16 Balancing Flexibility and Rigidity
Do Unions Make Sense in the On-Demand Economy?
Seth C. Oranburg and Liya Palagashvili
A Unionization: Background
Unionization – the organization o --- workers to act collectively to obtain higher wages or
better working conditions – has a storied history. While precursors to trade unions such as
guilds existed be --- ore the Revolutionary War, the modern union did not come into its
own until a --- ter the Industrial Revolution. As we move into a new economic era, the
necessity and use o --- unions is once again called into question.
1 History o --- Unionization
Precursors to trade unions first emerged in the 1800s as journeymen associations
attempted to set wage schedules --- or their trades, although early courts --- ound criminal
liability --- or these “conspiracies.”1 Unionization increased a --- ter the Civil War in part as a
means to keep nonwhites and women out o --- skilled trades.2 But unions did not truly
come into the main o --- American li --- e until the Great Depression, when labor unrest
compelled the --- ederal government to pass the National Labor Relations Act (NLRA).3
The NLRA authorized workers’ rights “to sel --- -organization, to --- orm, join, or assist labor
organizations, [and] to bargain collectively” and created the National Labor Relations
Board (NLRB) to en --- orce the regime.4 The NLRA, which announced that the o --- ficial
policy o --- the United States was “to encourage the practice and procedure o --- collective
bargaining,” was regarded as “perhaps the most radical piece o --- legislation ever enacted
by the United States Congress.”5 Workers staged waves o ---
actory occupations and 1 Commonwealth v. Pullis, 3 Doc. Hist. 59 (Pa. 1806), also known as the Philadelphia Cordwainers case. 2 James Gray Pope, A Brie — History o — United States Labor and Employment Law, The Ox — ord Inter- national Encyclopedia o — Legal History 477–486 (Stanley N. Katz ed.) (2009) (hereina — ter, “History o —
Labor Law”), at 480.
3
National Labor Relations Act o --- 1935 §§ 151–169 (2012). See also Mark Barenberg, The Political
Economy o --- the Wagner Act: Power, Symbol, and Workplace Cooperation, 106 Harv. L. Rev. 1379, 1389
(1993) (“[T] he opportunity --- or such a dramatic legislative initiative was generated by ‘mass politics’ in the
orm o — popular electoral realignment, populist political organization, and mass labor unrest . . . . That opportunity was seized by loosely interconnected networks o — political-technocratic entrepreneurs driven by progressive ideological commitment and ambition.”). 4 Kate Andrias, The New Labor Law, 126 Yale L. J. 2, 14 (2016). 5 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).
179
https://doi.org/10.1017/9781108610070.020 Published online by Cambridge University Press 180 The “Fissured” Workplace
sit-down strikes while President Roosevelt threatened to pack the Supreme Court with
new justices i --- they did not uphold the NLRA’s constitutionality.6 These unorthodox
tactics succeeded, and union membership in the manu --- acturing sector quadrupled in the
first decade o --- the NLRA.7
But World War II changed the economy --- orever. Be --- ore WWII, America’s role in the
global economy was as primarily a supplier o --- raw materials and manu --- actured goods.8
A --- ter the war, Europe was in shambles, and America (to whom the European nations
owed a great deal o --- money) supplanted Great Britain as the financial center o --- the west
as the dollar replaced the pound as the world’s reserve currency. Meanwhile, the --- ederal
government curtailed union rights9 and 27 states passed “right-to-work” laws that prohibit
mandatory union membership and dues.
The American labor --- orce changed dramatically this century, and unionization has
allen sharply in correlation with the shi — t — rom a goods to a services economy. In 1935, when the NLRA was passed, more American workers were employed in production than in services.10 In 2013, over 83 percent o — American workers were employed in the service sector.11 Union membership as a percent o — nonagricultural employment declined — rom a peak o — over 35 percent in 1954 to under 12 percent in 2003.12 Union membership in the private sector was only 6.7 percent in 2015.13
2 Unionization in the On-Demand Economy
Today, there is a debate regarding whether --- urther unionization is needed in various
sectors o --- the economy. Companies in the “on-demand economy”14 (i.e., Uber, Ly --- t,
Task Rabbit, Handy) have come under scrutiny --- or their employment practices.
Many workers supplying on-demand goods and services are classified as independent
contractors as opposed to traditional employees. Traditional employees receive an IRS
6
History o --- Labor Law, supra note 2, at 482.
7
Id.
8
Guiseppe Berlingieri, Outsourcing and the Shi --- t --- rom Manu --- acturing to Services, Winter 2013/14 Centre-
Piece 16, 16 (2014), http://cep.lse.ac.uk/pubs/download/cp413.pd --- .
9
For example, Congress passed the Labor Management Relations (Ta --- t–Hartley Act) in 1947, which
restricted some union activity and permitted states to enact “right-to-work” laws. In 1964, Congress passed
Civil Rights Act, which prohibited unions --- rom discriminating on the basis o --- race, religion, sex, or
national origin. Labor Management Relations Act o --- 1947 29 U.S.C. § 141 (2012).
10
U.S. Bureau o --- the Census, Series D 127-1411, Historical Statistics o --- the United States, Colo-
nial Times to 1970, Part 1 at 4 (2016) (Year: 1935; Total Non-Farm Workers: 27:035; Mining: 897;
Construction: 912; Manu --- acturing: 9069; Transportation: 2786; Trade: 5431; Finance: 1335; Other
Services: 3142; Government: 3481. Total Goods: 10,878; Total Services: 9908 (in thousands)).
11
Berlingieri, supra note 8.
12
Congressional Research Service, RL 32553, Union Members Trends in The United States (figure 1)
at 11 (2004).
13
Megan Dunn & James Walker, U.S. Bureau o --- Labor Statistics, Union Membership in The United States at
4 (2006), www.bls.gov/spotlight/2016/union-membership-in-the-united-states/pd --- /union-membership-in-
the-united-states.pd --- .
14
Mike Jaconi, The “On-Demand Economy” is Revolutionizing Consumer Behavior – Here’s How, Business
Insider Mag., www.businessinsider.com/the-on-demand-economy-2014-7 (accessed May 31, 2018), (The
definition used by the coalition o --- on-demand companies is as --- ollows: “The On-Demand Economy is
defined as the economic activity created by digital marketplaces that --- ulfill consumer demand via
immediate access to and convenient provisioning o --- goods and services.”). “Gig” and “sharing” economy
have also been used to describe this plat --- orm.
