The Hamilton Project • Brookings 1
Seth D. Harris and Alan B. Krueger
DISCUSSION PAPER 2015-10 | DECEMBER 2015
A Proposal for Modernizing Labor Laws for
Twenty-First-Century Work: The “Independent Worker”
2 Informing Students about Their College Options: A Proposal for Broadening the Expanding College Opportunities Project
MISSION STATEMENT
ADVISORY COUNCIL
MISSION STATEMENT
The Hamilton Project seeks to advance America’s promise
of opportunity, prosperity, and growth.
We believe that today’s increasingly competitive global economy
demands public policy ideas commensurate with the challenges
of the 21st Century. The Project’s economic strategy reflects a
judgment that long-term prosperity is best achieved by fostering
economic growth and broad participation in that growth, by
enhancing individual economic security, and by embracing a role
for effective government in making needed public investments.
Our strategy calls for combining public investment, a secure social
safety net, and fiscal discipline. In that framework, the Project
puts forward innovative proposals from leading economic thinkers
— based on credible evidence and experience, not ideology or
doctrine — to introduce new and effective policy options into the
national debate.
The Project is named after Alexander Hamilton, the nation’s
first Treasury Secretary, who laid the foundation for the modern
American economy. Hamilton stood for sound fiscal policy,
believed that broad-based opportunity for advancement would
drive American economic growth, and recognized that “prudent
aids and encouragements on the part of government” are
necessary to enhance and guide market forces. The guiding
principles of the Project remain consistent with these views.
A Proposal for Modernizing Labor Laws
for Twenty-First-Century Work:
The “Independent Worker”
Seth D. Harris
Cornell University
Alan B. Krueger
Princeton University
DECEMBER 2015
NOTE: is discussion paper is a proposal from the authors. As emphasized in e Hamilton Project’s
original strategy paper, the Project was designed in part to provide a forum for leading thinkers across the
nation to put forward innovative and potentially important economic policy ideas that share the Project’s
broad goals of promoting economic growth, broad-based participation in growth, and economic security.
e authors are invited to express their own ideas in discussion papers, whether or not the Project’s sta or
advisory council agrees with the specic proposals. is discussion paper is oered in that spirit.
The Hamilton Project • Brookings 1
Abstract
New and emerging work relationships arising in the “online gig economy” do not t easily into the existing legal denitions of
“employee” and “independent contractor” status. e distinction is important because employees qualify for a range of legally
mandated benets and protections that are not available to independent contractors, such as the right to organize and bargain
collectively, workers’ compensation insurance coverage, and overtime compensation. is paper proposes a new legal category,
which we call “independent workers,” for those who occupy the gray area between employees and independent contractors.
Independent workers typically work with intermediaries who match workers to customers. e independent worker and
the intermediary have some elements of the arms-length independent business relationships that characterize “independent
contractor” status, and some elements of a traditional employee-employer relationship. On the one hand, independent workers
have the ability to choose when to work, and whether to work at all. ey may work with multiple intermediaries simultaneously,
or conduct personal tasks while they are working with an intermediary. It is thus impossible in many circumstances to attribute
independent workers’ work hours to any employer. In this critical respect, independent workers are similar to independent
businesses. On the other hand, the intermediary retains some control over the way independent workers perform their work,
such as by setting their fees or fee caps, and they may “re” workers by prohibiting them from using their service. In these
respects, independent workers are similar to traditional employees.
Evidence is presented suggesting that about 600,000 workers, or 0.4 percent of total U.S. employment, work with an online
intermediary in the gig economy. Although there are probably many more workers who currently work with an oine intermediary
who would qualify for independent worker status than there are who work with an online intermediary, the number of workers
participating in the online gig economy is growing very rapidly.
In our proposal, independent workers — regardless of whether they work through an online or oine intermediary — would
qualify for many, although not all, of the benets and protections that employees receive, including the freedom to organize
and collectively bargain, civil rights protections, tax withholding, and employer contributions for payroll taxes. Because it is
conceptually impossible to attribute their work hours to any single intermediary, however, independent workers would not
qualify for hours-based benets, including overtime or minimum wage requirements. Further, because independent workers
would rarely, if ever, qualify for unemployment insurance benets given the discretion they have to choose whether to work
through an intermediary, they would not be covered by the program or be required to contribute taxes to fund that program.
However, intermediaries would be permitted to pool independent workers for purposes of purchasing and providing insurance
and other benets at lower cost and higher quality without the risk that their relationship will be transformed into an employment
relationship.
Our proposal seeks to structure benets to make independent worker status neutral when compared with employee status, as well
as to enhance the eciency of the operation of the labor market. By extending many of the legal benets and protections found in
employment relationships to independent workers, our proposal would protect and extend the social compact between workers
and employers, and reduce the legal uncertainty and legal costs that currently beset many independent worker relationships.
2 A Proposal for Modernizing Labor Laws for Twenty-First-Century Work: The “Independent Worker”
Table of Contents
ABSTRACT 2
CHAPTER 1. INTRODUCTION 5
CHAPTER 2. CHALLENGES AND BACKGROUND 6
CHAPTER 3. PRINCIPLES OF A NEW WORKER CLASSIFICATION 13
CHAPTER 4. LEGAL REFORM FOR “INDEPENDENT WORKERS” 15
CHAPTER 5. ARE INDEPENDENT WORKERS DIFFERENT FROM OTHER
THIRD-PARTY PLAYERS IN LABOR MARKETS? 22
CHAPTER 6. ECONOMIC ANALYSIS OF THE PROPOSAL 27
CHAPTER 7. CONCLUSION 33
CHAPTER 8. APPENDIX 28
AUTHORS, ACKNOWLEDGMENTS AND DISCLAIMER 34
ENDNOTES 35
REFERENCES 36
The Hamilton Project • Brookings 3
4 A Proposal for Modernizing Labor Laws for Twenty-First-Century Work: The “Independent Worker”
N
ew and emerging work relationships arising in
the “online gig economy” do not t the existing
legal denitions of “employee” and “independent
contractor” status. ese denitions determine which workers
are required to receive certain protections and benets from
their employers. Under the current legal framework, the
workers and intermediaries with whom they work in these
emerging relationships face unnecessary and excessive
uncertainty regarding a range of legal protections and benets
that employees receive. Legal uncertainty creates ineciencies
for all parties concerned. In addition, work-related benets
that could prove valuable to both workers and businesses—
such as intermediaries using their size and pooling advantages
to purchase low-cost life insurance for the independent
workers they engage—are sometimes eschewed to reduce the
risk that the law will impose an employment relationship and
the corresponding legal obligations because of the provision of
these benets. As a result, the emergence of new forms of work
that could benet workers, businesses, and consumers could
be slowed, or even stopped, by a legal regime for classifying
workers that does not accommodate these innovative
arrangements. A further concern with the current legal
framework is that companies working online and oine to
match workers to nal customers could organize work in such
a way as to classify jobs that were traditionally performed by
employees into independent contractor relationships to avoid
providing employees with benets that are a crucial part of the
social compact.
