Lowering the Stakes of the Employment Contract

Fordham Law School Fordham Law School

FLASH: The Fordham Law Archive of Scholarship and History FLASH: The Fordham Law Archive of Scholarship and History

Faculty Scholarship

2022

Lowering the Stakes of the Employment Contract Lowering the Stakes of the Employment Contract

Aditi Bagchi

Fordham University School of Law

, abagchi@law.fordham.edu

Follow this and additional works at: https://ir.lawnet.fordham.edu/faculty_scholarship

Part of the Contracts Commons, and the Labor and Employment Law Commons

Recommended Citation Recommended Citation

Aditi Bagchi,

Lowering the Stakes of the Employment Contract

, 102 B.U. L. Rev. 1185 (2022)

Available at: https://ir.lawnet.fordham.edu/faculty_scholarship/1199

This Article is brought to you for free and open access by FLASH: The Fordham Law Archive of Scholarship and

History. It has been accepted for inclusion in Faculty Scholarship by an authorized administrator of FLASH: The

Fordham Law Archive of Scholarship and History. For more information, please contact

tmelnick@law.fordham.edu.

1185

LOWERING THE STAKES OF THE

EMPLOYMENT CONTRACT

ADITI BAGCHI

ABSTRACT

Every country has to make hard choices about the distribution of entitlements.

But employers control the entitlements that individual Americans enjoy to a far

greater extent than those in other rich democracies. In this Essay, I argue that,

in the absence of the political consensus necessary to deliver state solutions to

political questions, employers here are assigned an exaggerated role in

employees’ lives. Government incentives for and directives to employers have

become a strategy of political deflection. The effect has been to raise the stakes

of employment well beyond the scope of those terms and conditions that relate

to attracting and extracting productive labor.

I consider three examples of political deflection to employers. Employers

control employees’ access to healthcare and free speech and, more recently,

have imposed public health measures such as vaccination mandates. Employers’

rules are backed by the threat of termination. This Essay argues that relying on

contractual authority rather than political authority to distribute social rights

and obligations in this way undermines political legitimacy. Governments might

see the private sector as a relatively apolitical refuge through which public

policies can be pursued stably over the course of political shifts. But removing

questions about who gets and owes what from politics and subjecting employees

to raw power, as we do when we raise the stakes of employment and lower the

stakes of politics, detracts from the rightful function of politics to mediate

disagreements among us.

Professor of Law, Fordham University School of Law. Many thanks to Rory Van Loo

for inviting me to participate in this Symposium and for comments received on an initial

presentation of this Essay at the accompanying conference.

1186 BOSTON UNIVERSITY LAW REVIEW [Vol. 102:1185

CONTENTS

INTRODUCTION ............................................................................................. 1187

I. VARIETY IN EMPLOYMENT SCOPE ..................................................... 1189

II. ASSIGNING PUBLIC FUNCTIONS TO EMPLOYERS ............................... 1195

A. Off-site Speech ........................................................................... 1195

B. Healthcare ................................................................................. 1202

C. Vaccine Mandates ..................................................................... 1207

III. DISTINGUISHING CONTRACTUAL AND POLITICAL AUTHORITY ........ 1209

A. Who Gets What .......................................................................... 1209

B. Who Decides .............................................................................. 1211

CONCLUSION ................................................................................................. 1213

2022] LOWERING THE STAKES OF EMPLOYMENT CONTRACT 1187

INTRODUCTION

As a child, one does not grasp why it matters which job you will end up with.

Children tend to talk a lot about what they will do and who they will be, and they

are encouraged to think about what they like or whom they admire. Only decades

later, if ever, are they encouraged to think about starting salaries, career ladders,

and job structure. As luck would have it, our early naïveté may be

inconsequential: the choices that we make tend not to matter as much as external

factors outside of our control. Although the factors that determine how we are

sorted into different jobs are not those that animate childhood whimsy, we were

right to intuit all along that the kind of job we end up with matters enormously.

Most obviously, the job one has—together with the job of one’s spouse—

drives the amount of resources available to a person to pursue her life projects.

Will she live paycheck to paycheck, eating better at the beginning of the month

than toward the end of it, or will she have disposable income to spend on luxury

goods and services? Will she travel the world or feel trapped by her zip code?

How many children can she afford and on what life path will she set them?

Some of her income may buy leisure time. The working hours associated with

her job will also directly determine the quantity of her leisure time. Employment

will determine stability and security, social status, potentially self-respect, and,

of course, how she experiences most of her waking hours as an adult.

Compensation and status are variable features of employment in every variety

of democratic capitalism.

Even security, leisure, and self-respect, though they

may vary less in other wealthy countries, are rarely uniform across the labor

market.

To the extent that these goods depend on employment to a greater

degree in the United States, this appears to be the result of negative rather than

affirmative public policies.

Neither state nor federal governments set out to

delegate to employers the task of distributing anxiety or prestige; the distribution

of those goods and burdens is incidental to the system we use for allocating

social resources like labor, capital, land, and other property. Most people in the

United States are persuaded that relying on the marketplace to allocate such

resources benefits us overall, even though the market rewards people differently

for their contributions.

See, e.g., David Rueda & Jonas Pontusson, Wage Inequality and Varieties of Capitalism,

52 WORLD POL. 350, 350-51 (2000) (noting increases in distribution of wages in capitalist

economies).

Omer Moav & Ofer Cohen, Who Needs Job Security?, AZURE, Autumn 2008, at 1, 8

(noting flexibility of U.S. labor market is due, in part, to variable wages—lack of extensive

income protection—and lack of uniform severance pay and job security).

Id. at 8 (stating U.S. employees do not have extensive income or welfare protection).

Hannah Hartig, Stark Partisan Divisions in Americans’ Views of ‘Socialism,’

‘Capitalism,’ PEW RSCH. CTR. (June 25, 2019), https://www.pewresearch.org/fact-

tank/2019/06/25/stark-partisan-divisions-in-americans-views-of-socialism-capitalism/

[https://perma.cc/XJH4-QCS5] (observing 65% of Americans have positive impression of

capitalism, wherein people rely on market for resources instead of government, in contrast to

42% for socialism).

1188 BOSTON UNIVERSITY LAW REVIEW [Vol. 102:1185

Remarkably, though, there are some basic social entitlements that the federal

government has intentionally delegated to private employers, even though

employers themselves might not otherwise have sought to control them.

That is, we have not simply accepted that people will be paid more or less,

enjoy greater or weaker job security, work harder or less hard, depending on

their labor market power. We have effectively upped the ante and expanded the

set of life consequences associated with labor market power beyond what a

market-based economy in principle entails by way of tax incentives and other

legal and social pressure on employers.

Without aiming to be exhaustive, I will discuss two disparate entitlements:

speech rights and healthcare. In the United States today, an individual’s ability

to speak and write freely and her ability to access healthcare are both critically

dependent on her employer.

We have not and probably could not agree on a

more direct government role for controlling access to speech and healthcare.

light of social divisions, ours has been a deliberate public policy of lowering the

stakes of political discourse in favor of raising the stakes of the employment

contract.

More recently, we have seen this go still further, beyond defining individual

entitlements and into the context of defining individual responsibility. In the face

of significant public health pressures, city, state, and federal governments have

sought to require individuals to get vaccinated against COVID-19.

But they

have not dared to impose a vaccine requirement on citizens qua citizens. Instead,

they have deflected political responsibility for mandates by delegating to

employers enforcement of a vaccine mandate. It is clear that there is no appetite

for a direct mandate on American residents and that, one way or the other,

management of public health will be delegated to employers to a considerable

degree.

See infra Part II.

See, e.g., Bradley Jones, Increasing Share of Americans Favor a Single Government

Program to Provide Health Care Coverage, PEW RSCH. CTR. (Sept. 29, 2020),

https://www.pewresearch.org/fact-tank/2020/09/29/increasing-share-of-americans-favor-a-

single-government-program-to-provide-health-care-coverage/ [https://perma.cc/H5GF-

RYPQ] (noting that 63% of U.S. adults who say government has responsibility to provide

health coverage for all, 36% say it should be via single national government program, and

26% say it should be provided through mix of private insurance companies and government

programs).

See, e.g., Exec. Order No. 14,043, 86 Fed. Reg. 50,989 (Sept. 9, 2021) (requiring

COVID-19 vaccination for federal government employees and contractors); COVID-19

Vaccination and Testing; Emergency Temporary Standard, 86 Fed. Reg. 61,402 (Nov. 5,

2021) (requiring employers of over 100 workers to develop, implement, and enforce

mandatory COVID-19 vaccination or testing policy), withdrawn, 87 Fed. Reg. 3,928 (Jan. 16,

2022) (to be codified at 29 C.F.R. pt. 1910); Erin Tiernan, Boston Mayor Michelle Wu

Announces Vaccine Mandate amid Jeers, BOS. HERALD (Dec. 23, 2021, 9:10 PM),

https://www.bostonherald.com/2021/12/20/boston-mayor-michelle-wu-announces-vaccine-

mandate-for-restaurants-gyms-sports-venues/ (discussing vaccine mandates of Boston, New

York City, and Los Angeles).

2022] LOWERING THE STAKES OF EMPLOYMENT CONTRACT 1189

The effect of using employer authority to compensate for a weak state is that

employment, more so than ever or elsewhere, is the site at which one’s

entitlements and obligations are set. And these entitlements and obligations are

set and controlled by employers, with government incentives and directives in

the background; they are imposed on employees, backed by the threat of

termination.

I will argue that relying on contractual authority rather than

political authority to distribute social rights and obligations beyond

compensation and other basic terms and conditions of employment undermines

political legitimacy even when it is intended to protect it. Governments might

see the private sector as a relatively apolitical refuge through which public

policies can be pursued stably over the course of political shifts. But removing

questions about who gets and owes what from politics and subjecting them to

raw power, as we do when we raise the stakes of employment and lower the

stakes of politics, detracts from the rightful function of politics to mediate

disputes among us about how we should live together.

