open questions this chapter has identi — ied are the places where those ques- tions lead next. Chapter 15: What Governance-Literate Law Would Look Like Five questions constitute the book’s analytical contribution. What is the shared problem? What governance institution addresses it — and is that institution — unctioning in the output-based sense, mean- ing its decision-making governs actual member behavior, its moni- toring generates conduct-governing in — ormation, its sanctioning responds to violations at observable — requencies, and its rules change when circumstances change? What does the legal rule under analysis do to that institution — and does the e —

ect trace to a spe- ci — ic governance dimension (monitoring, sanctioning, decision- making, adjustment) or operate across dimensions simultaneously? Is the e —

ect enabling, disciplining, or degrading — and i — disciplin- ing, does the remedy satis — y both scope calibration (targeting the governance — unction that — ailed) and intensity calibration (imposing governance costs proportionate to the governance bene — it o — correc- tion)? And is the law, in light o — those e —

ects, good, bad, or mixed — evaluated with recognition that the method’s quantitative preci- sion is highest in legislative and regulatory design settings, where in — ormational investment can be made ex ante, and primarily struc- 374 Law and Governance

tural in litigation settings, where the governance evaluation organ- izes the inquiry without — ully resolving the quantitative questions at Steps 4 and 5? These questions are simple to state and di —

icult to ask. They are di —

icult to ask not because they require exotic knowledge but because legal analysis is trained not to ask them. Le- gal training teaches lawyers to see disputes, identi — y parties, locate rights, and apply doctrine. It does not train lawyers to see govern- ance as an object o — analysis: the ongoing institutional architecture through which groups manage shared problems over time. The re- sulting blind spot is not a product o — bad — aith. It is built into the

ramework, and — rameworks with built-in blind spots produce sys- tematic errors. A legal system that incorporated governance analysis into its standard toolkit would reach di —

erent conclusions in speci — ic, iden- ti — iable cases, and those di —

erent conclusions would produce better outcomes by the legal system’s own criteria. The three cases that opened this book, Silver v. New York Stock Exchange, the Michigan State University abuse scandal, and the Israeli daycare experiment, illustrate what di —

erent conclusions would look like. Each case was resolved by legal actors who were competent, well-intentioned, and analytically sophisticated within the — ramework their training pro- vided. Each resolution produced governance — ailure as a predictable byproduct o — that — ramework’s inability to account — or what it was damaging. Error! Use the Home tab to apply Heeading 1 to the text that you want to appear here. 375

Chapter 15. Governance-Literate Law The most direct way to show what governance-literate law would produce is to revisit three cases that appeared in the book’s intro- duction — Silver, Michigan State, and the Israeli daycare experi- ment — and demonstrate how the — ramework changes the analysis in each. These are not hypothetical applications. Each case has a de- veloped doctrinal record, and in each case the governance — rame- work identi — ies institutional consequences that the actual legal anal- ysis missed.

Silver Revisited The Supreme Court in Silver v. New York Stock Exchange held that a stock exchange’s expulsion o — a non-member broker without notice or hearing violated the Sherman Act’s limitations on private re- straints o — trade. The Court was protecting Harold Silver — rom arbi- trary exclusion and ensuring that the exchange’s governance author- ity would not operate as an instrument o — commercial oppression. On the — acts be — ore it, the Court’s concern was legitimate. The NYSE had cut Silver’s wire connections without explanation, without prior notice, and without any procedure by which Silver could have identi — ied or contested the charges against him. A governance-literate court would have asked a di —

erent ques- tion — irst. Silver had not been treated — airly as a party. That much was clear. The harder question was what the NYSE’s other members would lose when exclusion decisions became subject to judicial re- vision. The answer is precise: every remaining member’s coopera- tive production o — market integrity depended on the governance in- stitution’s credible en — orcement capacity. That capacity was not a background condition. It was the mechanism through which the 376 Law and Governance

NYSE’s sel

-regulatory — unction generated value — or participants and, through market integrity, — or the investing public. Paul Ma- honey documented what its erosion cost: the exchange’s authority to discipline members quickly and credibly, the — eature that sus- tained market integrity, was systematically undermined by the pro- cedural overlay that Silver initiated. What the governance-literate court would have done is cali- brate the remedy to the identi — ied governance — ailure. The — ailure was speci — ic: no notice, no charges, no hearing. The remedy should have been equally speci — ic: require that the NYSE provide those pro- cedural protections — or expulsion decisions a —

ecting non-member access. What it should not have done is apply a broad holding that exchange disciplinary authority requires administrative due process compliance — or all en — orcement actions. The broad holding con- verted a remedy — or a speci — ic governance — ailure into a structural degradation o — the entire governance institution. The Securities Acts Amendments o — 1975 took that degradation to its logical con- clusion, requiring SEC pre-approval o — all exchange rule changes. The exchange, which had governed itsel —

or a century through swi — t internal authority, became a regulated entity. Mahoney’s research shows the governance cost was real and lasting. Governance-literate law would have imposed due process protections — or Silver while protecting the en — orcement architecture on which every other member’s cooperation depended. The lesson is about calibration, not about de — erence. Govern- ance-literate law is calibrated, intentional law that asks whether a legal remedy addresses the speci — ic governance — ailure or whether it destroys governance capacity while correcting an identi — ied abuse. The appropriate question is not whether courts should de — er to pri- vate institutions, but what remedy actually solves the problem at Error! Use the Home tab to apply Heeading 1 to the text that you want to appear here. 377

hand. Silver needed notice and a hearing requirement. It did not need the — ull procedural apparatus o — administrative adjudication. The gap between those two remedies is the space that governance-lit- erate law occupies.

Michigan State Revisited The Department o — Education’s response to the Michigan State Uni- versity abuse scandal illustrates governance blindness in the regula- tory context. The special investigative report had identi — ied a spe- ci — ic structural de — ect: athletic medical sta —

reported to the athletic department rather than to an independent medical authority, which meant complaints about a treating physician were evaluated by the department whose institutional reputation depended on suppress- ing them. The governance — ailure was a monitoring — ailure. The in- stitution’s architecture guaranteed that monitoring would produce

alse negatives. In — ormation about misconduct would not reach par- ties with authority to act on it because the parties with monitoring authority had a structural incentive to suppress what they moni- tored. The — ederal remedy required revised policies, new Title IX co- ordinators, mandatory training, and retrospective complaint re- view. That speci — ic structural — eature, the reporting line between athletic medicine and the athletic department that the investigative report had identi — ied as the source o — the — ailure, went unre — ormed. Within a year, the interim president appointed to signal institu- tional change was himsel —

orced to resign a — ter making statements publicly dismissive o — the victims’ experiences. The governance ar- chitecture that produced the original crisis was untouched. 378 Law and Governance

 Governance-literate  --- ederal en --- orcement would have required structural re --- orm as a condition o ---  the settlement. Not a particular outcome, not a mandated result, but a changed governance architec- ture. Athletic medical sta ---

should report to a medical authority in- dependent o — the athletic department. That requirement does not dictate any substantive decision by the medical authority. It removes the con — lict o — interest that made the monitoring system structurally incapable o — generating accurate in — ormation about misconduct. This is the governance-disciplining intervention that Chapter 7 dis- tinguishes — rom governance-degrading intervention: it targets the speci — ic identi — ied — ailure, the monitoring — ailure created by the re- porting structure, without dismantling the broader governance in- stitution. A compliance mandate that leaves the monitoring — ailure intact will produce compliance theater while the governance — ailure continues. A structural mandate that repairs the monitoring — ailure changes what the institution can produce. These are not the same remedy, and the di —

erence is not an administrative detail. Federal en — orcement against universities (Chapter 12) repeats the pattern. Settlement agreements that require policy adoption without governance re — orm (training requirements, additional co- ordinators, demographic reporting) treat governance — ailures as compliance de — icits. Governance — ailures are not compliance de — i- cits. They are structural conditions that produce predictable outputs regardless o — the stated policies in place. Harvard had policies against antisemitism be — ore the — ederal en — orcement campaign. The policies did not prevent systematic exclusion — rom institutional channels be- cause the governance architecture that determines who controls those channels was not addressed by the policies. Policies prescribe; governance structures determine who decides what the policy Error! Use the Home tab to apply Heeading 1 to the text that you want to appear here. 379

means and what counts as compliance. Federal en

orcement that re- quires better policies without re — orming governance structures will reach the same result that three decades o — corporate governance re-

orm reached: compliance activity that leaves the structural condi- tions unchanged.

The Israeli Daycare Revisited The Israeli daycare experiment does not involve litigation or legis- lation. Its lesson is about regulatory design. When Uri Gneezy and Aldo Rustichini studied the introduction o —

ines — or late pickup at ten daycare centers in Hai — a, they — ound that the — ine doubled the incidence o — tardiness and that the elevated tardiness rate persisted a — ter the — ine was removed. Behavioral economists cite the experi- ment as evidence that monetary incentives crowd out intrinsic mo- tivation. The governance analysis adds a dimension that the behav- ioral economics — raming leaves implicit. The daycare community had a governance institution be — ore the

ine. It was in — ormal and uncodi — ied, but it had all — our — unctional elements: a recognized norm governing parental behavior (pickup by closing time as a social obligation), a monitoring mechanism (teachers observing and — orming impressions o — parents who were regularly late), a sanctions mechanism (social disapproval, parental guilt, awkward interactions), and an adjustment mechanism (in — or- mal renegotiation when parents had legitimate di —

iculties). This governance institution was producing near-complete compliance. It was producing compliance through social rather than legal mecha- nisms, which is why it produced no — ormal records and why the re- searchers who introduced the — ine did not characterize it as a gov- ernance institution. But it was governing. 380 Law and Governance

The  --- ine did not supplement this governance institution. It re- placed it. The  --- ine converted parental guilt, the moral obligation that had sustained compliance, into a market transaction. Once par- ents understood that late pickup had a price, the social norm disin- tegrated. The moral community that had en --- orced the norm no longer had a norm to en --- orce. And unlike a regulation that an agency can simply repeal, a social norm destroyed by a monetary interven- tion does not reconstitute when the intervention ends. The re- searchers removed the  --- ines. The norm did not return. The govern- ance institution was gone permanently.
A governance-literate designer would  --- irst ask what institu- tional mechanism already governs the behavior, and only then ask how regulation can supplement rather than replace it. Identi --- ication leads directly to a di ---

erent design question: not “how do we add an en — orcement mechanism?” but “how do we supplement the existing governance institution without displacing it?” Possible answers in- clude measures that strengthen the existing social norm rather than replacing it: making late pickup more visible, providing public ac- knowledgment o — consistent compliance, creating social rein — orce- ment — or the existing norm rather than a market substitute — or it. Whether any o — these would have worked better than the — ine is an empirical question the researchers’ study does not answer. The gov- ernance question is prior: a regulatory designer who identi — ies the existing governance institution — irst will ask di —

erent questions than a designer who sees only a problem to be solved with an incen- tive. Error! Use the Home tab to apply Heeading 1 to the text that you want to appear here. 381

The Pattern o

Legal Blindness Three cases. Three institutions. Three governance — ailures pro- duced by legal actors who were doing their jobs within the — rame- work their training provided. The exchange court that — ocused on Harold Silver’s injury as a litigant be — ore the court. The — ederal agency that — ocused on policy compliance outputs. The researchers who introduced an incentive without identi — ying what it was sup- plementing. In none o — these cases was the legal actor incompetent. In all three, the legal actor was blind to the governance institution that the legal action was about to damage or destroy. The blindness — ollows a structure. Legal analysis is calibrated to party-centered relationships: a party with a grievance, an institution that caused it, a remedy that corrects the wrong. The governance institution, the organized system through which all the other par- ticipants manage their shared problem, is not a party. It appears, i —

at all, as the context within which the litigation arose, not as an ob- ject o — the analysis itsel — . Courts see the expelled member but not the remaining members whose cooperation depends on the expulsion authority. Agencies see the individual incidents o — abuse but not the reporting structure that systematically suppressed in — ormation about them. Researchers see the behavior they want to change but not the governance mechanism through which the behavior had been governed. Making the governance institution analytically visible integrates governance into the standard criteria by which law is evaluated, alongside rights, e —

iciency, and anti-discrimination en — orcement. The — ramework identi — ies governance as a value that legal analysis currently cannot weigh because it currently cannot see. Complete legal analysis requires visibility into all the values at stake. An anal- ysis that cannot weigh governance will consistently undervalue it. 382 Law and Governance

The result is not neutral omission; it is systematic governance deg- radation, institution by institution, case by case, over time.

A Prescriptive Agenda: Organizing Re

orm by Implementation Authority The book’s application chapters have embedded speci — ic re — orm pro- posals in their institutional analyses. What they have not done is assemble those proposals into a uni — ied prescriptive agenda. The agenda organizes around three principal tiers, di —

erentiated by the legal authority required to implement each. A crosscutting principle governs all three tiers: the governance method’s — irst and best appli- cation is in legislative and regulatory design, where the in — orma- tional investment that Steps 4 and 5 demand can be made ex ante through hearings, commissioned studies, and notice-and-comment proceedings. Its litigation application is structural — it organizes the governance inquiry so that courts ask the right questions about institutional e —

ects — rather than quantitative. Re — orm proposals that can be implemented through legislative or regulatory design should be, because the method is strongest in that setting. Where governance analysis enters litigation, courts should apply it at Steps 1 through 3 and Step 6, where legal characterization produces pre- cise — indings, and treat Steps 4 and 5 as structural considerations that in — orm the analysis without requiring the quantitative precision that only legislative-design settings can support. What courts can do today without legislative change. Courts can apply governance analysis as a variable in the standard multi- — actor balancing — rameworks that already govern many o — the doctrinal areas this book addresses. The antitrust rule o — reason al- ready requires courts to assess whether a cooperative arrangement Error! Use the Home tab to apply Heeading 1 to the text that you want to appear here. 383

serves legitimate purposes and whether any restraint on competi- tion is reasonably necessary to achieve those purposes; courts can incorporate governance — unction into the assessment o — legitimate purposes and calibrate the scope o — remedies to speci — ic governance

ailures rather than to the existence o — governance institutions as such. The antitrust inquiry into whether the alleged restraint is rea- sonably necessary to achieve the legitimate purpose shares the gov- ernance method’s structure: a restraint is reasonably necessary when it targets a speci — ic governance — ailure without imposing unneces- sary costs on the governance institution. Courts applying the rule o —

reason can make that analysis explicit rather than leaving it implicit in proportionality judgments. Courts applying Delaware’s review standards to corporate board decisions can make explicit what the Delaware Supreme Court’s doctrine already implies: that the business judgment rule — unctions as a governance subsidy reducing the private cost o — producing cor- porate governance, and that withdrawing the rule in circumstances where governance quality is at stake should be calibrated to correct the identi — ied — ailure rather than to impose governance costs across all board decisions. The business judgment rule operates as an un- targeted quantity subsidy: it protects every board decision at a — re- quency and cost that do not correspond to the variance in govern- ance quality across decision-making circumstances. Governance-lit- erate Delaware doctrine would calibrate the subsidy to the govern- ance risk. High-governance-risk decisions (con — licts o — interest, in- terested transactions, situations where board discretion is particu- larly likely to be captured by private interests) should receive less de — erential review than low-governance-risk decisions. The busi- ness judgment rule as currently applied does not make this distinc- tion. A governance-literate doctrine would. 384 Law and Governance

Courts reviewing nonpro --- it trustees' compliance with  --- iduciary duties should ask whether the institutional structures make compli- ance possible, not only whether  --- ormal compliance standards are satis --- ied, because governance architecture determines what compli- ance can mean in practice. A university trustee board that has no mechanism  --- or reporting governance  --- ailures to parties with au- thority to correct them cannot  --- ully comply with  --- iduciary duties even i ---  every trustee individually acts in good  --- aith. The governance structure matters not as a de --- ense to breach but as a constraint on whether breach can be detected and corrected.
None o ---  these applications requires a new legal rule. They re- quire asking a question that existing doctrine does not consistently ask: what does this legal action do to the governance institution whose decisions I am reviewing? That question is available to courts today and within the scope o ---  existing doctrine across antitrust, cor- porate, administrative, and nonpro --- it law.
What administrative agencies can do under existing au- thority. Three agency actions would produce substantial govern- ance improvements without new legislation.
The IRS has authority under existing doctrine to require that tax-exempt educational institutions satis --- y a governance-based op- erational test as a condition o ---  tax exemption, parallel to the com- munity bene --- it standard the IRS developed  --- or nonpro --- it hospitals through administrative interpretation rather than statutory change. A re --- ormed operational test would require educational institutions to demonstrate governance structures that prevent systematic iden- tity-based exclusion: speci --- ic and measurable commitments, stand- ing  --- or a ---

ected students and — aculty to report governance — ailures, and board composition rules that prevent inde — initely sel — -perpetu- Error! Use the Home tab to apply Heeading 1 to the text that you want to appear here. 385

ating control. The test would not prescribe outcomes or require in- stitutions to adopt particular viewpoints. It would require only that the governance structure be capable o — producing accountability. An institution could maintain its current policies while demonstrating that its governance architecture would permit modi — ication o — those policies in response to demonstrated exclusion o — protected constit- uencies. The operational test would make governance architecture a condition o — tax exemption without centralizing control o — insti- tutional content. The SEC retains authority under the Securities Exchange Act to revisit the regulatory — ramework — or proxy advisors, whose concen- tration (two — irms, Institutional Shareholder Services and Glass Lewis, control advisory services to the large majority o — institutional investors) created the adverse selection dynamic that Chapter 11 identi — ies as the clearest unintended consequence o — mandatory proxy voting rules. A rule requiring proxy advisors to disclose con-

licts o — interest, to permit issuer response to dra — t recommenda- tions be — ore publication, and to make their methodological assump- tions subject to periodic public comment would address the govern- ance — ailure that proxy advisor concentration has produced without requiring legislative revision o — Dodd-Frank’s say-on-pay — rame- work. The disclosure requirement would not eliminate proxy advi- sor in — luence; it would make that in — luence visible and subject to the kind o — scrutiny that characterization as a knowledge institution re- quires. FINRA retains rulemaking authority over expulsion procedures