To address these problems, we propose a new legal category of
workers, which we call “independent workers,” who occupy a
middle ground between traditional employees and independent
contractors. An archetypal example of independent workers
is for-hire drivers who work on the Ly or Uber platforms.
We refer to these companies, and others like them, as
“intermediaries” because they are the intermediary between
the independent worker and the ultimate customer. ese
independent worker arrangements bear some similarities to
independent contractors and some similarities to traditional
employees. On the one hand, the drivers can choose when and
whether to work, similar to independent contractors, but on
the other hand, drivers face restrictions that are imposed by
the intermediary on how much they charge customers. Other
online intermediaries that utilize a similar model include
TaskRabbit (for a variety of tasks) and Mechanical Turk (for
tasks completed online).
Technology is creating exciting new opportunities to link
workers who provide services directly to customers, with
potentially large gains in the quality, speed, and eciency of
service. From an economic and societal perspective, however,
it is important that, if these new intermediaries are to succeed
and expand, it is a result of their superior technology, eciency,
or service, not because their technology or business model
enables regulatory arbitrage. For example, if an intermediary
succeeds by displacing traditional employers who oer the
same service because the intermediary gains a cost advantage
by avoiding provision of certain legally mandated benets and
protections, then welfare is reduced by the innovation.
Below, we propose that Congress and, where appropriate,
state legislatures, enact legislation to dene and establish a
third legal category of workers: independent workers. is
legislation would clearly dene the protections and benets
that intermediaries would be required to provide to the workers
with whom they conduct business. ese protections and
benets would approximate the social compact guaranteed to
employees, albeit with important dierences that reect the
substantive distinctions between employment relationships
and independent worker–intermediary relationships. In
craing this legislation, Congress should abide by a set of
governing principles to identify these workers; we describe
those principles below. We also provide an analysis of the
size, growth, and business models used by an emerging set of
online intermediaries.
Chapter 1. Introduction
The Hamilton Project • Brookings 5
I
t is our view that labor and employment law has evolved
over time in the United States to reect a social compact
between employees and employers. is social compact
represents a synthesis between the desire to enhance the
eciency of the operation of the labor market (e.g., to overcome
information asymmetries and imperfections) and to ensure
that the employment relationship treats workers fairly in light
of the unequal bargaining power that typies most employee–
employer relationships. is social compact has served the
United States well and, in our view, should be preserved and
protected unless there are compelling reasons to alter it.
But workers participating in the growing online “gig economy”
are at risk of being excluded from this social compact. ese
are the workers who use an Internet-based app created by an
intermediary that matches customers to workers who will
perform personal services. Independent workers do not t into
either of the two legal statuses currently available under U.S.
labor, employment, and tax law: employees or independent
contractors. As noted, such workers have some similarities to
independent contractors and some similarities to traditional
employees. We oer a fuller discussion of these similarities
and dierences below. e resulting ambiguity in these
workers’ legal status leads to uncertainty and ineciency in
the labor market that are harmful to both the workers and the
intermediaries in several ways.
First, determining whether workers in the online gig economy
are employees or independent contractors will require, and
can be expected to continue to require, long, costly and
uncertain legal battles.
1
Some Western economies (e.g., Czech
Republic, Estonia, France [in selected circumstances], Mexico,
e Netherlands, Portugal) have statutory presumptions that
essentially establish “employee” status as a default condition
(OECD 2014). Absent a rebuttal of the presumption, there
is no uncertainty regarding a worker’s status. ere is no
default status in U.S. law, however. e resulting uncertainty
is costly to workers, who do not know the benets that
they will ultimately qualify for, and to intermediaries, who
face uncertain costs. Both parties face the prospect of high
transaction costs resulting from litigation or government
enforcement interventions.
Second, current labor and employment laws are not
harmonized or applied consistently. Workers and employers
must confront dierent tests across statutes for employee
status and independent contractor status. ese tests and
courts’ interpretations vary across statutes because the core
purposes of those statutes vary (e.g., tax law serves a dierent
purpose from occupational safety and health law). So, a
statute’s scope of coverage should be expected to best serve
that law’s purpose. Nonetheless, the classication of workers
as employees or independent contractors requires analysis of
several dierent tests that, at least theoretically, could lead
to dierent results. For example, a worker might be deemed
entitled to the minimum wage, but not to have her employer
pay half of her payroll taxes.
An even greater risk comes from the fact that these tests
are collections of factors for consideration rather than clear
thresholds or required elements. Labels applied in contracts
are irrelevant. Courts and administrative agencies oen warn
that no single factor governs, and the weighing of factors is
oen le to individual decision makers. As a practical matter,
in too many cases conclusions are driven by a predetermined
desired outcome rather than by objective analysis. As a result,
similarly situated workers, such as truck drivers, could be
employees under a statute in one jurisdiction, but independent
contractors under the same statute in a dierent jurisdiction.
2
Because they occupy a middle ground between employees
and independent contractors, independent workers and the
intermediaries with which they work are especially vexed by
this ambiguous system. As noted above, independent workers
satisfy dierent factors of both the employee and independent
contractor tests under most labor, tax, and employment
laws. Will courts and administrative agencies classify them
consistently across laws and jurisdictions? Can independent
workers and intermediaries predict how they will be treated
when the legal dust settles? is risk and uncertainty creates a
barrier to the continuation and creation of relationships that
can be benecial to all parties involved.
ird, many independent workers who are classied as
independent contractors may not have the means to secure
many of the protections and benets that are available to
traditional employees. Independent workers also face barriers
Chapter 2. Challenges and Background
6 A Proposal for Modernizing Labor Laws for Twenty-First-Century Work: The “Independent Worker”
to “pooling” that would increase their bargaining power
both in dealings with their intermediaries and in markets
for fringe benets that could provide them with many of the
same benets and protections that are legally mandated for
employees.
Finally, some employers may reorganize their work to classify
employees as independent contractors to avoid providing
required benets and protections under the social compact and
to gain an unfair advantage over their competitors. Equally
troubling, the uncertainty in this dichotomous classication
system facilitates both intentional and unintentional
misclassication of workers by employers, usually in the
direction of independent contractor status that deprives
workers of many important legal protections and benets.