My argument will proceed in three parts. In Part I, I will review the concept

of decommodification developed in the literature on comparative political

economy and propose an analogue concept about “employment stakes” for local

use that is not attached to Marxist theory. In Part II, I will show how employment

stakes are singularly high in the United States by reviewing the examples of

speech, healthcare, and vaccination requirements. In Part III, I will argue that

relying on contractual authority rather than political authority to decide

individuals’ rights and obligations is a perverse strategy that ultimately

undermines the quality of our democracy.

I. VARIETY IN EMPLOYMENT SCOPE

In 1990, Gøsta Esping-Andersen published his now classic study of welfare

states in the twentieth century, The Three Worlds of Welfare Capitalism.

Despite over two decades of debate about whether his typology of capitalist

regimes accurately or most productively captures the range of states that we

observe, Esping-Andersen’s three broad regime types remain an intuitive and

powerful lens through which to make sense of institutional diversity.

See, e.g., More than 150 Houston Methodist Hospital System Workers Fired or Quit After

Refusing to Get COVID-19 Vaccine, CBS NEWS (June 23, 2021, 5:01 AM),

https://www.cbsnews.com/news/houston-methodist-hospital-system-workers-fired-quit-

covid-19-vaccine/ [https://perma.cc/B3DT-N3EJ] (noting Houston Methodist Hospital

required employees to be completely vaccinated by June 7, 2021, and suspended 178

employees for two weeks without pay after they did not comply).

GØSTA ESPING-ANDERSEN, THE THREE WORLDS OF WELFARE CAPITALISM (1990)

(recasting methodology and concepts of political economy to evaluate welfare state).

See Wil A. Arts & John Gelissen, Models of the Welfare State, in THE OXFORD

HANDBOOK OF THE WELFARE STATE 569, 571-72 (Francis G. Castles, Stephan Leibfried, Jane

Lewis, Herbert Obinger & Christopher Pierson eds., 2010) (discussing widespread comment

on Esping-Andersen’s work spanning twenty years). While critics have offered modifications,

1190 BOSTON UNIVERSITY LAW REVIEW [Vol. 102:1185

Esping-Andersen proposed three ideal types of welfare regimes: liberal (or

neo-liberal or libertarian states, like the United States), conservative (or neo-

corporatist states, like Germany), and social-democratic states (most famously,

Sweden).

These states vary on many subtle dimensions,

and of course, no

one state conforms perfectly to its “type”;

I cannot do credit to his nuanced

study here. Roughly, liberal states are characterized by minimalist, means-tested

welfare assistance and individualized labor markets in which employment

contracts are set at the firm level, often without any labor union involvement.

Conservative regimes offer assistance that reinforces the family as the first unit

of social relief, as well as traditional gender roles within it.

Their social

insurance programs build on neo-corporatist labor market institutions and also

reinforce occupational status distinctions across the labor market through

segmented collective bargaining.

Finally, social-democratic regimes, such as

those in Scandinavia, offer universal and generous welfare schemes sponsored

by strong states that aim to decommodify citizens and reduce social

stratification.

Decommodification “occurs when a service is rendered as a

matter of right, and when a person can maintain a livelihood without reliance on

the market.”

Social-democratic states are associated with cross-class labor

confederations that coordinate across the economy.

Esping-Andersen is

Danish, but scholars working in all three regime types that have used his

typology tend systematically to reserve their accolades for the social-democratic

model.

some friendly and others critical, no competitive typology has displaced Esping-Andersen’s

as the dominant one. See, e.g., id. at 571-72, 580-83.

ESPING-ANDERSEN, supra note 9, at 26-28 (noting that the three types of regimes have

differentiated expectations).

Id. at 26 (recognizing there are qualitatively different arrangements between state,

market, and family across social rights and welfare-state stratification).

Id. at 28.

Id. at 42-43.

Id. at 27 (stating that corporatist regimes are typically shaped by traditional familyhood

and family benefits encourage motherhood).

Id. at 24 (recognizing neo-corporatist markets have distinct programs for different class

and status groups, each with its own set of distinct rights and privileges designed to

complement their “appropriate” station in life).

Id. at 27-28.

Esping-Andersen, supra note 9, at 21-22.

Arts & Gelissen, supra note 10, at 572-77.

See, e.g., ROBERT E. GOODIN, BRUCE HEADEY, RUUD MUFFELS & HENK-JAN DIRVEN,

THE REAL WORLDS OF WELFARE CAPITALISM 261-62 (1999) (“[O]ur analysis, however,

has . . . single[d] out the social democratic welfare strategy as strictly dominant over both the

others.”); Duane Swank, Social Democratic Welfare States in a Global Economy:

Scandinavia in Comparative Perspective, in GLOBALIZATION, EUROPEANIZATION AND THE

END OF SCANDINAVIAN SOCIAL DEMOCRACY? 85, 89, 95 (Robert Geyer, Christine Ingebritsen

& Jonathon W. Moses eds., 2000) (“[P]ublic provision of social services, including health and

employment services, is more developed in the Nordic model than in other welfare states.”).

2022] LOWERING THE STAKES OF EMPLOYMENT CONTRACT 1191

While most critiques of the original typology quibble with the number of

regime types or the labeling of particular countries,

the most potent critique

came from Clare Bambra and others, arguing that Esping-Andersen had

concentrated on the effects of a regime only on the male worker, neglecting the

extent to which “welfare state regimes[] facilitate female autonomy and

economic independence from the family.”

In later work, Esping-Andersen

largely conceded this point and attempted to incorporate the concept of

“defamilialization” by broadening his inquiry into the range of means by which

individuals manage social risks.

One of Esping-Andersen’s main insights is that capitalist countries with

developed welfare states do not all aim to achieve the same objectives, and in

fact, they do not produce the same results.

Most importantly, he emphasized

variation on the matter of whether citizens have rights by virtue of their

citizenship to social services and benefits or, alternatively, whether they must

rely on the labor market to secure basic necessities.

This is the question of

decommodification.

He also differentiated countries on the matter of where

individuals look for security in the face of risk: the state, the market, or family.

Finally, he evaluated states for their degree of social integration, or alternatively,

stratification by income or other status.

See, e.g., Peter A. Hall & David Soskice, Introduction to VARIETIES OF CAPITALISM: THE

INSTITUTIONAL FOUNDATIONS OF COMPARATIVE ADVANTAGE 1, 8 (Peter A. Hall & David

Soskice eds., 2001) (offering new two-regime typology); John Hudson & Stefan Kühner,

Analyzing the Productive and Protective Dimensions of Welfare: Looking Beyond the OECD,

46 SOC. POL’Y & ADMIN. 35, 38-39, 42-43 (2012) (contesting regime criteria and extending

application outside Organisation for Economic Co-operation and Development); Francis G.

Castles & Deborah Mitchell, Worlds of Welfare and Families of Nations, in FAMILIES OF

NATIONS: PATTERNS OF PUBLIC POLICY IN WESTERN DEMOCRACIES 93, 106 (Francis G. Castles

ed., 1993) (disaggregating liberal type).

Clare Bambra, Defamilisation and Welfare State Regimes: A Cluster Analysis, 16 INT’L

J. SOC. WELFARE 326, 327 (2007).

See GØSTA ESPING-ANDERSEN, SOCIAL FOUNDATIONS OF POSTINDUSTRIAL ECONOMIES

45-46 (1999) (noting that, for example, while large proportion of women are precommodified

in that their welfare derives from being in family, female independence necessitates

defamilializing welfare obligations).

ESPING-ANDERSEN, supra note 9, at 22.

Id. at 21 (“[I]f social rights are given the legal and practical status of property rights, if

they are inviolable, and if they are granted on the basis of citizenship rather than performance,

they will entail a de-commodification of the status of individuals vis-à-vis the market.”).

Id. at 21-22; see also Gøsta Esping-Andersen, Multi-Dimensional Decommodiﬁcation:

A Reply to Graham Room, 28 POL’Y & POL. 353, 353 (2000) (referring to decommodification

“a citizen’s relative independence from pure market forces”).

ESPING-ANDERSEN, supra note 9, at 26-28.

Id. (noting liberal regimes result in class stratification that is a blend of relative equality

of poverty among those receiving welfare and market-differentiated welfare among the rest,

conservative regimes emphasize upholding status differentials, and social-democratic states

offer the same rights regardless of status).

1192 BOSTON UNIVERSITY LAW REVIEW [Vol. 102:1185

While scholars of political economy have long been interested in levels of

inequality, Esping-Andersen’s emphasis on decommodification introduced a

different standard by which to evaluate capitalist economies. While inequality

speaks to integration or stratification and turns on the relative resources that

individuals have, decommodification asks about entitlements, not just outcomes.

A citizen is “decommodified” if she does not have to rely on the market to meet

basic needs and wants.

Esping-Andersen did not originate the concept of

decommodification; its origins are in Marxist theory.

As Esping-Andersen’s

peer Karl Polanyi explained, capitalism treats labor as a commodity available

for exchange.

A worker is commodified in the Marxist sense as long as he sells

his labor rather than the goods produced with it.

Esping-Andersen did not

incorporate the Marxist concept as such.

He does not focus on workers’

productive roles; he is not concerned, therefore, with ownership of the means of

production and whether workers are alienated from their labor. Instead, he is

interested in their consumption, that is, their access to goods and services.

reads history to deliver a clear verdict in favor of promoting workers’ returns to

labor by enhancing reservation wages rather than by intervening in structure of

production.

Nevertheless, Esping-Andersen sets another sweeping and overambitious

threshold for decommodification. He provides that the minimal definition of

decommodification entails that “citizens can freely, and without potential loss

of job, income, or general welfare, opt out of work when they themselves

consider it necessary.”

As Mark Kleinman observed, no capitalist country

Id. at 37.

See id. (noting that issue of commodification was central to Marx’s analysis of class

development).

See KARL POLANYI, THE GREAT TRANSFORMATION 73 (Beacon Press 1957) (1944). The

extent and scope of Polanyi’s concept of decommodification is itself unclear and contested.