or member broker-dealers. The Alpine Securities litigation illumi- nated the constitutional stakes o — private delegated authority, but the governance analysis points to a more immediate re — orm: FINRA’s expulsion procedures should be calibrated to distinguish 386 Law and Governance

the governance-discipline

unction (maintaining market integrity through credible en — orcement) — rom the governance-abuse — unction (using expulsion authority against member — irms — or competitive rather than regulatory purposes). The calibration mechanism is pro- cedural: di —

erentiating the procedural requirements — or expulsion decisions that a —

ect member — irms’ access to markets — rom expul- sion decisions that a —

ect member — irms’ relationships with speci — ic customers. The — ormer requires substantial procedural protection because it threatens the governance capital that member — irms have invested in market access. The latter requires less elaborate protec- tion because it does not threaten the — irm’s status as a market partic- ipant. This di —

erentiation re — lects the governance analysis directly: it targets protection to the margin where governance capital is most vulnerable to abuse while avoiding procedural requirements that do not correspond to the governance risk. What legislation would most improve governance out- comes. Three legislative re — orms address the structural — eatures that administrative and judicial action cannot reach. Stakeholder standing in nonpro — it law is the re — orm most di- rectly supported by the book’s analysis. State legislatures should amend nonpro — it corporation statutes to extend derivative standing to stakeholders with sustained institutional ties, currently enrolled students and currently employed — aculty, — or en — orcement o —

iduci- ary duties against educational institution boards. The corporate de- rivative action, adapted — or the nonpro — it context with procedural sa — eguards against harassment litigation, would create the account- ability mechanism that the attorney general en — orcement model cannot provide at scale. Every state that has created a meaning — ul private right o — action — or other governance — ailures has — ound that the threat o — en — orcement produces governance improvement even Error! Use the Home tab to apply Heeading 1 to the text that you want to appear here. 387

when litigation rates remain low. The deterrent e

ect o — standing is the governance-relevant e —

ect. Federal law could extend compara- ble standing under the civil rights statutes — or governance challenges based on institutional — ailure to prevent discrimination, creating a private right o — action without replacing the attorney general’s en-

orcement authority. A best-available-evidence doctrine — or empirical scholarship o — -

ered in litigation would address the governance — ailure in legal knowledge production that Chapter 13 identi — ies. Federal courts ap- plying Daubert to expert testimony already evaluate the reliability o —

the processes through which expert knowledge was produced. Ex- tending that analysis to legislative and adjudicative — acts presented through law review citations, asking whether the screening process that certi — ied the scholarship gives reasonable assurance o — its meth- odological reliability, would create incentives — or knowledge insti- tutions to improve governance quality without mandating any par- ticular organizational — orm. Scholarship certi — ied by institutions with robust screening would carry greater evidentiary weight; scholarship certi — ied by institutions with demonstrated governance

ailures would be subject to increased scrutiny. The market- — or-ep- istemic-goods problem that Chapter 13 identi — ies as a structural — ea- ture o — the current system responds to demand-side incentives: i —

courts treat governance-certi

ied scholarship di —

erently, knowledge institutions have reason to improve their governance. The imple- mentation would require — ederal courts to develop standards — or evaluating the screening process that produced scholarship o —

ered in support o — legislative and adjudicative — acts, a task that parallels the Daubert — ramework’s evaluation o — expert testimony. The spe- 388 Law and Governance

ci

ic criteria could be developed through notice-and-comment rule- making by the Judicial Con — erence or through common law devel- opment as courts address the question across cases. A governance impact assessment requirement, analogous to the environmental impact statement requirement under NEPA — or — ed- eral agency action with signi — icant environmental e —

ects, would re- quire agencies proposing rules with signi — icant e —

ects on private governance institutions to analyze those e —

ects be — ore the rule takes e —

ect. The assessment would not give governance analysis veto power over regulation. It would require that governance e —

ects be analyzed, documented, and part o — the administrative record availa- ble — or review. Environmental impact assessment did not eliminate

ederal agency action that damages the environment; it made envi- ronmental costs visible and required that they be weighed. Govern- ance impact assessment would do the same — or governance costs. The requirement would be limited to rules with identi — ied govern- ance e —

ects: rules a —

ecting the structure o — private governance in- stitutions, rules a —

ecting the sanctioning capacity o — private institu- tions, rules a —

ecting in — ormation — low within governance institu- tions. A rule addressing a wholly separate problem that happens to a —

ect a governance institution incidentally would not trigger the re- quirement. The core governance impact assessment questions — what private governance institutions do these rules a —

ect, what governance dimensions do they a —

ect, what are the measured gov- ernance outcomes in comparable circumstances — are answerable questions that an administrative record can support. Error! Use the Home tab to apply Heeading 1 to the text that you want to appear here. 389

The Research Agenda This Book Opens The — ramework is a beginning, not a terminus. Each application chapter identi — ied governance — ailures that the — ramework explains and re — orm targets that the — ramework suggests. The research agenda that — ollows — rom those explanations is substantial. The most pressing empirical question is whether governance degradation has the e —

ects on governance output that the theoreti- cal — ramework predicts. Chapter 10’s account o — the NYSE’s post- Silver trajectory is the most — ully documented case study: Mahoney traced the degradation o — en — orcement capacity through the multi- decade doctrinal arc, and Alpine Securities brought the constitu- tional stakes into — ocus as o —

1. But a comprehensive empirical account o — the relationship between legal intervention intensity and governance quality, across multiple institutional domains, would signi — icantly strengthen the — ramework’s prescriptive claims and could identi — y conditions under which governance is more or less resilient to legal degradation. The optimal calibration o — governance-disciplining law is in- completely developed. Chapter 7 distinguishes legal rules that ena- ble governance, degrade it, and discipline it, but the criteria — or dis- tinguishing discipline — rom degradation remain contested in speci — ic applications. When does a procedural requirement targeting a spe- ci — ic governance abuse cross the line into structural degradation o —

governance capacity? The answer is likely institution-speci

ic and context-dependent in ways that the — ramework does not yet make precise. Research on speci — ic doctrinal areas, including antitrust challenges to cooperative exclusion, judicial review o — pro — essional association membership decisions, and — ederal regulatory oversight o — sel — -regulatory organizations, would develop criteria — or calibra- tion that the — ramework’s current level o — generality cannot supply. 390 Law and Governance

 The relationship between governance and markets is less  --- ully analyzed than the relationship between governance and law. Private governance institutions exist in market environments that shape their evolution in ways that legal intervention does not always con- trol. Demutualization o ---  stock exchanges, concentration in the proxy advisory industry, and consolidation among arbitration ser- vice providers are market developments with governance implica- tions that the book has addressed but not  --- ully analyzed. Govern- ance-literate analysis o ---  market structure, asking how market con- ditions a ---

ect the production and distribution o — governance club goods, extends the — ramework into domains where legal analysis has not traditionally asked governance questions. The governance o — governance is the deepest remaining prob- lem. Several chapters have identi — ied cases in which governance in- stitutions — ail because governance o — the institution itsel — has been captured or disabled: the sel — -perpetuating board that excludes ac- countability, the law review editorial structure that cannot evaluate methodological quality, and the exchange whose disciplinary — unc- tion was converted into a supervised administrative process. In each o — these cases, the question is not only what law did to a governance institution but what governance — ailures produced governance — ail- ure in the institution itsel — . Ostrom’s design principles identi — y con- ditions associated with long-run institutional survival, but they do not — ully speci — y how governance institutions maintain their own governance quality over time, when they are embedded in legal sys- tems that may or may not support that maintenance. Error! Use the Home tab to apply Heeading 1 to the text that you want to appear here. 391

Governance as a Legal Variable Legal scholarship has de — ined rights. It has speci — ied e —

iciency. It has developed distributive principles. It has articulated legitimacy crite- ria. It has not de — ined governance in terms precise enough to carry doctrinal argument across — ields, to generate predictions about what legal rules will do to institutions, and to produce a method — or eval- uating law by its governance e —

ects. That is the gap this book — ills. The gap matters not because governance always takes priority over rights, e —

iciency, distribution, or legitimacy. It does not. The gap matters because a legal variable that cannot be precisely speci-

ied cannot be meaning — ully weighed. Courts that recognize govern- ance as a value but cannot de — ine it will weigh it imprecisely, incon- sistently, and o — ten not at all. The business judgment rule, the anti- trust rule o — reason, the FAA’s pro-arbitration policy, and common law de — erence to voluntary associations have — unctioned as intuitive governance subsidies — or decades, reducing the cost o — producing governance without naming what they were doing or explaining why it mattered. The NYSE cases, the corporate governance re — orm cycle, the university accountability crisis, and the law review screen- ing problem are all cases in which legal analysis that could name the governance stake could have reached more precise, more durable, and more e —

ective conclusions. Governance-literate law is not more de — erential law. It is not less activist law. It is more complete law: law that sees the governance institution among the values at stake, speci — ies what is at risk, and calibrates its intervention to what the identi — ied — ailure actually re- quires. Harold Silver deserved notice and a hearing. The NYSE’s governance in — rastructure deserved to survive Silver’s remedy. Both things were true. A court that could see only Silver could not reach a conclusion consistent with both. A governance-literate court can. Conclusion and Research Agenda This book began — rom a problem in legal analysis that is easy to state and di —

icult to correct. Law is very good at seeing disputes. It is much less good at seeing the institutions that make cooperation pos- sible be — ore disputes arise and that must continue — unctioning a — ter any particular dispute is resolved. The result is a persistent blind spot in legal cognition. Courts, regulators, and scholars o — ten ask whether a rule vindicates a right, deters misconduct, or — its within a doctrinal category. They ask less o — ten what that rule does to the governance institution through which a group manages a shared problem over time. This book’s central claim has been that the miss- ing question matters, and that legal analysis cannot judge law — ully until it can ask it. Making governance cognizable as a legal object— one that courts, agencies, and legislatures can see, describe, and weigh—is the — irst step toward judgment that accounts — or institu- tional e —

ects. Part I supplied the missing object o — analysis. Governance, as this book has used the term, is the organized system by which a group manages a shared problem over time. That de — inition was de- signed to travel across doctrinal — ields because the phenomenon it identi — ies does not belong to any one o — them. Governance is not corporate governance only, not administrative governance only, not Governance-Literate Law 393

commons governance only, and not merely a synonym

or regula- tion, management, adjudication, or social order. It is a distinct insti- tutional — orm with minimum — unctional requirements: decision- making, monitoring, sanctions, and adjustment. I — any one o — those elements is disabled, governance does not merely weaken. It — ails in a speci — ic and recognizable way. Part I also showed that governance does not — loat — ree o — law. Governance may arise without charter or statute, but law deter- mines whether governance can — orm, persist, adapt, and remain ac- countable in ways legal analysis can recognize. The six legal condi- tions identi — ied here, permission to organize, membership bounda- ries, internal decision procedures, external en — orceability, — iduciary structure, and stakeholder standing, are not su —

icient to guarantee success — ul governance, but they are necessary to sustain it in legally legible — orm. Without them, governance becomes — ragile, opaque, or practically impossible. That point matters because much o — legal doctrine proceeds as though governance were either naturally pre- sent or analytically irrelevant. It is neither. Part II developed the theoretical payo —

o — that de — inition. Gov- ernance institutions o — ten have the structure o — club goods: they de- pend on excludability, are — inanced and maintained by members, and generate bene — its whose — ull value members cannot capture — or themselves. That structure explains both why governance is chron- ically undersupplied and why law matters so much to its production. Some rules reduce the private cost o — producing governance and thereby enable it. Other rules disable the sanction mechanism, un- dermine adjustment, remove accountability, or substitute compli- ance — orm — or governance substance. And some rules do something more di —

icult and more valuable: they discipline governance. They 394 Law and Governance

correct a speci

ic governance — ailure without destroying the institu- tional capacity that makes governance possible in the — irst place. That distinction between enablement, discipline, and degradation is the book’s principal evaluative contribution. The seven-step method was built to make that contribution us- able. Identi — y the shared problem. Identi — y the governance institu- tion. Speci — y the legal conditions o — its existence. Describe the mem- ber bene — its it supplies. Assess the spillovers it generates. Analyze how law acts on it. Then evaluate the law as enabling, disciplining, degrading, or mixed. The method does not promise mechanical an- swers. It does something more realistic and more important. It

orces legal analysis to ask institutionally relevant questions in a sta- ble sequence, and it makes governance e —

ects visible where ordinary doctrinal analysis leaves them obscure. The application chapters showed why that visibility matters. In private trading networks, the legal system has repeatedly misread governance as mere party-centered exclusion, overlooking the shared en — orcement in — rastructure on which cooperation depends. In exchange regulation, Silver demonstrated how a legal interven- tion addressing a real abuse can nevertheless degrade the sanction- ing — unction on which market integrity rests. In corporate law, the

ramework clari — ied why some doctrines operate as governance sub- sidies while others create compliance activity without governance repair. In universities and nonpro — its, it identi — ied a standing and accountability void that policy mandates cannot solve because the governance architecture producing the — ailure remains unchanged. In knowledge institutions, the — ramework turned re — lexively on le- gal scholarship itsel — , asking what governance conditions make legal knowledge reliable enough — or courts, agencies, and scholars to Governance-Literate Law 395

trust. Across all o

these domains, the conclusion was not that gov- ernance should be protected — rom law. It was that law should be judged by what it does to governance. This book demonstrates one such judgment re — lexively in its own production. The book argues that law reviews screen legal knowledge poorly and that this epistemic — ailure matters to the reli- ability o — law review-certi — ied claims. Yet the book itsel — appears through a process that depends — undamentally on law review-style institutional certi — ication. The manuscript itsel — operates under the conditions it critiques. The de — ense is not that the critique does not apply here. The de — ense is that governance — ailures and governance value can coexist in the same institution. Law reviews per — orm real screening and supply real value to legal knowledge production. They also predictably degrade the quality o — that screening by placing ed- itorial authority in student hands and by making publication incen- tives rather than epistemic reliability the primary metric o — success. Acknowledging the contradiction is not weakness but a require- ment o — any institution-centered analysis. An institutional analysis that cannot see its own institutional location is not complete analy- sis. This book is subject to the same scrutiny it applies to other in- stitutions. I — its central claims are to be believed, readers must ask whether the governance conditions that produced this manuscript support or undermine the knowledge it attempts to convey. This book has identi — ied both the value o — governance analysis and its limits. Governance language can be used to launder private power. Exclusion can sustain cooperation, but exclusion has also been a central mechanism o — domination, cartelization, and discrim- ination. Governance institutions can produce valuable spillovers, but they can also impose coercion on members and outsiders. And 396 Law and Governance

the

ramework is not — ully determinate in hard cases. These chal- lenges do not undermine the — ramework; they identi — y where — uture scholarship must extend it. A — ramework that asks what law does to governance must also ask who governs, — or whose bene — it, at whose expense, and under what constraints o — rights, antitrust, anti-dis- crimination law, and democratic accountability. The mature version o — governance analysis will not be one that ignores these tensions. It will be one that can incorporate them without losing sight o — insti- tutional capacity itsel — . The prescriptive implication is modest in one sense and ambi- tious in another. It is modest because the — ramework supplements rights analysis, wel — are analysis, antitrust analysis, and — iduciary analysis by adding a dimension they presently miss. But it is ambi- tious because that additional dimension changes the evaluation o —

amiliar legal rules. A court that asks what a remedy does to a gov- ernance institution may choose a narrower and better-calibrated remedy than one — ocused only on the immediate plainti —

. A regula- tor that asks how a compliance mandate a —

ects internal monitoring and adjustment may design a di —

erent rule than one — ocused only on

ormal outputs. A legislature that sees governance as a legally con- stituted good may create enabling rules where none now exist. Gov- ernance analysis is an expansion o — legal judgment, not a substitute

or it. It is a way o — making law more institutionally literate. Conclusion 397

Conclusion The — ramework developed here is a beginning, not a terminus. Its value now depends on whether — uture work can speci — y, test, and extend it across doctrinal and empirical settings the present book could only open. Among the questions this — ramework opens, sev- eral deserve early attention. The — irst priority is to measure governance degradation empir- ically. The most immediate empirical question is whether the deg- radation mechanisms identi — ied here reliably reduce governance quality, and under what conditions. The NYSE a — ter Silver is the best-developed case study in this book, but it should not remain the only one. Future work must test the relationship between interven- tion intensity and governance output across exchanges, pro — essional associations, nonpro — its, digital organizations, and commons insti- tutions. The second task is to speci — y the calibration line between disci- pline and degradation. The hardest normative problem is not iden- ti — ying obvious enablement or obvious destruction. It is determin- ing when a remedy targeted at a real governance abuse becomes broad enough, costly enough, or rigid enough to impair the institu- tion’s core — unctions. That line is likely domain-speci — ic. Antitrust, nonpro — it standing, pro — essional discipline, sel — -regulatory organi- zations, and university oversight each require their own calibration criteria. The third project must develop disclosure tools — or governance quality. The — ramework depends on in — ormation that governance institutions o — ten possess privately and have little incentive to dis- close. I — courts or agencies are to evaluate governance claims seri- 398 Law and Governance

ously, scholars must develop disclosure metrics that make govern- ance quality observable: sanction — requency, monitoring outputs, rule-change histories, arbitration patterns, stakeholder challenge mechanisms, and comparable indicators. Without an in — ormational complement, governance analysis will remain strongest in theory and weakest in application. The — ourth task is to study the governance o — governance itsel — . Several o — this book’s most important — ailures are second-order — ail- ures: institutions — ail because the mechanisms that are supposed to maintain their own governance quality have themselves been cap- tured, hollowed out, or disabled. Sel — -perpetuating nonpro — it boards, law reviews unable to evaluate the scholarship they certi — y, and sel — -regulatory bodies trans — ormed into compliance bureaucra- cies are all examples. Future work must ask how governance insti- tutions preserve the quality o — their own decision-making, monitor- ing, sanctions, and adjustment over time. The — i — th need is to extend the — ramework comparatively and historically. The analysis here has been largely American in its legal

ocus, even when the institutional examples were not. Comparative work must examine how di —

erent legal systems supply or deny the six legal conditions, and whether some systems are more hospitable than others to governance — ormation and adaptation. Historical work must trace the long-run interaction between legal change and governance change, especially in periods when law reorganized whole sectors o — institutional li — e. The sixth priority is to integrate governance analysis with rights and equality more — ully. The relationship between governance au- tonomy and anti-discrimination law, labor rights, constitutional protections, and procedural — airness remains only partly developed here. Future work must not treat these as external exceptions to Conclusion 399

governance analysis. It must treat them as part o

the internal design problem: how to preserve governance capacity while preventing governance institutions — rom becoming vehicles o — exclusion, coer- cion, or domination. The seventh project should analyze market structure as a varia- ble in governance quality. This book has — ocused more on how law a —

ects governance than on how markets do. Yet demutualization, concentration, digital intermediation, and changes in exit options all alter the cost o — exclusion and the supply o — governance goods. A complete governance theory must connect legal conditions to mar- ket conditions and ask how both together shape the production, re- silience, and erosion o — governance. The eighth task must apply the — ramework to emerging institu- tional — orms. DAOs, plat — orm moderation systems, private accredi- tation bodies, transnational commercial networks, and AI-mediated rule systems all present governance structures that do not — it older doctrinal categories cleanly. These are not peripheral cases. They are likely to be the settings in which governance analysis proves either most use — ul or most inadequate. The next generation o — research must test the — ramework where institutional — orm is changing — ast- est. The broader claim that emerges — rom this research agenda is simple. Governance is a legal object worthy o — sustained study in its own right. Once governance is seen clearly, many disputes that once appeared to be simple two-party con — licts, episodic, or purely doc- trinal turn out to be contests over institutional capacity. The task ahead is not merely to protect governance, and not merely to criti- cize it. It is to learn how law can recognize, enable, discipline, and, when necessary, restructure governance without blinding itsel — to 400 Law and Governance

the institutional consequences o

its own interventions. That is the inquiry this book opens. It is also the one that must — ollow it. Legal analysis has learned to see rights, incentives, and remedies with great precision; its next task is to learn to see governance, and to judge law by what law makes o — it. Notes and Re — erences