https://doi.org/10.1017/9781108610070.020 Published online by Cambridge University Press Balancing Flexibility and Rigidity 181
tax Form W-2, while contractors receive a Form 1099.15 But the distinction between W-2
Employees and 1099 Contractors is much more than --- ormal. Employees and contractors
are treated di ---
erently under both tax and labor regulations. For example, employees enjoy protections under the Fair Labor Standards Act (FLSA), the Employment Retire- ment Income Security Act (ERISA), and the Family Medical Leave Act (FMLA), while contractors generally are not protected by these laws. Furthermore, and more importantly
or the purposes o — this paper, contractors do not have the same right as employees to unionize and bargain collectively. Despite the ostensible tax and employment classification o — on-demand workers as 1099 contractors, unions are now pushing — or the right to represent these workers. Supporters o — unionization claim this will allow on-demand workers to bargain — or better wages and benefits, and this is necessary because contractors do not enjoy the benefit o —
employment laws. As one recent example, Seattle passed an ordinance granting labor
unions the right to represent drivers --- or Uber (though Uber does not classi --- y its drivers as
employees). The U.S. Chamber o --- Commerce has sued to challenge the validity o --- this
ordinance on several grounds (e.g., preemption by NLRA, antitrust violations), and the
case is currently playing out.16 In New York City, a group o --- drivers steered by the
International Association o --- Machinists and Aerospace Workers created the “Independ-
ent Drivers Guild” in 2016. Uber has recognized and even --- unded the guild, but as part
o --- the deal, Uber required that Machinists do not attempt to unionize drivers --- or five
years.17 The Guild’s mission states: “The courts have thus --- ar restricted app-based drivers
rom organizing as a traditional union, but we believe that the power o — a union is workers banding together. With the Guild, we unite to win better working conditions and increase worker’s earnings, now.”18
B Unionization Economics
From an economic perspective, industry unionizations could have both large costs and
substantial benefits, depending on the various conditions o --- the industry. Economists
di ---
erentiate labor markets into cases that are monopsonistic (or oligopsonistic) versus cases that are competitive, and they draw di —
erent theoretical analyses — or union harm or benefit depending on these labor market conditions. Unions provide benefits to workers when employers have much more power than employees. By unionizing, otherwise powerless employees can collectively protect themselves — rom an otherwise dominant employer. Indeed, union rules were developed in the 1930s to protect — actory workers
rom monopolistic manu — acturing employers who would unilaterally impose unduly
15
See U.S. Internal Revenue Service, Forms and Associated Taxes --- or Independent Contractors (2017),
www.irs.gov/businesses/small-businesses-sel --- -employed/ --- orms-and-associated-taxes- --- or-independent-contractors;
Internal Revenue Service, Depositing And Reporting Employment Taxes (2017), www.irs.gov/businesses/small-
businesses-sel --- -employed/depositing-and-reporting-employment-taxes.
16
Chamber o --- Commerce v. Seattle, 890 F.3d 769 (9th Cir. 2018) (partially reversing grant o --- De --- endants’
motion to dismiss and remanding --- or --- urther proceedings only on antitrust claim).
17
Miranda Kantz, How Drivers Are Finally Out --- oxing Uber, Wired (accessed May 25, 2017),
www.wired.com/2017/05/how-drivers-are-finally-out --- oxing-uber/.
18
About the IDG, Independent Drivers Guild, www.drivingguild.org (Jun. 1, 2018).
https://doi.org/10.1017/9781108610070.020 Published online by Cambridge University Press 182 The “Fissured” Workplace
burdensome working conditions without commensurate pay or benefits. The original
goals o --- union rules was to provide minimum and maximum amounts o --- hours to be
worked so workers could not be exploited by power --- ul employers. Rigidity in hours and
days worked, promotions, benefits, and other conditions o --- employment are the sine qua
non o --- collective bargaining agreements (CBAs).
CBAs give workers certainty about these terms. Immutability is their nature and
purpose. The economic literature is replete with examples o --- how CBAs create rigid
rules --- or work, so this chapter will only briefly exempli --- y a --- ew o --- the rules that are
pertinent to work in the sharing economy.
While there are many di ---
erent types o — CBAs, we can best understand CBAs in general by looking at meta-analysis o — such contracts. The most comprehensive meta- analysis o — CBAs is — ound in the Bloomberg Law series entitled Basic Patterns in Union Contracts. Most recently in its — ourteenth edition (1995), this analysis o — over 400 CBAs
rom a variety o — industries, type and size o — unions, number o — employees covered, and geographical areas shows that there are — our provisions that are extremely common in traditional CBAs but particularly inapposite to the on-demand and gig economy.19 To begin, 97 percent o — CBAs limit discharge to “cause” or “just cause.” All other discharge may be wrong — ul and is subject to a proceeding that generally have three or — our steps. Almost all union CBAs require employers to provide li — e insurance, some — orm o —
medical coverage, and, anachronistically, pension plans. Retirement eligibility is almost
always stipulated at age 65. More than hal --- require “income maintenance” (required
annual pay raises --- or everyone), and almost all provide --- or overtime pay (generally time-
and-a-hal --- ) when work exceeds 8 hours per day or 40 hours per week, or when work
occurs on a holiday or weekend. Seniority is generally the dispositive --- actor in hiring,
firing, promotion, and layo ---
s. More than hal — o — union employees governed by CBAs get five weeks o — vacation per year, and over 80 percent enjoy — our weeks per year o — paid leave. These rigid rules may benefit workers in certain “monopsony” markets, but they also cause unemployment and higher consumer prices in other “competitive” markets.
1 Unionization in Monopsony
Traditionally, the argument --- or unionization has been strongest in the presence o --- lopsided
bargaining power (i.e. very power --- ul employers and powerless employees). In economic
theory, such “asymmetries” occur when labor markets have single buyers o --- labor – called
“monopsonies” (in contrast with monopolies, which are single sellers). The classic example
o --- such a monopsonistic labor market is when there is a large company in a small rural
town that can exploit workers because workers do not have many other opportunities --- or
employment. Related issues arise where a hand --- ul o --- employers that dominate a market are
somehow able to collude to depress wages through their “oligopsony.”
Generally, monopsony exists only --- or jobs in a narrow geographic region or that
require a highly specialized skill set, since close alternative occupations are limited.
The labor market --- or nurses, --- or example, has been cited in the literature as a classic case
o --- monopsonistic power because nurses typically have one buyer: hospitals. As described
by Link and Landon, labor markets --- or nurses approaches “classical oligopsony or
19
Although the terms gig economy and on-demand plat --- orms do not always overlap, in this chapter we use
on-demand and gig economy interchangeably when there is overlap.
https://doi.org/10.1017/9781108610070.020 Published online by Cambridge University Press Balancing Flexibility and Rigidity 183
monopsony. This was evidenced by the --- act that over 70 percent o --- the hospitals in the
United States are located in a one-hospital community while many other communities
are serviced by only a --- ew hospitals.”20 Other special industry examples approaching
monopsonistic power are university pro --- essors (because o --- the specialized skill set), coal
miners in the twentieth century, and baseball players subject to reverse clauses.21 Reverse
clauses bounded the player to the team, until the contract expired or unless the owner
wanted to “ --- ree” the player. In other words, players were not “ --- ree agents” with the power
to change teams. Other sources o --- monopsonistic power can arise i --- workers are ignorant
o --- alternatives or incur high costs by changing jobs, or i --- employees have developed other
ties to the employer or the geographic location.
Monopsony can depress worker’s wages, benefits, and working conditions. By definition,
monopsony means that employees are bound to a certain employer by skill, geography, or
other reasons. Workers’ dependence on a single employer gives that employer power to
lower wages, limit benefits, and otherwise exploit that power asymmetry.