See, e.g., Gareth Dale, Social Democracy, Embeddedness and Decommodification: On the

Conceptual Innovations and Intellectual Affiliations of Karl Polanyi, 15 NEW POL. ECON. 369,

386-87 (2010) (criticizing Polanyi’s understanding of decommodification based on his

misjudgments of economic policy regimes); CLAUS OFFE, CONTRADICTIONS OF THE WELFARE

STATE 14 (John Keane ed., 1984) (developing decommodification idea further while noting

contradiction in fact that various branches of welfare state are compelled to perform

commodification and decommodification, two incompatible functions).

See ESPING-ANDERSEN, supra note 9, at 21 (“Stripping society of the institutional layers

that guaranteed social reproduction outside the labor contract meant that people were

commodified.”).

Esping-Andersen, supra note 26, at 355 (noting he does not believe that “self-realisation

in working life is intrinsic to the commodity status of labour”).

Id. at 354.

Id. at 353-54 (arguing that, historically, welfare strategy, the need for workers to be

secure and resourceful before creating communal economy, demonstrated superiority over

orthodox strategy, which is characterized by altering means of production for equality).

ESPING-ANDERSEN, supra note 9, at 23.

2022] LOWERING THE STAKES OF EMPLOYMENT CONTRACT 1193

would aspire to such a standard.

No country sets out to make it possible for

people simply to choose not to work without suffering any kind of setback, and

most of us would reject such a standard in principle as outside the demands of

justice.

Perhaps the lesson is that decommodification should be reconceptualized

further. But given how far it has already come from its Marxist origins, a

different lesson can be that decommodification as such is not the best measure

of success for welfare state regimes. The idea is compelling because it is a proxy

for a more specific phenomenon than the theory behind it implies: the degree of

power that employers exercise over workers. Indeed, Esping-Andersen himself

observed that decommodification “strengthens the worker and weakens the

absolute authority of the employer.”

Others have more or less directly pointed to the short discussion of employers

in Esping-Andersen’s work as a significant limitation. Robert E. Goodin and

Martin Rein have suggested that we think more about who pays for and supplies

the services of the welfare state.

Peter A. Hall and David Soskice not only

reduced the three types to two clusters—liberal market economies and

coordinated market economies—but also shifted focus to employers as the

engine that determines which model prevails in a given country, as well as that

model’s evolution over time.

Combining the interest in employer control and employer provision, we arrive

at the following question to ask of any regime: To which social goods do

employers regularly control access? Or as I will suggest below, when are

employers the gatekeepers and allocators of goods for social purposes other than

labor compensation? In every capitalist society that Esping-Andersen and his

interlocutors considered, employers offer varied renumeration to employees for

their work.

In some countries, employers have more latitude than others about

the terms and conditions they may offer.

That latitude gets at the question of

how much turns on the particular job that an individual holds. Elevating the

social wage by mandating minimum wages or paid time off is one way of

MARK KLEINMAN, A EUROPEAN WELFARE STATE?: EUROPEAN UNION SOCIAL POLICY IN

CONTEXT 30 (2002).

ESPING-ANDERSEN, supra note 9, at 22.

Robert E. Goodin & Martin Rein, Regimes on Pillars: Alternative Welfare State Logics

and Dynamics, 79 PUB. ADMIN. 769, 770 (2001) (“The central organizing question in regime

discourse is ‘who gets what, and on what conditions?’ The ‘pillar’ perspective sees things

from the provider’s point of view. The central organizing question in pillar discourse is ‘who

pays, and who provides?’”).

Hall & Soskice, supra note 21, at 8.

See ESPING-ANDERSEN, supra note 9, at 43 (noting in liberal regimes, social insurance

pegs entitlements and benefits to varied employment, work performance, and contributions).

See id. at 23 (noting that, for example, in some countries, sickness insurance that

guarantees benefits equal to normal earnings, with minimal proof of medical impairment

requirements and for the duration that the individual deems appropriate, are usually reserved

for white-collar employees).

1194 BOSTON UNIVERSITY LAW REVIEW [Vol. 102:1185

reducing variation among employment contracts. Such variation speaks to the

level of inequality and stratification in a society.

However, the concept of commodification is powerful because it captures

something more than inequality at the aggregate level. It speaks to the power

dynamic in employment relationships at the micro level.

It does not, however,

distinguish between the various goods and services that employment contracts

might control. The concept did not seem to anticipate that employment contracts

would be a site for distribution of goods even beyond the “ordinary” terms of

renumeration for labor.

In the United States, employment determines the extent

to which employees have access even to goods that the employer would not

wield as compensation but for public policies that affirmatively assign

responsibility for those entitlements to employers. While the welfare state might

shield employees from dependence on employers for certain kinds of goods and

services, our institutions also rely on employers to allocate other goods and

services that employers would not use as compensation absent legal incentives.

One might immediately and rightly object that there is no natural line between

goods or entitlements that employers would offer as compensation and those that

are “artificially” introduced as consequences of employment. This is no doubt

true—and I would not attempt to draw a clear line. Nevertheless, it is useful to

distinguish in principle between terms of the employment contract that the

employer itself uses in order to attract and extract productive labor from terms

that the employer uses in order to avoid some legal liability. The most clear

example of a compensation-driven term are cash wages. Employers will pay

employees different amounts depending on their marginal productivity and the

available supply of substitutable labor.

By contrast, a hypothetical term that

offers employees free iPads because the tax code makes them cheaper for the

employer to provide them than for the employee to purchase them directly is

“artificial.” If the price difference for employers and employees is substantial,

employers would be gatekeepers for access to iPads. Whether one has an iPad

would then turn on what kind of job one has and who one works for, more so

than if iPads were purchased directly. If iPads were purchased directly, we

would expect people with more money to be more likely to have an iPad (or to

have multiple iPads or newer models), but the iPads that people have will also

reflect to a considerable degree the strength of their own preferences with respect

to iPads. Some high-income employees will choose not to buy one at all. Some

low-income employees will spend their first disposable dollars on the newest

model. Most importantly, an employee’s decision to work or remain with a

particular employer will not turn on her eligibility for an iPad at her place of

Graham Room, Commodification and Decommodification: A Developmental Critique,

28 POL’Y & POL. 331, 332 (2000).

See Kenneth Burdett & Dale T. Mortensen, Wage Differentials, Employer Size, and

Unemployment, 39 INT’L ECON. REV. 257, 258, 269 (1998) (noting employers pay wages

conditional on other workers’ behavior and that productivity impacts wages).

2022] LOWERING THE STAKES OF EMPLOYMENT CONTRACT 1195

employment. She will not, for example, be reluctant to leave her employer when

she is nearing the hypothetical six-month eligibility mark.

Is it terrible for employers to allocate iPads to employees directly? We might

not worry too much about this hypothetical role of employers if iPads are not

regarded as a critical resource for individuals. I will argue below that we should

worry about it when the stakes we have introduced into the employment

relationship are higher, as they are in the case of speech rights and healthcare.

II. ASSIGNING PUBLIC FUNCTIONS TO EMPLOYERS

In this Part, I will discuss three consequences of American employment that

are the result of legal incentives for employers rather than their desire to

optimally attract and extract productive labor. The first two are social goods,

albeit quite fundamental ones: speech rights and healthcare. I will then turn to a

more recent condition of employment, a potential duty to be vaccinated. In each

case, the state has effectively delegated to employers the task of allocating a

good or a burden. In each case, there are good reasons why policy evolved as it

did. But we are nevertheless left with an unsavory state of affairs, in which, far

from decommodifying employment, our institutions have artificially raised the

stakes of employment even beyond those that follow from employers’ business

interests.

A. Off-site Speech

Speech rights are at first blush universal in the United States. After all, the

First Amendment guarantees them for everyone.

Because the First

Amendment only limits government regulation of speech, however, it does not

ensure that individuals are permitted to say whatever they want without

penalty.

Nor should it. There are good reasons to think that individuals should face

consequences for certain kinds of harmful speech even when the government

ought not to regulate that speech. This is most obviously true of entirely private

speech: if your friend repeatedly says hurtful things to you, you probably should

cease to be her friend. Unfortunately, it is less clear who should be in the position

to “cancel” harmful speech that is public. Employees are just people, and they

often make false, irrational, dangerous, ridiculous, or vicious statements,

including racist or sexist comments—both at work and outside of it.

When

people make comments at work, it is easier to see that employers are both

U.S. CONST. amend. I.

Roth v. United States, 354 U.S. 476, 482, 486-87 (1957) (holding that certain categories

of speech are not protected, and the government can prohibit and punish those types of speech

without violating First Amendment).

See, e.g., Stella Chan, Augie Martin & Alicia Lee, Tech CEO Resigns After Video Shows

His Racist Rant Towards Asian-American Family at California Restaurant, CNN BUS. (July

13, 2020, 5:58 PM), https://www.cnn.com/2020/07/13/business/michael-lofthouse-resign-

racist-rant-trnd/index.html [https://perma.cc/R6K8-8GP7].

1196 BOSTON UNIVERSITY LAW REVIEW [Vol. 102:1185

appropriately responsible for the work environment and have a legitimate

interest in controlling its tenor.

But of course, employees make comments that are potentially wrongful

outside of work as well, and then it is less clear what the appropriate role of the

employer is in policing their speech.

After all, they also make comments that

are not wrongful but are simply unfavorable to the employer; the employer may

“idiosyncratically” judge such comments as wrongful.

Still, when speech

targets vulnerable members of society or operates to perpetuate social

hierarchies, it is doubtful that the speakers are entitled not to suffer any

consequence for their speech acts, even if the government ought not to directly

prohibit or penalize such speech. But what lesser, civil consequence should such

speech have, and equally important, who decides in each case? Should

irresponsible speakers simply lose their friends? Should they lose their jobs?

Should they be shunned by grocery stores and banks, common carriers and

utility companies? I do not purport to offer a solution to the problem here. My

point is that we not only disagree about what kinds of speech are harmful, but

there is also a further absence of public consensus on what the consequences of

harmful speech should be, if any, and who should decide them.

What has emerged is a default regime: employers are frequently called upon

to terminate employees when they are associated with offensive speech, and

employers frequently oblige.