THE “EMPLOYEE” VS. “INDEPENDENT
CONTRACTOR” DICHOTOMY
e dierence between the status of employees and independent
contractors is more than an issue of nomenclature. A sizable list
of protections and benets are at stake, depending on how the
relationship is classied. It is worth reviewing what is at stake.
Employees benet from contracts with their employers
that include signicant substantive terms that are imposed
by law. In essence, employees agree to be economically
dependent on their employers by relinquishing control over
many aspects of their work lives (and, to some extent, their
economic futures) and, in return, employers must provide
workers with a degree of economic security. Myriad laws at
the federal and state level require employers to pay employees
at least the minimum wage and overtime premium pay;
refrain from discriminating in hiring, ring, and the terms
and conditions of employment on the basis of race, sex, and
other selected personal characteristics; maintain safe and
healthy workplaces; contribute toward the payroll taxes that
make employees eligible for unemployment insurance, Social
Security, Disability Insurance, and Medicare; and provide
workers’ compensation insurance, among other protections.
Under the Patient Protection and Aordable Care Act (ACA),
many employers also will be required to provide employees
with health insurance or pay a penalty if they do not. Finally,
the Employee Retirement Income Security Act (ERISA)
requires covered employers to satisfy certain requirements if
they provide employees with a retirement savings plan.
Various laws apply tests to identify employees and their
employers who are covered by some or all of this social compact.
Key features of the determination of employee status include
the likelihood that the employment relationships will continue
indenitely, or at least beyond the completion of a given task,
even if only for a specied term, and whether the employer gives
the worker instructions about how to do the work. Employees are
also expected to have little control over their work hours, unless
their employers delegate such control to them. Table 1 provides
a summary of how the Fair Labor Standards Act (FLSA),
Internal Revenue Code (IRC), common law, and selected other
employment laws determine employee status. ese tests are
an imperfect and increasingly outdated means for determining
eligibility for coverage under the social compact.
Independent contractors, in contrast to employees, do
not relinquish control over their economic lives to others.
Generally speaking, they are independent businesses working
with multiple other businesses or clients without signicant
limitations, except those to which they may agree by contract
or laws that may pertain to businesses in their sector. Typically,
these relationships are not expected to last beyond the
completion of a particular task, activity, or deadline. In the past,
independent contractors have operated more at the periphery
of others’ businesses rather than performing more-integral
work, but the ssuring of work and business relationships and
the increasingly complex supply chains that have developed
over the past several decades in some industries have made this
consideration less important (Weil 2014).
Independent contractors control the methods and means of
the work they perform for others, make signicant capital
investments, possibly employ others, and retain the opportunity
for prot or loss. For these reasons, independent contractors are
expected to have some bargaining power—even if it is not equal
bargaining power—that allows them to enter into successful
arms-length contracts with other businesses and clients.
Existing law wrongly implies that employees and independent
contractors occupy the entire eld of work relationships in the
U.S. economy. is dichotomy is a vestige of the early law of
“masters” and “servants” that is as archaic as the words suggest.
Newly emerging “independent workers” participate in new
kinds of work relationships that occupy a space between these
two statuses.
Other countries have not clung to a dichotomous employee–
independent contractor categorization of work relationships.
Both Canada and Germany, for example, recognize a
“dependent contractor” status for some independent
contractors. is status becomes relevant when a contractor
has formed an essentially exclusive relationship (80% being
a “rule of thumb” for “exclusive” in Canada) over a lengthy
period of time with one client such that the contractor
is economically dependent on the continuation of that
relationship. In some Canadian provinces these dependent
contractors are treated like employees, at least with respect to
termination notications and eligibility for union membership
(Kennedy 2014). While dependent contractor status illustrates
that there is room for more than two legal statuses in the world
of work, it is worth noting that the dependent contractor
concept does not accurately correspond to the relationship
between intermediaries and independent workers because
The Hamilton Project • Brookings 7
independent workers typically have only eeting relationships
with their nal customers. For this reason, we see no evidence
that Canada, for example, has sought to apply the status to its
intermediaries and independent workers.
THE GRAY AREA
e heart of the challenge for independent workers is that they
do not resemble independent contractors or employees with
respect to their most fundamental characteristics. Independent
workers typically have little individual bargaining power and,
as a result, do not have the ability to negotiate contracts with
either intermediaries or their ultimate customers that could
secure for them the protections and benets that are available
to employees. ey are not true independent businesspeople in
that they do not have freedom to negotiate their compensation
or terms of service. But their relationships with intermediaries
are not so dependent, deep, extensive, or long lasting that we
should ask these intermediaries to assume responsibility for all
aspects of independent workers’ economic security. ey are
not true employees. us, the existing employee–independent
contractor dichotomy does not oer a satisfying or reliable
path in these new and emerging circumstances.
Forcing these new forms of work into a traditional employment
relationship could be an existential threat to the emergence
of online-intermediated work, with adverse consequences
for workers, consumers, businesses, and the economy. At the
same time, relying on the existing employee–independent
contractor dichotomy to classify workers whose circumstances
do not easily t either denition risks depriving those workers
of any benets or protections of the social compact, and risks
TABLE 1.
Denitions of “Employee” Under Selected Statutes
Role of
work:
Is the work
performed
integral to the
employer’s
business?
Skills
Involved:
Is the
work not
necessarily
dependent on
special skills?
Investment:
Does the
employer
provide the
necessary
tools and/or
equipment
and bear the
risk of loss
from those
investments?
Independent
Business
Judgment:
Has the
worker
withdrawn
from the
competitive
market to
work for the
employer?
Duration:
Does the
worker have
a permanent
or indefinite
relationship
with the
employer?
Control:
Does the
employer set
pay amount,
work hours,
and manner in
which work is
performed?
Benefits:
Does the
worker
receive
insurance,
pension plan,
sick days,
or other
benefits that
suggest an
employment
relationship?
Method of
Payment:
Does the
worker
receive a
guaranteed
wage or
salary as
opposed to a
fee per task?
Intent:
Do the parties
believe they
have created
a employer–
employee
relationship?
Fair Labor
Standards
Act
(Centered
on degree
of economic
dependence
on employer)
YES YES YES YES YES YES N/A N/A N/A
Internal
Revenue
Code (IRC)
(Centered
on control)
YES
1
YES YES
2
YES
3
YES YES YES YES N/A
Nationwide
Mut. Ins. v.
Darden
(ERISA and
other laws)4
YES YES YES N/A YES YES YES YES N/A
Common Law
(From
Restatement
Second of
Agency § 220)
YES YES YES YES YES YES N/A YES YES
Note: “Yes” contributes to a conclusion that the worker is an “employee”; “N/A” indicates the factor is not considered under the specified law.