1 Silver v. New York Stock Exchange, 373 U.S. 341 (1963). 2 See generally Jessica Gavora, Sexual Assault on Campus: A Frustrating Search — or Justice (Bombardier Books 2017) (documenting — ailures in Michigan State’s handling o — reports against Larry Nassar). 3 Uri Gneezy & Aldo Rustichini, A Fine is a Price, 29 J. Legal Stud. 1 (2000). 4 Silver v. New York Stock Exchange, 373 U.S. 341, 364-65 (1963). The Court held that antitrust immunity — or exchange sel — -regulatory action was available only where the exchange a —

orded procedural sa — eguards, in- cluding notice, an explanation o — charges, and an opportunity to be heard. Id. 5 Securities Acts Amendments o — 1975, Pub. L. No. 94-29, §§ 19(b), 19(d), 89 Stat. 97, 146-52 (codi — ied at 15 U.S.C. §§ 78s(b), 78s(d)) (requiring SEC approval o — all SRO rule changes prior to e —

ectiveness and mandating

air procedures in all disciplinary proceedings). The Senate Report ex- pressly identi — ies Silver as motivating the procedural — ramework. S. Rep. No. 94-75, at 28-30 (1975). 6 Paul G. Mahoney, The Exchange as Regulator, 83 Va. L. Rev. 1453, 1490- 1503 (1997) (documenting the erosion o — exchange sel — -regulatory capac- ity — ollowing Silver and the 1975 Amendments and arguing that the trans-

ormation o — exchanges into “regulated sel — -regulators” reduced rather than improved market governance quality). 7 Securities Acts Amendments o — 1975, Pub. L. No. 94–29, 89 Stat. 97 (codi — ied in scattered sections o — 15 U.S.C.). 8 Stoel Rives LLP & Husch Blackwell LLP, Report to the O —

ice — or Civil Rights: Michigan State University 14-23 (Sept. 2019) ( — inding that Nassar’s supervisory chain ran through the athletic department and that com- plaints were not routed to Title IX o —

icials). The $4.5 million Clery Act 402 Law and Governance

ine is documented in U.S. Dep’t o — Education, Press Release: Department o —

Education Fines Michigan State University $4.5 Million

or Clery Act Viola- tions (Sept. 5, 2019). 9 See Kathleen Gray, John Engler Resigns as Michigan State University’s In- terim President, N.Y. Times, Jan. 17, 2019. Engler’s resignation — ollowed his reported comments that Nassar’s survivors were “enjoying” media at- tention. Id. 10 Uri Gneezy & Aldo Rustichini, A Fine Is a Price, 29 J. Legal Stud. 1, 7-11 (2000). The study examined ten daycare centers in Hai — a, Israel over a twenty-week period. Late pickups doubled a — ter the — ine was introduced and remained elevated a — ter the — ine was removed, demonstrating irre- versible crowding out o — the prior social norm. Id. at — ig. 1. 11 Andrei Shlei — er & Robert W. Vishny, A Survey o — Corporate Governance, 52 J. Finance 737, 737 (1997). For the board-centered account, see Ste- phen M. Bainbridge, The New Corporate Governance in Theory and Practice 27-35 (Ox — ord University Press, 2008). 12 Orly Lobel, The Renew Deal: The Fall o — Regulation and the Rise o — Govern- ance in Contemporary Legal Thought, 89 Minn. L. Rev. 342, 344-45 (2004). The — oundational texts o — the new governance literature include Jody Freeman, Collaborative Governance in the Administrative State, 45 UCLA L. Rev. 1 (1997), and Michael C. Dor — & Charles F. Sabel, A Constitution o —

Democratic Experimentalism, 98 Colum. L. Rev. 267 (1998). 13 R.A.W. Rhodes, The New Governance: Governing without Government, 44 Pol. Stud. 652, 657 (1996). See also James N. Rosenau & Ernst-Otto Czempiel, eds., Governance Without Government: Order and Change in World Politics (Cambridge University Press, 1992). 14 Elinor Ostrom, Governing the Commons: The Evolution o — Institutions — or Collective Action 90-102 (Cambridge University Press, 1990). 15 On monitoring as a collective good requiring institutional supply ra- ther than spontaneous provision, see Ostrom, supra note 10, at 94-100; Michael Hechter, Principles o — Group Solidarity 157-60 (University o — Cali-

ornia Press, 1987). Conclusion and Research Agenda 403

16 Gneezy & Rustichini, supra note 6, at 7-11. The — ine trans — ormed a so- cial norm into a market transaction, and the norm did not recover when the — ine was removed. Id. at 10. For the theoretical — ramework, see Sam- uel Bowles, Policies Designed — or Sel — -Interested Citizens May Undermine “The Moral Sentiments,” 320 Science 1605, 1606-07 (2008) (synthesizing evi- dence that external incentives crowd out moral motivation when they re — rame situations as market exchanges or signal institutional distrust). 17 Elinor Ostrom, Roy Gardner & James Walker, Rules, Games, and Com- mon-Pool Resources 185-206 (University o — Michigan Press, 1994). Groups that could communicate and establish their own rules achieved 73-99% o —

the optimal net yield; groups subjected to external regulation with im- per — ect en — orcement progressively deviated toward overextraction. Id. at 167-73. For — ield experiment replication with actual resource users, see Juan Camilo Cárdenas, John Stranlund & Cleve Willis, Local Environmen- tal Control and Institutional Crowding-Out, 28 World Dev. 1719, 1728-32 (2000) ( — inding that externally regulated villagers in rural Colombia ex- tracted more than sel — -governing groups, with external regulation crowding out the trust and reciprocity that had sustained cooperation). 18 Ostrom, supra note 10, at 90-91 (Design Principle 3: most individuals a —

ected by the operational rules can participate in modi — ying them). For empirical review o — this principle across a large sample o — commons in- stitutions, see Michael Cox, Gwen Arnold & Sergio Villamayor Tomás, A Review o — Design Principles — or Community-Based Natural Resource Manage- ment, 15 Ecology & Soc’y Art. 38, at 7-9 (2010). 19 See Bernstein, Opting Out, supra note 20, at 119-25 (documenting how the DDC’s arbitration rules were modi — ied over time in response to dis- putes that revealed inadequacies in existing procedures, including adjust- ments to evidence standards and sanction gradations — ollowing contested arbitral decisions). 20 Securities Acts Amendments o — 1975, Pub. L. No. 94-29, § 19(b), 89 Stat. 97, 146-49 (codi — ied at 15 U.S.C. § 78s(b)) (requiring SROs to — ile proposed rule changes with the SEC and obtain approval be — ore imple- mentation). Commissioner Philip A. Loomis, Jr. described the periodic 404 Law and Governance

review

ramework at its inception: the Senate Committee intended SEC oversight to be “more — ormal and pervasive” than the pre-1975 regime. Philip A. Loomis, Jr., Address at the Joint Securities Con — erence: The Se- curities Acts Amendments o — 1975, Sel — -Regulation and the National Market System 4-7 (Nov. 18, 1975). 21 Elinor Ostrom, Understanding Institutional Diversity 58-63, 259-61 (Princeton University Press, 2005) (distinguishing operational-level, col- lective-choice, and constitutional-level rules and arguing that institu- tional resilience depends on the capacity to modi — y rules at all three lev- els). For DAO governance as an example o — adaptive adjustment, see Aa- ron Wright, The Rise o — Decentralized Autonomous Organizations: Opportu- nities and Challenges, 4 Stan. J. Blockchain L. & Pol’y 152, 168-74 (2021) (describing governance protocols with automated threshold-triggered revision mechanisms). 22 Robert C. Ellickson, Order Without Law: How Neighbors Settle Disputes 40-81, 123-36 (Harvard University Press, 1991). Ellickson’s taxonomy o —

social controllers ranges

rom individual sel — -help through third-party organizations to government; in — ormal community norms operate pri- marily through what Ellickson calls “third-party social — orces,” including gossip, reputational e —

ects, and graduated social pressure. Id. at 123-26. 23 Lisa Bernstein, Opting Out o — the Legal System: Extralegal Contractual Re- lations in the Diamond Industry, 21 J. Legal Stud. 115, 128-31 (1992) (docu- menting that resort to state courts is treated as a violation o — community norms and is itsel — subject to institutional sanction). 24 James M. Acheson, The Lobster Gangs o — Maine 64-89 (University Press o — New England, 1988). 25 See Chapter 13 in — ra. The — our statutes are Wyo. Stat. Ann. §§ 17-31-101 to -116 (2021, as amended 2022); Tenn. Code Ann. §§ 48-250-101 to -115 (2022); Utah Code Ann. §§ 48-5-101 to -116 (2023); N.H. Rev. Stat. Ann. §§ 301-C:1 to :28 (2024). Each conditions legal personality on identi — ied natural persons, amendable smart contracts, and dissolution when hu- man control lapses. Conclusion and Research Agenda 405

26 See Chapters 11 and 12 in — ra. The corporate governance and university governance materials develop this argument in detail. 27 Adol — A. Berle, Jr. & Gardiner C. Means, The Modern Corporation and Private Property (Macmillan, 1932; rev. ed. Harcourt, Brace & World, 1968), at 6-7, 112-16. 28 Andrei Shlei — er & Robert W. Vishny, A Survey o — Corporate Governance, 52 J. Fin. 737, 737 (1997). 29 Margaret M. Blair, Ownership and Control: Rethinking Corporate Govern- ance — or the Twenty-First Century (Brookings Institution Press, 1995), at 3. 30 G20/OECD Principles o — Corporate Governance 11 (OECD Publishing, 2023). 31 Stephen M. Bainbridge, Director Primacy: The Means and Ends o — Corpo- rate Governance, 97 Nw. U. L. Rev. 547, 550 (2003). See also Stephen M. Bainbridge, The New Corporate Governance in Theory and Practice (Ox — ord University Press, 2008). 32 Leo E. Strine, Jr., The Dangers o — Denial: The Need — or a Clear-Eyed Un- derstanding o — the Power and Accountability Structure Established by the Dela- ware General Corporation Law, 50 Wake Forest L. Rev. 761, 768 (2015). 33 Milton Friedman, The Social Responsibility o — Business Is to Increase Its Pro — its, N.Y. Times Mag., Sept. 13, 1970; Michael C. Jensen & William H. Meckling, Theory o — the Firm: Managerial Behavior, Agency Costs and Own- ership Structure, 3 J. Fin. Econ. 305 (1976). 34 R. Edward Freeman, Strategic Management: A Stakeholder Approach (Pit- man Publishing, 1984), at 46. 35 Margaret M. Blair & Lynn A. Stout, A Team Production Theory o — Corpo- rate Law, 85 Va. L. Rev. 247, 270-87 (1999). 36 Del. Code Ann. tit. 8, § 141(a) (2024). 37 William T. Allen, Our Schizophrenic Conception o — the Business Corpora- tion, 14 Cardozo L. Rev. 261 (1992). 38 Blasius Industries, Inc. v. Atlas Corp., 564 A.2d 651, 659 (Del. Ch. 1988). 39 Schnell v. Chris-Cra — t Industries, Inc., 285 A.2d 437 (Del. 1971). 40 Gerald F. Davis, The Twilight o — the Berle and Means Corporation, 34 Seat- tle U. L. Rev. 1121 (2011). 406 Law and Governance

41 Jody Freeman, Collaborative Governance in the Administrative State, 45 UCLA L. Rev. 1, 22-33 (1997). 42 Orly Lobel, The Renew Deal: The Fall o — Regulation and the Rise o — Govern- ance in Contemporary Legal Thought, 89 Minn. L. Rev. 342, 373-440 (2004). 43 Michael C. Dor — & Charles F. Sabel, A Constitution o — Democratic Experi- mentalism, 98 Colum. L. Rev. 267, 314-35 (1998). 44 Charles F. Sabel & William H. Simon, Minimalism and Experimentalism in the Administrative State, 100 Geo. L.J. 53, 55-65 (2011). 45 Ian Ayres & John Braithwaite, Responsive Regulation: Transcending the Deregulation Debate (Ox — ord University Press, 1992), at 35-39. 46 Bradley C. Karkkainen, “New Governance” in Legal Thought and in the World: Some Splitting as Antidote to Overzealous Lumping, 89 Minn. L. Rev. 471, 480-90 (2004). 47 Joanne Scott & David M. Trubek, Mind the Gap: Law and New Ap- proaches to Governance in the European Union, 8 Eur. L.J. 1, 5 (2002). 48 James N. Rosenau, “Governance, Order, and Change in World Poli- tics,” in James N. Rosenau & Ernst-Otto Czempiel (eds.), Governance with- out Government: Order and Change in World Politics 1, 4 (Cambridge Uni- versity Press, 1992). 49 Id. 50 R.A.W. Rhodes, The New Governance: Governing without Government, 44 Pol. Stud. 652, 653-60 (1996). 51 Gerry Stoker, Governance as Theory: Five Propositions, 50 Int’l Soc. Sci. J. 17, 17-28 (1998). 52 Jon Pierre & B. Guy Peters, Governance, Politics and the State (Macmil- lan, 2000), at 7. 53 World Bank, Governance and Development (World Bank, 1992). 54 Daniel Kau — mann, Aart Kraay & Massimo Mastruzzi, The Worldwide Governance Indicators: Methodology and Analytical Issues, 3 Hague J. on the Rule o — L. 220 (2011). Conclusion and Research Agenda 407

55 Gary Marks, “Structural Policy and Multilevel Governance in the EC,” in Alan W. Ca — runy & Glenda G. Rosenthal (eds.), The State o — the Euro- pean Community Vol. 2: The Maastricht Debates and Beyond 391 (Lynne Rienner, 1993); Liesbet Hooghe & Gary Marks, Unraveling the Central State, but How? Types o — Multi-Level Governance, 97 Am. Pol. Sci. Rev. 233 (2003). 56 Elinor Ostrom, Governing the Commons: The Evolution o — Institutions — or Collective Action 90-102 (Cambridge University Press, 1990). 57 Elinor Ostrom, Understanding Institutional Diversity (Princeton Univer- sity Press, 2005). 58 Daniel H. Cole, The Varieties o — Comparative Institutional Analysis, 2013 Wis. L. Rev. 383. 59 Lee Anne Fennell, Ostrom’s Law: Property Rights in the Commons, 5 Int’l J. Commons 9 (2011). 60 Oliver E. Williamson, The Mechanisms o — Governance (Ox — ord Univer- sity Press, 1996), at 12. 61 Oliver E. Williamson, Markets and Hierarchies: Analysis and Antitrust Im- plications (Free Press, 1975); Oliver E. Williamson, The Economic Institu- tions o — Capitalism (Free Press, 1985), at 17. 62 Oliver E. Williamson, Comparative Economic Organization: The Analysis o — Discrete Structural Alternatives, 36 Admin. Sci. Q. 269 (1991). 63 Oliver E. Williamson, The New Institutional Economics: Taking Stock, Looking Ahead, 38 J. Econ. Literature 595, 597 — ig.1 (2000). 64 Walter W. Powell, Neither Market Nor Hierarchy: Network Forms o — Or- ganization, 12 Res. Organizational Behav. 295 (1990). 65 Mark Granovetter, Economic Action and Social Structure: The Problem o —

Embeddedness, 91 Am. J. Soc. 481 (1985). 66 Henry B. Hansmann, The Role o — Nonpro — it Enterprise, 89 Yale L.J. 835, 838-40 (1980). 67 James J. Fishman & Stephen Schwarz, Nonpro — it Organizations: Cases and Materials (Foundation Press, 5th ed. 2010); James J. Fishman, Stephen Schwarz & Lloyd Hitoshi Mayer, Nonpro — it Organizations: Cases and Mate- rials (Foundation Press, 6th ed. 2015). 408 Law and Governance

