UNBUNDLING EMPLOYMENT:
    FLEXIBLE BENEFITS FOR THE GIG ECONOMY

                             Seth C. Oranburg*

                                   ABSTRACT    Federal labor law requires employers to give employees a rigid bundle o ---  bene --- its, including the right to unionize, unemployment insurance, worker’s compensation insurance, health insurance,  --- amily medical leave, and more. These bene --- its are not  --- ree—bene --- its cost about one-third o ---  wages—and someone must pay  --- or them. Which o ---

these bene

its are worth their cost? This Article takes a theoretical approach to that problem and proposes a — lexible bene — its solution. Labor law developed under a traditional model o — work: long-term employees depended on a single employer to engage in goods- producing work. Few people work that way today. Instead, modern workers are increasingly using multiple technology plat — orms (such as Uber, Ly — t, TaskRabbit, Amazon Flex, DoorDash, Handy, Moonlighting, FLEXABLE, PeoplePerHour, Rover, Snagajob, TaskEasy, Upwork, and many more) to provide short-term service- producing work. Labor laws are a bad — it — or this “gig economy.” New legal paradigms are needed. The rigid labor law classi — ication o — all workers as either “employees” (who get the entire bundle o — bene — its) or “independent contractors” (who get none) has led to many lawsuits attempting to rede — ine who is an “employee” in the gig economy. This issue grows larger as more than one- — i — th o — the work — orce is now categorized as an independent contractor. Ironically, the requirement to provide a rigid bundle o — bene — its to employees has resulted in — ewer workers receiving any bene — its at all.

* Associate Pro — essor, Duquesne University School o — Law; Research Fellow and Program A —

iliate Scholar, New York University School o — Law; J.D., University o — Chicago Law School; B.A., University o — Florida.

                                         1




   Electronic copy available at: https://ssrn.com/abstract=3135936

2 DREXEL LAW REVIEW [Vol. 11:1

This Article argues

or unbundling employment bene — its so workers in the gig economy can obtain a more optimal mix o — bene — its and wages. This Article also provides a — ramework — or a more — lexible system o — employee bene — its. It thus makes three contributions. First, this Article demonstrates how a rigid requirement o — employment bene — its can harm workers. Second, it shows how labor law should incorporate advances in economic theory that it has hereto — ore generally ignored. Third, this Article presents a — lexible — ramework to solve the re — ractory problem o — rigid worker categorization.

                                TABLE OF CONTENTS INTRODUCTION .................................................................................. 2 I. LABOR AND THE LAW, THEN AND NOW........................................ 8
   A. Origins o ---  Modern Labor Law............................................. 9
   B. The New Economy ............................................................. 14 II. PROBLEMS APPLYING OLD LABOR LAW TO THE NEW ECONOMY
   ................................................................................................ 20
   A. Employees and Independent Contractors .......................... 21
   B. The Joint Employment Doctrine ........................................ 35
   C. The Real Problem o ---  Monopsony ....................................... 41 III. THE GIG WORKER CLASSIFICATION .......................................... 46
   A. What Gig Workers Want................................................... 47
   B. The Gig Worker Framework............................................... 50
   C. Concerns and Criticisms ................................................... 55 CONCLUSION .................................................................................... 57 APPENDIX ......................................................................................... 59


                                    INTRODUCTION   The way people work has drastically changed, and labor law has not kept pace. This Article proposes necessary modernizations to labor law  --- or the new economy. The main problem is that the National Labor Relations Act o ---  1935 requires all workers to  --- it into one o ---  two categories: employees or independent contractors.1 Many scholars in law and economics have recognized tensions between this rigid

1. 29 U.S.C. §§ 151–169 (2018).




     Electronic copy available at: https://ssrn.com/abstract=3135936

2019] UNBUNDLING EMPLOYMENT 3

classi

ication and the — lexible way people work today.2 This Article contributes to that conversation, — irst, by elaborating on why old labor law concepts do not — it with the way people work in the new economy. Second, it points out how labor law — ails to account — or modern economic theory. It proposes that market concentration—a concept borrowed — rom antitrust law—could remedy this — ailure. Third, inspired by securities law, this Article proposes implementation o — a — lexible new labor classi — ication that works — or the new economy. Our current labor laws crystallized eighty years ago under the pressure o — the Great Depression, when over hal — the work — orce labored in — actories.3 Double-digit unemployment and sustained de — lation soon galvanized Congress to strictly regulate the nature o — work.4 Our labor laws were rigidly — orged in this crucible o — economic despair. Yet much has changed since the Great Depression. America — irst evolved — rom a manu — acturing economy to a predominately service economy. Then the gig economy emerged—re — lected by technology plat — orms like Handy, Uber, and Airbnb that allow people to o —

er untapped and underutilized labor to each other—which in turn moved the economy even — urther away — rom the manu — acturing model o — work. This new economy re — lects a modernized way o — working that does not — it with the old dichotomy o — sta —

ing up versus contracting out. More and more Americans now per — orm various gigs in the sharing economy.5 Regardless, labor law continues to bi — urcate workers into two categories: employees, who are entitled to the entire

1. See, e.g., Liya Palagashvili, Disrupting the Employee and Contractor Laws, 2017 U. CHI. LEGAL F. 379, 379–82 (2018) (recognizing that the current employee-independent contractor labor law structure does not accurately capture the status o — workers providing services in the on-demand economy).
2. Donald M. Fisk, American Labor in the 20th Century, 6 COMPENSATION & WORKING CONDITIONS 3, 5 (2001).
3. Robert A. Margo, Employment and Unemployment in the 1930s, 7 J. ECON. PERSPS. 41, 42, 45 (1993).

4. See Palagashvili, supra note 2, at 379–80 (noting the “rapid growth” o

   this new emerging economy).

   Electronic copy available at: https://ssrn.com/abstract=3135936 4 DREXEL LAW REVIEW [Vol. 11:1

bundle o

employment bene — its, and independent contractors, who are not entitled to any.6 This Article argues that neither category properly protects or even accurately describes how people work in the sharing economy. Moreover, merely adding a rigid third category will not solve the problem. This Article thus proposes a new and

lexible — ramework — or rede — ining worker classi — ications and unbundling the bene — its — rom work: the gig worker. Under this proposal, each — irm that creates technology plat — orms — or the sharing economy would be permitted to register its own de — inition o — “gig worker” with the Department o — Labor (DOL) by — iling a new type o — registration statement that this Article calls a “Form GW.” The Form GW would stipulate which employment bene — its (e.g., health insurance, retirement plan contribution, and unemployment insurance) would be o —

ered, and which bene — its would not. The DOL would review each Form GW to ensure that the proposed bene — its comply with laws that apply to all workers (such as minimum wage) and that the — orm is clear enough to be easily understood by a reasonable worker. Aside — rom these minimum standards, however, the DOL would not review the merits o — the proposal as o —

ering a good or bad work opportunity. Rather, the DOL would simply determine whether the de — inition complies with labor laws. Firms that success — ully register a Form GW would be able to o —

er gig worker jobs accordingly. The term “o —

er” is used deliberately here to suggest parallels between this new labor- law proposal and the established securities-law paradigm — or o —

ering stock. This Form GW process is designed to mirror the most success — ul elements o — the Securities Act o —

1. O

   ten called the “truth-in-securities law,” the 1933 Act has two primary objectives: — irst, providing necessary in — ormation to investors, and second, prohibiting deceit, misrepresentation, and — raud in the o —

ering o — investment opportunities.7 Under

1. 29 U.S.C. §§ 151–169.

2. The Laws that Govern the Securities Industry, U.S. SEC. & EXCHANGE COMMISSION, https://www.sec.gov/answers/about-lawsshtml.html (last modi — ied Oct. 1, 2013).

   Electronic copy available at: https://ssrn.com/abstract=3135936 2019] UNBUNDLING EMPLOYMENT 5

the 1933 Act,

irms must — ile a Form S-1 with the Securities and Exchange Commission (SEC) be — ore selling stock on a stock market through an initial public o —

ering (IPO).8 The SEC reviews the disclosures in that S-1 — or compliance with securities law and will hold o —

erors criminally liable — or — alse or misleading statements in the disclosures. It does not, however, determine the merits o — the o —

ering as a good or bad investment opportunity. Similarly, the goals o — the Form GW process would be, — irst, to ensure that potential workers would have the in — ormation they need to decide whether to participate on a sharing-economy plat — orm, and second, to hold the plat — orm liable — or — alse or misleading claims about work on its plat — orm or the bene — its obtained thereby. Unbundling the bene — its — rom work—allowing — irms to de — ine the set o — bene — its that gig workers on their plat — orms receive—will create competition among plat — orms to o —

er the mix o — pay and bene — its that is most desirable to gig workers. Competitive labor markets can provide power — ul solutions to many employment problems, but markets do not always

unction properly. In particular, the market condition o — labor monopsony—a market structure where a single buyer controls the labor market as the major or only purchaser o — a certain type o — labor services—can lead to lower wages and — ewer bene — its

or workers.9 It would be abominable i — the law were to — acilitate exploitation o — powerless workers by power — ul — irms. There — ore, the gig worker proposal also borrows concepts — rom antitrust and competition law: the DOL could deny a Form GW proposal based on particular market conditions, such as monopsony, collusion, or concentration. Other scholars have proposed that there should simply be a third category o — worker.10 These proposals are, however,

1. 15 U.S.C. § 77

   (e)(1) (2018).

2. COUNCIL OF ECON. ADVISERS, LABOR MARKET MONOPSONY: TRENDS, CONSEQUENCES, AND POLICY RESPONSES 2–3 (2016), https://obamawhitehouse.archives.gov/sites/de — ault/ — iles/page/ — iles/20161025_monopsony_lab or_mrkt_cea.pd — .

3. See generally Miriam A. Cherry & Antonio Aloisi, “Dependent Contractors” in the Gig Economy: A Comparative Approach, 66 AM. U. L. REV. 635 (2017) (concluding that the de — inition

   Electronic copy available at: https://ssrn.com/abstract=3135936 6 DREXEL LAW REVIEW [Vol. 11:1

undamentally insu —

icient because they do not account — or the myriad new ways people might work as the economy continues to change. Moreover, legislating a singular new category with strict criteria — or inclusion has been demonstrated to do little to solve the problems that currently exist in a two-category system.11 There — ore, we must do more than simply add a third category that — its within the existing labor law — ramework. Instead, we must reconceptualize labor law under a more

lexible paradigm that is better prepared — or the — uture o — work. The gig worker process o —

ered herein is intended to be such a paradigm. Some may be concerned about the costs o — instituting a — lexible process such as the one proposed here, but there is precedent in both securities law and antitrust law that shows how a proposal o — this nature might pay — or itsel — , and then some.12 At the outset, it is important to recognize that this gig worker proposal does not take a position regarding whether individuals should have — reedom o — contract. The argument — or

reedom o — contract is essentially that individuals should have the right to — orm any nonviolent contract without government restrictions.13 Some have argued that Article I, section 10 o — the U.S. Constitution prohibits states — rom impairing the obligations o — contracts that were — reely entered into.14 Others

o

“employee” should be expanded to include a hybrid category o — workers known as “dependent contractors”); Brett Harris, Uber, Ly — t, and Regulating the Sharing Economy, 41 SEATTLE U. L. REV. 269 (2017) (exploring regulatory issues concerning the “gig economy”); SETH D. HARRIS & ALAN B. KRUEGER, THE HAMILTON PROJECT, BROOKINGS INST., A PROPOSAL FOR MODERNIZING LABOR LAWS FOR TWENTY-FIRST-CENTURY WORK: THE “INDEPENDENT WORKER” 2 (2015), http://www.hamiltonproject.org/assets/ — iles/modernizing_labor_laws_ — or_twenty_ — irst_centur y_work_krueger_harris.pd — (suggesting the establishment o — an “independent worker” category).

1. See Cherry & Aloisi, supra note 10, at 675–76; Chelsea Fitzgerald, When Tech Startups Outgrow the 1099 Model: Moving Firms Out o — the Kiddie Pool, 18 VAND. J. ENT. & TECH. L. 629, 632–35 (2016); Mark J. Loewenstein, Agency Law and the New Economy, 72 BUS. LAW. 1009, 1036– 37 (2017).

2. See in

   ra Part IV.

3. Stephanie Drotar, Breaking “Too Darn Bad”: Restoring the Balance Between Freedom o

Contract and Consumer Protection, 59 N.Y. L. SCH. L. REV. 603, 604–05 (2014).

1. See, e.g., DAVID E. BERNSTEIN, FREEDOM OF CONTRACT 8 (2008), https://www.law.gmu.edu/assets/ — iles/publications/working_papers/08-

   Electronic copy available at: https://ssrn.com/abstract=3135936 2019] UNBUNDLING EMPLOYMENT 7

have challenged the market paradigm

rom a moral perspective by, — or example, asking whether people should be allowed to sell their organs or sexual — avors.15 But this Article does not opine on these broader issues. Rather, it more narrowly — ocuses on whether participants in sharing-economy plat — orms, who can have varied careers, jobs, and tasks available online and in person, — it into one o — two buckets: employees or independent contractors. The — indings show they do not, and this Article accordingly o —

ers a — lexible third classi — ication. This proposal also does not address tax issues. It is worth mentioning, however, that tax issues o — ten drive labor law classi — ications.16 For example, i — a worker is classi — ied as an employee, then the employer withholds payroll taxes. It is much easier — or the tax authorities to collect — rom a — ew large employers than — rom several individual employees. But this is a story o — the tail wagging the dog. Rather than — ocus on collecting taxes — rom work, we should — ocus on creating more jobs and better bene — its. First, we must unleash the potential o —

the sharing economy to help people

ind the work and services they need; then, the government can contemplate and implement an appropriate tax structure. The — undamental problem with labor laws is their in — lexible adherence to the outdated codi — ication o — all workers as either employees or independent contractors. As a result o — this rigidity, many workers’ bene — its and protections are suboptimal. However, this unhappy circumstance can be improved by unbundling the bene — its — rom work. This unbundling will allow — irms to compete — or workers by o —

ering them the best mix o — bene — its: the gig worker solution. To argue

or its gig worker solution, this Article proceeds as — ollows. First, the Article describes the Depression-era economy and the labor laws created in response thereto, and contrasts this with

51%20Freedom%20o

%20Contract.pd — (arguing Supreme Court cases throughout American history support this — inding).

1. E.g., MICHAEL J. TREBILCOCK, THE LIMITS OF FREEDOM OF CONTRACT 35–38 (1997).

2. Susan C. Allen, Employee or Independent Contractor?: The Worker Classi

   ication Dilemma, TAX ADVISER (Oct. 1, 2015), https://www.thetaxadviser.com/issues/2015/oct/employee-or- independent-contractor.html.

   Electronic copy available at: https://ssrn.com/abstract=3135936 8 DREXEL LAW REVIEW [Vol. 11:1

the new economy to show how labor law is generally ill-suited

or gig work. Second, it analyzes several particularly thorny problems that arise when trying to apply these old laws to the new economy. Finally, the Article expands on its gig worker solution to these problems and addresses some criticisms to its proposed approach.