1
The IRS looks at the role of the work as an indicator of control – if the work is “key” to employer’s business, the employer will likely have the right to direct or to control the work)
2
The IRS also specifically looks at whether the worker has a high degree of unreimbursed expenses.
3
The IRC does not use “business judgment” as a term, but does ask if the worker’s services are available to the market directly.
4
The Supreme Court draws its multi-factor test from Nationwide Mutual Ins. Co. v. Darden, 503 U.S. 318 (1992).
8 A Proposal for Modernizing Labor Laws for Twenty-First-Century Work: The “Independent Worker”
the erosion of the social compact for employees. If the dual
goals of labor and employment law are increased eciency
and protection of workers from the consequences of unequal
bargaining power, then the status quo serves neither goal in
the case of independent workers.
INDEPENDENT WORKERS
Independent workers operate in a triangular relationship:
they provide services to customers identied with the help of
intermediaries. e intermediaries create a communications
channel, typically an “app,” that customers use to identify
themselves as needing a service—for example, a car ride,
landscaping services, or food delivery. (An intermediary need
not utilize the Internet to match independent workers and
customers, but we initially focus on online intermediaries
because they have the greatest potential to disrupt working
relationships.) e intermediaries’ apps allow independent
workers to select which customers they would like to serve. e
intermediary does not assign the customer to the independent
worker; rather, the independent worker chooses or declines
to serve the customer (sometimes within broadly dened
limits). However, the intermediary may set certain threshold
requirements for independent workers who are eligible to use
its app, such as criminal background checks. e intermediary
may also set the price (or at least an upper bound on the price)
for the service provided by independent workers through its
app. But the intermediary exercises no further control over
how and whether a particular independent worker will serve
a particular customer. e intermediary is typically rewarded
for its services with a predetermined percentage of the fee paid
by the customer to the independent worker.
e best known of these kinds of triangular relationships are
drivers providing ride-sharing services to customers brokered
through online apps provided by Uber and Ly. Somewhat
less famous are the independent workers doing odd jobs like
landscaping, maid service, home repairs, and other tasks for
customers using similar apps provided by intermediaries such
as Taskrabbit and umbtack. ere are several especially
important aspects of these triangular relationships. First, the
independent worker provides personal services only when
she chooses to do so. e intermediary plays no role in this
decision. is is qualitatively dierent from an employment
relationship, even one in which a worker may be allowed to
work from home or to choose exible working hours. e
independent worker chooses when and whether to work at all.
e relationship can be eeting, occasional, or constant, at the
discretion of the independent worker.
BOX 1.
An Example
A simple hypothetical example illustrates a representative challenge posed by the existing employee–independent contractor
dichotomy. Imagine an independent worker driving around her city in her car. She has apps for Uber and Ly open on
separate electronic devices. She is waiting for a customer who is seeking a ride from the area in which she is driving to
another part of the city. Two questions arise: (1) Should the driver be compensated for this waiting time? And, if so, (2) who
should compensate her?
Under existing FLSA doctrine, and assuming the driver is an employee, whether the driver’s waiting time constitutes
compensable work hours turns on the question of whether the driver is “waiting to be engaged” or “engaged to wait.” is
distinction, in turn, depends on whether the driver can use the waiting time for her own personal purposes. If she can,
she is waiting to be engaged and does not qualify to be paid for the waiting time. If the employer controls the employee’s
movement during the waiting time, or there is too little time available for personal activity, the employee is engaged to wait
and entitled to compensation.
In this context, it seems the better argument is that the driver is waiting to be engaged. She can turn o the apps at will and
go to her traditional job, undertake another moneymaking activity, drive her children to school, or park by the side of the
road and take a nap. Even if she does not turn o either app, she is not obligated to pick up any particular customer. She can
wait for a customer of her choosing, or until aer she has completed her personal activities, whatever they might be.
Let’s assume for purposes of reaching the second question, however, that this legal conclusion is incorrect and that the driver
is deemed to be engaged to wait. She has two apps open: one for Uber and one for Ly. Who should pay the driver for this
waiting time? Both Uber and Ly? Whichever intermediary oers the ride that the driver ultimately accepts? Whichever
intermediary oers the most rides to this driver during that day? We should not pretend that existing FLSA doctrine
answers this question since there is no analogy in the employee–employer relationship to a driver with two simultaneously
open apps for dierent services. is situation is not joint employment. If anything, Uber and Ly are competitors for the
driver’s services, not co-employers. e best legal answer seems to be that there is not a good answer.
The Hamilton Project • Brookings 9
e independent worker may oer her services through multiple
intermediaries, or combine working with intermediaries and
employment with a traditional employer. Like traditional
employees, independent workers are integral to the business
of the intermediary. e intermediary’s business lives or dies
by the provision of services by independent workers. Ly
would not exist if no drivers were willing to provide car ride
services through the Ly platform.
ese relationships do not t neatly into the employee–
independent contractor dichotomy. Independent workers are
not employees for the following reasons: they do not make
themselves economically dependent on any single employer,
they do not have an indenite relationship with any employer,
and they do not relinquish control over their work hours or
the opportunity for prot or loss. Independent workers are not
independent contractors because some aspects of the methods
and means of work—including the price of their services—are
controlled by the intermediary and because they are integral to
the business of the intermediary. Independent workers are, in
some respects, like individuals working for others, and in other
respects are like independent businesses (e.g., they use their own
equipment and control their own hours). Hence we propose a
new legal and economic category of independent workers.
BOX 2.
The Scope and Growth of the Online Gig Economy
ere has been much speculation about the size and growth rate of the gig economy. We are particularly interested
in the number of workers participating in the “online gig economy,” because this sector is growing rapidly and oen
involves workers that fall in the gray area between employees and independent contractors. As we are dening it, the
online gig economy involves the use of an Internet-based app to match customers to workers who perform discrete
personal tasks, such as driving a passenger from point A to point B, or delivering a meal to a customer’s house. Note
that this denition excludes intermediaries that facilitate the sale of goods and impersonal services to customers, such
as TeacherPayTeachers.com, a Web site where teachers sell lesson plans and other nonpersonal services to other teachers,
and Etsy.com, a Web site where individuals sell handmade or vintage goods. It also excludes Airbnb, a Web site where
people can rent apartments, houses, and other accommodations.
Uber (left scale) ChaCha Fiverr GrubHub Handy
Mechanical Turk/
Amazon Turk
Sidecar
TaskRabbit Thumbtack Upwork/
Elance/oDesk
Lyft
Other
0
2
4
6
8
10
12
14
16
18
0
10
20
30
40
50
60
70
80
90
Index (Relative to Uber)
Index (Uber)
Jan. 2004 Jan. 2006 Jan. 2008 Jan. 2010
Jan. 2012 Jan. 2014 Jan. 2016
FIGURE 1.