Labor unions are thought to ameliorate some o --- monopsony’s harm to workers. Unions
are organizations, backed by governments, that negotiate with businesses or other
organizations on behal --- o --- their union members (the workers). I --- companies are unwill-
ingly to meet various standards that workers want, union members use strikes, sit-ins, and
other tools to --- orce the employer to meet such standards. Theoretically, these collective
actions can lead to better terms --- or workers where unions induce pressure --- or wages
increases, better benefits, and improved working conditions. Rees explains that unions
are able to do this because “their ability to impose costs on management through strikes,
slowdowns, or other pressure tactics, in the short run, are greater than the costs o --- the
wage increase provided through collective bargaining.22 In relatively monopsonistic labor
markets, empirically, it has --- ound that unionization provided greater wage and --- ringe
benefits with registered nurses, practical nurses, and hospital secretaries and house-
keepers.23 From one study, the wage e ---
ect o — unions on nurses is plus 8 percent, and plus 11–12 percent — or hospital secretaries and housekeepers.24
2 Unionization in Competitive Markets
On the other hand, when labor markets are relatively competitive, the case --- or unions is
“mixed” in terms o --- demonstrated benefits --- rom unions, and demonstrated harms.25
20
Charles Link & John Landon, Monopsony and Union Power in the Market --- or Nurses, 41 S. Econ. J. 649,
649 (1975).
21
For university pro --- essors, see, Michael R. Ransom, Seniority and Monopsony in the Academic Labor
Market, 83 Am. Econ. Rev. 221 (1993); --- or coal miners in the twentieth century, see William M. Boal,
Testing --- or Employer Monopsony in Turn-o --- -the-Century Coal Mining, 26 RAND J. Econ. 519 (1995); --- or
baseball players subject to the reverse clause, see Lawrence M. Kahn, The Sports Business as a Labor
Market Laboratory, 14 J. Econ. Perspectives 75 (2000) and Gerald W. Scully, Pay and Per --- ormance in
Major League Baseball, 64 Am. Econ. Rev. 915 (1974).
22
Albert Rees, The E ---
ects o — Unions on Resource Allocation, 6 J. Lab. & Econ, 69, 69 (1963). 23 Roger Feldman & Richard Sche — fler, The Union Impact on Hospital Wages and Fringe Benefits, 35 Indus. & Lab. Rel. Rev. 196 (1982); Charles Link & John Landon, Monopsony and Union Power in the Market
or Nurses, 41 S. Econ. J. 649 (1975). 24 Feldman & Sche — fler, supra note 22. 25 For a theoretical and empirical analysis on di —
erences o — demonstrated benefits between monopsony and competitive labor markets, see Kip Viscusi, Union, Labor Market Structure, and the Wel — are Implications o —
Quality Work, 1 J. Lab. Res. 175–192 (1990).
https://doi.org/10.1017/9781108610070.020 Published online by Cambridge University Press 184 The “Fissured” Workplace
With competitive labor market processes, economic theory suggests that employees are
generally neither substantially underpaid nor substantially overpaid. I --- , in a particular
circumstance, workers are underpaid, not only do they have an unusually strong incen-
tive to search --- or and accept better jobs elsewhere, but so, too, do profit-motivated
competing employers have strong incentives to seek out and recruit underpaid workers,
whether by locating near these workers, o ---
ering them better pay, or taking other cost- e —
ective measures to entice these workers. These processes o — adjustment by both workers and employers generally push employment arrangements and worker pay up to competi- tive levels. This is not to say that in all competitive markets, at any given moment, wages are precisely e — ficient. Rather, in labor markets that are characterized by competition, there is a tendency toward this outcome. Contrast this with the monopsony. In mono- psonistic labor markets, employers do not — ace strong incentives to increase worker pay, and they might decrease it. This is again to say that unions can benefit workers in conventional monopsonistic labor markets. But according to economic theory, when unions enter into an otherwise normal labor market, they can pressure firms to provide wages, benefits, and working conditions that are above the competitive levels. Although above-market wages might sound good — or workers, such market distortions, in theory, can harm nonunion workers, employers, and even consumers by creating dead weight loss to society. Unionization can lead to above-market wages because unions “gain market power by obtaining a legal monopoly on the provision o — labor services to a particular firm or industry.”26 The degree o — impact by unions on prices is directly correlated with unions’ power. Unions become more power — ul i — the firm does not have many alternatives to the labor supply or i — other labor supplies are restricted. The main legal instrument that unions use to restrict employment conditions is the CBA. CBAs typically limit working conditions like hours and days worked, require an annual cost-o — -living pay increase, limit firing to “cause,” and have proceedings — or employees who believe they were wrongly terminated. CBAs generally strengthen employee power. In a monopsonistic market, union-negotiated CBAs can restore power symmetry, but in competitive markets, CBAs can require wages above even the competitive levels.27 According to economic theory, one negative impact o — a union in otherwise normal labor markets is creating unemployment. This is because unions can hold wages above the competitive level, thereby creating a wage floor. Now — acing increased union wages, employers reduce their quantity demanded o — unionized labor. On the supply side, the wage floors lead to an excess o — the supply o — workers. The net e —
ect is the creation o —
unemployment.
Some economists also argue that higher unionized wages create lower wages --- or
nonunionized workers because the supply o --- nonunionized workers increases, thereby
depressing the wages in nonunionized sectors.28
26
Paul A. Samuelson & William D. Nordhaus, Economics 320 (Snehi Kumari ed., 19th ed. 2010).
27
Where unions are not power --- ul (i.e. unions cannot restrict the use o --- nonunion labor), then it is possible
(absent other countervailing --- actors) that competitive --- orces would drive those firms with above-
competitive wages out o --- the market. However, the use o --- legislation to preclude nonunion entry (i.e.
signi --- ying power --- ul unions) is one --- actor that would prevent some o --- these otherwise competitive --- orces
28
H. Gregg Lewis, The Labor Monopoly Problem: A Positive Paradigm, J. Pol. Econ. (1951); W. Mellow,
Unionism and Wages: A Longitudinal Analysis, Rev. Economics & Statistics (1981); For more recent
empirical work, see Bernd Fitzenberger, Karstan Kohn, & Alexander Lembcke, Union Densities and
https://doi.org/10.1017/9781108610070.020 Published online by Cambridge University Press Balancing Flexibility and Rigidity 185
CBAs almost always contain more than just wages. They include provisions on hiring
and firing, seniority, benefits, and other stipulations. Provisions that create --- rictions in the
mobility and churn over o --- labor ( --- or example, by making it more di --- ficult and costly to
fire an employee), can benefit the employees, but these restrictions can also decrease
productivity, and impose higher labor costs.29 Some provisions in the CBAs can thus
increase costs to employers which may be passed on to downstream consumers or limit
purchases --- rom upstream suppliers.