            I. LABOR AND THE LAW, THEN AND NOW   Labor law as it exists today was largely conceived and codi --- ied when over hal ---  o ---  Americans worked in manu --- acturing.17 It is vital to understand the zeitgeist o ---  the manu --- acturing economy and its most critical  --- ailure, the Great Depression, in order to understand the legislative history and political economy that gave rise to labor law. It then becomes apparent that the sharing economy—and, indeed, the entire social and political climate—is quite di ---

erent today. The manu — acturing economy is based on resource extraction: raw materials are mined — rom the earth, ore is smelted, and products, like cars, are assembled. In business terms, value in the manu — acturing economy moves — rom le — t to right, as each step in the manu — acturing process adds value to consumers.18 This is called a value chain.19 To understand the manu — acturing value chain, consider the manu — acture o — a car. First, coal is mined to make steel. This steel has more value than the coal in the ground did. Second, the steel is transported to a car assembly-line — actory, where it has more value as a car door panel. Third, the door panel is incorporated with other inputs — rom other upstream producers, such as glass windshields and electronic components, to create a — unctional car. A whole car that can drive is worth more than the sum o — its static parts. Fourth, the — inished product (our new car) is transported — rom a centralized manu — acturing — acility in

1. See in

   ra Section II.A.

2. Porter’s Value Chain: Understanding How Value Is Created Within Organizations, MINDTOOLS, https://www.mindtools.com/pages/article/newSTR_66.htm (last visited Dec. 16, 2018).

3. Id.

   Electronic copy available at: https://ssrn.com/abstract=3135936 2019] UNBUNDLING EMPLOYMENT 9

Indiana to retail auto dealers all over America, where it is more convenient — or prospective buyers to test and acquire that car. Fi — th, salespeople at those dealerships in — orm buyers about the car’s — eatures, help buyers secure — inancing, and teach them to use the technical — eatures on the vehicle. Sixth, independent a — termarket car maintenance and repair service providers help keep the car running. Each step in this process, which can be visualized as a river upon which inputs — low — rom upstream supply to downstream sales, adds value to the product. For a visual illustration o — the value chain in the manu — acturing economy, see Figure 1 in the Appendix. It is most important to recognize that the labor laws were — ormed with a manu — acturing economy, not a gig economy, in mind.

                    A. Origins o ---  Modern Labor Law   In 1929, only about 20% o ---  gain --- ully employed Americans worked in the service sector and 22% worked in agriculture,

orestry, and — ishing.20 Meanwhile, more than hal — o — all gain — ully employed Americans worked in or — or — actories, with jobs in extraction o — minerals, manu — acturing, construction, transportation, and trade.21 The unemployment rate was less than 1% o — gain — ul workers.22 Then, the Great Depression power — ully mani — ested on October 29, 1929. On this date, known as Black Tuesday, stock markets crashed. People made runs on bankrupt banks, and panic erupted in the streets. Surviving banks substantially curtailed their lending.23 Factories dramatically reduced output by over 30%, and total — actory productivity decreased by 18%.24

1. U.S. BUREAU OF THE CENSUS, HISTORICAL STATISTICS OF THE UNITED STATES, COLONIAL TIMES TO 1970, PART 1, at 137–38 (1975), https:// — raser.stlouis — ed.org/ — iles/docs/publications/histstatus/hstat1970_cen_1975_v1.pd — .
2. Id. at 138.
3. Id. at 135.
4. Charles W. Calomiris, Financial Factors in the Great Depression, 7 J. ECON. PERSP. 61, 69 (1993) (“Surviving banks substantially curtailed their lending, with loan-to-deposit ratios

alling — rom 0.85 in 1929 to a low o — 0.58 in January 1933.”).

1. See Lee E. Ohanian, Why Did Productivity Fall So Much During the Great Depression?, 91 AM. ECON. REV. 34, 34 (2007).

   Electronic copy available at: https://ssrn.com/abstract=3135936 10 DREXEL LAW REVIEW [Vol. 11:1

Millions o

Americans who worked in these — actories were laid o —

, and the unemployment rate spiked to over 22% — rom 1933 to 1935.25 The masses o — unemployed and underemployed Americans had less money to spend because they were not earning enough income.26 This put additional pressure on the remaining manu — acturers, who had to — urther decrease output in light o — decreased demand.27 The vicious cycle o — scarcity and in — lation disrupted the entire international economy and changed the way people — elt about work and the subsequent role o — government.28 Meanwhile, there were also thinkers coming up with new ideas about how to understand the economy. John Maynard Keynes argued that the cause o — this depression was insu —

icient spending power.29 His solution to this crisis and proposition to prevent similar ones in the — uture was — or the government to create policies that ensure the average person has more money in his or her pocket.30 In other words, Keynes believed the

ederal government could solve the insu —

icient spending power that led to the Great Depression by the government itsel —

spending more. He argued that government expenditures, especially on in — rastructure, would solve this problem.31 In 1932, at the height o — the Great Depression, when Franklin Delano Roosevelt ran against incumbent President Herbert Hoover, they both campaigned on an orthodox economic plat — orm.32 For example, FDR originally promised to balance

1. U.S. BUREAU OF THE CENSUS, supra note 20, at 135.
2. See id. at 146–61.
3. See id.
4. BARRY EICHENGREEN & KEVIN H. O’ROURKE, A TALE OF TWO DEPRESSIONS 1 (2009), http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.520.3990&rep=rep1&type=pd — (“The Great Depression was a global phenomenon.”).
5. See generally JOHN MAYNARD KEYNES, THE GENERAL THEORY OF EMPLOYMENT, INTEREST, AND MONEY (Harcourt Publ’g Co. 1964) (1953) (arguing that reduced purchasing power leads to diminished economic output).
6. Id.
7. Id.

8. A President’s Evolving Approach to Fiscal Policy in Times o

   Crisis, FDR LIBR., https:// — drlibrary.org/budget (last visited Dec. 15, 2018) [hereina — ter A President’s Evolving Approach] (“FDR began his 1932 campaign — or the presidency espousing orthodox — iscal belie — s. He promised to balance the — ederal budget, which Herbert Hoover had been unable to do.”).

   Electronic copy available at: https://ssrn.com/abstract=3135936 2019] UNBUNDLING EMPLOYMENT 11

the budget.33 Hoover was not that popular at the time, given that during his presidency — ormerly working-class people became dependent on — ood — rom notorious gangster Al Capone’s soup kitchens.34 FDR won the 1932 presidential election in a landslide, but he changed his economic approach once in o —

ice. Even though his 1932 campaign included both conventional and orthodox balanced budget promises, he said in his 1936 campaign that it would have been a crime against the American people to have balanced our budget in 1933, 1934, or 1935.35 Instead, FDR promised “a new deal with the American people.”36 FDR’s New Deal included many policies that provided the government with much more control over labor wages and the pricing o — goods and services. For example, the National Recovery Act authorized FDR, in his executive capacity, to regulate wages and prices directly.37 Such direct wage control by the — ederal government was unprecedented in American history. Many people were concerned by these policies, including economists. Even Keynes, the economist who argued that the government should spend money on in — rastructure to help America recover — rom the Great Depression, wrote a letter to the White House e —

ectively saying that FDR had gone too — ar with these policies.38 Regardless, FDR actually went — urther and

1. Id.

2. See, e.g., Ian Harvey, Al Capone Started One o

   the First Soup Kitchens During the Great Depression, VINTAGE NEWS (May 19, 2017) (“Al Capone’s soup kitchen . . . served over 120,000 meals to hungry people. The — ree soup kitchen kept regular working hours, serving break — ast, lunch, and dinner and — ed thousands every day despite only having a — ew employees.”).

3. See A President’s Evolving Approach, supra note 32 (“FDR answered in 1936 at a campaign speech in Pittsburgh: ‘To balance our budget in 1933 or 1934 or 1935 would have been a crime against the American people. To do so we should either have had to make a capital levy that would have been con — iscatory, or we should have had to set our — ace against human su —

ering with callous indi —

erence. When Americans su —

ered, we re — used to pass by on the other side. Humanity came — irst.’”).

1. Acceptance Speech to the 1932 Democratic Convention, FDR LIBR., https://

   drlibrary.org/dnc- curriculum-hub (last visited Dec. 15, 2018) (“I pledge you, I pledge mysel — to a new deal — or the American people.”).

2. See 15 U.S.C. § 703 (1934) (repealed 1935).

3. John Maynard Keynes, From Keynes to Roosevelt: Our Recovery Plan Assayed, N.Y. TIMES, Dec. 31, 1933, https://www.nytimes.com/1933/12/31/archives/ — rom-keynes-to-roosevelt-our-

   Electronic copy available at: https://ssrn.com/abstract=3135936 12 DREXEL LAW REVIEW [Vol. 11:1

created what some scholars have re

erred to as the most radical piece o — legislation in American — ederal history: the National Labor Relations Act o — 1935 (NLRA).39 The radical nature o — the NLRA is best understood in its historical context. At the time, there seemed to be a war brewing—not a war between nations, but a war between capitalism and communism. The NLRA was an olive branch, extended to the striking masses and their increasingly organized labor organizations to achieve an industrial peace.40 It o —

ered collective bargaining — or its “therapeutic impact on industrial con — lict.”41 Its express goal was to increase wages.42 By its plain terms, the NLRA “apparently accorded a governmental blessing to power — ul workers’ organizations that were to acquire equal bargaining power with corporations, accomplish a redistribution o — income, and subject the workplace to a regime o — participatory democracy.”43 In other words, the NLRA was one o — the most socialist pieces o —

legislation ever passed by the U.S. Congress. Critics even worried that the NLRA would “out-S[oviet] the Russian Soviets.”44 The NLRA thus emerged — rom populism and protest, applying Keynesian economic theory piecemeal and to its

recovery-plan-assayed-the-british.html (“I do not mean to impugn the social justice and social expediency o — the redistribution o — incomes aimed at by the NRA and by the various schemes

or agricultural restriction. The latter, in particular, I should strongly support in principle. But too much emphasis on the remedial value o — a higher price-level as an object in itsel — may lead to serious misapprehension o — the part prices can play in the technique o — recovery. The stimulation o — output by increasing aggregate purchasing power is the right way to get prices up and not the other way around.”).

1. Karl E. Klare, Judicial Deradicalization o

   the Wagner Act and the Origins o — Modern Legal Consciousness, 1937-1941, 62 MINN. L. REV. 265, 265 (1978) (“When passed, the National Labor Relations (Wagner) Act was perhaps the most radical piece o — legislation ever enacted by the United States Congress.”).

2. Id. at 281.
3. Id. at 282.
4. 29 U.S.C. § 151 (2018) (stating that unequal bargaining power “tends to aggravate recurrent business depressions, by depressing wage rates and the purchasing power o — wage earners”).
5. Klare, supra note 39, at 285.

6. Id. at 286 (quoting ARTHUR M. SCHLESINGER, JR., THE AGE OF ROOSEVELT: THE COMING OF THE NEW DEAL 405 (1961)).

   Electronic copy available at: https://ssrn.com/abstract=3135936 2019] UNBUNDLING EMPLOYMENT 13

socialist extreme. The NLRA established that the o

icial policy o — the United States was to promote labor unions.45 It did not just legalize union activity, but it actively created an a —

irmative duty on the part o — employers to bargain with union representatives.46 The NLRA also created the National Labor Relations Board (NLRB), which was an entity designed to protect workers, primarily by supporting unionization.47 It was not initially certain that such radical legislation would be — ound constitutional, but a pivotal opinion written by Supreme Court Chie — Justice Charles Evans Hughes, who was nominated by Herbert Hoover, validated the NLRA.48 In NLRB v. Jones & Laughlin Steel Corp.,49 the NLRB sought sanctions against Jones & Laughlin on the grounds that the company was discriminating against union employees who wanted to join the Amalgamated Association o — Iron, Steel, and Tin Workers o —

America, a labor organization.50 The NLRA prohibits discrimination against workers on the basis o — union membership, and the NLRB (whose administrative — unction is to en — orce the NLRA) ordered Jones & Laughlin to rehire and give back pay to ten employees who were — ired a — ter they voted to unionize.51 Jones & Laughlin re — used, arguing that the NLRA was unconstitutional.52 Writing — or the majority in a 5–4 decision, Justice Hughes wrote, “Although activities may be intrastate in character when separately considered, i — they have such a close and substantial relation to interstate commerce that their control is essential or appropriate to protect that

45. Introduction to the NLRB, NAT’L LAB. REL. BOARD, https://www.nlrb.gov/nlrb- introduction (last visited Dec. 15, 2018) (“Congress enacted the National Labor Relations Act . . . in 1935 to protect the rights o ---  employees and employers, to encourage collective bargaining, and to curtail certain private sector labor and management practices, which can harm the general wel --- are o ---  workers, businesses and the U.S. economy.”).
46. 29 U.S.C § 158(a)(5).
47. See id. §§ 153–156.
48. Charles            E.          Hughes,           OYEZ:          BODY           POLITIC, https://www.oyez.org/justices/charles_e_hughes (last visited Dec. 16, 2018).
49. 301 U.S. 1 (1937).
50. Id. at 22.
51. Id.
52. Id. at 25.




    Electronic copy available at: https://ssrn.com/abstract=3135936

14 DREXEL LAW REVIEW [Vol. 11:1

commerce

rom burdens and obstructions, Congress cannot be denied the power to exercise that control.”53 This expansive reading o — the Commerce Clause e —

ectively validated the NLRA and declared it constitutional.54 Justice James Clark McReynolds dissented, questioning Congress’s enhanced power under the Commerce Clause,55 but over the next — ew years an emboldened Congress passed additional New Deal legislation that expanded the — ederal government’s control over labor relations.56 The NLRA was succeeded by the Fair Labor Standards Act o —

1938 (FLSA).57 The FLSA legislated many popular sentiments. For example, the FLSA established the — orty-hour workweek.58 Employers must pay overtime i — a worker works — or more than

orty hours in one week.59 The contemporary concept o — time- and-a-hal — pay comes — rom the FLSA.60 Many scholars have questioned whether these populist Depression-era laws were help — ul to workers at the time,61 but it is even more doubt — ul that these policies, rigidly applied, provide optimal working conditions — or workers in our current gig economy.

                             B. The New Economy   The concerns that stimulated the development o ---  labor law are not major issues today. Politically, we no longer  --- ace a “Red Scare.” A violent socialist revolution does not seem to be on the horizon. Economically, the concerns raised by the advent o ---  the sharing economy are very di ---

erent — rom the problems with

1. Id. at 37.
2. Id. at 36–37, 43.
3. Id. at 76–78 (McReynolds, J., dissenting).

4. See, e.g., Fair Labor Standards Act o

   1938, 29 U.S.C. §§ 201–219 (2018).

5. Id.
6. Id. §§ 202, 207.
7. See id.
8. See id. § 207(a)(1).

9. See Julius Getman, The National Labor Relations Act: What Went Wrong; Can We Fix It?, 45 B.C. L. REV. 125, 126 (2003) (“The key provisions that led to such great hopes by unions and their supporters remain in — orce, but a — ter many years o — working with the NLRA, optimism has given way to cynicism and despair about the law’s ability to protect workers and enhance collective bargaining.”).