Because this social practice enables us to punish

those who engage in wrongful speech without running afoul of the First

Amendment, there is little initiative to question whether employers are the right

entities to decide to punish offensive speech when it is far removed from the

workplace. After all, as discussed further below, employers’ incentives as to how

to handle off-site speech turn on factors that do not systematically correspond to

the merits of that speech or its relevance to the workplace. Assigning the social

function of policing speech to employers also has consequences for many

employees who do not engage in speech that is offensive to public values: once

See, e.g., Lora Kolodny, Tesla Factory Workers Have Filed a Lawsuit Claiming

Widespread Racism, Unsafe Conditions, CNBC (Nov. 15, 2017, 11:48 AM),

https://www.cnbc.com/2017/11/14/tesla-class-action-lawsuit-alleges-racism-unsafe-factory-

conditions.html [https://perma.cc/2JJM-KND4] (stating that Tesla and Elon Musk personally

contributed to racist culture at Tesla by failing to stop factory workers from using racial

epithets, failing to take corrective action against harassers, and firing Black workers who

complained).

See, e.g., Nicholas H. Meza, Comment, A New Approach for Clarity in the

Determination of Protected Concerted Activity Online, 45 ARIZ. ST. L.J. 329, 329-30 (2013)

(noting employees regularly share opinions online, and employers struggle to determine when

discipline is appropriate).

See, e.g., id. at 351-52 (detailing incident where bartender was dismissed for criticizing

employer’s tipping policy on Facebook).

See, e.g., Jonah E. Bromwich, Amy Cooper, Who Falsely Accused Black Bird-Watcher,

Has Charge Dismissed, N.Y. TIMES (May 26, 2021), https://www.nytimes.com/2021/02

/16/nyregion/amy-cooper-charges-dismissed.html (noting that Cooper was fired after gaining

notoriety for her racist conduct at the park).

2022] LOWERING THE STAKES OF EMPLOYMENT CONTRACT 1197

employers are empowered to monitor and control employees’ off-site speech,

they can terminate employees for many kinds of speech that are not to their

liking, including speech that is simply critical of the employer and its goods or

services.

My limited purpose here is to observe that the consequence of our

disagreement about liability for speech, either the substantive standards of

propriety or the procedural question of who should apply those standards and

how, has made employers the gatekeepers for speech. In some cases, they are

reluctant gatekeepers; they are responding to the concurrent threats of legal

liability and consumer boycott. In other moments, they can abuse their

gatekeeper status to control speech with limited public consequence that is

simply critical of the employer itself.

Employers are subject to some limitations in their control of employee speech.

Most notably, the government itself is prohibited from using the threat of

dismissal from public employment as a means of inhibiting speech. In Pickering

v. Board of Education,

the Supreme Court balanced a citizen’s interest in

speaking about matters of public concern against the government-employer’s

interest in efficiently providing public services.

Although Pickering protection

is to some extent vulnerable to the ideologies of the government in power,

enjoys the robustness of constitutional grounding. But while the Pickering rule

is significant in its own domain, it does not apply to employees of private

employers, who are not subject to the constraints of the First Amendment.

However, employees of private employers are not altogether without

protection. At the federal level, protection is narrow: section 7 of the National

Labor Relations Act prevents employers from interference with concerted action

by employees.

This has been interpreted to encompass speech, including

speech outside of work.

If an employee’s speech is directed to another

employee, the Board considers a slew of factors to determine whether any

disciplinary measure taken against the employee was permissible, including

whether the employer maintained a specific rule prohibiting or otherwise

previously condemned the language used by the employee, whether the

employee’s statement was impulsive or deliberate, whether the discipline

imposed upon the employee was consistent with past discipline for similar acts,

See, e.g., Meza, supra note 49, at 351-52 (stating employee was dismissed not for hate

speech but for criticizing employer’s tipping policies).

See, e.g., id.

391 U.S. 563 (1968).

Id. at 565-67, 574.

See, e.g., Nancy M. Modesitt, The War on Whistleblowers, 6 U. PA. J.L. & PUB. AFFS.

603, 627 (2021) (describing Trump Administration’s hostile moves to weaken whistleblower

protection for federal employees).

See Pub. Utils. Comm’n v. Pollak, 343 U.S. 451, 461 (1952) (noting First and Fifth

Amendments only restrict federal government, not parties such as employers).

National Labor Relations Act § 7, 29 U.S.C. § 157.

Pier Sixty, LLC, 362 N.L.R.B. 505, 531 (2015).

1198 BOSTON UNIVERSITY LAW REVIEW [Vol. 102:1185

whether the disciplinary measures were directed at offensive speech rather than

protected activity, and whether the employer had a record of anti-union

hostility.

The Board has found itself increasingly charged with adjudicating cases in

which employees have posted comments on social media and has issued

guidance by way of examples of protected and unprotected conduct.

Employees’ disparaging comments about an employer are protected as long as

the comments are related to working conditions or other workplace grievances.

However, the Board recently pulled back under the Trump Administration by

requiring more evidence that an employee intended to “initiate, induce, or

prepare for group action.”

While some commentators would have the Board

adopt a still more restrictive but also more predictive rule,

others regard the

evolution of doctrine concerning social media as broadly consistent with long-

standing precedent.

While the interpretation of section 7 and satisfaction with it has varied (and

this observer would favor a more protective standard), on the whole, the generic

language of the statute has been broadly construed to ensure that speech

plausibly intended to mobilize employees toward concerted action is protected.

Id.

See Memorandum from Anne Purcell, Nat’l Lab. Rels. Bd. Assoc. Gen. Couns., to Reg’l

Dirs., Officers-in-Charge & Resident Officers (Aug. 18, 2011) (available at

https://apps.nlrb.gov/link/document.aspx/09031d458056e743) (“Recent developments in the

Office of the General Counsel have presented emerging issues concerning the protected

and/or concerted nature of employees’ Facebook and Twitter postings, the coercive impact of

a union’s Facebook and YouTube postings, and the lawfulness of employers’ social media

policies and rules.”); Memorandum from Anne Purcell, Nat’l Lab. Rels. Bd. Assoc. Gen.

Couns., to Reg’l Dirs., Officers-in-Charge & Resident Officers (Jan. 24, 2012) (available at

https://apps.nlrb.gov/link/document.aspx/09031d45807d6567) (collecting cases where

employees were terminated as result of social media activity).

Sierra Publ’g Co. v. NLRB, 889 F.2d 210, 216 (9th Cir. 1989) (“[A]ppeals to third

parties forfeit § 7 protection only if their connection to the employees’ working conditions is

too attenuated or if they are unrelated to any grievance which the workers may have.”).

Alstate Maint., LLC, 367 N.L.R.B. No. 68, 4, 8 (Jan. 11, 2019) (noting that “other

evidence that a statement made in the presence of coworkers was made to initiate, induce or

prepare for group action—such as an express call for employees to act collectively—would

also support a finding of concertedness”).

See, e.g., Meza, supra note 49, at 367 (arguing for a more bright-line but less protective

rule).

See, e.g., Ariana R. Levinson, Social Media and the National Labor Relations Board, in

RESEARCH HANDBOOK ON ELECTRONIC COMMERCE LAW 337, 337 (John A. Rothchild ed.,

2016) (“This chapter buttresses the claim that Board regulation of social media policies is

consistent with past practice and precedent by analogy to Board precedent governing

employer policies on solicitation and distribution and on the wearing of insignia, which are

similar to the social media policies currently being regulated.”).

Fresh & Easy Neighborhood Mkt., Inc., 361 N.L.R.B. 151, 167 (2014) (Miscimarra,

Member, concurring in part and dissenting in part) (“[T]he Board majority announces a broad

2022] LOWERING THE STAKES OF EMPLOYMENT CONTRACT 1199

However, it does not even aspire to limit employer discipline of any other kind

of employee speech intended only to complain about the employer, colleagues,

or customers, let alone matters unrelated to employment. Even when speech falls

within the bounds of section 7, the employer can still discipline or terminate the

employee if the speech was sufficiently “disloyal.”

State legislatures have offered protection beyond the context of collective

mobilization.

Many states, such as California, Colorado, North Dakota, and

Louisiana, have statutes that protect employees from dismissal for political

activity or specific categories of protected speech.

And many unionized

workplaces offer protections against arbitrary dismissal that arguably protect

speech more ably than more narrowly drawn protections that are speech-

specific.

But most workers are subject to at-will dismissal and do not benefit

from those procedural protections.

Legislative protections of speech are critical but fall short of what many

employees might intuitively expect given the vaulted status of “free speech”

principles in the popular American imagination.

These statutes do not embrace

the more holistic protection recommended by, for example, the Restatement

(Third) of Employment Law. The Restatement rule would protect employees

whenever they engage in “lawful conduct that occurs outside of the locations,

hours, and responsibilities of employment and does not refer to or otherwise

involve the employer or its business,” and allow employees to express “political,

moral, ethical, religious, or other personal beliefs” outside of work as long as it

holding that ‘an employee seeking the assistance or support of his or her coworkers in raising

a sexual harassment complaint is acting for the purpose of mutual aid or protection’”).

See Sierra Publ’g Co., 889 F.2d at 216-18.

Genevieve Lakier, The Non-First Amendment Law of Free Speech, 134 HARV. L. REV.

2299, 2339-42 (2021) (“Today, as a result of all this [state] legislative activity, a significant

body of . . . ‘quasi-first Amendment law’ protects the expressive autonomy of public- and

private-sector workers even when the First Amendment cases do not.”).

Andrew Melzer & Whittney Barth, Whether Employees Can Be Fired for Participating

in Peaceful Protests, 2020 U. ILL. L. REV. ONLINE 221, 228.

Cynthia L. Estlund, Free Speech and Due Process in the Workplace, 71 IND. L.J. 101,

102-03 (1995) (“[I]n the typical unionized workplace, employees enjoy much greater freedom

of expression—perhaps more than many public employees protected by the First

Amendment.”).

Id. at 102 (“[M]any employees in both the public sector and the private sector enjoy no

general legal protection against arbitrary discipline or discharge.”).