Google Trends: Four-Week Moving Average of Web Searches
Source: Google Trends analysis by authors.
Notes: “Other” includes agentanything, axiom law, clickworker, Eden McCallum, Gengo, Gocurb/TaxiMagic, hourly nerd, Instacart, medicast, Red Beacon,
Samasource, Shyp, Skillshare, trycaviar/caviar delivery, and Washio. Search period spans January 1, 2015- November 7, 2015. Google Trends normalizes the data for
each term specified relative to the total number of Google searches conducted in that week, so that the time period with the most searches for Uber equals 100.
10 A Proposal for Modernizing Labor Laws for Twenty-First-Century Work: The “Independent Worker”
e Appendix table lists 26 prominent companies that act as intermediaries in the online gig economy, the types of
services that they supply, and the nature of their business arrangements with workers. is list is meant to illustrate
examples of emerging opportunities in the gig economy, and not to necessarily classify their workers. We are not
advocating that every worker engaged with every intermediary in the online gig economy should be classied as an
independent worker. It is quite common that these businesses compensate workers who utilize their app on a commission
basis, with commissions taken by the intermediary typically in the 10 to 20 percent range, though commissions are
sometimes higher. Some of the intermediaries control the fee that workers can charge end customers for their services,
while others allow workers to propose a fee. Determining whether the workers who participate in these online markets
are independent workers would require a deeper analysis of their relationship with their intermediary.
Unfortunately, because almost all of these companies are privately held start-ups, little public information is available
regarding their size, growth rate, revenues, or protability. Nevertheless, we can obtain a rough estimate of their size and
growth rate from Google Trends (www.google.com/trends/). Google Trends enables users to track the relative frequency
of searches for various terms. Specically, we used Google Trends to compute the relative number of searches conducted
in the United States containing the names of each intermediary listed in the Appendix table; we then normalized the data
relative to searches for the term “Uber” each week. “Uber” is by far the most frequent term that arises in searches for the
intermediaries in the table.
Figure 1 summarizes the data, and shows a four-week moving average of the relative frequency of searches for each
intermediary. e exponential growth rate of Uber searches since 2013 matches the exponential growth rate of Uber
driver-partners reported in Hall and Krueger (2015). In addition to Uber, intermediaries Ly and Grubhub also exhibit
an exponential growth path. Searches for ChaCha, a search engine guided by humans, grew rapidly until 2012 and then
trailed o as the company encountered diculties.
Figure 2 provides a bar chart on the relative number of searches for each intermediary, and combines the data for
every week from January through early November 2015. e searches are indexed relative to searches for Uber. e
second-most-common intermediary that Internet users searched for was Grubhub. Searches for Grubhub were about
one-h as common as searches for Uber. (Note that Uber is shown on the scale on the le vertical axis, and all of the
0
20
40
60
80
100
10
30
50
70
90
Index (Relative to Uber)
Uber
GrubHub
Lyft
Upwork
Fiverr
Thumbtack
Mechanical Turk
TaskRabbit
Sidecar
Handy
ChaCha
Other
18.5
12.4
100
3.5 3.2
3.1
2.9
1.7
1.4
1.1
0.7
0
FIGURE 2.
Google Trends: Cumulative Web Searches in 2015
Source: Google Trends analysis by authors.
Notes: “Other” includes Agent Anything, Axiom, Caviar, Clickworker, Curb, Eden McCallum, Gengo, Hourly Nerd, Instacart, Medicast, Red Beacon, Samasource,
Shyp, Skillshare, and Washio. Search period spans January 1, 2015- November 7, 2015. Google Trends normalizes the data for each term specified relative to the
total number of Google searches conducted in that week, so that the period with the most searches for Uber equals 100.
The Hamilton Project • Brookings 11
others are shown on the scale on the right vertical axis.) Searches for Ly, Uber’s largest online competitor, were only
12 percent as frequent as were searches for Uber. e skewness of the distribution of searches for online intermediaries
is noteworthy. Only seven other intermediaries registered more than one percent as much search interest as Uber. e
other 25 intermediaries combined generated only about half (48.5 percent) as much search activity as did Uber.
We can derive a rough estimate of the size of the online gig economy as follows. First, note that in December 2014 Uber
had 162,000 active drivers in the United States, and the number of drivers more than doubled every six months from
2012 to 2014 (Hall and Krueger 2015). Assuming this pace continued into 2015, Uber had around 400,000 active driver-
partners in the fall of 2015. If the number of workers providing services through an intermediary is proportional to the
number of Google searches—an assumption that is quite plausible for Ly and less clear for other intermediaries—then
there would only be about 600,000 workers, or 0.4 percent of total employment in the United States, engaged with all of
the intermediaries in the Appendix table. If, however, Google searches translate into ve to ten times as many workers
per search incident for apps other than Uber and Ly—to make an extreme assumption—then there would be about 1.2
to 1.9 million workers engaged in the online gig economy. is gure is in the ballpark of McKinsey’s estimate that 1
percent of the U.S. working-age population participates in “contingent work that is transacted on a digital marketplace”
(McKinsey Global Institute 2015). ere are probably many more workers who work for traditional intermediaries (i.e.,
that do not use apps to match workers with customers) who would be classied as independent workers than there are
workers who work for emerging intermediaries (i.e., that use Internet-based apps) who would be classied as independent
workers.
Although precise estimates of the number of workers engaged in the gig economy are not available and must await
further research, these calculations suggest that independent workers operating in online markets make up a very small
share of total U.S. employment at present. However, it is clear that some intermediaries are growing rapidly, and creating
a rapidly expanding new segment of the workforce.
12 A Proposal for Modernizing Labor Laws for Twenty-First-Century Work: The “Independent Worker”
T
o identify independent workers and guide the
determination of the benets and protections for which
they should qualify, we oer three main principles:
immeasurability of work hours, neutrality, and eciency.
IMMEASURABILITY OF WORK HOURS
e boundary between work and nonwork for independent
workers is largely indeterminable. A worker in the online gig
economy could be primarily engaged in personal tasks while
one or more intermediaries’ apps are turned on. It would
stretch any reasonable denition of “work” to count this time
as work hours, as the example in box 1 illustrates. is fact
of the online gig economy creates an immediate problem for
implementing the social compact. Many benets included
in that compact, such as the minimum
wage, overtime pay, and ACA eligibility,
are tied to hours worked—and, even more
specically, hours worked for a particular
employer. Determining whether and
for whom an independent worker is
“working” is impossible or deeply
problematic in too many circumstances
for the concept of work hours to translate
into these emerging relationships.
ere are circumstances in which
independent workers are undeniably
working. For example, a landscaper is
working during the time she is mowing
a customer’s lawn or trimming hedges. A
driver is working while he has a customer
in his car and the car is under way to the customer’s destination.