Empirically, scholars --- ound that unions do increase unionized workers’ wages, bene-
fits, and working conditions.30 But on the cost side, they --- ound that greater unionization
increases unemployment in the country, and specifically among --- emales, young males,
and older individuals.31 Some studies --- ound that although unions benefit unionized
workers, this can come at the expense o --- harming nonunion members via the surplus o ---
labor e ---
ect.32 Other studies — ound that unions can help nonunion members through other mechanisms such as the threat o — a union: when one company believes that its workers may unionize because another company unionized, it increases worker condi- tions and wages to appease the threat o — unionization by workers.33 While some studies — ound that unions diminish productivity, economic e — ficiency, and labor market dynamism,34 countervailing studies — ound that unions increase e — ficiency by reducing quit rates and labor turnover, which in turn lowers the cost o — training new
Varieties o --- Coverage: The Anatomy o --- Union Wage E ---
ects in Germany, 66 Indus. & Lab. Rel. Rev. (2013) find both that union wages rise with union density in covered places, but also that higher union density is associated with lower wages in nonunionized firms. Fitzenberger, Kohn, and Lembcke discuss that this finding could either be because o — the downward pressure on wages — rom an increase in labor supply in nonunionized sectors, or because o — a decline in investment when union density increases. 29 On diminishing productivity, economic e — ficiency, and labor market dynamism, see Robert DeFina, Unions, Relative Wages, and Economic E — ficiency, J. Lab. Econ. (1983); on suboptimal deployment o —
labor through restrictive practices, see David Metcal --- Trade Unions and Economic Per --- ormance: The British
Evidence, LSE Quarterly 3 (1989); on limiting innovative and investment activities, see Connolly, Hirsh,
& Hirshey, Union Rent-Seeking, Intangible Capital, and Market Value o --- the Firm, Rev. Economics &
Statistics (1986); Grout, Investment and Wages in the Absence o --- Binding Contracts: A Nash Bargaining
Approach, Econometrica (1984); Hirsch and Link, Labor Union E ---
ects on Innovative Activity, 8 J. Lab. Econ 8 (1987); on unions having negative impact on profitability, see Barry Hirsch, Union Coverage and Profitability Among U.S. Firms, 73 Rev. Economics & Statistics (1991); Bronars, Deere, & Tracy, The E —
ects o — Unions on Firm Behavior: An Empirical Analysis Using Firm-Level Data, 33 Ind. Rel. (1994); Pasquale Laporta and Alexander Jenkins, Unionization and Profitability in the Canadian Manu — acturing Sector, 51 Ind. Rel. (1996); J. Machin and M. Stewart, Trade Unions and Financial Per — ormance, 48 Ox — . Econ. Papers (1996); Richard S. Ruback & Martin B. Zimmerman, Unionization and Profitability: Evidence — rom Capital Markets, 92 J. Pol. Econ. 1134, 1134 (1984). 30 Richard B. Freeman, The E —
ect o — Unionism on Fringe Benefits, 34 Indus. & Lab. Rel. Rev. 489, 509 (1981); Richard B. Freeman, Individual Mobility and Union Voice in the Labor Market, 66 Am. Econ. Rev. 361 (1976). 31 Giuseppe Bertola, Francine Blau, & Lawrence Kahn, Labor Market Institutions and Demographic Employment Patterns, J. Population Econ. (2007); Edward Montgomery, Employment and Unemploy- ment E —
ects o — Unions, J. Lab. Econ. (1989). 32 Lewis, supra note 27; Mellow, supra note 27; Fitzenberger, Karstan Kohn, & Alexander Lembcke, supra note 27. 33 Bruce Western & Jake Rosen — eld, Unions, Norms, and the Rise o — U.S. Wage Inequality, 76 Am. Soc. Rev (2011); Jake Rosen — eld, What Unions No Longer Do (2014). 34 Robert DeFina, Unions, Relative Wages, and Economic E — ficiency, J. Lab. Econ. (1983); David Metcal —
(1989), supra note 28; Connolly, Hirsh, and Hirshey (1986), supra note 28; Grout (1984), supra note 28;
Hirsh & Link (1987), supra note 28.
https://doi.org/10.1017/9781108610070.020 Published online by Cambridge University Press 186 The “Fissured” Workplace
workers and increases productivity.35 In sum, the empirical debate on unions and
productivity is vast, spanning over 60 years and including hundreds o --- studies with
conflicting findings.
In attempting to reconcile some o --- the mixed findings on unions and productivity, The
Economics o --- Trade Unions (2017) provides a comprehensive meta-analysis and sophisti-
cated meta-regressions on the vast literature pertaining to the question o --- unions and
productivity.36 It reveals that --- actors such as industry, country, and institutional context
matter --- or how unionization impacts productivity. The authors find that unionization has
been beneficial in the United States --- or the construction and education industries,37 that
unionization has harmed productivity in the UK regardless o --- industry, and that unions
harm productivity in more regulated labor markets than in less regulated labor markets
(particularly in the manu --- acturing sector).38 In other words, on this last --- actor, less
regulated labor markets can moderate the negative aspects o --- unions.
In analyzing the channels through which unionization impacts productivity, the
authors conclude that productivity is harmed through channels o --- technology and
investment.39 Specifically, they find a moderate negative e ---
ect o — unions on physical capital investment and a large negative impact o — unions on intangible capital (technol- ogy and adoption o — it, stock o — knowledge, and the know-how), concluding that “unions have an adverse association with productivity and productivity growth through the channel o — investment behavior o — firms.”40 In terms o — the labor channel, they conclude that less employee turnover is a mechanism by which unions have been — ound to improve firm per — ormance, though this result is less robust than others given the small number o — studies that have success — ully disaggregated this employee channel.41 As applied to the gig economy, however, reducing employee turnover may have a smaller impact on per — ormance o —
firms. By its very nature, the gig economy is designed to have low costs o --- onboarding and
o ---
boarding employees. There — ore, the improvement that labor unions tend to bring to firm per — ormance through lower employee turnover may have less impact in the gig economy. Overall, the authors find strong evidence that unions negatively impact most firms’ financial per — ormance. This can adversely a —
ect long-term investments and growth, to the extent that the lower financial per — ormance comes — rom the channel o — quasi-rents
rom long-lived assets, as Doucouliagos and Laroche (2009) have — ound.42 However,
35
Freeman (1976) supra note 29; Freeman & Medo ---
, What Do Unions Do? (1984); S.B. Vroman, The Union–Nonunion Wage Di —
erential and Monitoring Costs, 32 Econ. Letters (1990); C. Bauer and J. Lingens, Does Collective Bargaining Restore E — ficiency in a Search Model with Large Firms?, 113 Econ. J. 113 (2013); E. Barth, K. Moene, and F. Willumsen, The Scandinavian Model – An Interpretation, 117 J. Pub. Econ. (2014). 36 Hristos Doucouliagos, Richard Freeman, & Patrice Laroche, The Economics o — Trade Unions: A Study o — a Research Field and Its Findings (2017). 37 Though the authors indicate that studies that include technology in the regressions find less positive e —
ect o — unions and that when studies use physical output or e — ficiency as the “outcome variable” they find larger negative e —
ects. 38 Doucouliagos et al., supra note 35, at 40–71. 39 Id. at 86–109. 40 Id. at 109 41 Id. at 110–129. 42 Doucouliagos & Laroche, Unions and Profits: A Meta Regression Analysis, 48 Ind. Rel (2009).
https://doi.org/10.1017/9781108610070.020 Published online by Cambridge University Press Balancing Flexibility and Rigidity 187
even when the --- ocus is on firm per --- ormance, unions can have a positive impact. The case
or unions is arguably strongest when there is clear lopsided bargaining power in — avor o —
employers. Unions can restore equilibrium and lead to a more e --- ficient outcome. In
more competitive markets, there is more o --- a debate on how beneficial unions can be,
and there are large adverse consequences to consider.
C The On-Demand-Economy Is a Competitive Labor
Market, Not a Monopsony
The conditions that can lead to monopsonistic power do not exist in the twenty-first
century on-demand economy. As discussed above, the classic monopsony occurs where a
small rural town has only one employer or where a --- ew employers collude to depress
wages. This is virtually impossible in an economy that runs on the global Internet. Indeed,
the on-demand economy exists on top o --- the existing traditional economy. More aptly
named, the “gig economy” does not replace traditional work, but provides a competitive
alternative to the nine-to-five.
One o --- the interesting --- eatures o --- the gig economy is that it is not just one industry.