See, e.g., William R.H. Mosher, Speak No Evil – The Right to Limit Employee Speech

This Election Season and Beyond, FISHER PHILLIPS (Oct. 7, 2020),

https://www.fisherphillips.com/news-insights/speak-no-evil-the-right-to-limit-employee-

speech-this-election-season-and-beyond.html [https://perma.cc/4G3H-TMWS] (“[M]ost

state laws granting private employees the right to engage in free speech limit those rights in

clear and specific ways. . . . New York’s free speech law only protects employees’ actions if

they are legal, take place off the employer’s premises, without use of the employer’s

equipment or property, and do not create a material conflict of interest with the employer’s

business.”).

1200 BOSTON UNIVERSITY LAW REVIEW [Vol. 102:1185

is done “in a manner that does not refer to or otherwise involve the employer or

its business.”

While such protection for private speech is intuitively consistent

with the spirit of autonomy and free discourse implicitly endorsed by the First

Amendment, there is no blackletter law on the books that comes close to

realizing it.

To the contrary, instead of carving out a zone for private speech into which

employers may not intrude, public policies affirmatively push employers to

regulate employee speech. Again, the reason for relying on employers as

gatekeepers is compelling: we do not have a good alternative way to control

harmful speech.

One way employers are engaged as gatekeepers is by way of the threat of

liability. Employers are potentially liable when their employees post racist or

sexist comments on social media.

Employers also face market pressure to

discipline or terminate employees whose words or conduct have made them

infamous online.

In some cases, the employee’s notoriety is well deserved; in

other cases, it is not. In all cases, however, there is disagreement about whether

the employee deserves termination—and it is the employer that decides this

social question.

There is an important, attendant social cost to allocating such power to

employers. Risk-averse employees will avoid public expression that might ignite

the ire of either their employer or some other segment of the public that could

mobilize to pressure their employer. The chilling effect is not unlike the chilling

effect we could expect from the prospect of government sanctions; given the

immediate bread-and-butter implications of losing one’s job, the threat of

employer sanctions might be considerably greater.

The role of employers as

speech gatekeepers has expanded just as employees are more likely to engage in

controversial public speech. At least two phenomena might contribute to this

RESTATEMENT (THIRD) OF EMP. L. § 7.08 (AM. L. INST. 2014).

See Little v. Windermere Relocation, Inc., 301 F.3d 958, 966-69 (9th Cir. 2002) (holding

that employer can be liable for off-site speech); Roy v. Correct Care Sols., LLC, 914 F.3d 52,

73 n.4 (1st Cir. 2019) (finding that Facebook messages could be considered when determining

if employee’s harassment created hostile work environment, particularly when those

messages were about workplace conduct and sent by someone who worked with plaintiff);

EEOC v. Fed. Express Corp., No. 94-cv-00790, 1995 WL 569446, at *3 (W.D. Wash. Aug.

8, 1995) (stating that there is “no serious dispute that an employer may face liability for sexual

harassment of employees by non-employees inside the workplace” and expressing doubt that

location matters).

Vikram R. Bhargava, Firm Responses to Mass Outrage: Technology, Blame, and

Employment, 163 J. BUS. ETHICS 379, 379 (2018) (“When an employee’s off-duty conduct

generates mass social media outrage[, b]usinesses commonly respond by firing the

employee.”).

Timothy Zick, The Costs of Dissent: Protest and Civil Liabilities, 89 GEO. WASH. L.

REV. 233, 240 (2021) (“The chilling effect of these costs and liabilities on First Amendment

protest rights cannot be overstated.”).

2022] LOWERING THE STAKES OF EMPLOYMENT CONTRACT 1201

trend: First, social media has lowered the cost of access to a large audience.

Second, there may be a cultural trend that encourages people to “speak their

mind” about a broader range of social causes but especially ones connected in

some way to their employer.

While there is much to worry about concerning

the quality of contemporary public discourse, we should also worry about the

role of employers in regulating this discourse.

It is tempting to favor a simple solution to the gatekeeper problem, that is,

more expansive, categorical protections for employee speech—enacted by

statute or developed in case law.

The problem is that there is a legitimate social

interest in controlling or creating accountability for online speech, which can be

genuinely harmful to workers subject to hostile work environments.

The same

progressive forces that might otherwise have mobilized to protect a sphere of

employee autonomy will rightfully balk at eliminating accountability for certain

kinds of harmful speech. Public employees, including police officers, have been

terminated on many occasions for violating codes of conduct on social media by

posting racist comments.

The same logic of harm and accountability apply to

private workplaces. But while the government is at least constrained by the First

Amendment and the Pickering doctrine when it draws a balance between the

rights of employees subject to harmful speech and the rights of employees to

engage in harmful speech, private employers are free to draw their counterpart

lines by reference to their interests alone.

Each employer will draw it

See Eugene Volokh, What Cheap Speech Has Done: (Greater) Equality and Its

Discontents, 54 U.C. DAVIS L. REV. 2303, 2306 (2021) (observing that online speech has

democratized access to public discourse).

See Kelley Changfong-Hagen, “Don’t Be Evil”: Collective Action and Employee

Prosocial Activism, 5 COLUM. HUM. RTS. L. REV. ONLINE 188, 209 (2021) (discussing

“growing phenomenon of employee activism in the tech sector”).

See, e.g., Meredith McCaffrey, Comment, Public or Private? The Split over First

Amendment Protection of Union Speech by Public Employees, 60 B.C. L. REV. II.-274, II.-

288 (2019) (arguing Second Circuit was wrong for adopting a per se rule that any person who

speaks as a union member does so as a private citizen and thus is protected from retaliation

by the First Amendment).

See Tatiana Hyman, Note, The Harms of Racist Online Hate Speech in the Post-COVID

Working World: Expanding Employee Protections, 89 FORDHAM L. REV. 1553, 1588 (2021)

(“Racist hate speech can damage an employee’s psychological well-being and have adverse

effects on work performance both inside of the workplace and over the web.”).

See, e.g., Grutzmacher v. Howard County, 851 F.3d 332, 340 (4th Cir. 2017); Liverman

v. City of Petersburg, 844 F.3d 400, 405-06 (4th Cir. 2016); Venable v. Metro. Gov., 430 F.

Supp. 3d 350, 354 (M.D. Tenn. 2019).

See DAVID L. HUDSON, JR., FIRST AMENDMENT CTR., BALANCING ACT: PUBLIC

EMPLOYEES AND FREE SPEECH 2 (2002), https://www.freedomforuminstitute.org/wp-

content/uploads/2014/09/fac_firstreport_public-employees-free-speech.pdf

[https://perma.cc/BG5G-CTTE] (“While the private employer probably can fire an employee

whose speech he dislikes, the First Amendment governs the circumstances under which public

employers may discipline employees for their speech.”).

1202 BOSTON UNIVERSITY LAW REVIEW [Vol. 102:1185

differently—thereby raising the stakes of which employer an employee happens

to work for.

If the blunt approach of adopting a blanket zone of protection for employee

speech imposes unacceptable costs on a few and threatens to erode the

legitimacy of public discourse for us all, the narrower alternative would be

spelling out in greater detail—by statute or case law—what kinds of speech

employers may or must penalize and carving such speech out from an articulated

sphere of protected speech. It is difficult to see those steps in our future. We

might agree that it is desirable to reign in employer power with respect to

employee speech. We might even accept, as a pragmatic compromise, the idea

of legal standards that specify precisely which speech employers may discipline.

But such standards ultimately would have to be developed by state actors—

legislative, administrative, or judicial. These state actors would then have to

make substantive judgments about what kinds of speech are permissible and too

harmful to others to warrant protection. In the current state of public discourse,

it is hard to imagine productive political action in this direction. Even if a

political faction could enact a substantive standard, they may not trust judges

with enforcing it.

The vitriol of civic discourse makes the prospect of political

action dangerous. The impotence of politics reinforces our reliance on

employers to police speech, however imperfect guardians of civic discourse they

may be.

B. Healthcare

Healthcare is the most evident and entrenched area in which American

employers exercise surprising control over employees—surprising in

comparison to employers in other advanced industrialized democracies.

Nowhere else do employers control whether employees have access to a good

that many people regard in some form as a human right.

See William N. Eskridge, Jr., The Circumstances of Politics and the Application of

Statutes, 100 COLUM. L. REV. 558, 566 (2000) (book review) (explaining mistrust comes from

legislatures being unable to predict how courts will interpret statutes).

Aaron E. Carroll, The Real Reason the U.S. Has Employer-Sponsored Health Insurance,

N.Y. TIMES: THE UPSHOT (Sept. 5, 2017), https://www.nytimes.com/2017/09/05/upshot/the-

real-reason-the-us-has-employer-sponsored-health-insurance.html (“The basic structure of

the American health care system, in which most people have private insurance through their

jobs, might seem historically inevitable, consistent with the capitalistic, individualist ethos of

the nation.”).

Jeneen Interlandi, Why Doesn’t the United States Have Universal Health Care? The

Answer Has Everything to Do with Race., N.Y. TIMES (Aug. 14, 2019),

https://www.nytimes.com/interactive/2019/08/14/magazine/universal-health-care-

racism.html (“[T]he United States remains the only high-income country in the world where

such care is not guaranteed to every citizen.”).

2022] LOWERING THE STAKES OF EMPLOYMENT CONTRACT 1203

The history behind America’s employer-based health insurance regime is

quintessentially path dependent.

During World War II, wage controls made it

difficult for employers to attract employees.

Employers used benefit packages

to compete for workers because insurance and pension fund contributions were

not subject to the prevailing wage caps.

American labor unions have famously

operated in the voluntarist tradition, seeking to improve working conditions

through private firm-level bargaining rather than political action.

It was

consistent with this broad strategy that unions helped entrench employer-based

healthcare by advocating for health insurance as a benefit in the context of firm-

level collective bargaining.

Later, tax incentives made health insurance an even

more enticing form of compensation: while taxable income for an employee

does not include employer contributions to health insurance, employers can

deduct those contributions from their income.

The effect is a federal subsidy

that increases with the marginal tax rate to which an employee is subject. It is

the second-largest subsidy of the federal tax code after the mortgage interest

deduction.

Recent healthcare reform efforts have shown that individuals are attached to

their existing insurance and that any pivot away from the employer-based model

would be politically costly, if not impossible.