   Electronic copy available at: https://ssrn.com/abstract=3135936 2019] UNBUNDLING EMPLOYMENT 15

manu

acturing jobs during the Great Depression. Thus, the laws designed to address labor concerns do not adequately represent the interests o — employees today. Most Americans are not deeply concerned about power — ul employers in one- — actory towns because most Americans do not work in such — actories or live in such towns anymore. Looking back with 20/20 hindsight, the demand — or a — orty-hour workweek seems out o — place. Today people are no longer demanding a — orty-hour workweek: they are looking — or a — our-hour workweek.62 Nowadays, people hope to use the internet to leverage their labor productivity to earn more while working less.63 Studies show that reduced workweeks can result in increased productivity and increased happiness.64 Americans’ expectations about work have changed as the reality o — work has changed. The laws have not. Macro labor conditions changed radically a — ter the labor laws were enacted. In 1935, when labor laws were emerging, more than 50% o — America’s labor — orce was involved in the production o — goods.65 Over 90% o — these workers labored in

actories that manu — actured di —

erent commodities.66 Work meant making stu —

, o — ten on an assembly line in a centralized location like a — actory. But that is not how most people work today. In 2013, 83% o — total employment was in the service sector.67 The nature o — work—and the protections workers need thereby—has radically changed in the past seventy years. In

1. See generally TIMOTHY FERRISS, THE 4-HOUR WORKWEEK (2007) (advocating

   or a — our- hour workweek, as well as several other li — estyle changes).

2. Id.
3. Charlotte Graham-McLay, A 4-Day Workweek? A Test Run Shows a Surprising Result, N.Y. TIMES (Jul. 19, 2018), https://www.nytimes.com/2018/07/19/world/asia/ — our-day-workweek- new-zealand.html.

4. U.S. BUREAU OF THE CENSUS, supra note 20, at 137 (showing the

   ollowing labor statistics

or the year 1935: (1) 27,035,000 non- — arm workers, (2) 897,000 mining employees, (3) 912,000 construction employees, (4) 9,069,000 manu — acturing employees, (5) 2,786,000 transportation employees, (6) 5,431,000 trade employees, (7) 1,335,000 — inance employees, (8) 3,142,000 employees in other services, and (9) 3481 government employees).

1. Id.

2. Giuseppe Berlingieri, Outsourcing and the Shi

   t — rom Manu — acturing to Services, CENTREPIECE, Winter 2013–2014, at 16, http://cep.lse.ac.uk/pubs/download/cp413.pd — .

   Electronic copy available at: https://ssrn.com/abstract=3135936 16 DREXEL LAW REVIEW [Vol. 11:1

2017, only 8% o

the non- — arm work — orce (12.2 million o — 150.1 million) labored in manu — acturing,68 whereas 29.1% (14.3 million out o — 49.1 million) worked in manu — acturing in 1947.69 Likewise, — ar — ewer Americans today work in resource- extraction jobs: in 2017, less than 0.5% o — non- — arm workers (678,000) labored in mining and logging,70 whereas 2% o — non-

arm workers (976,000) worked in mining and logging in 1944. In other words, not only have the total number o — workers in mining, logging, and manu — acturing jobs decreased as an absolute number over the past seventy years, but so has the percent o — our population engaged in these jobs. Un — ortunately, while the nature o — work has changed, the nature o — worker protection laws has not kept pace. Seventy years ago, work — or most people meant making stu —

. By the turn o — the millennium, however, work — or most people meant per — orming services. Now, more and more Americans are participating in a new economy that is made possible by technological advancement. The new economy—o — ten called the gig economy or the sharing economy—is not based on resource extraction but upon resource reallocation. That resource can be human labor or capital. One might, — or example, have a car sitting in the garage — our days a week. Be — ore internet technology such as plat — orm apps reduced transaction costs, it was o — ten too expensive to — ind a short-term renter, so the car would sit unused — our-sevenths o — the time. Likewise, it is di —

icult in the traditional economy to repurpose underused labor. Now, internet technology has trans — ormed the way assets are utilized. The internet has substantially decreased the cost o —

inding goods and services.71 The internet has also decreased

1. Manu

   acturing: Employment, Hours, and Earnings, U.S. BUREAU LAB. STAT., https://data.bls.gov/timeseries/CES3000000006?amp%253bdata_tool=XGtable&output_view=d ata&include_graphs=true (last visited Dec. 15, 2018).

2. U.S. BUREAU OF THE CENSUS, supra note 20, at 137.
3. Mining and Logging: Employment, Hours, and Earnings, U.S. BUREAU LAB. STAT., https://data.bls.gov/timeseries/CES1000000001?amp%253bdata_tool=XGtable&output_view=d ata&include_graphs=true (last visited Dec. 16, 2018).

4. Fabio Ancarani, Pricing and the Internet: Frictionless Commerce or Pricer’s Paradise?, 20 EUR. MGMT. J. 680, 680 (2002) (“Not only do customers have lower search costs — or in — ormation about pricing, but — irms and retailers also have lower search costs — or in — ormation about their

   Electronic copy available at: https://ssrn.com/abstract=3135936 2019] UNBUNDLING EMPLOYMENT 17

the cost o

getting in — ormation about goods and services.72 The economist Michael C. Munger de — ines the cost o —

inding and getting in — ormation about goods and services as a “triangulation cost.” In his book Tomorrow 3.0: Transaction Costs and the Sharing Economy, Munger argues that web plat — orm app technologies greatly reduce triangulation costs and create new possibilities — or more e —

icient allocation o — labor and assets.73 Technology—speci — ically, plat — orm apps—has thereby unlocked the ability to economically trans — er short-term labor and short-term leases o — goods. This has created a new economy that reallocates underutilized labor and capital. Today’s technology-enhanced economy has many names. The most general term is simply the “new economy,” a term coined in the late 1990s to describe the dramatic increase in economic growth due to the internet.74 This Article — ocuses primarily on one internet-powered technology: online plat — orms. Online plat — orms are digital matchmakers or marketplaces, where people can o —

er goods or services. While “plat — orm economies” technically predate the internet—think about — lea markets where you can shop or rent a store — ront, or classi — ied ads in the newspapers where you o —

er or — ind work— the internet supercharges matchmaking. Critically, plat — orm economies are distinguishable — rom traditional economies because, in a traditional economy, the value chain moves — rom le — t to right, or — rom production to consumption; in a peer-to- peer economy, however, the value chain is triangular, with a hosting plat — orm at the apex and users on either side.75 It is also important to note that plat — orm economies are complementary to traditional economies. Plat — orm economies

customers.”); see also Leyland F. Pitt et al., Pricing Strategies and the Net, 44 BUS. HORIZONS 45, 45–47 (2001).

1. Florian Zettelmeyer et al., How the Internet Lowers Prices: Evidence

   rom Matched Survey and Automobile Transaction Data, 43 J. MARKETING RES. 168, 168–81 (May 2006).

2. See MICHAEL C. MUNGER, TOMORROW 3.0: TRANSACTION COSTS AND THE SHARING ECONOMY, at vii (2018).
3. James Surowiecki, The New Economy Was a Myth, Right?, WIRED (July 1, 2002), https://www.wired.com/2002/07/myth-2/.

4. Thomas Eisenmann et al., Strategies

   or Two-Sided Markets, 84 HARV. BUS. REV. 1, 2 (2006).

   Electronic copy available at: https://ssrn.com/abstract=3135936 18 DREXEL LAW REVIEW [Vol. 11:1

are based on resource reallocation.76 The resources that were extracted and sold in the traditional economy may be underused, but a matchmaking service can put those resources to better use. For example, consider a vacant home. This home is built — rom goods extracted via the traditional economy. The home was sold to someone who no longer has much use — or it, but it is not a good candidate — or resale due to tax or other reasons. This vacant home is an underused asset. The plat — orm economy provides technological solutions to make better use o — this asset; it could be rented out by the day or month via an online plat — orm. Thus, the plat — orm economy value chain moves both le — t to right and right to le — t, with a plat — orm at the center.77 For a visual illustration o — the value chain in the sharing economy, see Figure 2 in the Appendix.78 Plat — orm economies may be — urther divided into two sub- categories: sharing economies and gig economies. The di —

erence between these economies pertains to whether goods or services are being o —

ered and sought on the plat — orm. This can also be thought o — as the divide between capital and labor. I use the term “sharing economy” to indicate an internet plat — orm that o —

ers matchmaking services or a technological

ramework — or o —

ering the use o — underutilized goods. For example, Airbnb allows people to o —

er up their spare rooms — or rent.79 Turo and Getaround enable people to rent their cars to their neighbors.80 LendingClub allows people to lend each other

1. Nina Gass, What the Emergence o

   the Plat — orm Economy Means — or Business, DUE.COM (Sept. 8, 2017), https://due.com/blog/the-plat — orm-economy/.

2. Press Release, DHL, DHL Reveals the Sharing Economy Is Shaking Up Logistics, (May 9, 2017), http://www.dhl.com/en/press/releases/releases_2017/all/dhl_reveals_the_sharing_economy_is _shaking_up_logistics.html.
3. BEN GESING, DHL TREND RESEARCH, SHARING ECONOMY LOGISTICS: RETHINKING LOGISTICS WITH ACCESS OVER OWNERSHIP 3 (2017), https://www.logistics.dhl/content/dam/dhl/global/core/documents/pd — /glo-core-sharing- economy.pd — .
4. Earn Money as an Airbnb Host, AIRBNB, https://www.airbnb.com/host/homes (last visited Dec. 16, 2018).

5. How It Works, GETAROUND, https://www.getaround.com/tour (last visited Dec. 16, 2018); How Turo Works, TURO, https://turo.com/how-turo-works (last visited Dec. 16, 2018).

   Electronic copy available at: https://ssrn.com/abstract=3135936 2019] UNBUNDLING EMPLOYMENT 19

cash.81 The common

eature o — all these sharing economy plat — orms is that they — eature a rental-like model — or goods which are traditionally di —

icult to rent. Sharing economy plat — orms deal with the reallocation o — capital, typically on a short-term basis. “Gig economy” indicates an internet plat — orm that o —

ers matchmaking — or services (i.e., labor).82 The term has its roots in musical per — ormance, i.e., the band will play a gig tomorrow at Joe’s Bar. This connotes a short or temporary job. Gig is also an abbreviation — or gigabyte,83 and this term thus also has technological connotations that are appropriate — or describing internet plat — orms. Examples o — gig economy plat — orms abound today: Rover matches pet sitters with dog owners.84 Uber and Ly — t match drivers with riders.85 TaskRabbit matches handymen with homeowners.86 The common element in gig economy plat — orms is that they match someone who wants to o —

er up some o — her underused labor with another person who wants to pay the — ormer — or her e —

orts. Gig economies are the — ocus o — this Article. They are technological — rameworks — or o —

ering services; thus, they implicate labor laws. However, work on these plat — orms is quite di —

erent — rom work in a — actory in 1930. Recall that gig economy plat — orms primarily o —

er matchmaking services, so, strictly speaking, people do not work — or the plat — orm, but rather use the plat — orm to — ind work. In this way, plat — orms are like classi — ied ads. It would be silly to assert that someone “works

1. About LendingClub, LENDINGCLUB, https://www.lendingclub.com/company/about-us (last visited Dec. 16, 2018).
2. Gig Economy, INVESTOPEDIA, https://www.investopedia.com/terms/g/gig-economy.asp (last updated May 24, 2018).

3. Gigabyte, TECHTERMS, https://techterms.com/de

   inition/gigabyte (last updated Feb. 26, 2013).

4. About the Dog People, ROVER, https://www.rover.com/about-us/?re

   = — ooter (last visited Dec. 16, 2018).

5. How Uber Works, UBER, https://www.uber.com/ride/how-uber-works/ (last visited Dec. 16, 2018); Why Ly — t, LYFT, https://www.ly — t.com/driver/why-drive-with-ly — t (last visited Dec. 16, 2018).

6. Revolutionizing Everyday Work, TASKRABBIT, https://www.taskrabbit.com/about (last visited Dec. 16, 2018).

   Electronic copy available at: https://ssrn.com/abstract=3135936 20 DREXEL LAW REVIEW [Vol. 11:1

or” the Pittsburgh Post-Gazette because he — ound a roo — ing job by placing a classi — ied ad in that newspaper. However, gig economy plat — orms have extra — eatures that do make them appear more like employers. For example, i — you

ind a dog walker on Rover, you will also pay — or that service via Rover’s payment portal. And i — you have a complaint about the service rendered, you will use Rover’s dispute resolution process. Rover conducts background checks on potential dog walkers, and a dog walker who gets too many negative reviews will be removed — rom the plat — orm.87 This level o — control over the nature o — the work goes — ar beyond a mere classi — ied ad. But

lexible work on a gig economy plat — orm is also quite di —

erent

rom a nine-to- — ive job smelting ore in a steel — actory. Moreover, the work on these plat — orms is rapidly evolving, as the new economy continues to rapidly grow.88 Un — ortunately, these new plat — orms remain subject to old laws, which constrain their business models in unintended and undesirable ways.89

II. PROBLEMS APPLYING OLD LABOR LAW TO THE NEW ECONOMY Modern labor law mainly concerns employees, so the de — inition o — employee is — undamental to understanding modern labor law. Workers can be classi — ied as either employees or independent contractors; these are the only two options, and neither — its the gig economy model o — work. Some have argued that the use o — independent contractor labor is being abused by employers who want to avoid providing bene — its.90 Others point out that gig economy business models cannot — unction i — all laborers must receive all the protections

1. About the Dog People, supra note 84.
2. Adapting to the Gig Economy as Technology Continues to Evolve, MARKETPLACE PLATFORM (Mar. 14, 2018), https://www.marketplaceplat — orm.com/adapting-to-the-gig-economy-as- technology-continues-to-evolve/.
3. 29 U.S.C. §§ 151–169 (2018).

4. See generally SHERROD BROWN, WORKING TOO HARD FOR TOO LITTLE: A PLAN FOR RESTORING THE VALUE OF WORK IN AMERICA (2017), https://www.brown.senate.gov/download/working-too-hard- — or-too-little (arguing some employers pre — er classi — ication o — workers as independent contractors to reduce employment costs including payroll taxes, labor standards, and workers’ rights).

   Electronic copy available at: https://ssrn.com/abstract=3135936 2019] UNBUNDLING EMPLOYMENT 21

and bene

its that are needed by steel workers in one- — actory towns.91 As a result, some courts and legislators have attempted to eviscerate the utility o — independent contractors.92 Recent court decisions expanding the joint employer doctrine could render the independent contractor label meaningless.93 This Article argues that the tension is not resolvable in today’s economy because it emerges — rom outdated ideas about work and workers. Instead, a new and — lexible de — inition o — “gig worker” is necessary to resolve the tension. But — irst, this Article highlights contemporary problems courts have had in applying the distinction between employees and independent contractors to the new ways people work in the sharing economy. These cases reveal that the tensions between old law and new work cannot be resolved without creating a novel worker classi — ication.

             A. Employees and Independent Contractors   The  --- undamental tension between the labor laws and the sharing economy arises doctrinally in the legal distinction between an employee and an independent contractor. These two distinct categories o ---  workers emerge out o ---  necessity  --- rom the NLRA because many o ---  the protections a ---

orded by the NLRA only — low to employees and not to independent contractors. The NLRA de — ines the terms employment and employee extremely broadly.94 In — act, the legislative history indicates that

1. William M. Boal, Testing

   or Employer Monopsony in Turn-o — -the-Century Coal Mining, 26 RAND J. ECON. 519, 522 (1995).