68 Rob Atkinson, Obedience as the Foundation o — Fiduciary Duty, 34 J. Corp. L. 43 (2008). 69 Marion R. Fremont-Smith, Governing Nonpro — it Organizations: Federal and State Law and Regulation (Harvard University Press, 2004). 70 Evelyn Brody, Charity Governance: What’s Trust Law Got to Do with It?, 80 Chi.-Kent L. Rev. 641 (2005); Evelyn Brody, The Board o — Nonpro — it Organizations: Puzzling Through the Gaps Between Law and Practice, 76 Fordham L. Rev. 521 (2007). 71 Walter W. Powell, Neither Market Nor Hierarchy: Network Forms o — Or- ganization, 12 Res. Organizational Behav. 295, 295-97 (1990). 72 Joel M. Podolny & Karen L. Page, Network Forms o — Organization, 24 Ann. Rev. Soc. 57, 57 (1998). 73 Candace Jones, William S. Hesterly & Stephen P. Borgatti, A General Theory o — Network Governance: Exchange Conditions and Social Mechanisms, 22 Acad. Mgmt. Rev. 911, 914 (1997). 74 Lisa Bernstein, Beyond Relational Contracts: Social Capital and Network Governance in Procurement Contracts, 7 J. Legal Analysis 561, 565 (2015). 75 Matthew Jennejohn, Do Networks Govern Contracts?, 48 J. Corp. L. 1 (2022). 76 Keith G. Provan & Patrick Kenis, Modes o — Network Governance: Struc- ture, Management, and E —

ectiveness, 18 J. Pub. Admin. Res. & Theory 229 (2008). 77 Gerald R. Salancik, WANTED: A Good Network Theory o — Organization, 40 Admin. Sci. Q. 345 (1995). 78 Silver v. New York Stock Exchange, 373 U.S. 341 (1963). 79 Associated Press v. United States, 326 U.S. 1 (1945). 80 Barak D. Richman, Stateless Commerce: The Diamond Network and the Persistence o — Relational Exchange (Harvard University Press, 2017). 81 Lobel, supra note 16, at 344-45. Gillian K. Had — ield & Barry R. Weingast, What Is Law? A Coordination 82

Model o

the Characteristics o — Legal Order, 4 J. Legal Analysis 471 (2012). Conclusion and Research Agenda 409

83 Gunther Teubner, Substantive and Re — lexive Elements in Modern Law, 17 Law & Soc’y Rev. 239 (1983); Gunther Teubner, Law as an Autopoietic Sys- tem (Blackwell, 1993). 84 Edward Peter Stringham, Private Governance: Creating Order in Economic and Social Li — e (Ox — ord University Press, 2015); Bruce L. Benson, The En- terprise o — Law: Justice Without the State (Paci — ic Research Institute, 1990). 85 Brian Z. Tamanaha, Legal Pluralism Explained: History, Theory, Conse- quences (Ox — ord University Press, 2021); Sally Engle Merry, Legal Plural- ism, 22 Law & Soc’y Rev. 869 (1988); Paul Schi —

Berman, Global Legal Plu- ralism: A Jurisprudence o — Law beyond Borders (Cambridge University Press, 2012). 86 Morton J. Horwitz, The Trans — ormation o — American Law, 1780-1860 (Harvard University Press, 1977), at xv-xvi, 253-66. 87 Morton J. Horwitz, The Trans — ormation o — American Law, 1870-1960: The Crisis o — Legal Orthodoxy (Ox — ord University Press, 1992). 88 Howard Gillman, The Constitution Besieged: The Rise and Demise o — Loch- ner Era Police Powers Jurisprudence (Duke University Press, 1993); Howard Gillman & Cornell W. Clayton, eds., Supreme Court Decision-Making: New Institutionalist Approaches (University o — Chicago Press, 1999). 89 George I. Lovell, Legislative De — errals: Statutory Ambiguity, Judicial Power, and American Democracy (Cambridge University Press, 2003), at 1- 25. 90 Keith E. Whittington, Constitutional Construction: Divided Powers and Constitutional Meaning (Harvard University Press, 1999), at 1-19. 91 Keith E. Whittington, Political Foundations o — Judicial Supremacy: The Presidency, the Supreme Court, and Constitutional Leadership in U.S. History (Princeton University Press, 2007), at 3-25. 92 Karen Orren & Stephen Skowronek, The Search — or American Political Development (Cambridge University Press, 2004), at 123. 93 Karen Orren & Stephen Skowronek, The Policy State: An American Pre- dicament (Harvard University Press, 2017). 94 Rogers M. Smith, Civic Ideals: Con — licting Visions o — Citizenship in U.S. History (Yale University Press, 1997), at 1-39. 410 Law and Governance

95 Douglass C. North, Institutions, Institutional Change and Economic Per — or- mance (Cambridge University Press, 1990), at 3-10. 96 Geo —

rey M. Hodgson, How Economics Forgot History: The Problem o — His- torical Speci — icity in Social Science (Routledge, 2001), at 299-319. 97 Elinor Ostrom, Understanding Institutional Diversity (Princeton Univer- sity Press, 2005), at 13-22. See also Michael D. McGinnis, An Introduction to IAD and the Language o — the Ostrom Workshop, 39 Pol. Stud. J. 169 (2011). 98 Elinor Ostrom, Governing the Commons: The Evolution o — Institutions — or Collective Action 90-102 (Cambridge University Press, 1990). 99 Michael Cox, Gwen Arnold & Sergio Villamayor Tomás, A Review o —

Design Principles

or Community-Based Natural Resource Management, 15 Ecology & Soc’y Art. 38, at 7-12 (2010). 100 Walter W. Powell, Neither Market Nor Hierarchy: Network Forms o — Or- ganization, 12 Res. Organizational Behav. 295, 295-300 (1990). 101 Mark Granovetter, Economic Action and Social Structure: The Problem o —

Embeddedness, 91 Am. J. Soc. 481, 481-93 (1985). 102 NAACP v. Alabama ex rel. Patterson, 357 U.S. 449, 460, 462 (1958). 103 Sarcuni v. bZx DAO, 664 F. Supp. 3d 1100 (S.D. Cal. 2023); CFTC v. Ooki DAO, No. 3:22-cv-05416 (N.D. Cal. 2023); Samuels v. Lido DAO (N.D. Cal. 2024). 104 Northwest Wholesale Stationers, Inc. v. Paci — ic Stationery & Printing Co., 472 U.S. 284, 296 (1985). 105 Zechariah Cha — ee, The Internal A —

airs o — Associations Not — or Pro — it, 43 Harv. L. Rev. 993 (1930). 106 Blatt v. University o — Southern Cali — ornia, 5 Cal. App. 3d 935 (1970). 107 Pinsker v. Paci — ic Coast Society o — Orthodontists, 1 Cal. 3d 160 (1969) (Pins- ker I); Pinsker v. Paci — ic Coast Society o — Orthodontists, 12 Cal. 3d 541 (1974) (Pinsker II). 108 Guido Calabresi & A. Douglas Melamed, Property Rules, Liability Rules, and Inalienability: One View o — the Cathedral, 85 Harv. L. Rev. 1089 (1972). For the application to governance membership, see Henry E. Smith, Ex- clusion versus Governance: Two Strategies — or Delineating Property Rights, 31 J. Legal Stud. S453 (2002). Conclusion and Research Agenda 411

109 Silver v. New York Stock Exchange, 373 U.S. 341 (1963). For analysis o —

the governance consequences, see Paul G. Mahoney, The Exchange as Reg- ulator, 83 Va. L. Rev. 1453 (1997). 110 Lisa Bernstein, Merchant Law in a Merchant Court: Rethinking the Code’s Search — or Immanent Business Norms, 144 U. Pa. L. Rev. 1765 (1996). 111 Del. Code Ann. tit. 8, § 122(18) (e —

ective Aug. 1, 2024); West Palm Beach Fire — ighters’ Pension Fund v. Moelis & Company, C.A. No. 2023-0309- JTL (Del. Ch. Feb. 23, 2024). 112 Manhattan Eye, Ear & Throat Hospital v. Spitzer, 186 Misc. 2d 126, 715 N.Y.S.2d 575 (N.Y. Sup. Ct. 1999). On the duty o — obedience generally, see Rob Atkinson, Obedience as the Foundation o — Fiduciary Duty, 34 J. Corp. L. 43 (2008). 113 Marion R. Fremont-Smith, Governing Nonpro — it Organizations: Federal and State Law and Regulation 1-37 (Harvard University Press, 2004). 114 Restatement (Second) o — Trusts § 391 (1959). 115 Samuel P. King & Randall W. Roth, Broken Trust: Greed, Mismanage- ment and Political Manipulation at America’s Largest Charitable Trust (Uni- versity o — Hawaii Press, 2006). 116 In re Milton Hershey School, 911 A.2d 1258 (Pa. 2006); Jonathan Klick & Robert H. Sitko —

, Agency Costs, Charitable Trusts, and Corporate Control: Evidence — rom Hershey’s Kiss-O —

, 108 Colum. L. Rev. 749 (2008). 117 Press accounts o — the Getty Trust governance — ailures during Barry Munitz’s tenure; Cali — ornia Attorney General’s O —

ice records. See gener- ally reporting by major Cali — ornia media outlets documenting these — acts. 118 Evelyn Brody, The Limits o — Charity Fiduciary Law, 57 Md. L. Rev. 1400, 1401 (1998). See also Evelyn Brody, Charity Governance: What’s Trust Law Got to Do with It?, 80 Chi.-Kent L. Rev. 641 (2005). 119 Blasius Industries, Inc. v. Atlas Corp., 564 A.2d 651, 659 (Del. Ch. 1988). 120 Lucian A. Bebchuk, The Myth o — the Shareholder Franchise, 93 Va. L. Rev. 675 (2007). 121 In re Caremark International Inc. Derivative Litigation, 698 A.2d 959, 967 (Del. Ch. 1996). 122 Marchand v. Barnhill, 212 A.3d 805, 812-14 (Del. 2019). 412 Law and Governance

123 Erica E. Harris, Christine M. Petrovits & Michelle H. Yetman, The E — -

ect o — Nonpro — it Governance on Donations: Evidence — rom the Revised Form 990, 90 Acct. Rev. 579 (2015). 124 Louis Freeh et al., Report o — the Special Investigative Counsel Regarding the Actions o — The Pennsylvania State University Related to the Child Sexual Abuse Committed by Gerald A. Sandusky, July 12, 2012, at 14, 108. 125 U.S. Senate Permanent Subcommittee on Investigations, The Role o —

the Board o

Directors in Enron’s Collapse, S. Prt. 107-70, at 4 (July 8, 2002). 126 Lauren B. Edelman, Working Law: Courts, Corporations, and Symbolic Civil Rights (University o — Chicago Press, 2016). 127 Lauren B. Edelman & Shauhin Talesh, To Comply or Not to Comply? That Isn’t the Question: How Organizations Construct the Meaning o — Compli- ance, in Regulatory Encounters: Organizations and Compliance (David Levi-Faur ed., 2014); Lauren B. Edelman & Shauhin Talesh, Legal Ambi- guity and the Politics o — Compliance, 88 N.Y.U. L. Rev. (2013). 128 Kimberly D. Krawiec, Cosmetic Compliance and the Failure o — Negotiated Governance, 81 Wash. U. L.Q. 487 (2003). 129 Robert C. Ellickson, Order Without Law: How Neighbors Settle Disputes (Harvard University Press, 1991); Lisa Bernstein, Opting Out o — the Legal System: Extralegal Contractual Relations in the Diamond Industry, 21 J. Legal Stud. 115 (1992); Gillian K. Had — ield & Barry R. Weingast, What Is Law? A Coordination Model o — the Characteristics o — Legal Order, 4 J. Legal Analysis 471 (2012); Bruce L. Benson, The Enterprise o — Law: Justice Without the State (Paci — ic Research Institute, 1990). 130 Paul A. Samuelson, The Pure Theory o — Public Expenditure, 36 Rev. Econ. & Stat. 387 (1954); Paul A. Samuelson, Diagrammatic Exposition o — a The- ory o — Public Expenditure, 37 Rev. Econ. & Stat. 350 (1955); see generally James M. Buchanan, An Economic Theory o — Clubs, 32 Economica (n.s.) 1 (1965); Vincent Ostrom & Elinor Ostrom, Public Goods and Public Choices, in E.S. Savas, ed., Alternatives — or Delivering Public Services: Toward Im- proved Per — ormance 7 (Westview Press, 1977). Conclusion and Research Agenda 413

131 Seth C. Oranburg, Governance as a Club Good (manuscript on — ile with author). The argument that governance has club good structure was de- veloped in that prior work and is extended and applied here across the broader governance theory this book develops. 132 Elinor Ostrom, Understanding Institutional Diversity (Princeton Univer- sity Press, 2005), at 23-24. 133 Ostrom & Ostrom, supra note 1, at 12-14. 134 Id. at 12-17. The canonical — our-cell matrix with both dimensions made explicit appears in Ostrom, Understanding Institutional Diversity, supra note 3, at 24 tbl. 1.3. See also Elinor Ostrom, Roy Gardner & James Walker, Rules, Games, and Common-Pool Resources 6-7 (University o —

Michigan Press, 1994). 135 Buchanan, supra note 1, at 1. For the comprehensive treatment o — club theory and its relationship to the broader theory o — externalities and pub- lic goods, see Richard Cornes & Todd Sandler, The Theory o — Externalities, Public Goods, and Club Goods (Cambridge University Press, 1986; 2d ed. 1996); Todd Sandler & John T. Tschirhart, The Economic Theory o — Clubs: An Evaluative Survey, 18 J. Econ. Literature 1481 (1980). For the intellec- tual genealogy o — club theory and its relationship to Buchanan’s broader project, see Alain Marciano, Retrospectives: James Buchanan: Clubs and Al- ternative Wel — are Economics, 35 J. Econ. Persp. 243 (2021). 136 The marginal conditions — or optimal club size are developed — ormally in Buchanan, supra note 1, at 4-8. For the simpli — ication that marginal crowding cost equals average cost per member at the optimum, see Mar- tin C. McGuire, Group Segregation and Optimal Jurisdictions, 82 J. Pol. Econ. 112, 115-17 (1974). 137 On the disciplining role o — exit in competitive club markets, see Albert O. Hirschman, Exit, Voice, and Loyalty: Responses to Decline in Firms, Or- ganizations, and States (Harvard University Press, 1970), at 21-29. 138 On the — ailure o — competitive discipline when governance clubs cannot be replicated, see L. Lynne Kiesling, The Promise and Perils o — Exclusion: Using Institutional Design Principles and the Theory o — Clubs to Analyse Re- gional Transmission Organization Governance, 22 J. Institutional Econ. 414 Law and Governance

(2026) (demonstrating that when governance clubs are non-replicable, Ostrom’s design principles must supply the internal governance disci- pline that market pressure cannot). 139 Todd Sandler, Buchanan Clubs, 24 Const. Pol. Econ. 265, 268-69 (2013) (distinguishing congestible club goods — rom common-pool resources with intrinsic subtractability). The distinction between Buchanan’s con- gestion pattern and Ostrom’s subtractability pattern in the governance context is developed in Oranburg, supra note 2. 140 Lisa Bernstein, Opting Out o — the Legal System: Extralegal Contractual Relations in the Diamond Industry, 21 J. Legal Stud. 115, 128-37 (1992). 141 Barak D. Richman, Stateless Commerce: The Diamond Network and the Persistence o — Relational Exchange (Harvard University Press, 2017), at 180- 215. 142 Securities Acts Amendments o — 1975, Pub. L. No. 94-29, §§ 19(b), 19(d), 89 Stat. 97, 146-52 (codi — ied at 15 U.S.C. §§ 78s(b), 78s(d)). 143 Paul G. Mahoney, The Exchange as Regulator, 83 Va. L. Rev. 1453, 1490- 1503 (1997). 144 Elinor Ostrom, Governing the Commons: The Evolution o — Institutions — or Collective Action 90-92 (Cambridge University Press, 1990) (Design Prin- ciple 1). On the bundle-o — -rights — ramework identi — ying the exclusion right as the authority to determine who participates in a governance in- stitution, see Edella Schlager & Elinor Ostrom, Property-Rights Regimes and Natural Resources: A Conceptual Analysis, 68 Land Econ. 249, 252-55 (1992). 145 Ostrom, Governing the Commons, supra note 15, at 61-69 (Törbel), 69-82 (Spanish and Philippine irrigation systems). 146 Hirschman, supra note 8, at 30-43 (on the relationship between exit availability and voice as governance mechanisms; when exit is blocked, voice must substitute). 147 Guido Calabresi & A. Douglas Melamed, Property Rules, Liability Rules, and Inalienability: One View o — the Cathedral, 85 Harv. L. Rev. 1089, 1105- 10 (1972); Henry E. Smith, Exclusion versus Governance: Two Strategies — or Delineating Property Rights, 31 J. Legal Stud. S453, S461-65 (2002). Conclusion and Research Agenda 415

148 Id. 149 Mancur Olson, The Logic o — Collective Action: Public Goods and the Theory o — Groups 9-16, 48-52 (Harvard University Press, 1965). 150 Silver v. New York Stock Exchange, 373 U.S. 341, 357 (1963); Northwest Wholesale Stationers, Inc. v. Paci — ic Stationery & Printing Co., 472 U.S. 284, 295-96 (1985). See also Jonathan B. Baker, Exclusion as a Core Competition Concern, 78 Antitrust L.J. 527 (2013). 151 Oliver E. Williamson, The Economic Institutions o — Capitalism (Free Press, 1985), at 20-22. See also Carl J. Dahlman, The Problem o — External- ity, 22 J.L. & Econ. 141, 148 (1979) (classi — ying transaction costs as “re- source losses due to lack o — in — ormation”). 152 Lisa Bernstein, Opting Out o — the Legal System: Extralegal Contractual Re- lations in the Diamond Industry, 21 J. Legal Stud. 115, 119-122 (1992). 153 Id. at 124-125, 128-131, 153. 154 Robert Cooter & Janet T. Landa, Personal versus Impersonal Trade: The Size o — Trading Groups and the Law o — Contracts, 4 Int’l Rev. L. & Econ. 15, 15-17 (1984). 155 Thomas C. Schelling, The Strategy o — Con — lict (Harvard University Press, 1960), at 54-58. 156 George A. Akerlo — , The Market — or “Lemons”: Quality Uncertainty and the Market Mechanism, 84 Q.J. Econ. 488 (1970). 157 Bernstein, supra note 2, at 121. 158 Lisa Bernstein, Merchant Law in a Merchant Court: Rethinking the Code’s Search — or Immanent Business Norms, 144 U. Pa. L. Rev. 1765, 1772-73 (1996). 159 Data on commercial arbitration resolution timelines appear in Ameri- can Arbitration Association, Dispute Resolution Statistics (annual). Data on

ederal civil case disposition appear in Administrative O —

ice o — the U.S. Courts, Federal Court Management Statistics (2022). See also Soia Mentschiko —