Therefore, recent legislation like

See Jacob S. Hacker, The Historical Logic of National Health Insurance: Structure and

Sequence in the Development of British, Canadian, and U.S. Medical Policy, 12 STUD. AM.

POL. DEV. 57, 76 (1998) (describing path dependence in healthcare policy); see also Paul

Pierson, Increasing Returns, Path Dependence, and the Study of Politics, 94 AM. POL. SCI.

REV. 251, 251 (2000) (describing path dependence in social policy generally).

Stephen Mihm, Opinion, Employer-Based Health Care Was a Wartime Accident, CHI.

TRIB. (Feb. 24, 2017, 3:30 PM), https://www.chicagotribune.com/opinion/commentary/ct-

obamacare-health-care-employers-20170224-story.html (“As companies struggled to deal

with wartime labor shortages, the wage freeze left them in a serious bind: How could they

retain workers if they couldn’t give raises?”).

See David A. Hyman & Mark Hall, Two Cheers for Employment-Based Health

Insurance, 2 YALE J. HEALTH POL’Y L. & ETHICS 23, 25 (2001).

See WILLIAM E. FORBATH, LAW AND THE SHAPING OF THE AMERICAN LABOR MOVEMENT

6-8 (1991) (arguing American labor movement turned to voluntarism after political action

was repeatedly thwarted by judicial hostility); Derek C. Bok, Reflections on the Distinctive

Character of American Labor Laws, 84 HARV. L. REV. 1394, 1419 (1971) (arguing various

features of the American worker impeded class consciousness in favor of individualist

approach to labor reform).

Allison K. Hoffman, An Optimist’s Take on the Decline of Small-Employer Health

Insurance, 98 IOWA L. REV. BULL. 113, 114 (2013) (noting that “labor unions advocated for

health insurance as a benefit for workers”).

I.R.C. § 106 (excluding employer contributions to health insurance from employee

taxable income).

See Hyman & Hall, supra note 88, at 25-26.

Shefali Luthra, Do 160 Million Americans Really Like Their Health Plans? Kind Of,

KAISER HEALTH NEWS (Nov. 21, 2019), https://khn.org/news/do-160-million-americans-

really-like-their-health-plans-kind-of/ [https://perma.cc/9F2J-5S4U] (“Articulating his

1204 BOSTON UNIVERSITY LAW REVIEW [Vol. 102:1185

the Affordable Care Act (“ACA”) relies on the employer model to expand

access.

Most importantly, it imposes a mandate on large employers.

It creates

tax incentives for small employers to provide insurance, offering credits up to

35% of premium costs to small business that insure their employees.

While the problems with the employer-based model discussed below have led

some observers to predict its demise,

we have yet to see any retrenchment.

Instead, we have primarily continued down the path we started, attempting to

correct for specific problems of access and portability without facing down the

structural limitation of the model.

What are the limitations of the employer-based model? There are several.

Most fundamentally, relying on employers for coverage renders employees

vulnerable to retrenchment in coverage. Indeed, we have observed in recent

decades that employees’ share of healthcare costs is increasing as a fraction of

total costs and that the mean quality of their coverage has fallen.

The effect is

to shift the risk of both illness and medical costs to individuals from

proposal for health care reform, former Vice President Joe Biden emphasized the number of

Americans who, he said, were more than perfectly satisfied with the coverage they have.”).

Employer Shared Responsibility Provisions, IRS, https://www.irs.gov/affordable-care-

act/employers/employer-shared-responsibility-provisions [https://perma.cc/W8TC-D8AW]

(last updated Nov. 23, 2021) (“Under the Affordable Care Act’s employer shared

responsibility provisions, certain employers (called applicable large employers or ALEs) must

either offer minimum essential coverage that is ‘affordable’ and that provides ‘minimum

value’ to their full-time employees (and their dependents), or potentially make an employer

shared responsibility payment to the IRS.” (citation omitted)).

Affordable Care Act, 26 U.S.C. § 4980H(a), (c)(2).

Small Business Health Care Tax Credit and the SHOP Marketplace, IRS,

https://www.irs.gov/affordable-care-act/employers/small-business-health-care-tax-credit-

and-the-shop-marketplace [https://perma.cc/4KA6-ADDA] (last updated Feb. 24, 2022) (“If

you are a small employer, there is a tax credit that can put money in your pocket.”).

See, e.g., Warren Greenberg, Employer-Based Health Insurance: At the End of the

Line?, 20 HEALTH LAW. 38, 40 (2008) (“Since . . . employer-based health insurance has been

the main source of health insurance in the private sector for more than sixty years, realistic

policy proposals may have to encourage the demise of the employer-based system in a gradual

way.”); see also Amy B. Monahan & Daniel Schwarcz, Saving Small-Employer Health

Insurance, 98 IOWA L. REV. 1935, 1935 (2013) (“[A] number of interweaving provisions

embedded within the Affordable Care Act create strong incentives that, starting in 2014, will

tend to undermine these markets and, in the process, increase the fiscal cost of reform.”).

See David S. Caroline, Comment, Employer Health-Care Mandates: The Wrong

Answer to the Wrong Question, 11 U. PA. J. BUS. L. 427, 446 (2009) (“[T]he challenge of

increasing efficient access to health care requires shifting employers out of the business of

providing health care, and instead finding creative ways to enable individuals to purchase the

care they want and need in a market-based system.”).

Michael Supanick, Private Insurance and Universal Healthcare: How Can Private

Insurance Be Utilized Within a Universal Healthcare System in the United States?, 30 S. CAL.

INTERDISC. L.J. 551, 557-58 (2021) (discussing employees’ rising costs not matched by

increases in income).

2022] LOWERING THE STAKES OF EMPLOYMENT CONTRACT 1205

employers—a shift that employers can effectuate unilaterally at the contractual

level.

100

Even when employers do provide high-level coverage, it is not clear that the

kinds of policies they purchase would be those that employees would purchase

themselves.

101

I will discuss specific divergences between employer and

employee preferences below, but for now I point only to the basic agency

problems associated with an intermediary with imperfect information acting on

behalf of a heterogenous group of employees.

102

Still another famous deficiency of the employer-based model is the problem

of portability. Most employees do not remain with a single employer for any

significant duration of their working lives; job tenure is shrinking.

103

When an

employee leaves her job, she cannot take her insurance with her to her new

position. Even if she is able to purchase insurance for the interim at substantial

individual cost, transferring to another insurance network frequently requires a

change of healthcare provider.

104

The effect is the phenomenon known as “job

lock.”

105

Employees are artificially immobilized in the labor market,

exacerbating the labor market power of employers. While welfare measures

intended to promote healthcare access are ordinarily associated with employee

leverage and decommodification,

106

the power exercised by employers over

100

See generally JACOB S. HACKER, THE GREAT RISK SHIFT: THE ASSAULT ON AMERICAN

JOBS, FAMILIES, HEALTH CARE AND RETIREMENT AND HOW YOU CAN FIGHT BACK (2006)

(arguing that economic risk, previously borne by corporate sector and government, has shifted

to individuals).

101

Dayna Bowen Matthew, Controlling the Reverse Agency Costs of Employment-Based

Health Insurance: Of Markets, Courts, and a Regulatory Quagmire, 31 WAKE FOREST L. REV.

1037, 1040-41 (1996) (arguing that employers are poor agents for employees and their

insurance purchases do not reflect employee preferences well).

102

See Hyman & Hall, supra note 88, at 26-28 (describing agency problems). However,

Hyman and Hall contend that “employers do a reasonably good job reflecting their workers’

values and preferences,” id. at 30, and point out that employer-based insurance pools attenuate

the problem of adverse selection and facilitate cross-subsidization of high-cost consumer-

patients, id. at 32-33.

103

See Katherine V.W. Stone, The New Psychological Contract: Implications of the

Changing Workplace for Labor and Employment Law, 48 UCLA L. REV. 519, 540 (2001)

(describing increasing shift to temporary work).

104

See Hoffman, supra note 90, at 123 (“This means that when an individual leaves a job,

she cannot take her insurance with her, causing disruptions in coverage and often requiring

individuals to change their care providers.”).

105

See Jonathan Gruber & Brigitte C. Madrian, Health Insurance, Labor Supply, and Job

Mobility: A Critical Review of the Literature, in HEALTH POLICY AND THE UNINSURED 97, 98

(Catherine G. McLaughlin ed., 2004) (reviewing job lock phenomenon); see also Uwe E.

Reinhardt, Employer-Based Health Insurance: A Balance Sheet, 18 HEALTH AFFS. 124, 127

(1999) (“Because the employer-based system ties health insurance to a particular job, it can

induce employees to remain indentured in a detested job simply because it is the sole source

of affordable health coverage.”).

106

See supra Part I.

1206 BOSTON UNIVERSITY LAW REVIEW [Vol. 102:1185

employees in the United States is actually greater than what would be expected

absent the intervention of federal healthcare policy.

Besides the structural effects of the employer-based healthcare model, in

recent years we have seen a more specific set of challenges arise from the power

of employers to choose what kind of healthcare benefits they offer employees.

These judgments turn out not to be driven by economic considerations alone or

information that the employer has about employee preferences. They turn too on

judgments about appropriate medical care.

Among its many new requirements, the ACA provided that employer-

insurance must cover contraceptives for insured women. In Little Sisters of the

Poor Saints Peter & Paul Home v. Pennsylvania,

107

the Supreme Court

permitted the Trump Administration to grant exemptions from that requirement

to employers with moral or religious objections to contraceptive coverage.

108

This decision crystallized the limits to treating employers as neutral agents on

behalf of either the federal government (which subsidizes healthcare by way of

tax policy) or employees themselves. This decision was a culmination of a long-

term political and social movement in favor of recognizing autonomy interests

on the part of business-employers. The Religious Freedom Restoration Act of

1993 allowed individuals, including businesses, whose exercise of religion has

been burdened by a statute to sue.

109

About a decade later, in Burwell v. Hobby

Lobby Stores, Inc.,

110

the Supreme Court held that a contraceptive mandate could

impose a substantial burden on religious exercise for for-profit, closely held

corporations.

111

For some observers, these developments fairly vindicate the

idea that employers have genuine interests in what kind of health benefits they

purchase for their employees.