It is equally undeniable that technological developments have
made recording this time even easier than using a clock or
watch. Yet these independent workers are working for the
customer during these times, and not the intermediary. Once
the connection between customer and independent worker
has been made, the intermediary has no role except to collect
payment and transmit it to the independent worker. Even
under the broadest denition of “employ” in the law, which is
found in the FLSA, the intermediary cannot be said to “suer
or permit” this work.
If a worker works for two intermediaries at the same time,
as illustrated by the example of a driver who uses apps
for Ly and Uber simultaneously, it is unclear how the law
would or should apportion total work hours between the two
companies. Moreover, a worker could spend time at home with
her app turned on, waiting for a possible work opportunity,
while primarily performing work for another intermediary or
engaging in nonwork activities. Conceptually, workers’ hours
spent waiting to be engaged in work cannot be apportioned
to a specic employer. In this sense, independent workers are
working for themselves and working on their own time.
If work hours cannot be apportioned and measured for the
purposes of assigning benets or assessing hourly earnings,
we think it makes little sense to require intermediaries to
provide hours-based benets, such as overtime and the
minimum wage. Although employees have a hard-earned
right to these protections, independent workers can be viewed
as having traded these protections for the exibility that their
work arrangement aords.
NEUTRALITY
Creative destruction works to raise living standards when new
entrants gain an advantage because they provide better goods
or services, or the same goods or services more eciently.
However, when start-ups gain advantage because they skirt
certain worker or customer protections, and not because they
Chapter 3. Principles of a New Worker Classification
The Hamilton Project • Brookings 13
Creative destruction works to raise living
standards when new entrants gain an
advantage because they provide
better goods or services, or the same
goods or services more efciently.
have developed a better product or better way of producing
it, creative destruction is destructive to living standards.
It is therefore important that businesses do not organize
themselves to move workers into independent worker status
in order to gain an unfair advantage over other employers by
skirting legal protections and required benets. From society’s
perspective, it is important that businesses not choose to
structure their work relationships to meet the denitions of
independent workers or independent contractors in order to
free ride on other employers with respect to providing certain
benets, such as health insurance.
Neutrality also requires that workers in “old economy” jobs
who meet the denition of independent worker, as opposed
to independent contractor or employee, should be classied as
“independent workers.” For example, as argued in the analysis
below, many taxi drivers who are currently classied as
independent contractors could be deemed to be independent
workers, depending on their terms and conditions of work. In
this way, taxi drivers would be treated just like independent
workers who provide rides through the Uber and Ly
platforms.
EFFICIENCY
e independent worker contract should be ecient in the
sense that it enables workers and intermediaries to maximize
the joint surplus that their relationship produces. For
example, independent contractor status is currently inecient
for many intermediaries and their contract workers because
the intermediary avoids providing benets that would make
both the worker and the intermediary better o to reduce
the chances of the relationship being ruled an employment
relationship. We believe that legal uncertainty, and the
intentional and unintentional misclassication it facilitates,
are signicant contributors to this ineciency.
ese principles are of rst-order importance in guiding
the reform of labor, employment, and other laws concerning
independent workers, although we acknowledge that progress
in meeting one of the principles can conict with progress
in meeting another. For example, craing rules to ensure
neutrality could create uncertainty that, at least in the short
run, could reduce eciency. Nevertheless, explicitly specifying
the key objectives and recognizing the trade-os involved is a
rst step toward devising a more rational system.
14 A Proposal for Modernizing Labor Laws for Twenty-First-Century Work: The “Independent Worker”
I
n view of these principles, we propose that Congress, and
state legislatures where applicable, enact legislation that
would guarantee or permit the following benets and
protections for independent workers to ensure they can benet
from America’s social compact. It is worth noting that federal
law problems can be solved with a single act of Congress that
amends the applicable tax, labor, and employment laws, as
well as antitrust statutes, as appropriate. We acknowledge that
proposed legislation addressing multiple subjects oen faces
the dicult challenge of working its way through multiple
committees with dierent jurisdictions in each house of
Congress. Nonetheless, the only way to ensure inclusion of
all of the protections and benets we consider important to
independent worker status is a single omnibus bill. State law
changes may also be required to address workers’ compensation
and unemployment insurance issues. States with their own
antitrust and workplace laws may need to amend those statutes
to reconcile them with Congress’s amendments to federal law,
if federal law changes do not override state laws.
While an argument might be made that courts or administrative
agencies could use their existing authority to address a few of the
problems created by the emergence of independent workers, the
evolution of an entirely new third legal classication for workers
should not be le to judges or regulators. Our principal concern
is not the typical process argument around the propriety of
unelected judges and regulatory ocials making certain policy
decisions rather than the democratically elected branches of
government. Rather, our objection is that courts do not have the
power, on their own, to ensure that independent workers receive
their full and fair share of the social compact—that is, the full
complement of protections and benets that must be established
by statute. Moreover, courts do not have sucient authority
to ensure a fully ecient solution to the problems created by
the emergence of independent workers. Similarly, regulatory
agencies like the Internal Revenue Service (IRS) and the U.S.
Department of Labor do not have the authority to provide all of
these benets or an ecient outcome. A comprehensive solution
will necessarily require Congress taking action followed, where
necessary, by state legislatures.
We propose the following reforms.
FREEDOM TO ORGANIZE AND COLLECTIVELY
BARGAIN
Antitrust laws should be amended to allow independent
workers to organize for the purpose of aggregating their
individual bargaining power so they may bargain successfully
with their intermediaries over the terms and conditions of
their work. Collective action could address imbalances in
bargaining power between individual independent workers
and intermediaries and thereby give independent workers
some ability to inuence their compensation and benets
while providing them an opportunity to gain a voice in their
relationships with intermediaries. e ability to organize
would also make independent worker status more neutral
with respect to employee status.
Collective action by employees is protected by the National
Labor Relations Act (NLRA). In principle, the NLRA
safeguards employees’ “right to self-organization, to form, join,
or assist labor organizations, to bargain collectively through
representatives of their own choosing, and to engage in other
concerted activities for the purpose of collective bargaining
or other mutual aid or protection, and shall also have the
right to refrain from any or all of such activities” (29 U.S.C.