2. See supra Section II.B.
3. See supra Section II.B.

4. 29 U.S.C. § 152(3) (“The term ‘employee’ shall include any employee, and shall not be limited to the employees o — a particular employer, unless this subchapter explicitly states otherwise, and shall include any individual whose work has ceased as a consequence o — , or in connection with, any current labor dispute or because o — any un — air labor practice, and who has not obtained any other regular and substantially equivalent employment, but shall not include any individual employed as an agricultural laborer, or in the domestic service o — any — amily or person at his home, or any individual employed by his parent or spouse, or any individual having the status o — an independent contractor, or any individual employed as a supervisor, or any individual employed by an employer subject to the Railway Labor Act . . ., as amended

   Electronic copy available at: https://ssrn.com/abstract=3135936 22 DREXEL LAW REVIEW [Vol. 11:1

the goal o

the NLRA was to protect anyone who might be an employee.95 The Internal Revenue Service (IRS) has adopted a narrower de — inition o — employee — or tax purposes, which is also quite relevant in legal analysis even outside the domain o — tax.96 This Article will brie — ly address IRS categorizations o —

workers—employees who should

ile a Form W-2 versus independent contractors who — ile a Form 1099—because these tax categories have become intertwined and con — lated with their respective labor law categories. However, this paper is designed to — undamentally address the labor law issues with regard to employment bene — its, so it does not address tax consequences. Un — ortunately, the IRS de — inition o — employee is more di —

icult to understand and apply than the NLRA de — inition, because the IRS calls — or a twenty- — actor test plus an analysis, which requires arbiters to consider all — acts and circumstances on a case-by-case basis.97 The Supreme Court has said that there cannot be one test to determine employment.98 The task o —

classi

ying workers as employees is thus quite di —

icult, because there have been no clear and consistent tests promulgated by legislatures, agencies, or courts. This task becomes even harder and more uncertain when we try to apply eighty-year-old case law, statutes, agency opinions, and legislative history to the new sharing economy.

rom time to time, or by any other person who is not an employer as herein de — ined.” (citation omitted)).

1. Introduction to the NLRB, supra note 45 (“Congress enacted the National Labor Relations Act . . . in 1935 to protect the rights o — employees and employers . . . .”).
2. Employee (Common-Law Employee), IRS, https://www.irs.gov/businesses/small- businesses-sel — -employed/employee-common-law-employee (last updated Apr. 23, 2018).
3. JOINT COMM. ON TAXATION, PRESENT LAW AND BACKGROUND RELATING TO WORKER CLASSIFICATION FOR FEDERAL TAX PURPOSES 3–5 (2007), https://www.irs.gov/pub/irs-utl/x-26- 07.pd — .

4. United States v. Silk, 331 U.S. 704, 713 (1947) (“The problem o

   di —

erentiating between employee and an independent contractor or between an agent and an independent contractor has given di —

iculty through the years be — ore social legislation multiplied its importance. When the matter arose in the administration o — the National Labor Relations Act . . . we pointed out that the legal standards to — ix responsibility — or acts o — servants, employees or agents had not been reduced to such certainty that it could be said there was ‘some simple, uni — orm and easily applicable test.’” (citation omitted)).

   Electronic copy available at: https://ssrn.com/abstract=3135936

2019] UNBUNDLING EMPLOYMENT 23

While the NLRA de

ines employee one way, the NLRB takes another position, the IRS o —

ers a third (indeed, the IRS has taken di —

erent and even contradictory positions), and appellate courts in di —

erent circuits o —

er a — ourth, — i — th, sixth, and seventh approach, while the Supreme Court has held only that there cannot be any one test.99 In other words, attempting to de — ine “employee” under American law in the gig economy is a virtually impossible task. Regardless, it is not enough to say the law is a mess. It is better to try and sort it out. The twenty- — actor IRS test is probably the best starting point — or doing so, as it contains most, i — not all, o — the — actors that courts and agencies might consider when determining whether a worker is an employee. Since this Article invokes the IRS test, it will also use tax language to explain the consequences resulting — rom its application. I — the IRS classi — ies a worker as an “employee,” that person receives an IRS Form W-2.100 On the other hand, i — a worker is an “independent contractor,” such person receives an IRS Form 1099.101 This is why people re — er to employee versus independent contractor analysis alternatively as the W-2/1099 discussion. To round out our tax analysis, a partner in a partnership or a member o — a — low-through LLC receives an IRS Schedule K-1 (a result o — the entity’s — iling o — the Form 1065).102 There — ore, W-2, 1099, and K-1 are the three main tax classi — ications — or workers. However, K-1 partners are generally not pertinent to the matter at hand o — categorizing workers in

1. Id.

2. About Form W-2, Wage and Tax Statement, IRS, https://www.irs.gov/

   orms-pubs/about-

orm-w2 (last updated Nov. 1, 2018) (“Every employer engaged in a trade or business who pays remuneration, including noncash payments o — $600 or more — or the year (all amounts i — any income, social security, or Medicare tax was withheld) — or services per — ormed by an employee must — ile a Form W-2 — or each employee . . . .”).

1. About Form 1099-MISC, Miscellaneous Income, IRS, https://www.irs.gov/

   orms- pubs/about- — orm-1099-misc-miscellaneous-income (last updated Nov. 1, 2018) (“File Form 1099-MISC — or each person to whom you have paid during the year . . . [ — or] services per — ormed by someone who is not your employee . . . .”).

2. About Schedule K-1 (Form 1065), Partner’s Share o

   Income, Deductions, Credits, Etc., IRS, https://www.irs.gov/ — orms-pubs/about-schedule-k1- — orm-1065 (last updated Nov. 5, 2018) (“The partnership — iles a copy o — this schedule with the IRS to report your share o — the partnership’s income, deductions, credits, etc.”).

   Electronic copy available at: https://ssrn.com/abstract=3135936 24 DREXEL LAW REVIEW [Vol. 11:1

the sharing economy, as sharing economy plat

orms generally do not attempt to characterize workers as partners, and the workers likewise have not sought to be recharacterized as partners. How a worker is classi — ied has implications on how the worker gets paid. This is discussed in detail below, but — or the present purpose o — motivating what may otherwise seem like a dry overview o — a twenty- — actor test, it bears mentioning that employees receive many bene — its, such as time-and-a-hal —

overtime, Family Medical Leave Act (FMLA) protections, and the right to unionize.103 But 1099 independent contractors do not receive these bene — its.104 The recent spate o — lawsuits — rom workers seeking reclassi — ication — rom 1099 independent contractors to W-2 employees comes — rom their desire to get these bene — its.105 The IRS’s twenty- — actor test — or de — ining “employee” is best understood as a signpost — or what all the — acts and circumstances might be when a court or agency evaluates the status o — a worker.106 It is not a strict test per se. The twenty

actors involve: (1) instructions, (2) training, (3) integration, (4) personal services, (5) assistants, (6) continuing relationship, (7) set hours o — work, (8) — ull time required, (9) employer’s premises, (10) order or sequence test, (11) oral or written reports, (12) payment terms, (13) payment o — expenses, (14) tools and materials, (15) signi — icant investment, (16) pro — it or loss, (17) working — or multiple — irms, (18) services available to the public, (19) right to discharge, and (20) right to terminate.107 Addressing each o — these — actors in turn:

1. See Employee Rights, NAT’L LAB. REL. BOARD, https://www.nlrb.gov/rights-we- protect/employee-rights (last visited Dec. 16, 2018).
2. See Jessica Lee, Unionize Uber? Legal Fight Over Seattle Drivers Draws National Attention, SEATTLE TIMES (Mar. 29, 2017, 6:30 AM), https://www.seattletimes.com/seattle- news/transportation/unionize-uber-legal- — ight-over-seattle-drivers-draws-national-attention/.
3. Id.
4. JOINT COMM. ON TAXATION, supra note 97, at 1.

5. Id..

   Electronic copy available at: https://ssrn.com/abstract=3135936 2019] UNBUNDLING EMPLOYMENT 25

(1) The “instructions” element asks the question: does the person who is working receive instructions directly — rom the purported employer? Or do they have some — lexibility about what they do on a day to day basis? (2) “Training” asks whether the purported employer provides training sessions. (3) “Integration” asks how integral is the work? Is it a plug and play operation? Could you have one person doing the carpentry today and a di —

erent person doing the carpentry tomorrow with a similar result? Or is this something more — undamental? You cannot just swap out the CEO o — Coca-Cola and expect the company to run well. That, there — ore, has to do with the integration — actor. (4) “Personal service” means the workers are required to per — orm the work personally and cannot substitute someone else to do the work, which is a hallmark o —

  employment.   (5) “Assistants” means the worker can hire, supervise, and
  pay other assistants to the employer, which is evidence o ---

  employment.   (6) A “continuing relationship” means that someone who
  shows up at work every day  --- or an extended period is
  probably an employee under that  --- actor. By contrast,
  someone who calls in each day and says, “Are you looking

or me to come in tomorrow?” is more likely to be an independent contractor. (7) “Set hours o — work” means an employer controls when work is per — ormed. An independent contractor has more

reedom as to when the work is completed. (8) “Full time required” is an important — actor. Someone who is working — orty hours a week, nine to — ive, and — or a single

irm is probably an employee. (9) “Employer’s premises” is another important — actor. However, as employers and the — ederal government promote hoteling and the ability to work remotely, this

actor seems less relevant.

    Electronic copy available at: https://ssrn.com/abstract=3135936

26 DREXEL LAW REVIEW [Vol. 11:1

(10) “Order or sequence” relates to how much control the putative employer has over the work. An independent contractor generally has more control about deciding the process o — work than an employee does. (11) The “oral or written reports” — actor asks whether an employer or purported employer will ask — or regular status report calls in the o —

ice: “What’s going on this week?” Someone who is going to be called into the o —

ice weekly and talked to is more likely an employee. (12) “Payment terms” looks to how o — ten the worker is paid: i — the worker is being paid on an invoice basis — or — ourteen days, she is more likely an independent contractor. I — the worker receives a paycheck every second and — ourth Friday o — the month, she is more likely an employee. (13) For “payment o — expenses,” generally, employees get reimbursed — or business and travel expenses, although that practice has been diminishing in certain areas. Still, independent contractors are generally not reimbursed — or their business and traveling expenses. (14) For “tools and materials,” independent contractors generally bring their own tools and materials to the job. A worker who has a laptop provided by a putative employer is probably an employee. Likewise, a construction worker using a putative employer’s reciprocating saw, as opposed to the one she or he brought to work that day, evidences employment. Wearing a hard hat or a t-shirt that says the name o — the construction company or sta —

ing company also evidences employment. (15) Making a “signi — icant investment” to per — orm the work makes a worker seem more like an independent contractor. An employee is not expected to bear the risk o —

 loss  --- or an operation.  (16) The “pro --- it or loss”  --- actor is similar to the risk o ---  loss
 concept. A worker who realizes she or he is able to
 potentially lose money is more likely to be an independent
 contractor.




 Electronic copy available at: https://ssrn.com/abstract=3135936

2019] UNBUNDLING EMPLOYMENT 27

(17) Working

or just one company, especially one that requires exclusive work — or that company, evidences employment. On the other hand, a worker who works — or three di —

erent people will have a hard time saying they are all employers, although the joint employment doctrine is discussed below. (18) Providing services to the general public, such as a lawyer who “hangs a shingle,” where anyone who comes to that o —

ice can get legal advice, as opposed to solely being an adviser to a certain corporation, looks more like an independent contractor, whereas a person who works — or just one company is more likely an employee o — that one company. (19) I — the putative employer has the right to — ire a worker, or seems to have the right to terminate, that is another — actor generally evidencing employment. (20) Similarly, a worker who has the right to terminate the relationship at any time, without incurring any liability, is more likely an employee because we know employees cannot be — orced to work, whereas an independent contractor would breach a contract i — she or he — ailed to complete a project.108

In addition to these twenty

actors, the IRS will look generally at behavioral control, — inancial control, and the overall relationship between the parties.109 Some o — the twenty — actors are no longer as relevant as they once were in our economy. Other — actors are more relevant than they ever have been, and they are all based on individual circumstances. Litigation that relates to these — actors is generally initiated by an independent contractor who wants to be reclassi — ied as an employee. Such a worker wants to be classi — ied as an employee under this test to receive certain bene — its. It is better — or most

1. Id.

2. Independent Contractor (Sel

   -Employed) or Employee?, IRS, https://www.irs.gov/businesses/small-businesses-sel — -employed/independent-contractor-sel — - employed-or-employee (last updated Nov. 29, 2018).

   Electronic copy available at: https://ssrn.com/abstract=3135936 28 DREXEL LAW REVIEW [Vol. 11:1

people to be classi

ied as an employee, rather than an independent contractor. While independent contractors gain

lexibility, an employee is protected by a — orty-hour, — ive-day workweek with time-and-a-hal — pay — or overtime (unless exempt). Other bene — its required by law include workers’ compensation, part-time disability, and the FMLA.110 Employees usually get health, dental, and vision insurance

rom employers, whereas independent contractors have to pay

or that on their own. Employees are usually included in a retirement plan, o — ten with an employer’s matching contribution, such as a 401(k) or a 403(b). Employee pensions are now less common, but at one time they were quite common. Li — e insurance is not required by law, but employees o — ten receive it as part o — their package, plus paid vacation time. Those employee rights and bene — its are reasons why a person might be happier to be classi — ied as an employee as opposed to an independent contractor. On the other hand, the classi — ication o — “employee” comes with some restrictions. One o — those restrictions is that the employer can demand exclusive work. An employee cannot simultaneously work — or a competitor and could be restricted

rom doing any other work. The employer can demand an employee show up — or work at nine in the morning and stay until — ive in the evening. The employer will almost always retain all the intellectual property that the employee generates. This way o — working does not accommodate an increasing number o — workers in the modern economy. Those strings also pull in the other direction regarding tort liability. The doctrine o — respondeat superior (literally, “let the boss answer”) holds a principal vicariously liable — or torts committed by its agents.111 Courts generally apply either the bene — its test or the characteristics test to determine when an

1. Hire and Manage Employees, U.S. SMALL BUS. ADMIN., https://www.sba.gov/business- guide/manage-your-business/hire-manage-employees#section-header-6 (last visited Dec. 16, 2018).

2. RESTATEMENT (THIRD) OF AGENCY § 2.04 (AM. LAW INST. 2006) (“An employer is subject to liability — or torts committed by employees while acting within the scope o — their employment.”).

   Electronic copy available at: https://ssrn.com/abstract=3135936 2019] UNBUNDLING EMPLOYMENT 29

employer is vicariously liable

or the torts o — an employee.112 Respondeat superior, however, does not apply to independent contractors.113 O — course, under common law, there is yet another test — or determining whether a putative employee is an independent contractor — or liability purposes.114 The Third Restatement o — Torts pro —

ers an eleven-part test that is similar to, but not exactly the same as, the IRS test described above.115 This additional test — urther illustrates the rampant con — usion about the distinction between employee and independent contractor. The schisms in law around this distinction make it extremely di —

icult to avoid litigation and liability — or misclassi — ication. These potential issues present a virtually insurmountable challenge to sharing economy companies, as there is no precedent — or the way they hire and work. Uber will be exposed to less liability i — its drivers are considered independent contractors rather than employees. Although the laws o — agency and the laws o — tort are not exactly aligned with the laws o — employment, similar — actors are generally applied to determine whether tort liability will impute to a principal. Companies that hire independent contractors have — ewer responsibilities because generally: (1) they do not have to withhold independent contractors’ taxes, (2) they are not responsible — or their torts and contracts, (3) they are not liable — or illegal conduct like making kickbacks or bribes, and (4) they have no obligation to pay — or unemployment insurance and other worker bene — its.116 As we saw with the Uber — iasco in 2017, however, there are some reputational liabilities that cannot be avoided, and there are many reasons why Uber and similar plat — orms would pre — er that their workers be classi — ied as independent contractors. But the law cannot give sharing economy plat — orms like Uber

1. Respondeat Superior, CORNELL LEGAL INFO. INST., https://www.law.cornell.edu/wex/respondeat_superior (last visited Dec. 17, 2018).
2. Id.
3. Id.
4. Id.