, Commercial Arbitration, 61 Colum. L. Rev. 846, 856-58 (1961). Precise comparisons depend on case type and complexity; the — ig- ures cited represent central tendencies across all commercial cases rather than matched pairs. 416 Law and Governance

160 Barak D. Richman, Stateless Commerce: The Diamond Network and the Persistence o — Relational Exchange (Harvard University Press, 2017), at 180- 215; Barak D. Richman, An Autopsy o — Cooperation: Diamond Dealers and the Limits o — Trust-Based Exchange, 9 J. Legal Analysis 247, 254-67 (2017). 161 Paul G. Mahoney, The Exchange as Regulator, 83 Va. L. Rev. 1453, 1456- 63 (1997). 162 U.S. Government Accountability O —

ice, Financial Crisis Losses and Po- tential Impacts o — the Dodd-Frank Act, GAO-13-180, at 16-18 (Jan. 2013). The GAO estimate covers output losses relative to pre-crisis trend pro- jections, direct — iscal costs, and household wealth losses, and represents an upper-bound estimate that does not capture all categories o — social cost. 163 Ashwini Chhatre & Arun Agrawal, Trade-o —

s and Synergies between Carbon Storage and Livelihood Bene — its — rom Forest Commons, 106 Proc. Nat’l Acad. Sci. 17667, 17667-70 (2009). The study examined eighty — orest commons across ten countries using data — rom the International Forestry Resources and Institutions research program. The — inding that greater local rule-making autonomy was associated with higher carbon storage was statistically signi — icant at the p = 0.048 level. 164 A.C. Smith et al., Community Forest Management Led to Rapid Local For- est Gain in Nepal: A 29 Year Mixed Methods Retrospective Case Study, 126 Land Use Pol’y 106526 (2023) (documenting — orest cover increase — rom 3.88 million hectares in 1992 to 6.63 million hectares by 2016, with ap- proximately 22,000 community — orest user groups now managing ap- proximately 2.3 million hectares). The World Bank independently re- ports Nepal’s — orest cover increasing — rom twenty-nine percent in 1994 to over — orty-six percent in 2022. 165 Elinor Ostrom, Governing the Commons: The Evolution o — Institutions — or Collective Action 69-82 (Cambridge University Press, 1990); Thomas F. Glick, Irrigation and Society in Medieval Valencia (Harvard University Press, 1970). The Tribunal de las Aguas was inscribed on the UNESCO Representative List o — the Intangible Cultural Heritage o — Humanity in 2009. Conclusion and Research Agenda 417

166 Barak D. Richman, How Community Institutions Create Economic Ad- vantage: Jewish Diamond Merchants in New York, 31 Law & Soc. Inquiry 383, 404-10 (2006) (documenting the downstream — raud costs that spread through the diamond supply chain as DDC governance eroded, con — irm- ing that the spillover was a — unction o — governance quality rather than an incidental — eature o — the trading network). 167 Robert Costanza et al., The Value o — the World’s Ecosystem Services and Natural Capital, 387 Nature 253, 253-60 (1997). Costanza and colleagues estimated the aggregate value o — global ecosystem services at $16-54 tril- lion per year, a — igure that became visible only through the counter — ac- tual analysis o — what would be lost i — those services ceased. The methodo- logical parallel to governance surplus is direct: both are positive external- ities whose value is invisible while the producing institution — unctions and becomes quanti — iable only when the institution — ails. 168 A.C. Pigou, The Economics o — Wel — are (4th ed. Macmillan, 1932), at 172- 75. 169 Richman, supra note 10; Bernstein, supra note 2, at 119 n.8. 170 Richard Cornes & Todd Sandler, Easy Riders, Joint Production, and Pub- lic Goods, 94 Econ. J. 580 (1984); Richard Cornes & Todd Sandler, The Theory o — Externalities, Public Goods, and Club Goods (2d ed. Cambridge University Press, 1996), at 194-216. 171 Pigou, supra note 16, at 172-75. 172 Morris M. Kleiner, Licensing Occupations: Ensuring Quality or Restricting Competition? (W.E. Upjohn Institute, 2006), at 1-15; Executive O —

ice o —

the President, Occupational Licensing: A Framework

or Policymakers 3 (July 2015) (reporting growth — rom approximately — ive percent o — the work-

orce in the 1950s to over twenty- — ive percent by 2015). 173 Marketing General Inc., Membership Marketing Benchmarking Report (2025) (reporting that — i — ty-six percent o — surveyed associations saw membership plateau or decline over the preceding year; the report sur- veys associations across industry, pro — essional, and trade categories, and the — igures are sel — -reported). These data are indicative o — a general trend 418 Law and Governance

and should not be read as a precise measurement o

governance under- supply. 174 GAO-13-180, supra note 12. 175 The Pigouvian subsidy analysis o — de — erence doctrines is developed in Seth C. Oranburg, Governance as a Club Good (manuscript on — ile with au- thor). Arthur Cecil Pigou, The Economics o — Wel — are (4th ed. Macmillan, 1932), at 172-75. 176 Aronson v. Lewis, 473 A.2d 805, 812 (Del. 1984). For the BJR as absten- tion doctrine, see Stephen M. Bainbridge, The Business Judgment Rule as Abstention Doctrine, 57 Vand. L. Rev. 83 (2004). See also ALI Principles o —

Corporate Governance § 4.01(c) (1992). 177 Smith v. Van Gorkom, 488 A.2d 858 (Del. 1985). Del. Code Ann. tit. 8, § 102(b)(7) (enacted 1986). 178 Federal Arbitration Act, 9 U.S.C. §§ 1-16; Moses H. Cone Memorial Hospi- tal v. Mercury Construction Corp., 460 U.S. 1, 24-25 (1983). See also AT&T Mobility LLC v. Concepcion, 563 U.S. 333 (2011). For the governance — unc- tion o — arbitration en — orcement, see Stephen J. Ware, Private Ordering and Commercial Arbitration: Lasting Lessons — rom Mentschiko —

, J. Disp. Re- sol. 55 (2019); Christopher R. Drahozal, Private Ordering and International Commercial Arbitration, 113 Penn St. L. Rev. 1031 (2009). 179 Northwest Wholesale Stationers, Inc. v. Paci — ic Stationery & Printing Co., 472 U.S. 284, 296 (1985). For the antitrust treatment o — sel — -regulatory organizations, see Robert Pito — sky, Sel — -Regulation and Antitrust (FTC, Feb. 18, 1998); Herbert Hovenkamp, Federal Antitrust Policy: The Law o —

Competition and Its Practice (6th ed. 2020). National Cooperative Research and Production Act, 15 U.S.C. §§ 4301-4306. 180 Zechariah Cha — ee, Jr., The Internal A —

airs o — Associations Not — or Pro — it, 43 Harv. L. Rev. 993 (1930). See also Blatt v. University o — Southern Cali — or- nia, 5 Cal. App. 3d 935, 940 (1970). Applebaum v. Board o — Directors, 104 Cal. App. 3d 648 (1980). On the 181

contract theory o

association rules, see Cha — ee, supra note 6, at 1001-05. Conclusion and Research Agenda 419

182 Del. Code Ann. tit. 8, § 122(18) (e —

ective Aug. 1, 2024); West Palm Beach Fire — ighters’ Pension Fund v. Moelis & Company, C.A. No. 2023-0309- JTL (Del. Ch. Feb. 23, 2024). 183 Blatt v. University o — Southern Cali — ornia, 5 Cal. App. 3d 935 (1970). Guido Calabresi & A. Douglas Melamed, Property Rules, Liability Rules, and Inalienability: One View o — the Cathedral, 85 Harv. L. Rev. 1089, 1105- 10 (1972). 184 Silver v. New York Stock Exchange, 373 U.S. 341 (1963); Paul G. Ma- honey, The Exchange as Regulator, 83 Va. L. Rev. 1453, 1490-1503 (1997). Securities Acts Amendments o — 1975, Pub. L. No. 94-29, §§ 19(b), 19(d), 89 Stat. 97, 146-52 (codi — ied at 15 U.S.C. §§ 78s(b), 78s(d)). 185 Eric A. Posner, The Regulation o — Groups: The In — luence o — Legal and Non- legal Sanctions on Collective Action, 63 U. Chi. L. Rev. 133, 174 (1996). 186 Pinsker v. Paci — ic Coast Society o — Orthodontists, 1 Cal. 3d 160 (1969) (Pins- ker I); Pinsker v. Paci — ic Coast Society o — Orthodontists, 12 Cal. 3d 541 (1974) (Pinsker II). Potvin v. Metropolitan Li — e Insurance Co., 22 Cal. 4th 1060 (2000). Labor-Management Reporting and Disclosure Act, 29 U.S.C. § 101(a)(5). 187 Thomas O. McGarity, Some Thoughts on “Deossi — ying” the Rulemaking Process, 41 Duke L.J. 1385, 1400-28 (1992). Richard J. Pierce, Jr., Seven Ways to Deossi — y Agency Rulemaking, 47 Admin. L. Rev. 59 (1995); Jerry L. Mashaw & David L. Har — st, The Struggle — or Auto Sa — ety (Harvard Uni- versity Press, 1990), at 226-56. 188 Lisa Bernstein, Merchant Law in a Merchant Court: Rethinking the Code’s Search — or Immanent Business Norms, 144 U. Pa. L. Rev. 1765, 1779-97 (1996). 189 Lisa Bernstein, The Questionable Empirical Basis o — Article 2’s Incorpora- tion Strategy: A Preliminary Study, 66 U. Chi. L. Rev. 710, 760-62 (1999). 190 Robert E. Scott, The Case — or Formalism in Relational Contract, 94 Nw. U. L. Rev. 847, 851-53 (2000). Alan Schwartz & Robert E. Scott, Contract Theory and the Limits o — Contract Law, 113 Yale L.J. 541 (2003). 420 Law and Governance

191 Roberta Romano, The Genius o — American Corporate Law (AEI Press, 1993), at 1-25. Romano, The Sarbanes-Oxley Act and the Making o — Quack Corporate Governance, 114 Yale L.J. 1521 (2005). 192 Frank H. Easterbrook & Daniel R. Fischel, The Economic Structure o —

Corporate Law (Harvard University Press, 1991), at 34-36. Lucian Arye Bebchuk, The Debate on Contractual Freedom in Corporate Law, 89 Colum. L. Rev. 1395 (1989); Je —

rey N. Gordon, The Mandatory Structure o — Corpo- rate Law, 89 Colum. L. Rev. 1549 (1989). 193 Employee Retirement Income Security Act § 514(a), 29 U.S.C. § 1144(a). Elizabeth Y. McCuskey, State Cost-Control Re — orms and ERISA Preemption, Commonwealth Fund Issue Brie — (May 2022). 194 Evelyn Brody, The Limits o — Charity Fiduciary Law, 57 Md. L. Rev. 1400, 1401 (1998). Marion R. Fremont-Smith, Governing Nonpro — it Organiza- tions: Federal and State Law and Regulation (Harvard University Press, 2004), at 367-412; Brody, Whose Public?: Parochialism and Paternalism in State Charity Law En — orcement, 79 Ind. L.J. 937 (2004). 195 Private Securities Litigation Re — orm Act o — 1995, Pub. L. No. 104-67, 109 Stat. 737 (codi — ied at 15 U.S.C. § 78u-4). 196 Thole v. U.S. Bank N.A., 140 S. Ct. 1615 (2020). 197 Labor-Management Reporting and Disclosure Act § 402, 29 U.S.C. § 482; Trbovich v. United Mine Workers, 404 U.S. 528 (1972). 198 Kimberly D. Krawiec, Cosmetic Compliance and the Failure o — Negotiated Governance, 81 Wash. U. L.Q. 487, 491-96 (2003). 199 Lauren B. Edelman, Working Law: Courts, Corporations, and Symbolic Civil Rights (University o — Chicago Press, 2016), at 6-14. 200 Romano, Sarbanes-Oxley, supra note 17, at 1528-32, 1580-87. In re Care- mark International Inc. Derivative Litigation, 698 A.2d 959, 967 (Del. Ch. 1996); Marchand v. Barnhill, 212 A.3d 805 (Del. 2019). 201 Roberts v. United States Jaycees, 468 U.S. 609 (1984); Board o — Directors o —

Rotary International v. Rotary Club o

Duarte, 481 U.S. 537 (1987). Conclusion and Research Agenda 421

202 Boy Scouts o — America v. Dale, 530 U.S. 640, 648 (2000). Dale Carpen- ter, Expressive Association and Anti-Discrimination Law A — ter Dale: A Tripar- tite Approach, 85 Minn. L. Rev. 1515 (2001); John D. Inazu, Liberty’s Re — - uge: The Forgotten Freedom o — Assembly (Yale University Press, 2012). 203 Silver v. New York Stock Exchange, 373 U.S. 341, 361-67 (1963). The sim- ultaneous activation o — Mechanisms 1 and 2 is analyzed in detail in Chap- ter 10 in — ra. 204 Mahoney, supra note 8, at 1490-1503 (documenting the decline in NYSE en — orcement activity — ollowing Silver and attributing it to the com- bined e —

ects o — antitrust liability exposure and increased procedural costs). 205 Marchand v. Barnhill, 212 A.3d 805, 809-12 (Del. 2019). The cascading governance — ailure at Blue Bell Creameries is also analyzed through the corporate governance lens in Chapter 11 in — ra. 206 Albert O. Hirschman, Exit, Voice, and Loyalty: Responses to Decline in Firms, Organizations, and States (Harvard University Press, 1970), at 21-29, 76-105. 207 Aspen Skiing Co. v. Aspen Highlands Skiing Corp., 472 U.S. 585 (1985). 208 Verizon Communications Inc. v. Law O —

ices o — Curtis V. Trinko, LLP, 540 U.S. 398, 408-11 (2004). On the essential — acilities doctrine and its limita- tions, see Herbert Hovenkamp, Federal Antitrust Policy, supra note 5, at §§ 7.7-7.8. 209 Edward B. Rock & Michael L. Wachter, Islands o — Conscious Power: Law, Norms, and the Sel — -Governing Corporation, 149 U. Pa. L. Rev. 1619, 1679-91 (2001). 210 Neil K. Komesar, Imper — ect Alternatives: Choosing Institutions in Law, Economics, and Public Policy (University o — Chicago Press, 1994), at 5-11. Komesar, Law’s Limits: The Rule o — Law and the Supply and Demand o —

Rights (Cambridge University Press, 2001). Allied Tube & Conduit Corp. v. Indian Head, Inc., 486 U.S. 492 (1988). 211

Cass R. Sunstein, Incompletely Theorized Agreements, 108 Harv. L. Rev. 212

1733, 1735-40 (1995); Cass R. Sunstein, Incommensurability and Valuation 422 Law and Governance

in Law, 92 Mich. L. Rev. 779, 779-86 (1994). On value incommensurabil- ity more broadly, see Ruth Chang, ed., Incommensurability, Incomparabil- ity, and Practical Reason (Harvard University Press, 1997), at 1-34; H.S. Mather, Law-Making and Incommensurability, 47 McGill L.J. 357, 358-63 (2002). 213 George A. Akerlo — , The Market — or “Lemons”: Quality Uncertainty and the Market Mechanism, 84 Q.J. Econ. 488 (1970). 214 The three premises stated here are analytic, not moral. They assert that governance institutions produce value, that destroying that value has costs, and that those costs should be visible. They do not assert that governance value always outweighs competing legal values. The distinc- tion parallels the relationship between Calabresi and Melamed’s speci — i- cation o — property rules and liability rules — a conceptual — ramework — or identi — ying what a legal rule does — and the normative question o —

which rule type a legal system should adopt. Guido Calabresi & A. Doug- las Melamed, Property Rules, Liability Rules, and Inalienability: One View o —

the Cathedral, 85 Yale L.J. 1089, 1092-93 (1972). Lon Fuller’s account o

the internal morality o

law supplies a related methodological parallel: Fuller argued that law’s — ormal requirements (generality, promulgation, non-retroactivity, clarity, consistency, constancy, possibility, and con- gruence) are conditions o — legal e —

ectiveness, not moral ideals. Lon L. Fuller, The Morality o — Law 33-94 (rev. ed. Yale University Press, 1969). The governance — ramework’s three premises play an analogous role: they are conditions o — analytical e —

ectiveness, speci — ying what must be true

or the governance variable to enter legal analysis with precision. 215 Lon L. Fuller, The Forms and Limits o — Adjudication, 92 Harv. L. Rev. 353, 394-404 (1978). For extended analysis, see J.W.F. Allison, Fuller’s Analysis o — Polycentric Disputes and the Limits o — Adjudication, 53 Cambridge L.J. 367 (1994). 216 Lisa Bernstein, Opting Out o — the Legal System: Extralegal Contractual Re- lations in the Diamond Industry, 21 J. Legal Stud. 115 (1992); Lisa Bernstein, Merchant Law in a Merchant Court: Rethinking the Code’s Search — or Imma- nent Business Norms, 144 U. Pa. L. Rev. 1765 (1996). Conclusion and Research Agenda 423