Companies --- rom a variety o --- industries including transportation, housing, --- ood and
dining and retail are all in the gig economy. On-demand plat --- orms like Uber, Ly --- t,
Airbnb, TaskRabbit, Handy, etc. generally do not require extensive on-boarding or
particular pre-requisite skills, so workers can easily enter and exit these plat --- orms – and
they do.
Gig workers typically do not work --- or just one on-demand plat --- orm. Perhaps because
labor entry and exit barriers are so low, many companies in this space within the same
industries and across industries compete --- or similar workers. In the ride-sharing space,
drivers are typically simultaneously working with Uber, Ly --- t, Juno, and Via. For example,
in a brie --- filed on behal --- o --- Ly --- t, a random survey o --- 10,000 Cali --- ornia Ly --- t drivers --- ound
that over hal --- the drivers have driven with another ridesharing company.43 Moreover,
83 percent o --- drivers reported driving with Ly --- t and another ridesharing company in the
same week, and o --- those driving with another ridesharing company in the same week,
75 percent reported they drove with another ride sharing company besides Ly --- t in the
same hour.44 An online survey o --- 1,200 ridesharing drivers, which was not produced on
behal --- o --- any ride sharing company, --- ound that nearly 80 percent o --- Uber drivers have
signed up --- or two or more services with companies such as Ly --- t, Postmates, DoorDash,
UberEats, and Amazon Flex.45
This sort o --- “multi-homing” (where a user --- requently connects to more than one
network)46 is the norm in the gig economy. Workers can sign on up on Handy or on
competing on-demand plat --- orms like Amazon Home Services, TaskRabbit, and Fiverr as
well as with traditional, local management companies that sta ---
cleaners (i.e. those not in
43
Declaration o --- Simona A. Agnolucci in Support o --- Ly --- t Inc.’s Brie --- Regarding Preliminary Approval o ---
Class Action Settlement (Case No. 3:13-cv-04065-VC) at 2.
44
Id. (emphasis added).
45
See “The Ride Sharing Guy,” https://therideshareguy.com/2018-uber-and-ly --- t-driver-survey-results-the-ride
share-guy/.
46
See Jean J. Gabszewicz & Xavier Y. Wauthy, Two-Sided Markets and Price Competition with Multi-
Homing, (CORE discussion Paper 2004/30, May 4, 2004), http://webdoc.sub.gwdg.de/ebook/serien/e/
CORE/dp2004_30.pd --- .
https://doi.org/10.1017/9781108610070.020 Published online by Cambridge University Press 188 The “Fissured” Workplace
on-demand space). Multi-homing indicates competition between on-demand plat --- orms
and between on-demand plat --- orms and traditional bricks-and-mortar employers.
On-demand companies not only contend with multi-homing; they also compete with
each other --- or labor, and with the traditional (non on-demand) companies, too. Handy is
competing with the local companies --- or the same housekeepers and handymen (and --- or
customers); in some markets Uber is competing with taxicabs --- or the same drivers;
Cleanly is competing with local wash-and- --- old service providers; Soothe and Urban
Massage are competing with each other and the local salons and massage spas --- or
massage therapists. Competition is --- ostered where gig workers might switch to other
on-demand plat --- orms or to traditional non-gig --- ull-time employment.
A recent report --- ound that as the traditional labor market has strengthened, this has
narrowed the pool o --- likely gig economy participants – indicating there is competition --- or
workers between traditional and gig economy workers.47 The gig economy is thus anti-
monopsonistic with regards to the traditional economy.
In --- act, by virtue o --- not having a physical location where workers have to “clock-in” to
work, the on-demand economy can ease the instances where large companies in small
towns exercise monopsony power. This is because on-demand economy opportunities
present an alternative option --- or workers without workers having to physically move. For
example, in a monopsonistic “company town,” where workers are tied to a particular
location because moving costs are high or because o ---
amily and community reasons, on- demand companies present alternative work arrangements — or them without workers having to physically move. This will exert pressure on employers in the “company town” to increase wages because they now must compete with on-demand companies. That is, the presence o — on-demand economy opportunities in a particular city — unction as i — a new company has opened up. On-demand plat — orms introduce competition in otherwise monopsonistic labor markets. In this way, the sharing economy lessens the benefits o —
unions in such markets.
Monopsonistic power can arise when workers have a specialized skillset, but this
condition does not describe the sharing economy in general. Gig jobs usually pertain
to driving (Uber, Ly --- t, Juno, etc.), cleaning or other tasks (Handy, TaskRabbit, Cleanly),
dog walking (Wagz), valet services (Luxe), and delivery (Postmates, GrubHub, Cleanly,
InstaCart, etc.). These are typically considered low-skilled labor, and the consensus in the
economics literature is that labor markets in the United States, especially those --- or low- to
medium-skilled workers, are quite competitive.48 There are some companies, such as
Catalant, that create an on-demand --- eature --- or services o --- specialized consultants, but
they also operate in a competitive labor market --- or consultant services, with companies
such as McKinsey, Bain, Boston Consulting Group, and Deloitte as the largest firms in
this space. Since gigs generally do not require specialized labor, and unspecialized labor
generally operates in a competitive market, the gig economy is less likely to be mono-
psonistic or oligopsonistic.
47
Diana Farrell & Fiona Greig, The Online Plat --- orm Economy: Has Growth Peaked?, JP Morgan Inst. 2, 3
(Nov. 2016), www.jpmorganchase.com/corporate/institute/document/jpmc-institute-online-plat --- orm-econ-
brie --- .pd --- .
48
Peter Kuhn, Is Monopsony the Right Way to Model Labor Markets? A Review o --- Alan Manning’s
Monopsony in Motion, 11 Int’l J. Econ. Bus. 369 (2004); Tyler Cowen, Tanner Lecture Comment on
Elizabeth Anderson, Feb. 17, 2015.
https://doi.org/10.1017/9781108610070.020 Published online by Cambridge University Press Balancing Flexibility and Rigidity 189
In discussion o --- monopsonies, there are have been e ---
orts to use job search costs as a source o — monopsony power. The argument is that even i — the geographic or specialized skill set conditions are not present, monopsonistic power can arise when it is di — ficult — or employees to know about and find other jobs.49 In a similar vein, finite turnover and hiring rate, or other aspects o — high costs o — changing a job, are all sources o — potential monopsonistic power. However, none o — these are present in the sharing economy. High turnover rate and high hiring rates are what define many o — these industries. A J.P. Morgan Chase study — ound significantly high churn — or gig economy workers: one in six workers in every given month is new, and three in six gig economy workers exit within one year.50 Contrast this with the Bureau o — Labor Statistics showing that traditional employees’ median length o — time with a current employer is over — our years.51 At Uber alone, Jonathan Hall and Alan Krueger report that there are thousands upon thousands o — new active drivers added each month: in November 2014, there were 32,000 new active drivers added; in December 2014, there were almost 40,000 new active drivers added.52 In total, — rom a base o — near zero in 2012, there were more than 160,000 active drivers by 2015 in the United States.53 While there were no global numbers in this report, the — ormer CEO o — Uber has discussed how Uber adds “hundreds o — thousands” o — drives globally each month.54 Hall and Krueger also report that about 45 percent o — Uber drivers exit a — ter one year o — starting.55 Lastly, another conception o — monopsonistic labor market conditions comes — rom Alan Manning (2003), who argues that the existence o — job search costs in finding employ- ment implies that there is a monopsony even when employers are small relative to the labor market. Several scholars have pointed out problems in Manning’s model and the assumptions upon which it rests.56 However, even this condition would not apply to the twenty-first century economy, and certainly not to working in the sharing economy given the countless examples o — the competition and e —
orts spent among on-demand companies to recruit workers.57 As one example, Uber spent $86.6 million on “driver
49
Alan Manning, Monopsony In Motion: Imper --- ect Competition In Labor Markets (2003).