5. Id.; Independent Contractors, WORKPLACE FAIRNESS, https://www.workplace — airness.org/independent-contractors#2 (last visited Dec. 16, 2018).

   Electronic copy available at: https://ssrn.com/abstract=3135936 30 DREXEL LAW REVIEW [Vol. 11:1

certainty about when a worker will be classi

ied as an independent contractor. Employers who wrongly classi — y employees as independent contractors are subject to a number o — penalties.117 In 2011, the DOL and the IRS signed a Memorandum o — Understanding to work together to try to scourge out misclassi — ication o —

employees.118 Many states also

erret out misclassi — ications.119 The penalties attached depend on the egregiousness o — the misclassi — ication.120 There are three categories o — worker misclassi — ication: unintentional misclassi — ication, intentional misclassi — ication, and — raudulent misclassi — ication.121 Fraudulent misclassi — ication might subject an employer to criminal penalties up to $1000 per misclassi — ied worker and a $50 — ine per year.122 The amount o —

tax that should have been withheld is owed with penalties, in the — orm o —

ines and interest.123 The DOL launched its misclassi — ication initiative in 2010.124 In 2014, the DOL awarded $10.2 million to nineteen state attorney general o —

ices to assist in this initiative.125 These states created an interagency task — orce, and there are now thirty-seven states

1. Christopher J. Cox et al., Weil Discusses Risks o

   Classi — ying Employees as Independent Contractors, CLS BLUE SKY BLOG (Oct. 19, 2017), http://clsbluesky.law.columbia.edu/2017/10/19/weil-discusses-risks-o — -classi — ying-employees- as-independent-contractors/.

2. Press Release, U.S. Dep’t o

   Labor, Labor Secretary, IRS Commissioner Sign Memorandum o — Understanding to Improve Agencies’ Coordination on Employee Misclassi — ication Compliance and Education (Sept. 19, 2011), https://www.dol.gov/newsroom/releases/whd/whd20110919.

3. Id.

4. Improperly Classi

   ying Employees as Independent Contractors: What Are the Penalties?, JUSTWORKS: JUSTBLOG (Dec. 28, 2017), https://justworks.com/blog/consequences-misclassi — ying- workers-independent-contractors.

5. Id.
6. Id.
7. Id.
8. See DAVID WEIL, U.S. DEP’T LABOR, IMPROVING WORKPLACE CONDITIONS THROUGH STRATEGIC ENFORCEMENT: A REPORT TO THE WAGE AND HOUR DIVISION 1–4 (2010), https://www.dol.gov/whd/resources/strategicEn — orcement.pd — .

9. Press Release, U.S. Dep’t o

   Labor, $10.2M Awarded to Fund Worker Misclassi — ication Detection, En — orcement Activities in 19 State Unemployment Insurance Programs (Aug. 17, 2014), https://www.dol.gov/newsroom/releases/eta/eta20141708.

   Electronic copy available at: https://ssrn.com/abstract=3135936

2019] UNBUNDLING EMPLOYMENT 31

that have state laws against worker misclassi

ication.126 Misclassi — ication can — urther result in owing money to the

ederal and state governments in addition to — ines. There are also numerous agencies that have authority to come a — ter employers. But while much rides on being able to distinguish between employees and independent contractors, it is not that easy to apply the IRS twenty- — actor test conclusively to gig economy jobs. Consider the gig economy plat — orm TaskRabbit. TaskRabbit advertises handymen — or small tasks.127 For example, someone might like to shop at IKEA, but hate putting the — urniture together. Another person might enjoy putting together IKEA — urniture. One person, a “tasker,” goes on TaskRabbit and o —

ers her — ree time to put together IKEA

urniture — or $28 an hour. Another person goes on TaskRabbit and decides that paying $28 an hour to avoid putting together IKEA — urniture is a good deal. TaskRabbit is an intermediary plat — orm allowing these people to connect. One tasker named John has a 98% approval rating. He charges $28 per hour. He has a — ive-out-o — - — ive star rating — rom TaskRabbit. Is he an employee o — TaskRabbit or is he an independent contractor to TaskRabbit? Generally, John looks like a contractor. He can re — use the job: he does not have to do that work. TaskRabbit is not going to tell him how to put that IKEA — urniture together. But John receives instructions — rom TaskRabbit on where and when to do the job. It is hard to say whether he is integrated with TaskRabbit services: he has earned “elite” status on that plat — orm, and he can charge more per hour because he is elite. Other — actors are even more di —

icult to apply. Is John o —

ering personal services to the public when he only o —

ers IKEA building services

1. LISA A. TAVARES ET AL., VENABLE LLP, FOCUS ON MISCLASSIFICATION—ARE YOUR WORKERS “EMPLOYEES” OR “INDEPENDENT CONTRACTORS?” 1 (2011), https://www.venable.com/ — iles/Publication/435 — 765d-a52e-4786-b22e-

45225aa10d0/Presentation/PublicationAttachment/5 — 52 — 7d2-b875-4871-9e96-

b0 — 101b3a52/EBEC_Alert_3-11.pd — .

1. How It Works, TASKRABBIT, https://www.taskrabbit.com/how-it-works (last visited Dec. 16, 2018).

   Electronic copy available at: https://ssrn.com/abstract=3135936 32 DREXEL LAW REVIEW [Vol. 11:1

through the TaskRabbit website? Is this a continuing relationship where John has — orty-nine reviews that enable him to charge more than an average tasker? Does he incur a risk o —

loss where he has to bring his own tools and drive his own car to the job site? The analysis o — whether drivers — or Uber, the ride-sharing service, are employees or independent contractors is even more complicated than TaskRabbit. Uber does not tell you to work eight hours per day or at any particular time, but Uber o —

ers a lot o — incentives to get drivers on the road. The company o —

ers certain bonuses a — ter — our hours and extra pay — or working during certain time periods.128 A driver can re — use to make any pick up, but that will lower the driver’s score, which makes it harder to get additional rides. I — the score goes low enough, that driver can be automatically removed — rom the plat — orm entirely. Does that constitute the ability o — Uber to constructively discharge drivers? Drivers generally bring their own cars, but Uber now o —

ers a driver car leasing program. Uber can deactivate drivers, and Uber must activate drivers to participate in the — irst place.129 Uber leasing is particularly problematic. A driver who is driving an Uber car, and not her own personal vehicle, starts to look a little di —

erent than a typical sharing-economy participant. The sharing economy reallocates underutilized assets, whereas the traditional economy requires obtaining new assets to o —

er a service. Renting a car that is sitting unused in someone’s driveway getting rusty is di —

erent than deciding to lease or purchase a new vehicle. Further, litigation on W-2/1099 issues is heating up. One attorney in particular is leading the charge against what she pejoratively calls the “1099 economy.” Shannon Liss-Riordan recently negotiated a $100 million settlement — rom her employee characterization suit against Uber, although a — ederal

1. How Surge Pricing Works, UBER, https://www.uber.com/drive/partner-app/how-surge- works/ (last visited Dec. 16, 2018).

2. Uber Driver Requirements: Do You Quali

   y to Drive?, RIDESHARING DRIVER (May 14, 2018), https://www.ridesharingdriver.com/uber-driver-requirements-quali — y/.

   Electronic copy available at: https://ssrn.com/abstract=3135936

2019] UNBUNDLING EMPLOYMENT 33

judge later ruled that deal to be un

air.130 Her — irm sued GrubHub.131 GrubHub is a — ood delivery service that matches up three parts o — a network: hungry people who want to eat, restaurants who want to sell — ood, and people who want to drive that — ood to hungry people.132 Her — irm sued Amazon, representing its “ — lex” delivery drivers who want to be reclassi — ied as employees.133 In all these cases, the analysis under traditional tests is complicated and inconclusive. Amazon — lex drivers do not work speci — ied hours, but they do have to meet many standards imposed by Amazon. A — lex driver must wear a sa — ety vest that says Amazon, drive a white van provided by Amazon with a sticker that says Amazon on it, and deliver Amazon packages to speci — ied locations. Those are — actors that counsel toward — lex drivers having employee status. On the other hand, — lex drivers do not expect a continuing relationship with Amazon, they can work — or multiple people at once, and they can select whether to work on any given day. These — actors make the classi — ication much less clear. The NLRB has also been involved in these lawsuits. Notably, the NLRB sued Handy Technologies — or its worker classi — ications.134 Handy provides a plat — orm — or home cleaning

1. See Michael Liedtke, Judge Rejects $100 Settlement in Key Case with Uber Drivers, GLOBE & MAIL (Aug. 19, 2016), https://www.theglobeandmail.com/report-on-business/international- business/us-business/judge-rejects-100-million-settlement-in-key-case-with-uber- drivers/article31462763/.
2. See Jon Steingart, GrubHub Wage Case May See New Delivery: Worker Status Test Redo, BLOOMBERG BNA (Jan. 25, 2018), https://www.bna.com/grubhub-wage-case-n73014474636/ (“Liss-Riordan is the lawyer — or — ormer GrubHub — ood delivery driver Rae — Lawson, who says the online — ood ordering company incorrectly classi — ied him as an independent contractor and that it owes him overtime and reimbursement — or business expenses he’d be entitled to under state law i — he had been classi — ied correctly as an employee.”).

3. See About Us, GRUBHUB, https://about.grubhub.com/about-us/overview/de

   ault.aspx (last visited Dec. 16, 2018).

4. See Angel Gonzalez, Amazon Delivery Drivers Sue Company over Job Status, SEATTLE TIMES (Oct. 5, 2016, 4:58 PM), https://www.seattletimes.com/business/amazon/amazon-delivery- drivers-sue-company-over-job-status/ (“The complaint, — iled late Tuesday in U.S. District Court in Seattle, was brought — orth by Shannon Liss-Riordan, the attorney who led two class-action lawsuits by discontented drivers against Uber.”).

5. See Josh Eidelson, U.S. Labor Board Complaint Says On-Demand Cleaners Are Employees, BLOOMBERG (Aug. 31, 2017, 12:07 PM), https://www.bloomberg.com/news/articles/2017-08- 31/u-s-labor-board-complaint-says-on-demand-cleaners-are-employees.

   Electronic copy available at: https://ssrn.com/abstract=3135936 34 DREXEL LAW REVIEW [Vol. 11:1

services.135 While the market dictates the upper limit o

prices

or services on Handy, Handy never pays less than $18 an hour; they set a minimum rate and then let the market adjust it — rom there. Handy generally engages with low-skilled, low-income, low-education workers, and part o — its corporate mission is to help these people climb the economic ladder. To do this, Handy encourages its workers to get bank accounts.136 The NLRB claimed this was providing training, which aided its case — or worker misclassi — ication against Handy.137 Most scholars probably recognize that having a bank account is better than putting the money under your mattress: workers can be paid more quickly, it helps people budget and understand the assets that they have, it builds credit, it is sa — er, etc. However, encouraging workers to do this, plus o —

ering them job training, language skills, and other opportunities are — actors that led the NLRB to reclassi — y Handy workers as employees.138 Handy would reasonably respond by o —

ering — ewer services and less training — or workers to avoid this risk o — reclassi — ication. The IRS, DOL, NLRB, state attorneys general, and plainti —

s’ attorneys are also — ighting — or reclassi — ication o — workers in the sharing economy. As we see with Handy, many o — these — irms would rather help their employees lessen that risk o —

reclassi

ication and all that comes with it. Gig economy — irms argue that their business model does not work i — these workers are employees and not independent contractors.139 And in today’s economy there are many workers who depend on these — irms. There are workers in the sharing economy who drive — or Uber and Ly — t, per — orm tasks — or TaskRabbit, and clean — or Handy. Such people cobbled together a — ull-time employment li — estyle — or themselves out o — working through three or — our o — these gig-based companies. What will

1. About Us, HANDY, https://www.handy.com/about (last visited Dec. 15, 2018).

2. NYU School o

   Law, Tech Entrepreneurs and the Regulatory State: Keynote, YOUTUBE (Apr. 5, 2016), https://www.youtube.com/watch?v=Q — 33 — bnUOQg& — eature=youtu.be.

3. Id.
4. Id.
5. Id.

   Electronic copy available at: https://ssrn.com/abstract=3135936

2019] UNBUNDLING EMPLOYMENT 35

happen to them i

there is reclassi — ication? Will they be able to maintain the pro — its they earn and enjoy? On the other hand, what about people that are driving — or Uber sixty hours a week, are totally dependent on that — irm — or sustenance, and might not have recourse i — they are deactivated? The problem with using the NLRA to protect sharing economy workers is that the sharing economy does not look like the traditional economy that existed when these laws were created. In the traditional economy, people were mining coal out o — the ground, using that coal to produce energy, using that energy to produce glass, and selling that glass.140 The traditional value chain moves — rom le — t to right.141 In the sharing economy there is a value chain that moves in both directions toward a plat — orm in the middle, — orming a triangle.142 Work, employment, productivity, and value creation are

undamentally di —

erent now.

                    B. The Joint Employment Doctrine    The joint employment doctrine, which was developed in the 1930s to prevent employers  --- rom circumventing the NLRA,143 is likewise being stretched and distorted to cover the innovative ways people work today. Analysis under this doctrine is like the analysis  --- or classi --- ying a worker as an employee or an independent contractor: a worker must  --- irst and  --- oremost be an employee in order  --- or joint employment to attach, so it is susceptible to all the problems with applying the worker classi --- ication that were described in the prior section.144 Additionally, the traditional analysis o ---  joint employment is challenged by the change in how people work. Instead o ---

having jobs, many people do jobs, and this makes joint employment analysis even harder.

1. See supra Part II.
2. Id.
3. Id.

4. See generally 29 C.F.R. § 791.2 (2018) (discussing the reasoning

   or introducing the joint employment doctrine under the Fair Labor Standards Act).

5. See supra Section III.A.

   Electronic copy available at: https://ssrn.com/abstract=3135936 36 DREXEL LAW REVIEW [Vol. 11:1

The joint employment doctrine is designed to prevent an employer — rom chiseling an employee out o — overtime by

orming two — irms and having the employee work — or each: work thirty-nine hours — or one, thirty-nine hours — or the other.145 I — two nominally separate but actually similar or related

irms employ a worker in this way, the putative joint employer may be liable — or thirty-eight hours o — overtime (time-and-a- hal — ) pay and other — ull-time employment bene — its.146 The term “joint employment” — irst appeared in the July 1939 DOL Interpretive Bulletin Number 13, which makes it clear that the DOL’s policy rationale was to prevent an end-run around labor laws that protect — ull-time employees, as evidenced by the 1930s parlance o — “wage chiseling.”147 There are two distinct types o — joint employment that have emerged — rom the common law: horizontal joint employment and vertical joint employment. Horizontal joint employment is older and simpler. Imagine that Capital Co. owns 51% shares in two di —

erent hotels, Hotel A and Hotel B. Hotel A hires Larry Labor to work thirty- — ive hours a week — or $10 per hour, and Hotel B also hires Larry to work thirty- — ive hours a week — or $10 per hour. Larry earns $700 per week and does not receive — ull- time employee bene — its like health insurance. Larry sues Hotel A, Hotel B, and Capital Co. — or $225 per week in unpaid overtime plus the value o — employee bene — its. Does Larry prevail?

1. See generally 29 C.F.R. § 791.2 (explaining how joint employment is dealt with

   or both associated and disassociated employers).