217 Bernstein, Opting Out, supra note 4, at 119-25. 218 Sunstein, Incompletely Theorized Agreements, supra note 1, at 1735-40. For the related argument that legal methods can structure analysis with- out determining outcomes, see Henry M. Hart, Jr. & Albert M. Sacks, The Legal Process: Basic Problems in the Making and Application o — Law (William N. Eskridge Jr. & Philip P. Frickey, eds., Foundation Press, 1994), at 143- 52; Neil K. Komesar, Imper — ect Alternatives: Choosing Institutions in Law, Economics, and Public Policy (University o — Chicago Press, 1994), at 3-11. 219 Roberts v. United States Jaycees, 468 U.S. 609 (1984). 220 Paul G. Mahoney, The Exchange as Regulator, 83 Va. L. Rev. 1453, 1456-63 (1997). 221 The escalating remedial structure draws on analogous doctrines in corporate law and administrative law. See Reves v. Ernst & Young, 507 U.S. 170, 179-83 (1993) (speci — ying evidentiary burdens in securities litigation that account — or institutional context). For declaratory relie — as a — irst- stage remedy in institutional disputes, see Samuel L. Bray, The Myth o — the Mild Declaratory Judgment, 63 Duke L.J. 1091, 1093-98 (2014) (arguing that declaratory relie — has more coercive — orce than commonly assumed, but noting its lower institutional disruption relative to injunctive reme- dies). 222 The corporate compliance monitor model is codi — ied in the U.S. De- partment o — Justice Principles o — Federal Prosecution o — Business Organi- zations, U.S. Attorneys’ Manual § 9-28.000 (rev. 2015), which speci — ies that monitors should be imposed only when “the potential bene — its o — a monitor are not clearly outweighed by the projected costs and burdens.” See also Vikramaditya Khanna & Timothy L. Dickinson, The Corporate Monitor: The New Corporate Czar?, 105 Mich. L. Rev. 1713, 1714-20 (2007) (analyzing the governance costs and bene — its o — external monitors in the corporate context). 223 Lisa Bernstein, Opting Out o — the Legal System: Extralegal Contractual Relations in the Diamond Industry, 21 J. Legal Stud. 115, 119–25 (1992). 224 Id. at 119–21. Bernstein — ound that DDC members almost uni — ormly submit to binding DDC arbitration rather than court proceedings, and 424 Law and Governance

that the DDC arbitration system resolves the overwhelming majority o

disputes without resort to

ormal legal mechanisms. 225 Avner Grei — , Contract En — orceability and Economic Institutions in Early Trade: The Maghribi Traders’ Coalition, 83 Am. Econ. Rev. 525, 525–30 (1993). 226 Janet T. Landa, A Theory o — the Ethnically Homogeneous Middleman Group: An Institutional Alternative to Contract Law, 10 J. Legal Stud. 349, 349–62 (1981). 227 Paul R. Milgrom, Douglass C. North & Barry R. Weingast, The Role o —

Institutions in the Revival o

Trade: The Law Merchant, Private Judges, and the Champagne Fairs, 2 Econ. & Politics 1, 1–23 (1990). 228 Elinor Ostrom, Governing the Commons: The Evolution o — Institutions — or Collective Action 58–102 (Cambridge Univ. Press, 1990). 229 Lisa Bernstein, Private Commercial Law in the Cotton Industry: Creating Cooperation Through Rules, Norms, and Institutions, 99 Mich. L. Rev. 1724, 1724–40 (2001). 230 Oliver E. Williamson, Calculativeness, Trust, and Economic Organization, 36 J.L. & Econ. 453, 453–86 (1993). 231 Lisa Bernstein, Beyond Relational Contracts: Social Capital and Network Governance in Procurement Contracts, 7 J. Legal Analysis 561, 563–65 (2015). 232 Grei — , supra note 3, at 525–30. 233 Lisa Bernstein, Contract Governance in Small-World Networks: The Case o — the Maghribi Traders, 113 Nw. U. L. Rev. 1009, 1012–25 (2019). 234 Barak D. Richman, An Autopsy o — Cooperation: Diamond Dealers and the Limits o — Trust-Based Exchange, 9 J. Legal Analysis 247, 248–65 (2017); see also Barak D. Richman, Stateless Commerce: The Diamond Network and the Persistence o — Relational Exchange 195–230 (Harvard Univ. Press, 2017) (providing extended treatment o — the — ive structural — orces eroding DDC governance capacity). Lon L. Fuller, The Forms and Limits o — Adjudication, 92 Harv. L. Rev. 235

353, 394–404 (1978). Conclusion and Research Agenda 425

236 See Chapter 8, supra (applying the seven-step method to Silver v. New York Stock Exchange to demonstrate the method’s mechanics and identi — y- ing the property-to-liability-rule conversion as Mechanism 1). 237 Silver v. New York Stock Exchange, 373 U.S. 341, 370–71 (1963) (Stewart, J., dissenting). 238 James M. Buchanan, An Economic Theory o — Clubs, 32 Economica 1 (1965). 239 Mancur Olson, The Logic o — Collective Action: Public Goods and the The- ory o — Groups 2–22 (Harvard Univ. Press, 1965). 240 Iris Bohnet, Bruno S. Frey & Ste —

en Huck, More Order with Less Law: On Contract En — orcement, Trust, and Crowding, 95 Am. Pol. Sci. Rev. 131, 131–44 (2001). 241 S. Nageeb Ali & David A. Miller, En — orcing Cooperation Among Ostra- cists, 1 Am. Econ. J.: Microeconomics 143 (2009); S. Nageeb Ali & David A. Miller, Ostracism and Forgiveness, 106 Am. Econ. Rev. 2329 (2016). 242 U.C.C. § 1-303(c) (Am. Law Inst. & Uni — . Law Comm’n 2022). 243 Bernstein, supra note 1, at 115–25; Bernstein, supra note 7, at 1724–40. 244 Barrow-Shaver Res. Co. v. Carrizo Oil & Gas, Inc., 590 S.W.3d 471 (Tex. 2019) (holding that jury could determine whether oil and gas industry custom incorporated a reasonableness standard into an express consent clause — or well assignments; multiple industry groups — iled amicus brie — s o —

ering con — licting accounts o — the governing custom, illustrating how judicial incorporation o — industry norms requires resolving contested collective interpretations). 245 9 U.S.C. § 2 (2018); Moses H. Cone Mem’l Hosp. v. Mercury Constr. Corp., 460 U.S. 1, 24–25 (1983) (establishing the liberal policy — avoring arbitra- tion agreements under the FAA). 246 See GCG, Chapter 5, supra (developing the Pigouvian subsidy analysis o — judicial de — erence doctrines). The Pigouvian subsidy concept applies here as — ollows: governance institutions produce positive externalities — or non-members (the downstream market participants who rely on govern- ance quality) but capture only member bene — its in their production deci- sions. This produces systematic underproduction o — governance relative 426 Law and Governance

to the social optimum. Legal rules that reduce governance production costs, speci — ically the FAA and the antitrust rule o — reason, move govern- ance supply closer to the social optimum by reducing the private cost o —

production,

unctioning as a subsidy correcting — or the externality. 247 Nw. Wholesale Stationers, Inc. v. Pac. Stationery & Printing Co., 472 U.S. 284, 293–96 (1985). 248 Id. at 288 & n.8 (noting the “sparse” — actual record and characterizing the Ninth Circuit’s description o — Paci — ic’s losses as an “inaccurate charac- terization”). 249 Id. at 295. 250 Id. at 288. 251 GCG, Chapter 5, supra. 252 Klor’s, Inc. v. Broadway-Hale Stores, Inc., 359 U.S. 207, 212 (1959). 253 See generally Note, Developments in the Law: Judicial Control o — Actions o —

Private Associations, 76 Harv. L. Rev. 983, 987–1011 (1963) (surveying the common law standards — or judicial review o — private association mem- bership decisions). 254 MFS Sec. Corp. v. SEC, 380 F.3d 611, 614–16 (2d Cir. 2004). 255 Alpine Sec. Corp. v. Fin. Indus. Regulatory Auth., 121 F.4th 1314 (D.C. Cir. 2024), cert. denied, No. 24-904 (U.S. June 2, 2025). 256 Sel — -Regulatory Organizations; Notice o — Filing o — Proposed Rule Changes, 90 Fed. Reg. 25689 (June 17, 2025) (FINRA proposed rule changes implementing SEC review requirement prior to expulsion order e —

ectiveness). 257 Higgins v. Am. Soc’y o — Clinical Pathologists, 51 N.J. 191, 200–02, 238 A.2d 665, 670–71 (1968). 258 Steven Tadelis, Reputation and Feedback Systems in Online Plat — orm Mar- kets, 8 Ann. Rev. Econ. 321, 322–25 (2016). 259 Kate Klonick, The New Governors: The People, Rules, and Processes Gov- erning Online Speech, 131 Harv. L. Rev. 1598, 1606–14 (2018). 260 Daphne Keller, The DSA’s Industrial Model — or Content Moderation, Ver-

assungsblog (Feb. 24, 2022), https://ver — assungsblog.de/the-dsas-indus- Conclusion and Research Agenda 427

trial-model-

or-content-moderation/ (arguing that DSA procedural re- quirements create incentives — or plat — orms to moderate less rather than better, because the cost o — compliance with appeal obligations at scale makes narrowing en — orcement scope economically rational). 261 Evelyn Douek, Content Moderation as Systems Thinking, 136 Harv. L. Rev. 526, 534–40 (2022). 262 Bernstein, supra note 7, at 1766–72 (demonstrating that within the cotton industry, members deliberately avoid incorporating in — ormal gov- ernance norms into legal contracts precisely to preserve the governance system’s — lexibility and to maintain the distinction between relationship- preserving norms and endgame norms en — orceable in litigation). 263 See Chapter 8, supra (analyzing Silver v. New York Stock Exchange, 373 U.S. 341 (1963), through the seven-step governance method and identi — y- ing the wire-termination decision as activating Mechanism 1 (the prop- erty-to-liability-rule conversion) at the governance level o — the entire ex- change membership system). 264 Paul G. Mahoney, The Exchange as Regulator, 83 Va. L. Rev. 1453, 1453–60 (1997) (developing the theory o — exchange governance as private ordering under reputational discipline); William A. Birdthistle, M. Todd Henderson & Onnig H. Dombalagian, Becoming a Fi — th Branch, 95 Cornell L. Rev. 464, 464–71 (2010) (documenting the trans — ormation o — SROs

rom private associations into quasi-governmental entities); Benjamin P. Edwards, Supreme Risk, 76 Fla. L. Rev. 1 (2024) (documenting constitu- tional threats to the SRO model — rom Appointments Clause and nondele- gation doctrine); Rohit A. Na — day, From Sense to Nonsense and Back Again: SRO Immunity, Doctrinal Bait-and-Switch, and a Call — or Coherence, 77 U. Chi. L. Rev. 847, 847–53 (2010) (critiquing doctrinal incoherence in Standard Investment Chartered). 265 Silver v. New York Stock Exchange, 373 U.S. 341, 348–49 (1963) (“We hold that the Exchange’s collective action in cutting o —

the wire services . . . constituted a per se violation o — § 1 o — the Sherman Act, unless the Ex- change is required to justi — y this action under the Securities Exchange Act.”). 428 Law and Governance

266 Id. at 375–80 (Stewart, J., dissenting) (arguing that the majority’s nar- row holding le — t exchanges without clear guidance on what procedures would satis — y its requirements and what standards would govern sub- stantive review o — exchange governance decisions). 267 Silver v. New York Stock Exchange, 373 U.S. 341, 361-64 (1963) (requir- ing “notice, a statement o — the charges, and a reasonable opportunity to be heard”). The accommodation threshold — process su —

icient to satis — y due process concerns without converting exclusion authority to a liabil- ity rule — is implicit in the Court’s speci — ication o — the procedural mini- mum. See Lon L. Fuller, The Forms and Limits o — Adjudication, 92 Harv. L. Rev. 353, 364-70 (1978) (distinguishing the procedural requirements ap- propriate to party-centered adjudication — rom the requirements appro- priate to polycentric institutional decisions). 268 Securities Acts Amendments o — 1975, Pub. L. No. 94-29, § 19(d), 89 Stat. 97, 149-52 (codi — ied at 15 U.S.C. § 78s(d)) (requiring SROs to pro- vide — air procedures in disciplinary proceedings and subjecting sanctions to SEC review). See Gordon v. New York Stock Exchange, Inc., 422 U.S. 659, 691-92 (1975) (holding that comprehensive SEC oversight substi- tutes — or antitrust liability as the accountability mechanism — or exchange governance decisions). 269 Philip A. Loomis, Jr., Address at the Joint Securities Con — erence: The Securities Acts Amendments o — 1975, Sel — -Regulation and the National Market System 4-7 (Nov. 18, 1975) (reporting the Senate Committee’s view that Commission oversight should be “more — ormal and pervasive” and that SROs exercise “delegated governmental powers” whose exercise requires structured procedural accountability). 270 Mahoney, supra note 2, at 1479–88 (describing the shi — t — rom private contract model to regulated SRO governance as a response to the Silver ruling and the 1975 Amendments); Douglas Michael, The Use o — Audited Sel — -Regulation as a Regulatory Technique, 1993 Ann. Rev. Reg. (ACUS) 171, 172–78 (1994) (de — ining audited sel — -regulation as delegation to a non-governmental entity with powers o — review and independent action Conclusion and Research Agenda 429

retained by a

ederal agency, and identi — ying due process and plenary agency control as necessary conditions — or the model’s legitimacy). 271 Securities Acts Amendments o — 1975, Pub. L. No. 94-29, 89 Stat. 97 (1975) (codi — ied as amended at scattered sections o — 15 U.S.C.) (amending §§ 6, 15, 17, 19, and other provisions o — the Securities Exchange Act o —

1934). 272 15 U.S.C. § 78s(b) (requiring SROs to — ile proposed rule changes with the Commission and prohibiting implementation without Commission approval except — or designated categories o — rule changes). 273 15 U.S.C. § 78s(d) (requiring SROs to — ile notice o —

inal disciplinary sanctions, denials o — membership, and bars — rom association, and provid- ing — or Commission review o — such actions). 274 15 U.S.C. § 78s(e) (authorizing the Commission to abrogate, add to, or delete provisions o — SRO rules where necessary or appropriate to ensure the Exchange Act’s purposes are achieved). 275 Silver, 373 U.S. at 364 (noting that the SEC lacked statutory authority to review the NYSE’s wire-termination decision a —

ecting non-members, which was one reason the Court concluded that no alternative regulatory remedy was available to provide the accountability mechanism Silver re- quired). 276 Gordon v. New York Stock Exchange, Inc., 422 U.S. 659 (1975) (decided June 26, 1975; Securities Acts Amendments o — 1975 signed June 4, 1975; case argued March 25–26, 1975). The case was argued ten weeks be — ore the June 4 enactment date and decided twenty-two days a — ter the enact- ment. 277 Id. at 689–91 (holding that — ixed commission rates under active SEC oversight were immune — rom antitrust challenge and that “to interpose the antitrust laws, which would bar — ixed commission rates as per se vio- lations o — the Sherman Act, in the — ace o — positive SEC action . . . would unduly inter — ere with the intended operation o — the Securities Exchange Act”). The Court’s analysis proceeded under pre-amendment § 19(b), which authorized the SEC to supervise exchanges “in respect o — such matters as . . . the — ixing o — reasonable rates o — commission.” Id. at 670–71. 430 Law and Governance

The timing o

Gordon is crucial because it proves the immunity principle did not depend on the 1975 Amendments themselves: the Court held that pre-amendment SEC oversight was su —

icient — or immunity, establishing that the Amendments were statutory institutionalization o — a pre-exist- ing practice rather than the source o — the immunity doctrine. This timing also shows that the Amendments responded to the doctrinal environ- ment Silver created — the antitrust courts’ nascent oversight — rather than to Gordon speci — ically. The temporal relationship is Securities Act Amendments o — 1975 (June 4) → Gordon decision (June 26), not the re- verse. 278 Id. at 689–91; see also Chapter 9, supra (discussing the accountability- substitution pattern in which alternative governance oversight mecha- nisms can restore governance capacity when they adequately substitute

or mechanisms removed by courts). 279 See Mahoney, supra note 2, at 1494–98 (analyzing post-1975 develop- ment o — exchange governance accountability as SEC oversight expanded); Birdthistle, Henderson & Dombalagian, supra note 2, at 476–82 (describ- ing 1975 Amendments as the statutory basis — or courts’ subsequent ex- pansion o — SRO immunity). Edwards, supra note 2, at 15–22 (treating the 1975 expansion o — SEC oversight as a doctrinal — oundation — or absolute immunity crystallization in subsequent decades). 280 Philip A. Loomis, Jr., The Securities Acts Amendments o — 1975, Sel — -Regu- lation and the National Market System, Address at the Joint Securities Con-

erence 1975 (sponsored by NASD, Boston Stock Exchange, and SEC), Boston, Mass. (Nov. 18, 1975) (on — ile with author) (reporting the Senate Committee’s position that the Commission’s oversight under the 1975 Amendments should be “more — ormal and pervasive” and describing SRO authority as “delegated governmental powers” whose exercise required more open and accountable procedures). 281 Austin Mun. Sec., Inc. v. Nat’l Ass’n o — Sec. Dealers, Inc., 757 F.2d 676, 692 (5th Cir. 1985) (holding that “NASD is entitled to absolute immunity — or its role in disciplining its members and associates” when per — orming del- egated regulatory — unctions). Conclusion and Research Agenda 431

282 D’Alessio v. New York Stock Exchange, Inc., 258 F.3d 93 (2d Cir. 2001). 283 Id. at 106 (internal quotation marks omitted). 284 See Chapter 6, supra (developing the analysis o — governance spillovers, including positive externalities — or non-members such as market integ- rity, reliable price discovery, and the trust in — rastructure enabling ex- change among parties who cannot individually monitor all counterpar- ties). See also Chapter 7, supra (identi — ying judicial de — erence doctrines as Pigouvian subsidies that correct — or underproduction o — governance caused by positive externalities that governance producers cannot cap- ture through membership — ees). 285 In re NYSE Specialists Sec. Litig., 503 F.3d 89 (2d Cir. 2007). 286 Id. at 96–97. 287 Birdthistle, Henderson & Dombalagian, supra note 2, at 466–68 (argu- ing that post-1975 SROs had become a “Fi — th Branch” o — governmental regulation, private in — orm but governmental in — unction, wielding regu- latory authority equivalent to that o — administrative agencies without the constitutional accountability structures those agencies require). 288 Mahoney, supra note 2, at 1467–75 (arguing that the original incentive alignment between exchange owners and governance quality was dis- rupted by regulatory capture and the shi — t toward member-governance models in which the regulated are also the regulators, with consequences

or the private contract model’s continued viability a — ter demutualiza- tion). 289 In re Facebook, Inc., IPO Sec. & Derivative Litig., 986 F. Supp. 2d 428, 447–50 (S.D.N.Y. 2013). 290 City o — Providence v. BATS Glob. Mkts., Inc., 878 F.3d 36, 43–47 (2d Cir. 2017). 291 Standard Inv. Chartered, Inc. v. Nat’l Ass’n o — Sec. Dealers, Inc., 637 F.3d 112, 115–16 (2d Cir. 2011) (explaining that courts apply “a — unctional test”

or SRO immunity examining “the nature o — the — unction per — ormed, not the identity o — the actor who per — ormed it,” with the party asserting im- munity bearing the burden o — demonstrating that the challenged conduct constitutes delegated quasi-governmental activity). 432 Law and Governance