§ 157). e NLRA seeks to enforce these rights by deeming
certain employer and union behaviors that infringe them to
be “unfair labor practices” that may be remedied by order of
the National Labor Relations Board (NLRB) (29 U.S.C. § 158).
e NLRB is the administrative agency created by the NLRA
to which Congress delegated responsibility over private sector-
labor relations. In addition to prosecuting and adjudicating
unfair labor practices, the NLRB administers elections
that determine whether a majority of an identied group of
employees wants to be represented by a union. If a union wins
a representation election, or secures voluntary recognition
from an employer with majority support within a group of
employees, then the union is the exclusive representative of all
employees in that “bargaining unit” (29 U.S.C. § 153(b)).
Because they are not employees, in our proposal independent
workers would not be covered by the NLRA and, therefore,
would not have access to the NLRB and its processes, or to any
of the NLRA’s remedies (29 U.S.C. § 152(3)). In some regards,
this may be benecial for independent workers’ organizing
Chapter 4. Legal Reform for “Independent Workers”
The Hamilton Project • Brookings 15
prospects. e NLRA has been long derided as ossied,
ineective, and lacking in eective remedies for violations
of employees’ rights to organize and bargain collectively
(Estlund 2002; Weiler 2009). Many unions have migrated away
from organizing workers through NLRB elections to private
“neutrality and card check” agreements with employers that
operate outside the scope of the NLRB (Brudney 2005). In part
because the NLRA does not eectively safeguard workers’
freedom to choose a union, the private sector union density
rate in the United States has declined from a high of 37 percent
in 1955 to below 7 percent in 2014 (Bureau of Labor Statistics
2015; Kleiner 2001; Lui 2013).
e advent of many of the same technologies that make
intermediaries possible has reduced the transaction costs of
organizing independent workers. Mass organizing on Twitter,
Facebook, Snapchat, and other social media platforms is in its
early stages, but opportunities may exist for creative organizers
to build signicant power for independent workers if they are
not subject to the detailed and burdensome requirements
of a private sector labor law designed for dierent kinds of
work relationships and workplaces. To facilitate organizing
eorts, intermediaries could even be required to provide
organizations seeking to represent independent workers
with the contact information of independent workers
who work with the intermediary. Nascent organizing eorts
by some independent worker groups have already begun
(www.coworker.org; Hudnall 2015).
e main legal challenge for independent workers’ organizing
activity is federal antitrust law. Section 1 of the Sherman
Antitrust Act (Sherman Act) establishes that “every contract,
combination . . . or conspiracy, in restraint of trade or
commerce among the several States, or with foreign nations,
is declared to be illegal” (15 U.S.C. § 1). Section 2 of the
Sherman Act makes it a misdemeanor to “monopolize, or
attempt to monopolize, or combine or conspire with any
other person or persons, to monopolize any part of the
trade or commerce among the several States, or with foreign
nations” (15 U.S.C. §2). Similarly, Section 3 holds that “every
contract, combination..., or conspiracy, in restraint of trade
or commerce” is illegal (15 U.S.C. §3). e purpose of these
provisions is to protect free and unfettered competition in
product and service markets from untoward eorts to “x”
the competition for the benet of particular competitors or
for competitors against consumers.
e risks these provisions create for independent workers
seeking to organize are illustrated by the challenges that
independent physicians encountered in the late 1990s
when they organized for the purpose of negotiating with
health maintenance organizations and managed care
organizations regarding fees, patient care, and other issues.
In several cases involving nonemployee physicians, the U.S.
government alleged antitrust violations under the Sherman
Act. Settlements prohibiting any such organizing eectively
foreclosed the physicians’ eorts.
4
Employees represented by unions do not face the same
antitrust limitations. Unions benet from a “labor exemption”
from antitrust law when they engage in core activities such
as organizing, bargaining with employers, or administering
collective bargaining agreements. e exemption ows from
very broad language in section 6 of the Clayton Antitrust Act:
e labor of a human being is not a commodity or article of
commerce. Nothing contained in the antitrust laws shall be
construed to forbid the existence and operation of labor . . .
organizations, instituted for the purposes of mutual help . . . or
to forbid or restrain individual members of such organizations
from lawfully carrying out the legitimate objects thereof; nor
shall such organizations, or the members thereof, be held
or construed to be illegal combinations or conspiracies in
restraint of trade, under the antitrust laws. (Clayton Antitrust
Act, 15 U.S.C. § 17)
Further reinforcing the labor exemption, and to cure courts’
inability to resist interventions against unions in disputes
with employers during the rst third of the twentieth century,
both section 20 of the Clayton Antitrust Act (Clayton Act) and
section 4 of the Norris-LaGuardia Act prohibit injunctions
that would limit employees’ ability to organize into unions
and bargain with their employers (29 U.S.C. § 101).
e “labor exemption” is generally available only when a bona
de labor organization is promoting legitimate labor interests
rather than entrepreneurial or other interests unrelated to
the employer–employee relationship. e labor organization
also must act independently of any nonlabor group. ese
limitations seek to ensure that labor unions focus their
organizing and bargaining eorts on the labor market rather
than on disrupting free competition in product and service
markets.
e Supreme Court held in Columbia River Packers Assn.,
Inc. v. Hinton (315 U.S. 143 1942), that the exemption is not
available to associations of independent contractors because
they cannot form a bona de labor organization under labor
law or negotiate over an employment relationship that does
not exist. Hinton and its progeny may doom any hope that
the Clayton Act can be interpreted to protect independent
workers’ organizing from antitrust attacks, despite the fact
that independent workers principally sell their own labor, as
contemplated by section 6’s broad declaration. It is possible
that an argument could be made that independent workers
are dierent from independent contractors and, as a result,
that Hinton should not govern. Yet a better approach seems to
be for Congress to cra an “independent workers exemption”
from any antitrust laws that might infringe upon their eorts
16 A Proposal for Modernizing Labor Laws for Twenty-First-Century Work: The “Independent Worker”
to organize and bargain through the imposition, for example,
of court injunctions or other judicial remedies.
To advance antitrust law’s interest in protecting product
and service markets from illicit restraints on competition,
this independent workers exemption could include the
same limitations to which the labor exemption is subject:
independent worker organizations could be required to
organize around and bargain over their equivalents of wages,
hours, and the terms and conditions of their contractual
relationships with the intermediaries. Collective eorts to
set the prices paid by customers, for example, or to otherwise
dene product or service market outcomes should not
be exempted. Another option would be to simply include
independent workers under the NLRA.