2. Id.
3. Marc Linder, The Small-Business Exemption Under the Fair Labor Standards Act: The “Original” Accumulation o — Capital and the Inversion o — Industrial Policy, 6 IOWA L. REV. 403, 408 (1998). The Department o — Labor originally — ormulated its position with regard to joint employment in its Interpretive Bulletin No. 13, which dealt with the “Determination o — Hours

or Which Employees Are Entitled to Compensation.” Under the rubric “Employees Having More than One Job,” the Wage and Hour Administrator illustrated the diametrically opposite outcomes that joint employment triggers under the minimum wage and overtime provisions o —

FLSA: whereas each joint employer can take credit

or the wages paid by the other toward the minimum wage, each joint employer is made liable — or overtime where the aggregation o — hours worked — or each exceeds the statutory threshold. U.S. Dep’t o — Labor, Wage & Hour Div., Interpretive Bulletin No. 13 (May 3, 1939) at 16–17.

   Electronic copy available at: https://ssrn.com/abstract=3135936

2019] UNBUNDLING EMPLOYMENT 37

According to the DOL Wage and Hour Division, Larry will prevail on his claim o — horizontal joint employment i — he can show the hotels are “su —

iciently related to or associated with each other.”148 Courts will look at all the — acts and circumstances, including but not limited to: who owns the putative joint employers, whether the employers have overlapping directors or managers, whether the employers share control over operations, whether one employer supervises the work o — the other, whether the employers share authority over the employee, whether the employers treat employees as a pool o — workers available to both o — them, whether the employers share clients or customers, and whether there are any agreements between the employers.149 Applying this to Larry’s hypothetical, i — Hotel B asks Hotel A to send over Larry speci — ically when Hotel B is understa —

ed, or i — Hotel B commonly asks Hotel A to provide workers in general, that is evidence that the hotels are joint employers. The doctrine o — vertical joint employment developed more recently and has a di —

erent — ocus. Not only is vertical joint employment di —

icult to analyze — actually, but the law is in a con — using state o —

lux. In January 2016, the DOL issued Administrator’s Interpretation (AI) 2016-01.150 This non-binding statement included the — irst instance where the DOL took an administrative position that distinguished between “horizontal” joint employment and “vertical” joint employment.151 Previously, this distinction was made only by certain courts. Scholars saw this as a shi — t in the DOL’s — ocus toward prosecuting vertical joint employers more vigorously. AI 2016-01 also set — orth the DOL’s “economic realities” test — or vertical joint employment, which is described below.152 But, on

1. Fact Sheet #35: Joint Employment and Independent Contractors Under the Migrant and Seasonal Agricultural Worker Protection, U.S. DEP’T LABOR, https://www.dol.gov/whd/regs/compliance/whd — s35.htm (last updated July 2008).
2. See id.
3. Administrator Interpretations Letter—Fair Labor Standards Act, U.S. DEP’T LABOR, https://www.dol.gov/WHD/opinion/adminIntrprtnFLSA.htm# — oot (last visited Sept. 5, 2018).
4. See id.

5. See id.

   Electronic copy available at: https://ssrn.com/abstract=3135936 38 DREXEL LAW REVIEW [Vol. 11:1

June 7, 2017, the DOL issued a three-sentence press release withdrawing AI 2016-01.153 Scholars saw this move as a shi — t in the DOL’s approach back to a more traditional theory o —

employment relationships.154 Obviously, it is hard to build a solid — oundation on shi — ting sand. This con — usion is disruptive

or employers and makes li — e di —

icult — or the lawyers who advise them. Moreover, courts can and do consider all the — acts and circumstances in evaluating vertical joint employment claims. However, there is some DOL guidance as to what economic reality — actors to consider: (1) whether the putative employer directs, controls, or supervises the work, (2) whether the putative employer has the power to hire or — ire the employee or to change rate or method o — pay, (3) permanence or length o —

the relationship between the putative employer and the employee, (4) whether the employee is per — orming low-skill (easily replaceable) services or per — orming tasks that require substantial training and integration, (5) whether the work is per — ormed on the putative employer’s premises, and (6) whether the putative employer generally per — orms — unctions that would ordinarily be per — ormed by employees.155 These

actors evidence that the employee is “economically dependent” on the putative employer. Imagine next that Louise Labor is employed by Smart Sta —

ing Services (SSS). SSS could send Louise to any job site, but in reality, SSS has sent Louise to Data Entry Inc.’s (DEI) o —

ices every working day — or — ive years, where DEI tells her what data to enter and how to use its systems. DEI pays SSS — or Louise’s services, and SSS pays Louise’s salary, — or nine months. Then one day, SSS closes down suddenly, without paying Louise.

1. Press Release, U.S. Dep’t o

   Labor, U.S. Secretary o — Labor Withdraws Joint Employment, Independent Contractor In — ormal Guidance (June 7, 2017), https://www.dol.gov/newsroom/releases/opa/opa20170607.

2. Michael J. Lotito & Ilyse Schuman, DOL Withdraws Joint Employer and Independent Contractor Guidance, LITTLER (June 7, 2017), https://www.littler.com/publication- press/publication/dol-withdraws-joint-employer-and-independent-contractor-guidance.
3. See 29 C.F.R. 500.20(h)(4)(i)–(vi) (2018).

   Electronic copy available at: https://ssrn.com/abstract=3135936

2019] UNBUNDLING EMPLOYMENT 39

Can Louise sue DEI

or back wages because DEI is really her employer? Applying the DOL — actors to Louise, it may seem obvious that DEI is her vertical joint employer: she works on DEI premises, per — orms the tasks it gives her (which are similar to the work that DEI employees do), and has done so every day — or nine months. However, this is also the norm in sta —

ing agency relationships. To call Louise an employee o — DEI would mean also reclassi — ying millions o — people who are similarly engaged in work. The test traditionally applied by most courts is — rom Bonnette v. Cali — ornia Health & Wel — are Agency.156 The Bonnette — our-part economic reality test is “whether the alleged employer (1) had the power to hire and — ire the employees, (2) supervised and controlled employee work schedules or conditions o —

employment, (3) determined the rate and method o

payment, and (4) maintained employment records.”157 But on January 25, 2017, the Fourth Circuit declined to apply the Bonnette test, and instead created a new test that dramatically increases the liability — or putative vertical employers.158 Some scholars say this new test renders the independent contractor concept meaningless.159 In Salinas v. Commercial Interiors, Inc., the Fourth Circuit considered six

actors to determine “whether two or more persons or entities ‘are not completely disassociated’ with respect to the worker.”160 The Salinas test thereby makes vertical joint employment even easier to — ind than horizontal joint employment. Horizontal joint employment exists where the employee has two employment relationships with two or more employers, and where the employee can prove the employers are su —

iciently associated or related such that courts impute

1. See 704 F.2d 1465, 1470 (9th Cir. 1983) (describing the

   our — actors to consider when determining i — workers are “employers” o — chore workers).

2. Id.
3. Salinas v. Commercial Interiors, Inc., 848 F.3d 125, 129 (4th Cir. 2017).
4. See, e.g., N. Peter Lareau, 10 LABOR & EMPLOYMENT LAW § 261.06, LexisNexis (database updated 2018).

5. 848 F.3d at 141.

   Electronic copy available at: https://ssrn.com/abstract=3135936 40 DREXEL LAW REVIEW [Vol. 11:1

ownership o

both to one. Vertical joint employment under Salinas seemingly shi — ts the burden o — persuasion to employers who will have to prove they are completely dissociated.161 In many traditional cases where vertical joint employment was not

ound, complete dissociation would have been very hard to prove.162 On January 8, 2018, the Supreme Court denied certiorari and re — used to review Salinas.163 With a di —

erent test in almost all o — the — ederal circuits, the state o — the vertical employer doctrine is thus very much in — lux in the courts today. Legislators are also showing interest in strengthening the vertical joint employer doctrine. Senator Sherrod Brown o —

Ohio introduced the Fair Playing Field Act in 2015 to “crack down on employers who misclassi — y workers and cheat them out o — earned bene — its.”164 Brown’s 2017 plan — or restoring the value o — work in America, titled “Working Too Hard — or Too Little,” would make it harder — or employers to use independent contractors, as “[w]orkers are marginalized when they are hired as temps through a sta —

ing agency instead o — as direct employees.”165 Clearly, there is a lot o — tension and con — usion about employment today. While independent contractor relationships are becoming more prevalent, they are also becoming more contentious. The sharing economy has changed employment relationships, and people’s expectations about work seem to be changing as well. Battle lines are being drawn as workers want to be classi — ied as employees, and — irms, especially on the sharing economy plat — orm, want to obtain labor through independent contractors. Many scholars have suggested how the de — inition o — each, or the test — or both, could or should be changed. This Article makes a di —

erent argument: neither employee nor independent contractor is the proper label

1. See id. at 133–35.
2. Id. at 137.
3. See Vin Gurrieri, High Court Won’t Review 4th Circ. Joint Employer Test, LAW360 (Jan. 8, 2018, 11:27 AM), https://www.law360.com/articles/999254/high-court-won-t-review-4th-circ- joint-employer-test.

4. Fair Playing Field Act o

   2015, S. 2252, 114th Cong. (2015).

5. BROWN, supra note 90, at 16.

   Electronic copy available at: https://ssrn.com/abstract=3135936

2019] UNBUNDLING EMPLOYMENT 41

or the way people work in the sharing economy. We need new de — initions o — work and worker that — it this new economy.

                   C. The Real Problem o ---  Monopsony   The distorted and contorted legal doctrines necessitated by the NLRA and FLSA suggest that there is a more  --- undamental problem with labor law. This next section argues that the NLRA and FLSA  --- undamentally misunderstand a core problem in labor markets—monopsony—and thereby also  --- undamentally

ail to solve labor problems. Thanks to modern economic thinking, we now understand that labor markets can be more or less — air depending on competitive pressures. Indeed, we can use econometrics to describe the competitiveness—or, its antithesis, monopsony—o — labor markets. However, in 1938, the FLSA’s supporters did not appreciate this critical nuance, and instead passed a bill that envisions a uni — orm standard in all industries and across all regions.166 This is a crucial oversimpli — ication error that has led to many harms in labor markets. This is the regulatory equivalent o — using a sledgehammer to crack a nut, which causes many problems beyond those which it was intended to solve. This section explains the problem o — monopsony and brie — ly describes how other areas o — law (notably, antitrust) have taken a more nuanced approach to this concern. Law has many ways to remedy market — ailures, but when these remedies are applied to healthy markets, maladies o — ten result. Consider the human body as an analogy. I — a person has cancer, unpleasant chemotherapy drugs can cure that person. Indeed, medical philosophers have argued that using experimental drugs in cancer therapy is a moral imperative.167

1. John S. Forsythe, Legislative History o

   the Fair Labor Standards Act, 6 LAW & CONTEMP. PROBS. 464, 466 (1939). Forsythe writes “that [Roosevelt] did not envisage a uni — orm raising o —

standards in all industries and

or all regions.” Id. This also shows how radical the labor laws o — the 1930s were; not only did President Roosevelt go — urther than recommended by economist Keynes, but the FLSA itsel — went even — urther than President Roosevelt would have pushed it. See id.

1. Kenneth C. Micetich & David C. Thomasma, The Ethics o

   Patient Requests in Experimental Medicine, 34 CA: CANCER J. FOR CLINICIANS 118, 119 (“A physician is not justi — ied

   Electronic copy available at: https://ssrn.com/abstract=3135936 42 DREXEL LAW REVIEW [Vol. 11:1

Sometimes, moral philosophers even argue

or the use o —

experimental treatments on children, even when such treatments might produce serious side e —

ects, where there is hope that doing so will treat the child’s disease.168 Conversely, it is morally repugnant to thrust such unpleasant medicines upon a healthy body, which would become sick. Indeed,

oisting experimental drugs on a healthy person violates the

undamental axiom o — medicine: do no harm.169 The economy, like our bodies, exists in a state o — equilibrium. Opposing — orces being in check with each other produces a state o — economic health. When a — orce gets out o — control, intervention may be required to restore a state o — health. However, when an intervention is needlessly applied, it can have deleterious e —

ects. We have developed potent medicines

or intervening in economic maladies. We have also developed good assessments — or when an intervention is required. The problem with labor law, however, is that it does not consider economic determinants o — malaise when deciding whether to intervene in the labor market at large. NLRA and FLSA remedies apply regardless o — whether there is that malignant condition o — monopsony. Monopsony is the key economic concept — or understanding how di —

erent labor markets require di —

erent regulations. In short, a labor monopsony is a speci — ic circumstance where there is only one — irm employing a certain type o — labor.170 The classic example is a mining town, where the company who owns the

in withholding experimental therapy, even that which he

eels would not help in any measurable way or be any better — or the patient than ‘standard’ therapy, provided that the patient will not su —

er signi — icant harm.”).

1. Timothy F. Murphy, The Ethics o

   Research with Children, 5 VIRTUAL MENTOR 253, 253– 54 (2003); see also Paul Ramsey, The En — orcement o — Morals: Nontherapeutic Research on Children, 6 HASTINGS CTR. REP. 21, 21–22 (1976).

2. EDMUND D. PELLEGRINO & DAVID C. THOMASMA, A PHILOSOPHICAL BASIS OF MEDICAL PRACTICE: TOWARD A PHILOSOPHY AND ETHIC OF THE HEALING PROFESSIONS 170–91 (1981).

3. This concept comes

   rom a more general understanding o — monopsony as a market structure where there is only one buyer — or a certain type o — product. In general, that product can be either goods or services; but, — or present purposes, this Article — ocuses on services because it regards labor laws.

   Electronic copy available at: https://ssrn.com/abstract=3135936

2019] UNBUNDLING EMPLOYMENT 43

mine is the only purchaser o

mining services.171 Miners who work in that town labor under conditions o — monopsony to the extent that they cannot change the nature or location o — their work.172 Under these conditions, the mine can set low wages (and low bene — its) because the mine does not — ace any competition — or labor. When presented with the Hobson’s choice o — either being underpaid or not working at all, many miners would choose to work despite un — air conditions.173 In this situation, the mine pro — its while society su —

ers, because some able workers choose not to work instead o — working — or low wages while others are under-paid — or their e —

orts. This creates two problems. First, the mine uses its monopsony hiring power to extract wel — are — rom workers. Second, since a number o — able-bodied workers do not work, their labor represents a deadweight loss to society writ large. Monopsony, there — ore, is a type o — theoretical market

ailure.174 In a monopoly, the sole seller o — a good o —

ers the good at a high price in order to extract more money — rom the buyers who are willing to pay — or it while cutting o —

the less eager buyer — rom the market.175 Similarly, a monopsony allows — or both wealth trans — er — rom the powerless to the power — ul and an according diminution o — wealth to society.176 Economist Joan Robinson, a member o — the Cambridge school o — economics and a post-Keynesian thinker, highlighted this problem o —

monopsony in her 1933 book The Economics o

Imper — ect Competition, wherein she coined the term monopsony.177 Today, even anti-Keynesian economists generally agree that regulation is appropriate to prevent the problems o — monopoly and

1. See generally JOAN ROBINSON, THE ECONOMICS OF IMPERFECT COMPETITION (2d ed. 1969) (discussing the problem o — monopsony and — irst using the term in economic academia by explaining it in relatable terms).
2. Id.
3. Id.
4. See James Murphy Dowd, Oligopsony Power: Antitrust Injury and Collusive Buyer Practices in Input Markets, 76 B.U. L. REV. 1075, 1082 (1996) (stating that monopsony is “the — ailure o — a market to operate at a competitive equilibrium”).
5. See ROBINSON, supra note 171, at 47–52.
6. Id.