50
Farrell & Greig, supra note 36.
51
U.S. Bureau o --- Labor Statistics, USDL-16-1867, Employee Tenure In 2016 (2016) www.bls.gov/news
.release/pd --- /tenure.pd --- .
52
Jonathan V. Hall & Alan B. Krueger An Analysis o --- the Labor Market --- or Uber’s Driver-Partners in the
United States, 71 I.R.L. Rev. 705, 707 (2017).
53
Id.
54
Ellen Huet, Uber is Adding Hundreds o --- Thousands o --- New Drivers Every Month, Forbes (Jun. 3, 2015
11:05 PM), www. --- orbes.com/sites/ellenhuet/2015/06/03/uber-adding-hundreds-o --- -thousands-o --- -new-drivers-
every-month/#58375b79655e.
55
Hall & Krueger, supra note 41, at 708.
56
Kuhn supra note 37, at 376, points to a number o --- deficiencies in Manning’s model and illustrates how the
entire monopsony claim regarding job search costs is predicated on unrealistic assumptions about
diminishing returns to scale in recruiting workers. Kuhn also argues that, even i --- these assumptions about
a monopsony might hold in the very short run, the e ---
ect disappears in the medium to long run. He explains that the empirical evidence suggests it is quite unreasonable to claim that any individual firm in a labor market as large as those o — the United States and United Kingdom will have monopsonistic characteristics in the long run. Kuhn (at 376) concludes that the presence o — a monopsony in U.S. and U.K. labor markets is highly unlikely “unless one — ocuses on workers with very specific skill types in very defined geographical areas.” 57 For example, see case study o — how Uber employed controversial tactics to “steal” Ly — t drivers: Casey Lewton, This Is Uber’s Playbook — or Sabotaging Ly — t, The Verge, (accessed Aug. 26, 2014), www.theverge.com/2014/8/26/6067663/this-is-ubers-playbook- — or-sabotaging-ly — t.
https://doi.org/10.1017/9781108610070.020 Published online by Cambridge University Press 190 The “Fissured” Workplace
incentives” in 2014, and $130.1 million driver incentives in the first hal --- o --- 2015. Uber
also o ---
ers extensive sign-up bonuses, and re — erral bonuses – as one extreme example, in Singapore, Uber o —
ered $2,50058 — or a single re — erral.59 Furthermore, Manning’s book was published as Internet job search tools and recruit- ment were taking o —
. In the years since, search costs and — rictions have been reduced, and no credible claims have been made that the trend will be reversed. There — ore, the conditions that give rise to monopsony power are virtually nonexistent in the on-demand space. The discussion here does not indicate that workers on the on- demand plat — orm — ace no problems. They do — ace problems, but these are a — undamen- tally di —
erent set o — problems and conditions than what labor unions are set out to solve. Indeed, the gig economy solves some traditional problems in the economy. By introdu- cing competition, on-demand plat — orms obviate the need — or unions. There — ore, unions may not only be unnecessary, but actually harm — ul, in gig economy.
D The On-Demand Economy Requires Flexibility
Unions may provide many beneficial work arrangements, though they can come at the
cost o --- rigidi --- ying labor markets. But this rigidity is antithetical to the flexible on-demand
gig economy. Gig jobs are by nature flexible, and they provide a competitive alternative
to nine-to-five work that also imposes competitive pressure on otherwise monopsonistic
labor markets. In other words, by its very definition, the on-demand economy requires
flexibility.
The need --- or flexible labor supply in the on-demand economy is best illustrated by its
price system. The most unique --- eature o --- on-demand business models is that the
companies generally utilize an algorithm that matches supply and demand. To do this,
companies require a flexible labor supply because when demand is high, they need
quantity supplied to instantaneously increase to meet the greater demand. In markets and
in the algorithm o --- the business models, this is done through the price system. When
demand is high, the algorithm generates a higher price per unit to incentivize an increase
in the quantity supplied o --- the good. This strategy is called “dynamic pricing.”60
Take Uber --- or example. Within the algorithm, there is a built-in incentive structure
that increases the cost to riders and the pay to drivers (i.e. surge pricing) to encourage
drivers to provide rides when that service is needed most, and this instantaneous supply-
demand matching requires a flexible labor model (along with a nonstandard wage and
compensation structure). That is the core o --- Uber’s business model, and it is --- unctionally
di ---
erent to traditional businesses, such as taxicabs, even i — taxicabs use an app (as they do in NYC). In — act, this di —
erence is precisely why economists Judd Cramer and Alan Krueger
ound that the capacity utilization rate – measured by “the — raction o — time a driver has a
are-paying passenger in the car while he or she is working, and by the share o — total miles
58
Driver Re --- erral Program in Singapore, Uber, www.uber.com/en-SG/drive/resources/re --- errals/ (accessed
Jun. 13, 2018).
59
This is not to say that workers have always benefited --- rom the decisions to sign-up with companies such as
Uber and Ly --- t, rather that the presence o --- these programs indicates there is competition and e ---
ort directed to recruit workers. 60 Jonathan Hall & Chris Nosko, Dynamic Labor Supply in the Sharing Economy, (U. Chicago Working Paper, 2016), www.sole-jole.org/16433.pd — .
https://doi.org/10.1017/9781108610070.020 Published online by Cambridge University Press Balancing Flexibility and Rigidity 191
that drivers log in which a passenger is in their car”61 – is significantly higher --- or UberX
drivers than --- or taxi drivers.62 That is, “UberX drivers spend a significantly higher --- raction
o --- their time, and drive a substantially higher share o --- miles, with a passenger in their car
than do taxi drivers.”63 The paper concludes that part o --- the reason Uber drivers are able
to do so is because Uber’s flexible labor model and surge pricing more closely match
supply with demand throughout the day.64 Thus, even “cabs with an app” are still
unctionally di —
erent than Uber (or Ly — t) because cabs and other traditional — or-vehicles have workers drive around during a given period o — time in a given location, waiting — or an app or dispatcher to connect them. Furthermore, in another paper studying how drivers respond to surge pricing, M. Keith Chen and Michael Sheldon conclude that: “Uber partners both drive at times with higher demand — or rides, and dynamically extend their sessions when surge pricing raises earnings.”65 Uber drivers are able to respond to the surge pricing because their smartphone inter — ace “allows them to know current prices and session statistics like cumulative earnings, time, and trips.”66 Competitor companies such as Ly — t and Juno also include this — eature. Companies outside o — ride-sharing, such as Handy, also have a “peak pricing” algorithm that incentivizes more Handy cleaners to supply their services at popular times.67 Economists Jonathan Hall and Chris Nosko provide another investi- gation o — the impact o — Uber’s dynamic pricing strategy. They find that when demand increases, dynamic pricing allows — or quantity-supplied to meet the demand, and when they model a hypothetical case with no surge pricing, this matching e —
ect no longer holds. Hall and Nosko then quanti — y the resulting ine — ficiencies in a world with no surge pricing.68
E Union Rigidity Is at Odds with On-Demand Flexibility
Because the on-demand economy requires a flexible labor supply to match supply and
demand, union rules about when and where work may occur are at odds in this space
since they impose greater rigidity on labor. Moreover, a key reason workers participate in
the on-demand economy is to set their own schedules, evidencing that rigidity is opposed
to their interests.