112

If businesses are the kinds of entities that can

have religious liberty interests, the government can hardly regulate choices that

implicate those religious commitments.

113

But as Elizabeth Sepper has argued, “courts [have] draw[n] incorrect

conclusions about the legal and moral responsibility of employers for the

contents of their employees’ insurance plans.”

114

If employers are independent

moral agents making profound and intimate choices about appropriate medical

care, it is not clear how we can justify their involvement in those choices for

107

140 S. Ct. 2367 (2020).

108

Id. at 2386.

109

42 U.S.C. § 2000bb-1(c).

110

573 U.S. 682 (2014).

111

Id. at 726 (“[T]he mandate clearly imposes a substantial burden on those beliefs.”).

112

See, e.g., Robin Fretwell Wilson, The Calculus of Accommodation: Contraception,

Abortion, Same-Sex Marriage, and Other Clashes Between Religion and the State, 53 B.C. L.

REV. 1417, 1429-32 (2012) (defending religious exemptions).

113

See Elizabeth Sepper, Free Exercise Lochnerism, 115 COLUM. L. REV. 1453, 1456-58

(2015) (describing tenets of business religious liberty movement).

114

Elizabeth Sepper, Contraception and the Birth of Corporate Conscience, 22 AM. U. J.

GENDER SOC. POL’Y & L. 303, 304-05 (2014) (“Employee use of benefits no more burdens

employers than does their use of wages.”).

2022] LOWERING THE STAKES OF EMPLOYMENT CONTRACT 1207

other moral agents. Even if employers do have some kind of interest in which

benefits they proximately fund, their interest ought not to override the rights of

employees to make their own heath care choices.

115

The effect of first delegating

to employers the task of insuring employees and then deferring to their deeply

personal judgments about what kind of insurance to buy has been to vault

employers to almost lord-like status over employees.

116

Our unusual reliance on

employees to manage the social need for healthcare—a social policy problem

that every developed country has answered in some form—has made employees

dependent on employers for access to particular kinds of medical care. The

stakes of the employment contract are higher under this model than Esping-

Andersen ever contemplated.

C. Vaccine Mandates

My two examples of artificially elevated stakes of employment have

concerned the allocation of social entitlements: rights to speech and rights to

healthcare. In these contexts, employers operate as gatekeepers that control

individual employees’ access to public discourse and healthcare, respectively.

More recently, vaccine mandates have raised the specter of employers

controlling another kind of fundamental social allocation: the burdens of

responding to a national public health crisis.

The United States may be at one extreme in this respect, but its policies are

not so aberrant in the spectrum of industrialized democracies as are its policies

in respect of speech and healthcare. While there are few countries in which the

state has directly imposed vaccine mandates on the population, in most

countries, mandates are tied in some way to employment.

117

Here, we go further,

and employers not only manage the mandates but are responsible for their

enforcement.

Employer mandates preceded any governmental orders that required them.

The federal Equal Employment Opportunity Commission issued guidance early

on in the pandemic to the effect that employers were permitted to require their

employees to be vaccinated as a condition of employment as long as employers

made reasonable accommodations for disabilities and religion.

118

Private

115

See Leo E. Strine, Jr., A Job Is Not a Hobby: The Judicial Revival of Corporate

Paternalism and Its Problematic Implications, 41 J. CORP. L. 71, 75 (2015) (criticizing Hobby

Lobby’s extension of employers’ rights “beyond the workplace and into the employee’s

doctor’s office”).

116

Id. at 78 (“Corporate paternalism was not an incidental aspect of the scheme: paying

workers in scrip, and controlling where they could live, enabled employers to police all

aspects of their workers’ lives.”).

117

Robert Hart, Europe Kicks Off Some of the World’s Most Sweeping Vaccine Mandates

and Fines, FORBES (Jan. 17, 2022, 9:52 AM), https://www.forbes.com/sites/roberthart/2022

/01/17/europe-kicks-off-some-of-the-worlds-most-sweeping-vaccine-mandates-and-fines/.

118

What You Should Know About COVID-19 and the ADA, the Rehabilitation Act, and

Other EEO Laws, U.S. EEOC, https://www.eeoc.gov/wysk/what-you-should-know-about-

1208 BOSTON UNIVERSITY LAW REVIEW [Vol. 102:1185

initiative proved inadequate; thus, local, state, and federal governments began

imposing vaccine mandates on employees in particular sectors, especially the

healthcare sector.

119

These culminated in the most sweeping mandate of all, a

federal executive order requiring that all large employers impose a vaccine

mandate on their employees unless those employees would not have exposure

to other employees.

120

Most targeted vaccine mandates at the state level have

survived legal challenge.

121

On the federal level, the Supreme Court recently

struck down the Biden order as an unauthorized exercise of agency power.

122

The question that vaccine mandates raise for this discussion is not whether

they are a good idea as a general matter but whether the requirement to which

employees are subject should come from employers or from a governmental

body. Governments appear reluctant to directly order people to do anything with

their bodies. Vaccine mandates indeed implicate a fundamental liberty interest

and the specter of government coercion here is alarming, not unlike the manner

in which the specter of government regulation of speech offends the core

anxieties of the First Amendment. The problem is that relying on employers to

regulate vaccination does not make the coercive element of the mandate go

away. It merely switches the power behind the mandate. If vaccines mandates

are justified on public health grounds, they must be justified whether the

mandate comes most immediately from an employer or the government. The

prospect of individual employees avoiding the coercive effect of an employer

mandate by leaving their jobs promotes the illusion of free choice. If it is wrong

to force people to be vaccinated, it is wrong to force them to forego their

livelihoods to avoid vaccination. But if it is appropriate in an emergency to

require vaccination, then the public reasons that justify vaccination justify the

invocation of public authority to back the mandate. Relying on employers to

pursue the public objective expands the contractual authority of employers over

employees in an unprecedented and insidious way. There is a high bar before

citizens can demand vaccination of each other, but it is a demand we can make

covid-19-and-ada-rehabilitation-act-and-other-eeo-laws [https://perma.cc/43DT-Y6NV] (last

updated Mar. 14, 2022) (“The federal EEO laws do not prevent an employer from requiring

all employees physically entering the workplace to be fully vaccinated against COVID-19,

subject to the reasonable accommodation provisions of Title VII and the ADA and other EEO

considerations . . . .” (citation omitted)).

119

See, e.g., Tiernan, supra note 7.

120

Fact Sheet: Biden Administration Announces Details of Two Major Vaccination

Policies, WHITE HOUSE (Nov. 4, 2021), https://www.whitehouse.gov/briefing-

room/statements-releases/2021/11/04/fact-sheet-biden-administration-announces-details-of-

two-major-vaccination-policies/ [https://perma.cc/4VJU-6GLY] (“All covered employers

must ensure that their employees have received the necessary shots to be fully vaccinated –

either two doses of Pfizer or Moderna, or one dose of Johnson & Johnson – by January 4th.”).

121

See, e.g., We the Patriots USA, Inc. v. Hochul, 17 F.4th 266, 296 (2d Cir.), opinion

clarified, 17 F.4th 368 (2d Cir. 2021); Doe v. Mills, No. 1:21-cv-00242, 2021 WL 4783626,

at *17-18 (D. Me. Oct. 13, 2021).

122

Nat’l Fed’n of Indep. Bus. v. Dep’t of Lab., OSHA, No. 21A244, slip op. at 1-2 (Jan.

13, 2022) (per curiam) (holding mandate exceeded OSHA’s statutory authority).

2022] LOWERING THE STAKES OF EMPLOYMENT CONTRACT 1209

under exigent circumstances. It is still more rarely the kind of demand employers

should have the power to make on us.

III. DISTINGUISHING CONTRACTUAL AND POLITICAL AUTHORITY

Why does it matter whether employers decide whether to provide certain

goods or impose certain burdens if the alternative is that the state will decide?

We might think that what really matters is the substance of the entitlements and

burdens that individuals bear. We might even think that it is in principle better

for important value judgments to be made privately rather than by the state.

I will argue to the contrary that the role of employers in allocating entitlements

and burdens is significant at multiple levels. First, the substance of who gets

what can be expected to be different when employers play a mediating role.

Second, the source of the authority behind an act of power is not only important

to the legitimacy of that act but, in a political community, important to the

legitimacy of political authority. Allocative judgments that are constrained by

principles of liberty and/or principles of distributive justice should be made by

public authorities, or if delegated, remain regulated to ensure adherence to those

public principles.

A. Who Gets What

Most goods are allocated by employers as a set on the basis of features of

individual employers and of individual employees. In the case of healthcare, for

example, we would expect the kind of healthcare package offered by a given

employer (including, possibly, no healthcare package, if a smaller firm not

subject to the ACA mandate), to turn on the firm’s productivity, its size, the costs

of insuring its employee pool, and its perception of its workers’ preferences for

health insurance as compared to other modes of compensation. The employer

might then charge individual employees differently depending on the

compensation they are prepared to offer that employee, or the expected cost of

insuring them (including the number of their dependents). One fundamental

question raised by the employer model is whether the criteria that drive the

allocation of an entitlement are the right ones. All the criteria named above, with

the possible exception of the expected cost of insurance (especially if driven by

the number of insured), are morally arbitrary. No one deserves better health

insurance either because the firm that employs them is more competitive than a

rival or because the employee herself is more productive than a rival.

123

However, in a market society, almost no goods and services, not least income,

are allocated on the basis of individual moral desert.

124

The money we make in

123

See Ronald Dworkin, Comment on Narveson: In Defense of Equality, 1 SOC. PHIL. &

POL’Y 24, 29 (1983) (rejecting view that people deserve what they earn based on their talents).

124

There is a vast literature discussing the relationship between earnings and desert. See

generally, e.g., FRIEDRICH A. HAYEK, 2 LAW, LEGISLATION AND LIBERTY: THE MIRAGE OF

SOCIAL JUSTICE (1976); JOHN RAWLS, A THEORY OF JUSTICE (1971); Samuel Scheffler, Justice

1210 BOSTON UNIVERSITY LAW REVIEW [Vol. 102:1185

our jobs, like the jobs themselves, reflect facts about the relative demand for our

labor and the particular skills we happen to have, and our basic structure assumes

that we are all better off for allowing differential rewards to labor even if those

differences do not align with desert.