7. Id.

   Electronic copy available at: https://ssrn.com/abstract=3135936 44 DREXEL LAW REVIEW [Vol. 11:1

monopsony.178 The area o

law that addresses these concerns is called antitrust. Antitrust recognizes that regulations are appropriate— indeed, regulations are necessary—where market power leads to abuse and social harm. For example, antitrust law prohibits competitors — rom colluding to agree to raise prices or lower wages.179 In — act, antitrust conspiracy is a — ederal crime.180 Such collusion creates an unnatural monopoly, whereby competitors extract high prices or pay low wages, which in turn creates a deadweight loss to society. This conspiracy decreases social wel — are without producing any countervailing social bene — it. Thus, we intrinsically understand that antitrust conspiracy is “cheating” and ought to be a crime. Additionally, antitrust law sometimes intervenes in private ordering where market conditions could allow one — irm, acting alone, to exert market power. This is not conspiracy or collusion, which necessarily require more than one actor; rather, monopolization occurs when one actor has enough power to o —

er high prices or low wages without — ear that another — irm could enter the market and o —

er lower prices or higher wages.181 Such durable monopolies and monopsonies are actually quite rare. Rarely does one — irm corner an entire market and hold its 100% share — or an inde — inite time.182 Rather, markets are said to be relatively concentrated when there are a

ew power — ul actors with similar interests who might tacitly collude even i — they do not outright agree not to compete.183 Antitrust theory thus gets much more complicated when applied to the nuances o — power dynamics in real markets over

1. See Je

rey L. Harrison, Antitrust Policy and Monopsony, 76 CORNELL L. REV. 297, 309 (1991).

1. Sherman Antitrust Act o

   1890, 15 U.S.C. §§ 1–7 (2018).

2. Id.
3. Id.
4. See generally ADAM SMITH & BRUCE YANDLE, BOOTLEGGERS AND BAPTISTS: HOW ECONOMIC FORCES AND MORAL PERSUASION INTERACT TO SHAPE REGULATORY POLICIES (2014) (positing that the only situation where — irms are likely to hold onto durable monopolies or monopsonies — or an inde — inite period o — time is when government regulation has eviscerated competitive — orces).
5. Id.

   Electronic copy available at: https://ssrn.com/abstract=3135936

2019] UNBUNDLING EMPLOYMENT 45

time. Antitrust law has thereby evolved (in ways that labor law has not) to recognize the spectrum o — competitive dynamics.184 Antitrust laws recognize that monopoly and monopsony are not binary—not merely existing or absent—but, rather, they — all on a spectrum, which is commonly re — erred to as market concentration. In concentrated markets, only a — ew sellers or buyers wield immense market power. In competitive markets, there are many buyers and sellers. Market concentration is enumerated by its Her — indahl-Hirschman Index (HHI).185 An HHI ranges — rom 0 to 10,000 points, calculated by the squared sums o — the market shares o —

irms in a geographic industry, while giving more weight to larger — irms.186 According to the DOJ Antitrust Division, an HHI below 100 indicates a highly competitive industry, 100 =< HHI < 1500 indicates an unconcentrated industry, 1500 =< HHI < 2500 indicates modest concentration, and HHI >= 2500 indicates high concentration or market power.187 But labor law—and statutorily mandated employment bene — its—generally apply regardless o — whether there is high concentration or market power amid buyers o — labor. For example, the legal right to unionize is the same — or accountants,188 whose industry HHI score is 725189 (unconcentrated), as — or healthcare support sta —

such as nurses, whose HHI is 7500190 (highly concentrated).

1. See generally Thomas M. Jorde & David J. Teece, Innovation, Dynamic Competition, and Antitrust Policy, 13 REG. 35 (1990) (discussing how antitrust laws have changed over the past 150 years).

2. Her

   indahl-Hirschman Index, U.S. DEP’T JUST., https://www.justice.gov/atr/her — indahl- hirschman-index (last updated July 31, 2018).

3. Id.
4. U.S. DEP’T OF JUSTICE & FED. TRADE COMM’N, HORIZONTAL MERGER GUIDELINES 19 (2010), https://www. — tc.gov/sites/de — ault/ — iles/attachments/merger-review/100819hmg.pd — .

5. See, e.g., Union o

   Accountants and Auditors, IFAC, https://www.i — ac.org/about- i — ac/membership/members/union-accountants-and-auditors (last visited Dec. 16, 2018) (listing accountants as unionized).

6. The Her

   indahl-Hirschman Index and Occupational Concentration by Industry, U.S. BUREAU LAB. STAT. (Mar. 5, 2014), https://www.bls.gov/opub/ted/2014/ted_20140305.htm?view_ — ull.

7. AUDREY WATSON, BUREAU OF LABOR STAT., MEASURING OCCUPATIONAL CONCENTRATION BY INDUSTRY 5 (2014), https://www.bls.gov/opub/btn/volume- 3/pd — /measuring-occupational-concentration-by-industry.pd — .

   Electronic copy available at: https://ssrn.com/abstract=3135936 46 DREXEL LAW REVIEW [Vol. 11:1

Labor law’s economic oversight is particularly problematic when applied to the gig economy, because the gig economy provides alternative ways to work and thus reduces concentration and lowers market power in labor markets. Taxi cab drivers, — or example, can now work — or Uber i — they are not paid enough by Yellow Cab.191 Handymen can work — or themselves via Handy instead o — being employed by a general contracting — irm.192 In addition, i — a driver or handyman is not happy with Uber or Handy, she can seek work on the Ly — t or TaskRabbit plat — orm instead.193 In other words, the gig economy has the power to ameliorate the market ailment o — monopsony. Indeed, the real power o — the gig economy is to provide workers with alternative ways to work and derive bene — its thereby. A worker classi — ication system that rigidly lumps these gig workers into either employee or independent contractor classi — ications — ails to unlock the true potential o — this new economy. There — ore, a new and — lexible third classi — ication o —

worker is necessary.

              III. THE GIG WORKER CLASSIFICATION   To e ---

ectuate meaning — ul change — or workers in the gig economy, one must think beyond legacy notions o — employee and independent contractor. These terms, and the legal

ramework that developed considering them, do not properly govern workers on plat — orm economies, which this Article has termed “gig workers.” That is not to say the legacy system is

ailing those that — it into the traditional mold, but that the old

ramework does not work — or this new economy. There — ore, this Article proposes a — ramework — or unbundling the bene — its o —

work. This bene

it unbundling would allow plat — orms to compete — or gig workers by o —

ering them the best mix o — pay and bene — its. First, the proposed — ramework begins — rom an understanding o — what bene — its gig workers want to receive

1. How Uber Works, supra note 85.
2. About Us, supra note 135.

3. Why Ly

   t, supra note 85; How It Works, supra note 127.

   Electronic copy available at: https://ssrn.com/abstract=3135936 2019] UNBUNDLING EMPLOYMENT 47

rom work today. Second, this Article more — ormally articulates the structure o — this — ramework. Third, this Article addresses some potential criticisms.

                      A. What Gig Workers Want    As the above section makes clear, legally classi --- ying gig workers as employees or independent contractors is an exercise in jamming round pegs into square holes. This rigmarole is not only  --- utile  --- or the courts, but also harm --- ul to workers. It turns out that gig workers are not necessarily looking  --- or the complete bundle o ---  employment bene --- its that the NLRA and FLSA mandated in the 1930s,194 but this does not mean that they do not want or need any employment bene --- its at all. For instance, so long as health insurance is tied to work, this bene --- it in particular may be essential  --- or many (but not all) gig workers, and some gig workers might reasonably pre --- er a larger paycheck instead o ---  a retirement plan contribution. Thus, a more  --- lexible classi --- ication could help gig workers obtain a more optimal balance o ---  pay and bene --- its  --- rom work.    A major problem with the employee versus independent contractor debate is that it  --- ails to ask what gig workers actually want. This Article next explores this critical question, and it

inds that variability and — lexibility are essential — or the gig worker. When drivers on multiple ride-sharing and delivery plat — orms (including Uber, Ly — t, Postmates, DoorDash, UberEats, and Juno) were surveyed by an independent third party about what matters most to them, 53.5% responded “pay” and 38.4% responded “ — lexibility.”195 O — course, taking advantage o — pay incentives such as surge pricing requires

lexibility, so to some extent pay is dependent on user

1. See National Labor Relations Act o

   1935, 29 U.S.C. §§ 151-69 (2018), Fair Labor Standards Act o — 1938, 29 U.S.C. § 203 (2018).

2. Harry Campbell, RSG 2017 Survey Results: Driver Earnings, Satis

   action and Demographics, RIDESHARE GUY (Jan. 17, 2017), https://therideshareguy.com/rsg-2017-survey- results-driver-earnings-satis — action-and-demographics/.

   Electronic copy available at: https://ssrn.com/abstract=3135936 48 DREXEL LAW REVIEW [Vol. 11:1

lexibility. Only 1.1% responded “bene — its (health insurance, unemployment, etc.).”196 Contrast this with the — act that almost 100% o — union contracts in the manu — acturing sector require employers to provide li — e insurance and some sort o — medical coverage.197 Obviously, li — e and health insurance are not — ree, so employers will have to provide less pay i — they have to provide more bene — its. The Bureau o — Labor Statistics — ound the average cost o — employee bene — its in September 2017 was $11.31 per hour worked, while the average wages were $24.33 per hour worked.198 Why should Uber be required by law to allocate some 31.7% o — employee compensation to bene — its when only 1.1% o — surveyed gig workers consider this to be the most important — actor and 53.5% consider pay to be the most important — actor? One reason why sharing economy workers such as Uber drivers pre — er pay over bene — its is because these workers typically work — or several plat — orms. For example, 67.7% o —

Uber drivers work

or two or more on-demand driving or delivery services.199 These workers may also work — or additional sharing plat — orms in other sectors, and they may also work traditional jobs in the mainstream economy. While the incremental value o — each additional dollar earned is obviously quite high (otherwise there would be no rational reason — or people to work more at all), i — a worker already has health insurance — rom one employer, the incremental value o — a second, third, and — ourth health insurance policy — rom additional employers has quickly diminishing returns. Whatever the reason, sharing economy workers may have quite di —

erent pre — erences — or the mix o — pay, — lexibility, and bene — its provided — rom work than traditional economy workers. In — act, it seems quite apparent that gig economy

1. Id.

2. Medical Care Bene

   its: Access, Participation, and Take-Up Rates, U.S. DEP’T LAB., https://www.bls.gov/news.release/ebs2.t01.htm (last modi — ied July 20, 2018).

3. Press Release, U.S. Dep’t o

   Labor, Bureau o — Labor Stat., Employer Costs — or Employee Compensation—September 2017 (Dec. 14, 2017), https://www.bls.gov/news.release/pd — /ecec.pd — .

4. How Uber Works, supra note 85.

   Electronic copy available at: https://ssrn.com/abstract=3135936

2019] UNBUNDLING EMPLOYMENT 49

workers pre

er what gig economy plat — orms are o —

ering — rom the simple — act that they are — reely choosing to work — or these plat — orms! Accordingly, the sensible regulatory — ramework must allow people to make the — ree and in — ormed choice to work in the way they want to work, not to impose rigid work standards — rom a bygone era on a new generation o — workers. Additionally, the ethics o — imposing certain bene — its has recently come into question. For example, economist Todd Zywicki has asked whether it is ethical to nudge Americans to spend less and save more.200 His empirical analysis indicates that, contrary to popular opinion,201 there is actually a “tendency toward oversaving — or retirement.”202 While it might seem obvious that saving is good and one cannot do too much o — a good thing, the more complex reality is that there is an optimal amount o — savings. As Zywicki points out, “We all say we want to save more. But we also want more time with our

amilies, a — unctioning re — rigerator, and less debt—what i — these goals are in tension?”203 Thus, labor laws that require or even nudge employees to save — or retirement might actually harm the very workers they are intended to help. The ethics o — nudging workers and employers to provide certain bene — its is even more problematic when you consider that di —

erent people have di —

erent pre — erences. To put that another way, “The optimal intertemporal allocation o — work, leisure, and consumption will vary among di —

erent people depending on their subjective pre — erences.”204 For example, all else being equal, a one-hundred-year-old man has a shorter

uture li — e expectancy than a twenty-year-old man. The ethics o —

orcing or even nudging the older man to save — or retirement

1. Todd J. Zywicki, Do Americans Really Save Too Little and Should We Nudge Them to Save More? The Ethics o — Nudging Retirement Savings, 14 GEO. J.L. & PUB. POL’Y 877, 886 (2017).
2. See MELISSA A. Z. KNOLL, SOC. SEC. ADMIN., THE ROLE OF BEHAVIORAL ECONOMICS AND BEHAVIORAL DECISION MAKING IN AMERICANS’ RETIREMENT SAVINGS DECISIONS 2 (2010), https://www.ssa.gov/policy/docs/ssb/v70n4/v70n4p1.pd — (reporting survey results showing that 60% o — respondents do not think they are saving enough — or retirement).
3. Zywicki, supra note 200, at 888.
4. Id.

5. Id. at 892.

   Electronic copy available at: https://ssrn.com/abstract=3135936 50 DREXEL LAW REVIEW [Vol. 11:1

instead o

spending his money now is clearly dubious. The younger man also has a much lower risk o — getting ill, which also raises serious questions about the ethics o — requiring him to have health insurance. But that is exactly what the current, rigid binary labor law classi — ication does. Under our current system, an employee will receive less pay and more bene — its than an independent contractor, regardless o — which bene — its that employee actually needs or wants. The ethics o — this rigid binary system are on shaky ground. There — ore, a — ramework — or a new and — lexible third classi — ication is necessary.

               B. The Gig Worker Framework    The sharing economy is valued by consumers and gig workers alike  --- or its  --- lexibility. Shared work must likewise be regulated both  --- lexibly and with certainty. The current state o ---

uncertainty about worker classi

ication is only good — or lawyers and academics who can generate — ees and papers, while workers and plat — orms are harmed. But a blanket classi — ication o — all gig workers as either employees, independent contractors, or even the addition o — a third category (thereby applying rigid standards to all workers regardless o — skill, industry, and pre — erence) would be — olly. The DOL, IRS, and other regulators who govern employment should create a new and — lexible de — inition o — “gig worker” that is subject to an opt-in regime o —

disclosure and oversight. The opt-in gig worker system would — irst require a sharing economy plat — orm to submit a proposed de — inition o — “gig worker” under that plat — orm. This public — iling with government agencies would de — ine the mix o — pay, bene — its, and

lexibility that the plat — orm will o —

er. For example, Uber may propose that its gig workers will receive maximum pay and minimum bene — its, while Ly — t (its competitor) may propose a di —

erent blend that o —

ers lower pay but more certainty o —

income. Handy could o

er a minimum wage o — $18 per hour and require attendance at monthly training sessions, while TaskRabbit could pay gig workers solely according to a market price and put no additional demands on them. This system

 Electronic copy available at: https://ssrn.com/abstract=3135936

2019] UNBUNDLING EMPLOYMENT 51

could

urther promote competition, which in turn may lead to even better bene — its — or workers, as each — irm competes — or workers’ time and e —

ort on its plat — orm instead o — working on other plat — orms. Under this proposal, each — irm that creates technology plat — orms — or the sharing economy would be permitted to register its own de — inition o — “gig worker” by — iling a Form GW. The Form GW would stipulate which employment bene — its would be provided and which would not. The DOL would review each Form GW to ensure that it complies with laws that apply to all workers (such as minimum wage) and that it is clear enough to be easily understood by a reasonable worker. Aside

rom these minimum standards, however, the DOL would not review the merits o — the proposal as creating a good or bad work opportunity; rather, the DOL would simply determine whether the de — inition complies with labor laws. This Form GW process is designed to mirror the most success — ul elements o — the Securities Act o —

1. O

   ten called the “truth in securities law,” the 1933 Act has two basic objectives: — irst, providing necessary in — ormation to investors, and second, prohibiting deceit, misrepresentation, and — raud in the o —

ering o — investment opportunities.205 Under the 1933 Act,

irms must — ile a Form S-1 with the SEC be — ore selling stock on a stock market through an IPO.206 The SEC reviews the disclosures in that S-1 — or compliance with securities law, but it does not determine the merits o — the o —

ering as a good or bad investment opportunity.207 It holds o —

erors criminally liable — or

alse or misleading statements in the disclosures.208 Similarly, the goals o — the Form GW process would be, — irst, to ensure that potential workers would have the in — ormation necessary to decide whether to participate in a gig-economy plat — orm, and

1. The Laws that Govern the Securities Industry, U.S. SEC. & EXCHANGE COMMISSION, https://www.sec.gov/answers/about-lawsshtml.html (last modi — ied Oct. 1, 2013).