1 On-Demand Is Not Nine-to-Five
Consider the paid vacation time that most CBAs require. On-demand companies do not
require workers to work at any time. On-demand workers are generally --- ree to take
61
Judd Cramer & Alan B. Krueger, Disruptive Change in the Taxi Business: The Case o --- Uber, 106
Am. Econ. Rev. 177, 177 (2016).
62
Id. at 179.
63
Id. at 177.
64
Id.
65
M. Keith Chen & Michael Sheldon, Dynamic Pricing in a Labor Market: Surge Pricing and Flexible Work
on the Uber Plat --- orm 13, (UCLA Anderson Working Paper, 2015), www.anderson.ucla.edu/ --- aculty/keith
.chen/papers/SurgeAndFlexibleWork_WorkingPaper.pd --- .
66
Id. at 15.
67
What Is Peak Pricing?, Handy, https://help.handy.com/hc/en-us/articles/219851127-What-is-Peak-Pricing-,
(last visited Jun. 13, 2018).
68
Hall & Nosko, supra note 48.
https://doi.org/10.1017/9781108610070.020 Published online by Cambridge University Press 192 The “Fissured” Workplace
50 days o --- vacation, to never work a holiday, to never work overtime, or to work every
Sunday. How would a rule --- or time-and-a-hal --- on Sundays distort Uber’s surge pricing
model? I --- a CBA were to require that on-demand companies --- orbid workers --- rom working
on vacation days, then this would deteriorate the on-demand --- eature o --- on-demand
companies. The main --- eature o --- the industry is that consumers can get a ride, cleaning
services, --- ood, laundry services, or dog walking at any time they need it; they get it “on
demand.” A union rule --- orbidding workers to voluntarily work on holidays would harm
the essence o --- the on-demand services. And more generally, this type o --- rule is not
applicable to cases where these on-demand companies do not set requirements --- or their
workers on how many hours or how many days a week they need to supply their services.
This rule is a better fit in a context where employees were --- orced to work holidays, --- or
long hours, and with little vacation time o ---
. O — course, it is possible that gig economy “unions” could have an entirely di —
erent
ocus — rom traditional unions. For example, the Independent Driver’s Guild, a quasi- union — or Uber drivers, has organized campaigns to permit tipping and the ability to opt out o — Uber Pool (a service that allows shared rides with other passengers). But while these ambitions are unequally tied to Uber’s business model, they essentially amount to the traditional priority o — seeking higher pay. Other IDG e —
orts more clearly reflect agenda items similar to traditional union priorities, such as trying to get health care coverage — or all drivers. Moreover, traditional unions such the New York Taxi Workers Alliance and the Amalgamated Transit Union are vying — or the opportunity to represent Uber drivers alongside their representation o — workers in the traditional economy.69 There — ore, it seems that gig-economy unions can be analyzed as i — they would seek collective bargaining agreements similar to those sought by unions in traditional economies. The 8-hour workday and 40-hour workweek do not make much sense because on- demand workers have the — reedom to work any hours they like, whenever they want to work them. They can take time o —
or long periods o — time. In — act, most drivers working — or Uber (85 percent) work less than thirty-five hours a week — or Uber.70 And according to a Benenson Strategy Group survey — or Uber drivers, 69 percent o — drivers have other — ull-time or part-time work and hal — o — Uber drivers use the plat — orm — or less than ten hours a week.71 Data — rom the recent Ly — t Cali — ornia lawsuit also indicate that less than 1 percent o — Cali — ornia Ly — t drivers (755 out o — 150,000) worked thirty or more hours in at least hal — o — the weeks they drove with Ly — t.72 And, over 100,000 people (more than two-thirds o — the class members) “have driven less than sixty hours in total
or Ly — t.”73
69
Will Bredderman, The Other Uber Fight: Unions Brawl to Rep Drivers, Crain’s New York Business
(Jun. 14, 2018).
70
Jonathan v. Hall & Alan B. Krueger, An Analysis o --- the Labor Market --- or Uber’s Driver-Partners in the
United States 18 (Princeton U. Working Paper No. 22843, 2015).
71
Uber: The Driver Roadmap 2.0, Benensen Strategy Group, www.bsgco.com/insights/uber-the-driver-road
map, (last visited Jun. 14, 2018).
72
Cotter v. Ly --- t Inc., 176 F. Supp. 3d 930, 939 (N. D. Cal. 2016) (Order Den. Mot. --- or Prelim. Approval o ---
Class Action Settlement).
73
Id.
https://doi.org/10.1017/9781108610070.020 Published online by Cambridge University Press Balancing Flexibility and Rigidity 193
2 On-Demand Workers Pre --- er Flexibility
Moreover, one o --- the main reasons workers join the on-demand plat --- orm is precisely
because they desire the flexibility o --- the li --- estyle.74 They may desire to escape the rigidity
o --- traditional labor arrangements. For example, survey data o --- Uber drivers shows that
73 percent o --- the drivers said they would rather have a job where they choose their own
schedule and where they are their own boss, rather than the traditional model o --- a steady
nine-to-five job with some benefits and a set salary.75 Furthermore, 63 percent o --- Uber
drivers said they specifically use Uber to have more flexibility so they can balance work
and --- amily.76 Drivers --- or Ly --- t look similar. A survey o --- 3,100 Ly --- t drivers --- ound that
82 percent o --- Ly --- t drivers agreed or strongly agreed with the statement “I like being an
independent contractor” and 99 percent o --- Ly --- t drivers agreed with the statement “I like
to choose when I work.”
Furthermore, women particularly benefit --- rom the flexibility o --- the on-demand econ-
omy.77 This type o --- li --- estyle is especially important --- or women, who o --- ten have --- amily
care responsibilities, which they need to balance within the confines o --- a traditional nine-
to-five workday. In a survey o --- 2,000 US-based --- emale gig workers, the study --- ound that
32 percent o --- women indicated that they le --- t their --- ull-time jobs --- or the gig economy
because they wanted more flexibility, and another 28 percent indicated that they needed
more time to care --- or a child, parent, or relative.78 When women were asked to compare
the top benefits o --- being a gig worker, 96 percent placed “flexibility” as the primary
benefit.79 The qualitative answers in the report also indicate the --- reedom women see in
this type o --- work and importance in having opportunities to earn income outside o ---
traditional workplace.
3 On-Demand Strikes
One o --- the main “sticks” unions carry is the ability to strike and put pressure on
employers to agree to certain terms or to change working conditions and pay. Strikes
impose large costs on employers as they halt production. The strike pressures employers
to acquiesce to union demands. But strikes may be less e ---
ective on the Internet than in person. First, striking in the on-demand space is o —
set by the pricing algorithm: when many workers leave the plat — orm, prices and wages increase, attracting more workers until an equilibrium is reached. Second, strikes are more e —
ective in person, where striking
74
Elka Torpey & Andrew Hogan, Working in a Gig Economy, Career Outlook, U.S. Bureau o --- Labor
Statistics, www.bls.gov/careeroutlook/2016/article/what-is-the-gig-economy.htm.
75
See supra note 57 (“I --- both were available to you, at this point in your li --- e, would you rather have a steady
9-to-5 job with some benefits and a set salary or a job where you choose your own schedule and be your
own boss?”).