292 Na — day, supra note 2, at 879–84 (documenting the doctrinal uncer- tainty generated by the regulatory-commercial distinction and arguing that the resulting burden o — predicting immunity outcomes diverts ex- change resources — rom substantive governance to de — ensive legal posi- tioning). 293 Id. at 869–77 (arguing that the Second Circuit’s standing analysis in Standard Investment Chartered was internally inconsistent with the 1975 Amendments’ — ramework — or member participation in SRO rulemaking, which was itsel — part o — the accountability structure that justi — ied the ab- solute immunity courts had extended to SRO regulatory — unctions). 294 Alpine Sec. Corp. v. Fin. Indus. Regulatory Auth., 121 F.4th 1314 (D.C. Cir. 2024), cert. denied, No. 24-904 (U.S. June 2, 2025). 295 Id. at 1320–22 (describing Alpine’s Appointments Clause and nondele- gation challenges to FINRA’s authority to conduct expedited disciplinary proceedings that could result in immediate market exclusion pending completion o — Commission review). 296 Alpine Sec. Corp., No. 24-904 (U.S. June 2, 2025) (cert. denied without noted dissent). 297 Edwards, supra note 2, at 1–12, 30–45 (documenting the doctrinal tra- jectory — rom Lucia v. SEC through West Virginia v. EPA that made consti- tutional challenges to SRO authority credible and arguing that the Su- preme Court’s recent structural constitution jurisprudence creates an ex- istential threat to the SRO model as currently organized). 298 Lucia v. SEC, 585 U.S. 237 (2018) (holding that SEC administrative law judges are “O —

icers o — the United States” whose appointments must comply with the Appointments Clause). 299 West Virginia v. EPA, 597 U.S. 697 (2022) (articulating the major ques- tions doctrine and requiring clear congressional authorization be — ore agencies exercise authority over questions o — vast economic and political signi — icance). 300 MFS Sec. Corp. v. SEC, 380 F.3d 611, 614–16 (2d Cir. 2004). The 6.5-year interval between the initiation o — disciplinary proceedings and the Sec- ond Circuit’s resolution o — the SEC review process produced documented Conclusion and Research Agenda 433

market consequences

or a —

ected parties who could not determine the validity o — FINRA’s regulatory authority during the pendency o — the con- stitutional proceedings. See Chapter 9 Research Report (con — irming the 6.5-year timeline as documented in MFS Securities and the in — erence that governance-uncertainty costs during that interval are the chapter’s own analytical contribution). 301 Sel — -Regulatory Organizations; Notice o — Filing o — Proposed Rule Changes, 90 Fed. Reg. 25689 (June 17, 2025) (FINRA proposed rule changes implementing SEC review requirement prior to expulsion order e —

ectiveness, — iled in response to post-Alpine regulatory uncertainty about the constitutional basis — or FINRA’s authority to impose immedi- ate market exclusion pending SEC review). 302 Birdthistle, Henderson & Dombalagian, supra note 2, at 480–86 (docu- menting how demutualization trans — ormed SRO governance — rom a member-controlled non-pro — it model to a shareholder-driven — or-pro — it model, creating the structural con — licts between commercial interests and governance — unctions that the regulatory-commercial distinction at- tempts to manage). 303 Seth C. Oranburg, Corporate Governance as a Club Good: The Shareholder Rights Movement’s Category Error __ (manuscript on — ile with author) (“SFCG”) (documenting median S&P 500 CEO compensation at approxi- mately $7.5 million in 2011 and $17 million in 2024, a 127% increase over the say-on-pay period). The 2011 baseline — igure and the resulting per- centage increase are sensitive to the speci — ic compensation metric and data source. See Equilar & Associated Press, AP/Equilar CEO Pay Study (annual) (reporting median S&P 500 CEO total compensation). Compare Equilar, CEO Pay Trends: S&P 500 2024 (2025) (reporting median total compensation — or S&P 500 CEOs using all-in grant-date — igures, which may di —

er — rom realized-pay or summary-compensation-table measures), with Associated Press/Equilar, CEO Pay Study (2012) (reporting S&P 500 median CEO total compensation — or — iscal year 2011 at approximately $9.1 million under realized-compensation methodology). The directional claim, that executive compensation increased substantially during the 434 Law and Governance

say-on-pay period, is consistent across sources; the precise percentage in- crease varies by method. 304 SFCG, supra note 1, at __ (compiling say-on-pay — ailure rates by year

rom 2011 through 2024 and demonstrating that no year exceeded — our percent — ailure, meaning shareholders approved the overwhelming ma- jority o — packages presented to them). See Semler Brossy, Say on Pay: An- nual Report (2024) (reporting Russell 3000 — ailure rate peak o — 3.7% in 2022, declining to record low o — 1.3% in 2024). 305 James M. Buchanan, An Economic Theory o — Clubs, 32 Economica 1, 4–6 (1965) (de — ining the club good as a good with voluntary membership, ex- cludability, nonrivalry up to congestion, and quality dependence on membership composition). See Chapter 5, supra (developing the — ull club- good — ramework — or governance goods, including the distinction be- tween member bene — its and positive externalities — lowing to non-mem- bers). 306 Stephen M. Bainbridge, The Business Judgment Rule as Abstention Doc- trine, 57 Vand. L. Rev. 83, 87–90 (2004) (characterizing the business judgment rule as an institutional abstention doctrine through which courts preserve board authority by declining to review the merits o —

board decisions satis

ying the rule’s conditions). See SFCG, supra note 1, at __ (adopting the Bainbridge characterization and arguing that absten- tion is the method, not the purpose, o — the rule). 307 See Chapter 6, supra (developing the analysis o — governance spillovers and showing that governance producers who capture only member bene-

its systematically underproduce governance quality relative to the social optimum). 308 Jill E. Fisch, Darius Palia & Steven Davido —

Solomon, Is Say on Pay All About Pay? The Impact o — Firm Per — ormance, 8 Harv. Bus. L. Rev. 101, 103– 06 (2018) ( — inding that say-on-pay votes re — lect dissatis — action with — irm per — ormance rather than compensation levels and that the provision has

ailed to constrain CEO pay). See SFCG, supra note 1, at __ (summarizing the academic record o — say-on-pay’s — ailure to reduce compensation Conclusion and Research Agenda 435

growth). The U.S. record contrasts with

indings — rom other jurisdic- tions where say-on-pay preceded the proxy advisor intermediation pat- tern that de — ines U.S. governance. See Fabrizio Ferri & David Maber, Say on Pay Votes and CEO Compensation: Evidence — rom the UK, 17 Rev. Fin. 527, 530–48 (2013) ( — inding that the United Kingdom’s 2003 advisory vote requirement produced signi — icant compensation restraint at — irms with the most egregious prior pay practices, in a regulatory context with lower proxy advisor intermediation and greater institutional investor en- gagement capacity than the U.S. regime); Ricardo Correa & Ugur Lel, Say on Pay Laws, Executive Compensation, Pay Slice, and Firm Valuation Around the World, 88 J. Fin. Econ. 543, 545–60 (2016) ( — inding say-on-pay associated with lower compensation and smaller CEO pay slices in a cross-country panel, with e —

ect sizes varying by institutional environ- ment). These international — indings do not re — ute the U.S. results; they speci — y them. Say-on-pay may constrain compensation when institu- tional investors engage directly with compensation committees. In the U.S., the proxy advisor intermediation layer — created by the mandatory voting rule — inserts a standardized mechanical — ilter between institu- tional investors and compensation decisions, attenuating any disciplining e —

ect that shareholder engagement might otherwise produce. 309 Institutional Shareholder Services, U.S. Proxy Voting Guidelines (2024); Glass Lewis, Proxy Guidelines (2024). See SFCG, supra note 1, at __ (docu- menting combined ISS-Glass Lewis market share exceeding ninety per- cent and describing the — irms as operating “under acknowledged con — licts o — interest in markets — or consulting services they sell to the companies they rate”). 310 SFCG, supra note 1, at __ (citing robo-voting statistics showing auto- matic adoption o — ISS recommendations growing — rom approximately seven percent o — ISS customers in 2007 to twenty-three percent by 2021). SFCG, supra note 1, at __ (citing passive — und governance capacity data, 311

based on stewardship team

igures available circa 2019, at which point major index — und stewardship teams were signi — icantly smaller than by 436 Law and Governance

2024). See BlackRock, 2024 Stewardship Report (2024) (reporting steward- ship team o — approximately sixty- — ive to seventy pro — essionals responsi- ble — or engagement across global port — olio); Vanguard Investment Stew- ardship, 2024 Annual Report (2024) (reporting approximately sixty invest- ment stewardship team members); State Street Global Advisors, 2024 Stewardship Report (2024) (reporting approximately twelve-person Asset Stewardship team). Even at 2024 sta —

ing levels, the ratio o — port — olio companies to stewardship pro — essionals at each — irm produces coverage constraints that preclude meaning — ul deliberation at the — irm-speci — ic level. 312 In re Caremark Int’l Inc. Derivative Litig., 698 A.2d 959, 967 (Del. Ch. 1996) (Allen, V.C.) (holding that directors may be held liable — or — ailure to implement adequate in — ormation and compliance systems, but charac- terizing such claims as “possibly the most di —

icult theory in corporation law upon which a plainti —

might hope to win a judgment”). 313 Marchand v. Barnhill, 212 A.3d 805, 822–24 (Del. 2019) (holding that Caremark requires boards to implement compliance systems — or the mis- sion-critical risks most likely to threaten the company’s business model, and that — ailure to implement such systems is not excused by attention to peripheral compliance concerns). 314 In re McDonald’s Corp. S’holder Derivative Litig., 289 A.3d 343 (Del. Ch. 2023). In companion decisions issued March 1, 2023, the court dismissed the Caremark oversight claims against the directors but sustained a Care- mark-based claim against a — ormer o —

icer — or the o —

icer’s own conduct in the human capital management domain. The director dismissal limits the holding’s direct precedential reach against boards; the o —

icer survival nonetheless marked an extension o — oversight liability to corporate cul- ture and human capital management, domains where — irm-speci — ic ex- pertise is most valuable and where standardized judicial criteria are least suited to substitute — or institutional judgment. See SFCG, supra note 1, at __ (analyzing the McDonald’s litigation as crossing — rom governance quality criteria en — orcement into congestion introduction). Conclusion and Research Agenda 437

315 Adol — A. Berle & Gardiner C. Means, The Modern Corporation and Pri- vate Property 112-16 (rev. ed. Harcourt, Brace & World, 1967); Lucian A. Bebchuk, The Case — or Increasing Shareholder Power, 118 Harv. L. Rev. 833, 836-43 (2005) (arguing that increasing shareholder power improves cor- porate governance by reducing agency costs, without examining whether the shareholder participation the proposal would produce satis — ies the output-based — unctionality criteria the governance — ramework requires). The governance — ramework’s disagreement with Bebchuk is not that agency costs are unreal — they are — but that optimizing — or share- holder participation as the corrective mechanism treats corporate gov- ernance as a public good (more access is better) rather than as a club good (access beyond the congestion threshold degrades quality). 316 Aronson v. Lewis, 473 A.2d 805, 812 (Del. 1984), overruled in part on other grounds by Brehm v. Eisner, 746 A.2d 244 (Del. 2000) (establishing the business judgment rule’s conditions: disinterestedness and in — ormed business judgment). 317 Unocal Corp. v. Mesa Petroleum Co., 493 A.2d 946 (Del. 1985); Revlon, Inc. v. MacAndrews & Forbes Holdings, Inc., 506 A.2d 173 (Del. 1986). 318 Weinberger v. UOP, Inc., 457 A.2d 701 (Del. 1983) (establishing entire

airness review — or controlling stockholder transactions, requiring both

air dealing and — air price). 319 See Chapter 7, supra (identi — ying Mechanism 7 as governance degrada- tion through mandating open membership that disables screening, the mechanism by which legal rules eliminate the quality control that makes club governance valuable). 320 Delaware Senate Bill 21, enacted Mar. 25, 2025 (codi — ied at Del. Code Ann. tit. 8, §§ 144, 220) (creating a business judgment rule sa — e harbor — or con — licted transactions approved by in — ormed disinterested directors or stockholders, and narrowing shareholder inspection rights). 321 In re Tesla, Inc. Derivative Litig. (Tornetta v. Musk), C.A. No. 2018- 0408-KSJM (Del. Ch. Jan. 30, 2024) (ordering rescission o — Musk’s com- pensation package as a remedy — or inadequacy in the compensation-set- ting process), rev’d in part, a —

‘d in part, Nos. 120, 121, 2024 (Del. Dec. 19, 438 Law and Governance

2025) (a

irming the trial court’s liability — indings while reversing the re- scission remedy, holding that disinterested shareholder rati — ication pro- vided a business judgment rule sa — e harbor, and reducing attorney — ees

rom $345 million to approximately $54 million on a quantum meruit basis). 322 Glass Lewis, State o — U.S. Reincorporation 2025 (2025) (reporting that 64.3% o — U.S. companies with pending reincorporation proposals in the 2025 proxy season proposed leaving Delaware). During the period be- tween the trial court’s January 2024 Tornetta order and the Supreme Court’s December 2025 reversal, Tesla, Dropbox, and TripAdvisor rein- corporated out o — Delaware. 323 Texas Senate Bill 29, 89th Leg., R.S. (Tex. 2025) (codi — ying the busi- ness judgment rule by statute and expanding director authority, compet- ing with Delaware — or incorporation business on the dimension o — gov- ernance de — erence). See SFCG, supra note 1, at __ (analyzing the Dela- ware-Texas competition as evidence that states compete on governance excludability rather than shareholder rights, consistent with the club- good — ramework’s predictions). 324 Robert S. Harris, Tim Jenkinson & Steven N. Kaplan, Private Equity Per — ormance: What Do We Know?, 69 J. Fin. 1851, 1851–82 (2014) (docu- menting buyout — und net annualized returns exceeding public market equivalents by roughly — ive percentage points over the two decades — ol- lowing 2000, with EBITDA margin expansion averaging 370 basis points as the documented operational mechanism). See SFCG, supra note 1, at __ (using PE governance as a natural experiment comparing out- comes under concentrated-ownership governance with outcomes under dispersed-participation governance). 325 Lucian A. Bebchuk & Kobi Kastiel, The Untenable Case — or Perpetual Dual-Class Stock, 103 Va. L. Rev. 585, 589–96 (2017) (acknowledging the governance rationale — or dual-class structures at IPO while challenging perpetual sunset provisions). See SFCG, supra note 1, at __ (presenting the club-good case — or dual-class structures as internal excludability engi- neering designed to maintain governance quality by restricting voting Conclusion and Research Agenda 439

authority to investors whose economic interests align with long-term value creation). 326 Lucian A. Bebchuk & Jesse M. Fried, Pay Without Per — ormance: The Un-

ul — illed Promise o — Executive Compensation (2004); Lucian A. Bebchuk, Alma Cohen & Allen Ferrell, What Matters in Corporate Governance?, 22 Rev. Fin. Stud. 783 (2009) (documenting the relationship between board entrenchment indices and — irm valuation). See SFCG, supra note 1, at __ (acknowledging Bebchuk’s documentation o — real entrenchment prob- lems while contesting his proposed corrective). 327 Lucian A. Bebchuk, Alon Brav & Wei Jiang, The Long-Term E —

ects o —

Hedge Fund Activism, 115 Colum. L. Rev. 1085, 1089–96 (2015) (

inding operating per — ormance improvements at hedge — und activist targets per- sisting — or — ive years post-intervention). See Ed deHaan, David F. Larcker & Charles McClure, Long-Term Economic Consequences o — Hedge Fund Activ- ist Interventions, 24 Rev. Acct. Stud. 536, 537–42 (2019) (reanalyzing the Bebchuk data using value-weighted returns and — inding no signi — icant long-term per — ormance improvement — or the typical shareholder, with results concentrated in the smallest twenty percent o — targeted — irms). 328 Seth C. Oranburg, Exclusive Inclusion __ (manuscript on — ile with au- thor) (“ExI”) (documenting the 2025 — ederal en — orcement campaign against university antisemitism and analyzing the settlements’ — ailure to require governance re — orm). See Holding Campus Leaders Accountable and Con — ronting Antisemitism: Hearing Be — ore the H. Comm. on Educ. & the Work-

orce, 118th Cong., Serial No. 118-31 (Dec. 5, 2023), https://www.con- gress.gov/118/chrg/CHRG-118hhrg56239/CHRG-118hhrg56239.pd —

(testimony o

Dr. Claudine Gay, Ms. Liz Magill, and Dr. Sally Kornbluth; pivotal exchange with Rep. Elise Ste — anik in which each president gave equivocal answers when asked whether calling — or the genocide o — Jews violated university conduct codes; Magill: “It is a context-dependent de- cision, Congresswoman”; Gay: “It can be, depending on the context”). See Press Release, U.S. Dep’t o — Educ., O —

ice — or Civil Rights, U.S. Department o — Education’s O —

ice — or Civil Rights Sends Letters to 60 Universities Under In- vestigation — or Antisemitic Discrimination and Harassment (Mar. 10, 2025) 440 Law and Governance