2. See 15 U.S.C. § 77

   (2018).

3. See id. § 77h.

4. Id. § 77t.

   Electronic copy available at: https://ssrn.com/abstract=3135936 52 DREXEL LAW REVIEW [Vol. 11:1

second, to hold the plat

orm liable — or — alse or misleading claims about work on its plat — orm or the bene — its obtained thereby. O — course, workers can only make an in — ormed decision about whether to participate in a plat — orm i — there is indeed a choice to be made. Sometimes, working — or a particular — irm is a Hobson’s choice—that is, a choice o — taking what is o —

ered or taking nothing at all. Indeed, the un — ortunate circumstance o —

exploitive employers amid the rampant unemployment o

the Great Depression is what prompted the NLRA, which created the rigid bundle o — employment rights.209 It would be abominable i — the law were instead to — acilitate exploitation o —

powerless workers by power

ul — irms. However, this Article previously argued that the gig economy increases competition

or workers in the economy overall. Furthermore, the DOL could grant or deny a Form GW proposal based on conditions o — market — ailure such as monopsony—where there is only one — irm employing a certain type o — labor—or an otherwise highly concentrated market where the employer has substantial market power.210 Based on evidence o — market power, the DOL might not permit employers to o —

er a substantively un — air suite o — bene — its. For this analysis, the DOL could look to over one-hundred years o —

antitrust jurisprudence that has likewise sought to encourage competitive markets. Some may be concerned about the cost o — implementing this Form GW process, but there is precedent showing how this proposal might pay — or itsel — , and then some. For example, the median IPO in 2016 was $94.5 million.211 The median registration — ee — or a Form S-1 in 2016 was thus $11,765.25,

1. See 29 U.S.C. § 151 (2018).
2. COUNCIL OF ECON. ADVISERS, LABOR MARKET MONOPSONY: TRENDS, CONSEQUENCES AND POLICY RESPONSES 2–4 (2014), https://obamawhitehouse.archives.gov/sites/de — ault/ — iles/page/ — iles/20161025_monopsony_lab or_mrkt_cea.pd — .
3. David A. Westenberg, 2016 Year-End IPO Market Review, WILMERHALE (Feb. 2, 2017), https://www.wilmerhale.com/en/insights/blogs/the-road-to-ipo-legal-and-regulatory-insights- into-going-public/2016-year-end-ipo-market-review.

   Electronic copy available at: https://ssrn.com/abstract=3135936

2019] UNBUNDLING EMPLOYMENT 53

according to the SEC’s

ee calculator.212 Antitrust merger review generates even more — ees, ranging — rom $45,000 to $280,000, depending on the size o — the transaction.213 In short, a — irm that wants non-standard de — initions — or its workers can pay — or the regulatory apparatus necessary to do so. Government agencies would thus be tasked with reviewing the plat — orm’s proposed de — inition o — gig worker according to basic principles o —

airness and compliance, at least with the minimum standards that independent contractors should expect to receive under the law. Through a review and comment period, agencies would work with plat — orms to re — ine the de — inition and clari — y it so it would be easily understood by a person o — ordinary skills. Other requirements, such as translating the gig worker de — inition into multiple languages and posting it on a public website, could also be imposed on the plat — orm by the government agencies. Once approved, the plat — orm could be required to maintain the published standards

or gig workers unless the government agency approves an amendment, and — raud liability could result i — the plat — orm — ails to maintain the published standards.214 The gig worker de — inition — unctions as a quasi-public contract between the plat — orm and society that government agencies as well as private individuals would have rights to en — orce. There is precedent — or such a plain-language requirement in consumer protection law. The European Union, — or instance, recently passed the General Data Protection Regulation (GDPR),215 requiring — irms who handle personal data to use “clear and plain language” to make people “aware o — the risks, rules, sa — eguards[,] and rights in relation to the processing o —

1. Registration Fee Estimator, U.S. SEC. & EXCHANGE COMMISSION, https://www.sec.gov/o — m/registration- — ee-estimator.html# (last modi — ied Nov. 9, 2018).
2. Id.
3. This is similar to how laws require companies to create privacy policies and make them responsible — or breaches and violations o — those policies. See, e.g., Privacy Policies Are Legally Required, PRIVACYPOLICIES.COM, https://privacypolicies.com/blog/privacy-policies-legally- required/ (last updated Nov. 20, 2018).

4. Council Regulation 2016/679, General Data Protection Regulation, 2016 O.J. (L 119/1) 1.

   Electronic copy available at: https://ssrn.com/abstract=3135936 54 DREXEL LAW REVIEW [Vol. 11:1

personal data.”216 A declaration o

consent — or the use o —

personal data must “be provided in an intelligible and easily accessible — orm, using clear and plain language”; otherwise, consent will not be regarded as in — ormed.217 To make the terms clear, visualization should be used where appropriate, such that even children can easily understand the consent.218 I — the language o — the consent — orm is — ound not to be clear and plain, then the consent — orm “constitutes an in — ringement o — [the GDPR] and shall not be binding.”219 Similar requirements o —

clear and plain language could apply to the Form GW to ensure that ordinary workers o — average capacity can easily understand the bene — its they will receive — rom work. The Form GW could thus give gig workers a clear choice as to what bene — its they want to receive — rom work. So long as there is vigorous competition in the market — or gig workers, plat — orms will constantly innovate and compete to o —

er workers the best mix o — bene — its to attract and retain the top talent. This will preserve the virtue o — the sharing economy as a

lexible and innovative working environment while pressuring companies to o —

er an optimal mix o — bene — its to gig workers and value to consumers. In return — or promulgating a gig worker de — inition, government agencies must put compliant plat — orms in a sa — e harbor where they are not at risk o — employee reclassi — ication. This would incentivize plat — orms to engage in this exercise o —

social contracting and curtail the deadweight loss to most o

society that results

rom legal uncertainty and litigation. Employers who participate in the new opt-in gig worker system would bene — it — rom the corresponding reduction in legal risk.

1. Id. at 39.
2. Id. at 42.
3. Id. at 58.

4. Id. at art. 7.

   Electronic copy available at: https://ssrn.com/abstract=3135936 2019] UNBUNDLING EMPLOYMENT 55

                   C. Concerns and Criticisms   While the gig worker solution proposed by this Article could solve many problems that exist today under the rigid binary worker classi --- ication system, it could also create other challenges that would need to be addressed.   Tax law is the elephant in the room regarding this proposal. As mentioned in the Introduction, this Article does not address the tax law issues. Nevertheless, it must be acknowledged that the worker classi --- ication system is also a tax classi --- ication system. It may be possible to divorce labor law and tax law, such that gig workers whose  --- irms provide tax withholding services  --- ile a Form W-2, just as employees do, while gig workers who receive their  --- ull compensation up  --- ront  --- ile a Form 1099, just as independent contractors do. The merits o ---
   

our current employment tax law are best addressed in another paper. However, as a matter o — principle, there is no particular reason why employment bene — its and tax status need to be intertwined. Additionally, some scholars might criticize this proposal by claiming that the gig economy is just a smokescreen to hide exploitation o — workers in the new economy. While it is not possible to anticipate and address all the potential concerns about exploitation here, the reader should be reminded that competitive and e —

icient markets are thought to produce the highest quality product at the lowest price.220 With speci — ic application to labor markets, where those markets are highly competitive (i.e., unconcentrated), market — orces will drive employers to o —

er competitive pay and bene — its.221 Employers o —

ering an optimal package o — worker bene — its will attract the best workers — or the job who have the lowest propensity to

1. See HARRIS & KRUEGER, supra note 10, at 17.

2. David Lee & Emmanuel Saez, Optimal Minimum Wage Policy in Competitive Labor Markets, 96 J. PUB. ECON. 739, 741 (2012) (“I — there is no minimum wage [or any other taxes, mandatory bene — its, or other trans — ers], combining the demand side and the supply side de — ines the competitive equilibrium.”); see also Michael Rothschild & Joseph Stiglitz, Equilibrium in Competitive Insurance Markets: An Essay on the Economics o — Imper — ect In — ormation, 90 Q.J. ECON. 629, 629 (1976).

   Electronic copy available at: https://ssrn.com/abstract=3135936 56 DREXEL LAW REVIEW [Vol. 11:1

quit.222 The gig worker classi

ication proposal will create a new way — or employers to compete — or workers by o —

ering them the most optimal mix o — pay and bene — its, which can increase both employer and employee wel — are. Moreover, this allows — or people with diverse needs to select which bene — its they want most — rom work. However, this proposal is limited to competitive markets, so it cannot be used as a means o —

exploitation where workers have no real choice but to work

or a certain employer. Some scholars might object to this proposal based on their general distrust o — gig economy plat — orms. These plat — orms are usually created by startups, whose corporate culture o — ten leaves much to be desired.223 An inhumane tone — rom upper management can lead coders to treat plat — orm workers as “animals,” as a recent Vanity Fair article suggested when describing the way Uber’s coders manipulate bonuses to trick drivers.224 Accordingly, some might criticize this Form GW proposal on the basis that it will be exploited by juvenile startups, who will in turn leverage it to gain an un — air advantage over traditional employers. But this criticism misses the main purpose o — this proposal, which is to make gig

1. Joanne Salop & Steven Salop, Sel

   -Selection and Turnover in the Labor Market, 90 Q.J. ECON. 619, 620 (1976).

2. Insu

erably sexist conditions and discriminatory hiring practices have right — ully given tech startups an unsavory and untrustworthy reputation. For example, Uber’s — ounder and

ormer CEO, Travis Kalanick, took his team o —

ive Uber managers to an escort bar in Seoul, South Korea. See Ryan General, Uber’s CEO Once Went to an Escort Bar in South Korea with a Group o — Workers, NEXTSHARK, https://nextshark.com/ubers-ceo-went-escort-bar-south-korea-group- workers/ (last visited Dec. 16, 2018). One o — the managers who attended was — emale, and she complained that “it made [her] — eel horrible.” Id. Un — ortunately, this unacceptable behavior was not uncharacteristic — or Kalanick, who also issued the in — amous 2013 “sex memo.” See Marco della Cava, Uber’s Travis Kalanick O —

ered Sex Rules — or 2013 Party: Report, USA TODAY (June 8, 2017), https://www.usatoday.com/story/tech/talkingtech/2017/06/08/ubers-travis-kalanick- o —

ered-sex-rules-2013-party-report/102646050/. In other work, I have argued that this misogynistic tone — rom the top contributed to a “ — rat boy” culture and intolerable behavior by middle management, such as the sexist and discriminatory treatment reported by Susan Fowler. See Susan Fowler, Re — lecting on One Very, Very Strange Year at Uber, SUSAN FOWLER: BLOG (Feb. 19, 2017), https://www.susanj — owler.com/blog/2017/2/19/re — lecting-on-one-very-strange-year- at-uber.

1. Susan Fowler, ”What Have We Done?”: Silicon Valley Engineers Fear They’ve Created a Monster, VANITY FAIR (Aug. 9, 2018), https://www.vanity — air.com/news/2018/08/silicon-valley- engineers- — ear-they-created-a-monster.

   Electronic copy available at: https://ssrn.com/abstract=3135936

2019] UNBUNDLING EMPLOYMENT 57

plat

orms more transparent and accountable to their workers.225 Presently, these plat — orms make concerted e —

orts to avoid characterizing their workers as employees so that their workers have the minimum set o — bene — its under the law.226 I — the law recharacterizes one plat — orm’s independent contractors as employees, the other plat — orms will most likely change their practices to ensure their workers are not characterized as employees. Some, like Handy, have even said that they will close down i — the law requires that all participants on the plat — orms must be characterized as employees, and surely workers are better o —

having the option to work on Handy’s plat — orm instead o — having no option at all.227 Instead o — this cat- and-mouse game, the Form GW proposal would encourage plat — orms to publish a clear and en — orceable public contract — or their workers, e —

ectively reducing con — usion and opportunities

or abuse. This proposal encourages startups to be more transparent, mature, and accountable, which will improve conditions — or workers who are currently con — used and subject to manipulation. Further, the Form GW proposal hopes to provide a vehicle — or more humane and egalitarian startups to enjoy more success than their less enlightened peers.

                              CONCLUSION   Sharing economy plat --- orms are distinct  --- rom traditional employers in the way that they operate. Plat --- orms are not “employers” in the conventional sense. Rather, they are matchmakers: they provide an on-demand base o ---  workers willing to provide a service to consumers and then encourage consumers to use those services. Labor rules based on traditional manu --- acturing work and centralized production o ---

goods

or power — ul or monopsonistic employers are a bad — it

or the decentralized plat — orm service economy, especially where plat — orm economies enhance competition — or labor.

1. See supra Section IV.B.
2. See supra notes 103–05 and accompanying text.

3. NYU School o

   Law, supra note 136.

   Electronic copy available at: https://ssrn.com/abstract=3135936 58 DREXEL LAW REVIEW [Vol. 11:1

There

ore, labor law should adopt concentration measures and apply legal remedies such as statutory bene — its only where they are merited by economic conditions. Moreover, labor law should recognize that matchmaking plat — orms in the new economy are di —

erent — rom, and somewhat competitive with, traditional — irms. The gig worker opt-in — ramework proposed in this Article would leverage the competitive pressures to improve working conditions — or American workers, consumers, and — irms in the new economy.

 Electronic copy available at: https://ssrn.com/abstract=3135936

2019] UNBUNDLING EMPLOYMENT 59

                                             APPENDIX


              Value Chain in Traditional Economics

Capital Input ⛰ # %# ⚙ )⚙ + Labor Input ⛏$&(* - VALUE

Output # %# ⚙ )⚙ .❤ / Resource Inbound Manu- Outbound Marketing A — termarket Extraction Logistics
— acturing Logistics & Sales Services In traditional economics, the value added — lows — rom le — t to right, which is to say that value increases at each stage in this process. © 2019 Seth C. Oranburg

                Value Chain in Sharing Economics

                                                  !
                                                  Plat --- orm




        Service Provider                          Rating                  Service Consumer

Under-Used “ & # Available

               $                                                                %   Asset                                                                                         Cash



  In sharing economics, the value chain is triangular and bi-directional:
  value is created when under-used assets are shared via plat --- orms.
                                                                            © 2019 Seth C. Oranburg




              Electronic copy available at: https://ssrn.com/abstract=3135936