76
Id.
77
Natasha Singer, In the Sharing Economy, Workers Find Both Freedom and Uncertainty, N.Y. Times
(Aug. 16, 2014) www.nytimes.com/2014/08/17/technology/in-the-sharing-economy-workers-find-both- --- ree
dom-and-uncertainty.html; Paul Merrion & Fareeha Ali, Making Inroads: Women Cabbies on the Rise,
Crain’s Chicago Business, (Sept. 27, 2014), www.chicagobusiness.com/article/20140927/ISSUE01/
309279976/making-inroads-women-cabbies-on-the-rise.
78
Hyperwallet, The Future o --- Gig Work is Female, Hyperwallet Ecommerce Marketplaces, 1, 9, (May 3,
2017), www.hyperwallet.com/app/uploads/HW_The_Future_o --- _Gig_Work_is_Female.pd --- .
79
Id at 15.
https://doi.org/10.1017/9781108610070.020 Published online by Cambridge University Press 194 The “Fissured” Workplace
and picketing is visible. But the on-demand economy --- unctions via apps and web sites,
not at physical locations. There --- ore, unions’ most power --- ul stick, striking, can be
masticated by decentralized work, a hallmark o --- the gig economy.
Again returning to Uber as an example, i --- some workers went on strike, that would
incentivize other workers to supply more hours driving. When there is a decrease in the
supply o --- workers ( --- rom those striking), this would increase the price per ride (i.e. surge
pricing), and thus encourage other drivers to go out and work. Theoretically, i --- surge-
pricing is prolonged, this would incentivize new drivers to enter the plat --- orm.
Furthermore, as Judge Richard Posner points out, one o --- the unique --- eatures about
strikes is that the --- orm o --- picketing deters other employees --- rom working during strikes.80
In a picket strike, employees protest outside o --- a physical location. Everyone can see who
is striking and who is not. This pressures other colleagues to strike as well because, once
the striking workers return, they will be working side to side with these same people who
may have decided not to strike. This social mechanism endogenous to picketing allows
or more success — ul strikes. History confirms that picketing outside a physical location can encourage con — ormity among worker-strikers. However, picketing is absent in the on-demand economy because there is no physical location workers must be, which weakens this traditional sanctioning mechanism. In other words, whether workers are striking or not is not “public” in — ormation, since the striking picketers do not see other non-striking colleagues — ace-to- — ace when they’re entering the building to work, as would be the case with a traditional job.81 This anonymity reduces the pressure — or workers in the on-demand space to strike, and they can continue going to work while a “strike” is happening. There — ore, the strike is weakened as an e —
ective tool — or change since the firms retains a greater portion o — their contractors still working — or them. Thus, taken together, traditional union agreements not only would harm the on- demand economy, but also weaken the en — orceability and strength o — these rules in on- demand economy space.
F Conclusion
Unions are a power --- ul --- orce --- or workers’ rights where an overly power --- ul employer could
otherwise --- orce workers to work around the clock, --- or too little pay, or in unsa --- e
conditions. Unions can also empower workers to exercise their opinions vis-à-vis man-
agement. At the same time, power --- ul unions that restrict competition and use o --- non-
union labor can cause labor markets to become more rigid and harm labor mobility and
flexibility. Since unions can be help --- ul or harm --- ul, depending on various market condi-
tions, it is important to consider the market conditions o --- the gig economy when thinking
about rights --- or such workers to unionize.
The nature o --- work in the gig economy is --- undamentally di ---
erent — rom work in the traditional economy, and many o — the problems that traditional unions sought to solve do not exist in the same nature as in the gig economy. For example, — or most o — the gig economy jobs, work is at-will and on-demand, and concerns about guaranteed eight-hour days or mandatory overtime hours are inapposite. Elaborate protocols — or hiring and
80
Richard Posner, Some Economics o --- Labor Law, 51 U. Chi. L. Rev. 988, 1005–1006 (1984).
81
Though one area where there could be picketing is by airport waiting areas.
https://doi.org/10.1017/9781108610070.020 Published online by Cambridge University Press Balancing Flexibility and Rigidity 195
firing that may protect skilled workers may instead inter --- ere with the low barriers o --- entry
and exit --- or gig economy work, where it is valuable to easily “sign-up” or “deactivate”
rom the various di —
erent plat — orms and work with many di —
erent plat — orms at the same time.82 Concerns about wage rates are less in gig economy jobs where workers set their own “wages” and rates to charge — or the various services they provide (i.e. TaskRabbit, Thumbtack); indeed, a union requiring all “Taskers” to receive the same pay — or a given service would reduce the incentives to gain elite status on those plat — orms. Furthermore, gig economy workers are more heterogenous than traditional workers in that while some o — them might be working — ull-time — or one plat — orm, others are multi-homing on various plat — orms or using a plat — orm — or supplemental income, so homogenous and mandatory benefits (which come at the expense o — higher pay) — or all gig workers would make it hard
or non-traditional workers to benefit — rom the gig economy. There — ore, the traditional solution o — unionization and the traditional aspects o — CBAs may not be applicable as the way — orward — or this new economy. This does not mean that gig economy workers — ace per — ect conditions. For example, some gig economy workers do not have an e —
ective channel or avenue to express dissatis — action, or — ace problems o — “instant deactivation” techniques,83 or lack adequate knowledge and up — ront in — ormation about the cost-side o — their gig work (i.e. some Uber drivers have expressed that they did not — ully understand the cost/expense o — driving and car amortization). O — course, gig economy workers also do not receive health and retirement benefits as most traditional employees do. These sets o — problems require a set o — solutions that should likely come outside o —
traditional unionization techniques84 and traditional CBA agreements because the gig
economy is predicated on having a flexible labor supply that adjusts to real-time demand.
Such adjustment is not possible where union collective bargaining agreements restrict
wage changes, hiring and firing, maximum and minimum hours worked, layo ---
s, promo- tions, etc. It is certainly possible that CBAs — or gig economy workers will look — undamen- tally di —
erent than traditional CBAs. However, to the extent that unionization and CBAs in the gig economy operate on a similar — ramework o — creating rigid rules that o — ten involve restricting prices or hours-worked and restricting labor mobility via hiring and firing conditions, then it will likely threaten the existence o — the gig economy, which needs flexibility o — labor supply. This can harm the value o — the on-demand economy and reduce the opportunities — or people to earn a living or engage in a “side-hustle” on these plat — orms. Solutions in this new economy should look beyond traditional unionization techniques and should be open to more innovative approaches that can minimize harm while still addressing the needs o — gig economy workers.
82
However, there was a problem that in the early days o --- Uber, drivers signed up --- or Uber under potentially
alse advertising claims, and bought new cars specifically — or a job with Uber. These investments and high- fixed costs did create a barrier to exit — or Uber drivers. 83 In some cases, this is warranted, as sexual harassment claims instantly ‘deactivate’ a driver to prevent this particular driver to be matched with another rider. 84 For example, there is discussion o — re — orming labor law to move toward portable benefits solutions that are not tied to employment. And, interestingly, portable-benefits are already beginning to arise in the market- place place — or retirement benefits. Companies such as Honest Dollar and Nestana are providing competitive individualized 401K benefits plans to contractors and — reelancers.
https://doi.org/10.1017/9781108610070.020 Published online by Cambridge University Press https://doi.org/10.1017/9781108610070.020 Published online by Cambridge University Press