125

However, what we have observed is that the gatekeeper role that employers

play is not fully explained by the general benefits of relying on a market to

allocate social resources.

126

Their role is enhanced by federal policy that pushes

them to play that role. The moral arbitrariness of the criteria that employers use

might be mitigated by specific federal laws that alter their incentives (for

example, by penalizing so-called Cadillac health plans as the ACA originally

did),

127

but expanding the employer’s role in the provision of basic social goods

simultaneously expands the role of morally arbitrary criteria. Moreover, in the

case of intimate goods like healthcare, relying on employers as an intermediary

empowers employers to impose their idiosyncratic private judgments on

employees, raising still further the morally arbitrary consequences of a job.

The problem is present with respect to the other two examples of employment

stakes reviewed here. It might be less apparent that our speech rights depend on

morally arbitrary criteria because those criteria are not the familiar criteria that

drive demand for our labor. Instead, speech rights enjoyed by individuals will

turn on the sensitivity of an employer to the prospects of either legal liability or

retail accountability. A small manufacturing employer with a cohesive and

homogenous workforce might not find it worthwhile to police its employees’

social media presence. If the individual owner of the firm shares the cultural and

political sensibilities of her employees, she may have even less reason to care

what her employees are saying. These employees may enjoy great latitude and

not experience the threat of termination or discipline at work as constraining at

all in the context of speech. By contrast, even a low-level employee of a large

retail firm with a sophisticated human resources department may be subject to a

finely articulated social media policy; she may know that any misstep on social

media could easily be reported to her employer by any ill-wisher and it would

be used as grounds for termination—without severance, without notice, and

without appeal. She has every reason not to speak out about any topic that might

and Desert in Liberal Theory, 88 CALIF. L. REV. 965 (2000); ERIC RAKOWSKI, EQUAL JUSTICE

(1993); DAVID MILLER, PRINCIPLES OF SOCIAL JUSTICE (1999); Richard J. Arneson,

Liberalism, Distributive Subjectivism, and Equal Opportunity for Welfare, 19 PHIL. & PUB.

AFFS. 158, 174-83 (1990); G.A. Cohen, On the Currency of Egalitarian Justice, 99 ETHICS

906, 914 (1989). I adopt here the “egalitarian” view that while earnings may reflect some

desert-relevant criteria, such as effort, earnings do not track desert given the significant role

played by the accidents of birth.

125

See RAWLS, supra note 124, at 75-83 (articulating “difference principle,” which holds

that inequalities in basic resources are only justified to extent they benefit least advantaged).

I do not address here whether it is empirically true that the inequalities we observe in our own

society in fact operate to the benefit of the least advantaged.

126

See supra Part II.

127

I.R.C. § 4980I(e)(1) (2012) (repealed Dec. 20, 2019).

2022] LOWERING THE STAKES OF EMPLOYMENT CONTRACT 1211

upset anyone. These two employees enjoy the same First Amendment

protection, of course. But at least some of the principles of autonomy and free

discourse that animate the First Amendment are undermined by a regime in

which individuals’ speech is regulated so unevenly by private employers, and

on grounds that do not track the criteria that might justify consequences for

wrongful speech acts.

Similarly, the desirability of vaccine mandates should be separated from the

question of whether it is desirable to rely on employers to impose and enforce

them. Not everyone is subject to a vaccine mandate. Most obviously, if the

executive order requiring that large employers impose them had stood,

employees of small businesses outside of certain essential or public-facing

sectors would not have been subject to the mandate.

128

One might think that the

uneven application of the federal mandate cuts in its favor, as employees who

feel strongly against vaccination can gravitate to the employers that are not

subject to the directive. But not all employees are capable of switching jobs in

this way. Within the set of those individuals who are against vaccination,

whether one is effectively coerced into vaccination will depend on her ability to

get a job with an employer that does not impose a vaccine mandate. Even those

of us who would advocate more categorical public mandates recognize the

seriousness of demanding, subject to threat, that a person accept a chemical

injection into their bodies. The idea that people will or will not be subject to this

burden depending on their labor market power—which has nothing to do with

the justification for such a mandate and everything to do with its political costs—

is alarming. Employers are never a simple conduit for government policy. They

are not assigned their gatekeeping role for the sake of administrative

convenience. Their role is substantive. And it skews the all-important matter of

who gets what.

B. Who Decides

Even if employers distributed entitlements and burdens according to optimal

criteria on the merits, the fact that employers decide who gets what can be

problematic in itself. I do not refer here to the social and personal costs of

employer authority per se. Instead, I largely take for granted for purposes of the

present discussion, that employers should control (subject to legal constraints)

basic pay, working hours, and the substance of what employees are asked to do.

When we rely on the market to allocate social resources, we rely on individual

firms to buy and use labor based on their ground-level information about how it

can be most productively utilized. This reliance on the market has harsh

consequences and so we have a range of employment and other protections

intended to limit the human toll of the market as an institution. But I am not

challenging here the power that employers enjoy in principle.

128

COVID-19 Vaccination and Testing; Emergency Temporary Standard, 86 Fed. Reg.

61,402 (Nov. 5, 2021), withdrawn, 87 Fed. Reg. 3,928 (Jan. 16, 2022) (to be codified at 29

C.F.R. pt. 1910).

1212 BOSTON UNIVERSITY LAW REVIEW [Vol. 102:1185

My concern is more narrow. When we collectively allow employers to decide

how much money a person takes home at the end of the week, we do so because

we think that there are social gains to be had from allocating that power to

employers. We have developed a collective habit, however, of allocating power

to employers not because we think they have the best information or incentives

to make decisions that ultimately operate to the public benefit but because the

state does not have the wherewithal to exercise its relevant power directly. We

have adopted an employer-based health insurance model not because we

concluded that it made sense to tie healthcare to employment but because we

have been unable to decouple them. We have not empowered—indeed,

expected—employers to control employee speech because we think they are

especially wise about the optimal boundaries of free speech or because we think

their interests mirror those of the public. Their power has emerged from a regime

incapable of articulating a zone of protected speech, given the fraught nature of

the boundary-drawing exercise. And finally, we have not allocated to employers

the job of getting people vaccinated because infection occurs primarily at work

or because its toll falls primarily on employers. We rely on employers because

the state dares not invoke its own authority; because we regard employers’

power over employees as less controvertible in this time and place.

Once an employer’s authority over an employee is not justified by reference

to a theory about market efficiency, however, it is not justified except by virtue

of the immediate consent of the employee. The employer’s authority is then

nothing but contractual authority. Contractual authority, though, is contingently

rather than deeply justified. That is, when A agrees to be subject to some power

of B through contract, the legitimacy of the authority B gains through contract

depends on the legitimacy of background institutions, and in particular, the

initial entitlements that led A and B to contract on the terms they ended up with.

In any individual case, we rarely have reason to assess the legitimacy of the

backdrop to a contract, such as any given employment contract. If the market

power that B has ex ante is legal, she (mostly) may extract what she will with it

from other market actors. But when a type of contract term is not specific to any

two parties but is a patterned response to background legal incentives, we should

ask whether and why we should set up so many As to the sweeping authority of

Bs. The fact that any one individual agrees to this arrangement does not justify

the social deference to contract as the mechanism by which to settle the matter

covered by that term. This is all the more so because, once the matter is “settled”

by contract by way of an allocation of discretion, the resulting allocation across

employers is opaque and difficult to reconstruct. Both for the individual

employee and from the standpoint of the public, there is no accountability for

the allocations made by “employers” as a set when the social outcome is the

unplanned result of millions of micro decisions. Again, for certain kinds of

outcomes, we have an affirmative theory—traditional economic theory—that

leads us to be optimistic about aggregate outcomes. This is not true when the

matter (of healthcare or speech access, etc.) is only within the realm of the

employment contract by virtue of legal interventions.

2022] LOWERING THE STAKES OF EMPLOYMENT CONTRACT 1213

Unlike contractual authority, which is literally purchased and reflects relative

market power, political authority in a well-organized state is deeply justified.

Political authority in a democracy has higher pedigree than other kinds of

contingent authority. The judgments of the state about what to require of citizens

are often wrong, and rightly subject to constitutional limitation, but they are as

close as we can get to autonomous decision-making because they are literally

the requirements we collectively impose on ourselves. This Essay cannot fully

defend the justice of political authority, either in general or in the United States

today. I invoke instead the prevailing intuition, held on the basis of widely

disparate theories of political authority, that democratic governments have the

duty to provide and protect certain social goods; the right to impose burdens as

necessary to fulfill that affirmative duty, and; further, that it is the responsibility

of state actors to make hard judgments about these open-ended questions even

though they will always get them wrong in the eyes of many people.

If an employer’s prerogatives are too capacious, the employer is lordly. The

employment relationship no longer resembles an economic exchange because its

terms are not explicable by reference to the economic dynamic of the

relationship. Because the state has handed employers the task of making

allocative decisions for which the state should bear responsibility, employers

dominate employees more completely than even their market power alone would

enable. This is the opposite of what the state’s role in the employment

relationship should be.

CONCLUSION

My purpose in this Essay has been primarily to elucidate the way in which

political choices, or perhaps, political inertness, has raised the stakes of the

employment contract. We tend to justify the morally arbitrary differential

rewards of employment by reference to the function of the labor market in

helping to allocate social resources efficiently. We accept the moral costs of

inequality and hope tax and transfer regimes will mitigate its injustice. But

relying on employment to regulate the allocation of both goods and even burdens

in society has become a political strategy of deflection. It is an alternative to

making hard political choices about who gets what. The effect has been to raise

the stakes of employment even beyond the scope of those terms and conditions

that relate to attracting and extracting productive labor. To lower the stakes of

employment, we would need to take steps seemingly distant from the sphere of

employment narrowly construed. We would need to revitalize political discourse

to the point where we can rely on politics to deliver authoritative social

judgments. We would then no longer need to delegate those judgments to private

employers that lack the moral authority to make those judgments for the rest of

us.