(

ive directed investigations under Executive Order o — January 29, 2025, and — i — ty- — ive complaint-based investigations). The Columbia settlement totaled approximately $200 million, including $179 million to resolve

ederal investigation and a $21 million EEOC component. See Columbia Univ. O —

ice o — the President, Our Resolution with the Federal Government (July 23, 2025); Press Release, EEOC, In Largest EEOC Public Settlement in Almost 20 Years, Columbia University Agrees to Pay $21 Million to Resolve EEOC Antisemitism Charges (July 23, 2025). See also Brown University, University Update on Federal Matters (July 30, 2025) (reporting a $50 mil- lion, ten-year commitment to state work — orce development organiza- tions in Rhode Island, structured as policy compliance without govern- ance structural change). 329 ExI, supra note 1, at __ (developing the concept o — the “sovereign char- ity” as an institution that wields public in — luence, receives public subsidy, and per — orms public — unctions while governing itsel — as a private corpo- ration answerable to no external constituency). The $200 billion aggre- gate comprises — ederal research grants o — $64.6 billion (FY 2024), see Nat’l Ctr. — or Sci. & Eng’g Statistics, NSF, Higher Education Research and Devel- opment Survey, Fiscal Year 2024 (2025); — ederal student loan disbursements o — $81.3 billion and Pell Grants o — $37.9 billion (2024-25 preliminary), see Fed. Student Aid, Federal Student Aid Posts Updated Reports to FSA Data Center (Aug. 21, 2025); and higher-education tax bene — its averaging ap- proximately $25 billion per year, see Margot L. Crandall-Hollick, Cong. Rsch. Serv., R41967, Higher Education Tax Bene — its: Brie — Overview and Budgetary E —

ects (updated May 26, 2021). 330 The sovereign charity — ramework is developed more — ully in Seth C. Oranburg, Sovereign Charities (manuscript on — ile with author) (analyzing how elite private universities came to wield public power without public accountability). This chapter applies that structural analysis to the spe- ci — ic governance — ailure o — exclusive inclusion. 331 ExI, supra note 1, at __ (tracing the origin o — sel — -perpetuating board governance to colonial-era university charters and documenting its per- Conclusion and Research Agenda 441

sistence in contemporary governance, including at Harvard, whose Cor- poration has governed by internal appointment since 1650). See Charter o — the President and Fellows o — Harvard College (1650), reprinted in Char- ters and Legislative Acts Relating to the Governance o — Harvard, 1650– 1814, UAI 15.100, Harvard University Archives (establishing a Corpora- tion o — seven persons with sel — -perpetuating authority and “perpetual succession”; the Corporation expanded — rom seven to thirteen members in 2010 but retains its sel — -perpetuating structure, see Harvard Corpora- tion Governance Review Committee, Report to the University Community (Dec. 6, 2010)). For Yale’s hybrid structure, see By-Laws o — Yale Univer- sity § 1 (approved by the Corporation, Sept. 30, 2023) (speci — ying nine- teen members: three ex o —

icio, ten Successor Trustees who elect their own successors — or two six-year terms, and six Alumni Fellows elected by eligible degree-holders — or staggered six-year terms). 332 See Alan R. Palmiter, The Duty o — Obedience: The Forgotten Fiduciary Ob- ligation, 55 N.Y.L. Sch. L. Rev. 457, 457–62 (2010–2011) (documenting the historical presence o — the duty o — obedience in nonpro — it law and its e —

ective elimination through modern nonpro — it practice and board de — - erence by state attorneys general). 333 Robertson v. Princeton Univ., No. C-99-02 (N.J. Super. Ct. settled Dec. 10, 2008) (resolving a six-year litigation over whether Princeton had di- verted Robertson Foundation endowment — rom its stated mission o —

training graduates

or — ederal government service; settlement required structural separation and trans — er o — $50 million to an independent en- tity, demonstrating that speci — ic mission commitments are justiciable when dra — ted with su —

icient precision). 334 ExI, supra note 1, at __ (documenting the attorney-general-only stand- ing rule and arguing that the rule, which re — lected an era when charities were small and local, produces structural impunity when applied to insti- tutions with billion-dollar endowments and governance decisions a —

ect- ing millions o — people). The under-resourcing o — attorney general en-

orcement is well documented. See Cindy M. Lott et al., State Regulation and En — orcement in the Charitable Sector 8, 33 (Urban Inst. & Columbia 442 Law and Governance

Law Sch., Sept. 2016) (

inding only 355 — ull-time-equivalent charity regu- lators nationwide across 48 reporting jurisdictions, with 31% o — jurisdic- tions devoting less than one FTE to charities oversight; concluding that resources are “minuscule compared with the oversight they are expected to provide”); Marion R. Fremont-Smith, Governing Nonpro — it Organiza- tions: Federal and State Law and Regulation 318–20 (Belknap Press o — Har- vard Univ. Press 2004) (documenting that only approximately ten states actively en — orced — iduciary duties o — charitable — iduciaries); James J. Fish- man, Improving Charitable Accountability, 62 Md. L. Rev. 218, 262 (2003) (“Sta —

ing problems and a relative lack o — interest in monitoring nonpro — - its has made attorney general oversight more theoretical than deter- rent.”); Evelyn Brody, Whose Public? Parochialism and Paternalism in State Charity Law En — orcement, 79 Ind. L.J. 937, 947–48 (2004) (documenting the political dimensions o — en — orcement selectivity). 335 Turner v. Victoria, S271054 (Cal. 2023) (holding, under Cali — ornia non- pro — it en — orcement statutes, that a director who loses her board position a — ter — iling suit retains standing as a director to pursue the charitable trust breach claim; the opinion addressed director standing speci — ically and did not create or endorse standing — or other constituencies such as students, — aculty, or donors). See ExI, supra note 1, at __ (noting Turner v. Victoria as evidence that Cali — ornia courts remain open to legislative ex- pansion o — standing in the nonpro — it context, while acknowledging that the case itsel — provides no direct authority — or non-director stakeholder en — orcement). 336 ExI, supra note 1, at __ (introducing the concept o — “exclusive inclu- sion” and de — ining it as “the governance state in which an institution claims to protect all members while structurally ensuring that some are unprotected, because sel — -perpetuating boards, hollow — iduciary duties, and the absence o — stakeholder standing create an accountability void that organized — actions exploit to exclude dis — avored identity groups”). Conclusion and Research Agenda 443

337 ExI, supra note 1, at __ (presenting the disentanglement matrix as a di- agnostic tool — or identi — ying exclusive inclusion, sorting contested insti- tutional acts along two axes: who is acting (individual or institution) and what is targeted (perspective or identity)). 338 ExI, supra note 1, at __ (documenting that at UCLA, 115 academic boy- cott endorsers included 20 — aculty with chairs or leadership roles; at UC Berkeley, 171 endorsers included 19 in academic leadership positions; at UC Santa Cruz, 55 endorsers included an Associate Campus Provost and

our residential college provosts). Survey data shows 57.3% o — Jewish university members reported direct experiences o — antisemitism — ollow- ing October 7, 2023. 339 ExI, supra note 1, at __ (citing research documenting “race-avoidant processes” within computer science departmental governance at — our public universities, — inding that departmental DEI structures prioritized gender while systematically avoiding race, producing systematic exclu- sion o — Women o — Color). 340 ExI, supra note 1, at __ (citing research documenting that a — acially neutral behavioral pact at a predominantly White liberal arts college pro- duced systematic racial inequality through di —

erential en — orcement, with Black students — acing higher stakes — or violations, reduced access to in-

ormal rule-breaking, and intensi — ied surveillance). 341 ExI, supra note 1, at __ (citing empirical research on legal academia documenting signi — icant underrepresentation o — conservative and liber- tarian scholars and identi — ying hiring committee processes consistent with ideological screening; survey evidence showing only 6% o — social and personality psychologists identi — y as conservative; and — indings that 47% o — conservative — aculty sel — -censor compared to 19% o — liberal — ac- ulty). 342 ExI, supra note 1, at __ (documenting how accommodation systems that centralize medical veri — ication and grant discretionary power to pro- cedural adjudicators trans — orm ADA entitlements into contingent ad- ministrative privileges, producing systematic exclusion through govern- ance architecture rather than individual hostility). 444 Law and Governance

343 The research — unding — reeze operated through multiple mechanisms. See OMB, Memorandum M-25-13, Temporary Pause o — Agency Grant, Loan, and Other Financial Assistance Programs (Jan. 27, 2025) (directing all agencies to temporarily pause all — ederal — inancial assistance; — ormally re- scinded January 29, 2025, a — ter court orders). NIH imposed a 15% indi- rect cost rate cap estimated to cut $4 billion per year. See NIH, Notice NOT-OD-25-068, Supplemental Guidance to the 2024 NIH Grants Policy Statement: Indirect Cost Rates (Feb. 7, 2025) (enjoined, see preliminary in- junction, D. Mass. Mar. 5, 2025; permanent injunction Apr. 4, 2025). By August 2025, the administration had placed nearly $6 billion on hold at nine universities. See Liam Knox, Trump’s Funding Freezes Against Universi- ties in 5 Charts, Inside Higher Ed (Aug. 28, 2025). Harvard alone — aced ap- proximately $2.2 billion in — rozen grants, see President & Fellows o — Harvard Coll. v. U.S. Dep’t o — Health & Hum. Servs., No. 25-cv-11048-ADB, slip op. at 2 (D. Mass. Sept. 3, 2025), and Columbia had $400 million cancelled. 344 ExI, supra note 1, at __ (analyzing the Columbia, Brown, and other 2025 university settlements and — inding that every settlement addressed compliance outputs without requiring governance structural re — orm; “Not one settlement mandated changes to board composition or selection procedures. Not one created stakeholder standing — or students to chal- lenge department-level exclusion.”). This conclusion is based on review o — all publicly available settlement and resolution agreements — rom 2024– 2026 regarding campus antisemitism, including Trump-era — ederal agreements (Columbia, Brown, Northwestern, Cornell), Biden-era OCR voluntary resolution agreements (University o — Michigan, CUNY, La — a- yette, Drexel, Temple, Occidental, UC System, Johns Hopkins, UIUC), and private litigation settlements (Harvard, UCLA/UC, Barnard). Sev- eral settlements contain con — idential terms, but no publicly available agreement required changes to board composition, trustee selection pro- cedures, or stakeholder standing mechanisms. 345 President & Fellows o — Harvard Coll. v. U.S. Dep’t o — Health & Hum. Servs., No. 25-cv-11048-ADB, slip op. (D. Mass. Sept. 3, 2025) (Burroughs, J.), consolidated with Am. Ass’n o — Univ. Pro — essors–Harvard Faculty Chapter v. Conclusion and Research Agenda 445

U.S. Dep’t o

Justice, No. 25-cv-10910-ADB (D. Mass. — iled Apr. 11, 2025) (consolidated 84-page opinion granting summary judgment — or Harvard,

inding the $2.2 billion — ederal — unding — reeze violated the APA as arbi- trary and capricious and violated the First Amendment; concluding the administration had “used antisemitism as a smokescreen — or a targeted, ideologically-motivated assault on this country’s premier universities”). The opinion consolidated the two cases be — ore the same judge who had presided over Students — or Fair Admissions v. Harvard. A separate AAUP ac- tion involving Columbia, AAUP v. DOJ, No. 1:25-cv-02429 (S.D.N.Y.

iled Mar. 25, 2025) (Vyskocil, J.), was dismissed — or lack o — standing. Ap- peal to the First Circuit — rom the D. Mass. opinion was — iled December 18, 2025, and remains pending. See ExI, supra note 1, at __ (analyzing Harvard’s judicial victory as evidence that — ederal coercion — ails structur- ally: the APA ruling eliminated an unlaw — ul — unding — reeze but le — t the governance void that produced the original campus crisis intact). 346 Lauren B. Edelman, Legal Ambiguity and Symbolic Structures: Organiza- tional Mediation o — Civil Rights Law, 97 Am. J. Soc. 1531, 1531–41 (1992) (developing the legal endogeneity thesis: organizations shape legal com- pliance standards by constructing symbolic structures that courts de — er to as evidence o — compliance, without the behavioral compliance the law was designed to produce). See Chapter 7, supra (identi — ying compliance substitution as Mechanism 6: law requiring — ormal governance structures that organizations satis — y symbolically while leaving governance behav- ior unchanged). 347 The $37.4 billion — igure is a conservative lower bound — or 2021 data. See Elizabeth Plummer, Mariana P. Socal & Ge Bai, Estimation o — Tax Ben- e — it o — US Nonpro — it Hospitals, 332 JAMA 1732 (2024) (analyzing Medicare Cost Reports — or 2,927 nonpro — it hospitals and computing total tax bene-

its comprising — ederal income tax ($11.5B), sales tax ($9.1B), property tax ($7.8B), state income tax ($3.7B), charitable contributions ($3.2B), bond

inancing ($2.1B), and — ederal unemployment tax ($0.2B); characterizing the estimate as a “conservative lower bound” excluding excise taxes on utilities, communications, and — uel). Rev. Rul. 69-545, 1969-2 C.B. 117 446 Law and Governance

(conditioning nonpro

it hospital tax exemption on promoting “the health o — a broad class o — individuals in the community served”). See Pa- tient Protection and A —

ordable Care Act, Pub. L. No. 111-148, § 9007, 124 Stat. 119, 855–61 (2010) (codi — ied at I.R.C. § 501(r)) (requiring nonpro — it hospitals to conduct triennial community health needs assessments with community input and to adopt implementation strategies as conditions o — tax exemption). See ExI, supra note 1, at __ (using the hospital commu- nity bene — it standard as a model — or university operational test re — orm and noting that the standard was adopted through IRS rulemaking rather than congressional legislation). 348 See, e.g., Tooley v. Donaldson, Lu — kin & Jenrette, Inc., 845 A.2d 1031, 1033 (Del. 2004) (distinguishing direct — rom derivative claims by whether the injury is to the corporation or to the shareholder individually). The cor- porate derivative action permits shareholders to en — orce the corpora- tion’s rights against — aithless — iduciaries while maintaining the board’s authority over ordinary business decisions. The mechanism is accounta- bility without participation: the shareholder checks governance quality without entering the governance process. 349 Aronson v. Lewis, 473 A.2d 805, 811-14 (Del. 1984) (establishing the de- mand requirement — or derivative actions and speci — ying that shareholders must demonstrate that the board — aces a substantial likelihood o — liability be — ore equity excuses demand and permits the shareholder to proceed). The demand requirement is the procedural mechanism that separates en-

orcement standing — rom governance participation: it requires the share- holder to demonstrate that the board’s own governance process cannot address the — ailure be — ore permitting external en — orcement. 350 ExI, supra note 1, at __ (analyzing the constitutional constraints on op- erational test re — orm and concluding that content-neutral governance re- quirements occupy di —

erent constitutional ground than viewpoint-based conditions on institutional expression; the proposed operational test “targets governance structures, not viewpoint or speech”). 351 See, e.g., Gregg v. Georgia, 428 U.S. 153 (1976); Glossip v. Gross, 576 U.S. 863 (2015). Conclusion and Research Agenda 447

352 See, e.g., New York State Ri — le & Pistol Ass’n v. Bruen, 597 U.S. 1 (2022) (gun regulation); Citizens United v. FEC, 558 U.S. 310 (2010) (campaign — i- nance); United States v. Booker, 543 U.S. 220 (2005) (sentencing). 353 Motor Vehicle M — rs. Ass’n v. State Farm Mut. Auto. Ins. Co., 463 U.S. 29 (1983). 354 Friedrich A. Hayek, The Use o — Knowledge in Society, 35 Am. Econ. Rev. 519, 526-28 (1945) (arguing that the price system — unctions as a mecha- nism — or communicating dispersed in — ormation and coordinating action without centralized authority). Market governance in Hayek’s sense op- erates through the price mechanism rather than through internal deci- sion-making, monitoring, and sanctioning — which means the — our-ele- ment governance test o — Chapter 1 does not apply to market coordination as such. Market discipline is real, but it is not governance in the — rame- work’s sense. See Albert O. Hirschman, Exit, Voice, and Loyalty: Responses to Decline in Firms, Organizations, and States 15–47 (1970) (distinguishing between exit as a market mechanism and voice as an institutional mecha- nism — or organizational response to decline). 355 On the distinction between algorithmic coordination and governance, see Lawrence Lessig, Code: And Other Laws o — Cyberspace, Version 2.0 121-37 (Basic Books, 2006) (arguing that code constrains behavior through ar- chitecture rather than through the deliberative, sanctioning, and adjust- ment processes that governance institutions employ). Algorithmic sys- tems may incorporate — eatures that resemble governance elements — au- tomated monitoring, programmatic sanctions — but without the mem- ber participation in collective decision-making and adjustment that the

our-element test requires, these — eatures are closer to regulation by de- sign than to governance in the — ramework’s sense. 356 Charles M. Tiebout, A Pure Theory o — Local Expenditures, 64 J. Pol. Econ. 416, 418-20 (1956) (arguing that competition among local jurisdic- tions — or mobile residents produces e —

icient provision o — local public goods). Tiebout’s insight depends on voluntary exit: residents dissatis — ied with local governance can move. Mancur Olson, The Logic o — Collective Action: Public Goods and the Theory o — Groups 33-36 (Harvard University 448 Law and Governance

Press, 1965) (analyzing the conditions under which groups can overcome collective action problems, with group size and excludability as the deci- sive variables). Both Tiebout and Olson identi — y structural — eatures — voluntary exit and excludability — that are present in club-good govern- ance institutions and absent in state governance, which is why the

ramework’s analytical tools are calibrated — or the — ormer and not the lat- ter. 357 See, e.g., Nat’l Soc’y o — Pro — ‘l Eng’rs v. United States, 435 U.S. 679 (1978) (trade association’s ethical canon prohibiting competitive bidding vio- lated § 1 o — the Sherman Act); FTC v. Ind. Fed’n o — Dentists, 476 U.S. 447 (1986) (dental association’s policy o — withholding x-rays — rom insurers constituted un — air method o — competition). 358 See N.C. State Bd. o — Dental Exam’rs v. FTC, 574 U.S. 494 (2015) (state li- censing board composed o — market participants was not entitled to state- action immunity when excluding non-dentists — rom teeth-whitening market). 359 Roberts v. United States Jaycees, 468 U.S. 609 (1984) (holding that Min- nesota’s Human Rights Act compelling the Jaycees to admit women did not abridge — reedom o — intimate or expressive association). 360 See Chapter 9, supra; Lisa Bernstein, Opting Out o — the Legal System: Ex- tralegal Contractual Relations in the Diamond Industry, 21 J. Legal Stud. 115 (1992). 361 Seth C. Oranburg, Market Power and Governance Power: A Dual-Metric Framework — or Antitrust En — orcement in the DAO Era, CPI Antitrust Chron., Feb. 2026 (proposing that antitrust analysis o — decentralized gov- ernance institutions requires measuring both market concentration through the Her — indahl-Hirschman Index and governance concentration through the Nakamoto coe —

icient and Gini coe —

icients applied to token- voting distributions; developing a — our-quadrant classi — ication — Decen- tralized Titan, Algorithmic Leviathan, Commons Ideal, Captive Plat — orm — — or institutions that combine market power with varying degrees o —

governance decentralization).