ABILITY TO POOL
ere are potentially large eciency advantages if intermediaries
are able to pool their independent workers for the purpose of
purchasing or directly providing or administering certain
benets for workers. e benets of risk pooling in insurance
markets is well known. In essence, pooling
helps to reduce adverse selection in the
take-up of insurance that could render
insurance policies prohibitively expensive
and cause the entire insurance market
to cease to exist (Rothschild and Stiglitz
1976). In addition, by pooling employees
and jointly purchasing and administering
certain benets in bulk, intermediaries
and their workers could benet from
scale economies and superior bargaining
power that are unavailable to them as
individuals. As a result, prices almost
certainly would be signicantly lower for
independent workers and intermediaries
than for individual purchasers, and both
services and products could be greater in
quality, quantity, or both.
Intermediaries could use their scale and pooling opportunities
to oer independent workers a range of insurance services, tax
preparation assistance, and nancial services. Products and
services that likely would be oered include auto insurance,
disability insurance, health insurance and health care, banking
and savings products, retirement products, and liability insurance.
Currently, however, intermediaries are loath to take advantage of
pooling eciencies because oering benets to workers would
raise the risk that their work relationships would be adjudged
employment by a court or administrative agency. To overcome
this inecient predicament, we propose that intermediaries be
covered by a safe harbor provision such that pooling independent
workers for purposes of providing benets would not be legally
interpreted as an indication of employee status.
Pooling is a common feature of employment relationships,
and so the ability of intermediaries to pool their independent
workers to purchase goods and services would approximate
neutrality between the two statuses. Employers and employees
would continue to have some advantages, including favorable
tax treatment for retirement products and health insurance,
which would provide some incentive to establish employment
relationships over independent worker relationships. Yet
our proposal would give independent workers a greater
opportunity to participate in the social compact than would be
available were they to be classied as independent contractors.
CIVIL RIGHTS PROTECTIONS
Expanding workplace antidiscrimination protections to
include independent workers will help make that status
neutral compared with employee status, extend a key aspect of
the social compact to independent workers, and help the labor
market to operate more eciently.
Employees benet from protections provided by a broad,
well-developed, and reasonably eective battery of federal
employment antidiscrimination statutes. ese statutes
prohibit almost all employers from discriminating against
their employees or job applicants on the basis of race, national
origin, color, sex, religion, age, and disability.
5
Prohibited
discriminatory acts may relate to hiring, ring, promotions,
compensation or training decisions, job shi assignments,
or almost any decision aecting an employee’s terms and
conditions of employment. Additional statutes and presidential
executive orders add groups of employees to those protected
from discrimination by federal contractors, including LGBT
Americans and many veterans, and impose armative action
obligations on the contractors.
6
The Hamilton Project • Brookings 17
Intermediaries could use their scale and
pooling opportunities to offer independent
workers a range of insurance services, tax
preparation assistance, and nancial services.
Workers who are not employees, including independent
contractors, do not have access to these federal statutory
antidiscrimination protections. Under existing law,
independent workers also would not receive such protections.
In federal law, only section 1981 of the Civil Rights Act of
1866 (42 U.S.C. § 1981) addresses discrimination in the
relationships formed between employers and independent
contractors or independent workers, and it is a starkly
limited and inadequate tool when compared with employees’
protections. In particular, section 1981 prohibits only race
discrimination,
7
although “race” is dened broadly to include
ancestry and some ethnic characteristics.
8
Nonetheless,
independent contractors and independent workers could not
bring federal claims if intermediaries discriminate on the
basis of sex, disability, or age, for example.
Furthermore, section 1981 guards against only intentional
discrimination, or “disparate treatment” in civil rights
parlance, and not “disparate impact”—that is, particular
practices that produce discriminatory results whether
intended or not. Accordingly, if 98 percent of the relationships
entered into by an intermediary were with white independent
workers because the intermediary limits its recruitment to
wealthier geographic areas in which white workers are grossly
overrepresented, black and Latino independent workers
could not bring a successful claim absent some evidence of
discriminatory intent. Otherwise, however, section 1981
claims can be used to seek relief for many of the same kinds
of discriminatory acts prohibited by the federal employment
discrimination laws.
Section 1981 diers from federal employment discrimination
laws in some ways that may benet independent workers,
however. For example, remedies under section 1981 are more
expansive and generous than those available under the federal
employment discrimination laws. Section 1981 claims are
led directly in court without a requirement of exhausting
the administrative process at the U.S. Equal Employment
Opportunity Commission (EEOC).
9
Finally, section 1981’s
statute of limitations is longer.
e most obvious and straightforward solution to inadequate
antidiscrimination protections for independent workers
would be to include them within the protections of the federal
employment discrimination laws. is approach ensures
neutrality between employment relationships and independent
worker relationships while providing more-expansive
protection against inecient discriminatory acts in the
workplace and labor market. With this solution, intermediaries
would derive the benets of political compromises that have
limited damages recoveries and force claims into the EEOC
processes for mediation and dispute resolution.
Of course, this approach would require congressional action to
amend these laws. While civil rights laws have been traditionally
contentious topics in Congress, we believe that amending the
federal employment discrimination laws is more likely than a
wholesale rewriting of section 1981 both to expand the list of
protected groups and to include disparate impact claims. It is
also substantially more likely than craing and enacting a new
and targeted antidiscrimination statute dedicated exclusively to
the protection of independent workers.
TAX WITHHOLDING AND THE FEDERAL INSURANCE
CONTRIBUTIONS ACT
Withholding taxes for employees began during World War
II as a measure to raise revenues to fund the war eort.
10
A
withholding tax is an advance payment toward an employee’s
nal tax liability. In essence, employers deduct a certain
amount of income from an employee’s weekly or monthly
paycheck, and remit the money to the IRS as an advance
payment of income and payroll taxes. If the amount of taxes
withheld exceeds an employee’s ultimate tax liability, then
the excess is refunded by the IRS. Tax withholding helps
employees to smooth their aer-tax income throughout the
year and facilitates revenue collection by the IRS.
Absent their employers withholding their taxes and
transferring them to the IRS, employees would be responsible
for making quarterly payments to the IRS on their own,
or saving sucient funds to be able to pay their entire tax
liability when they le their income tax return. Independent
contractors are responsible for their own tax payments. is
can be burdensome and create tax penalties if it is not done
properly, and also can cause uctuations in consumption and
asset allocations because independent contractors may be
required to make a large tax payment when ling their income
taxes. Because of these added burdens of complying with tax
laws, there is reason to believe that independent contractors
are less likely than employees to pay their full tax liabilities
(Gandhi 1994). So, tax withholding also can be expected to
produce increased tax compliance, and greater revenues for
the federal government and the states.
Tax withholding by intermediaries would reduce workers’
administrative burden of paying income and social insurance
taxes. Given economies of scale, withholding services
provided by intermediaries would also be economically
ecient and improve compliance with tax laws. We propose
that intermediaries would be required to provide withholding
services for income and social insurance taxes owed by all
independent workers with whom they work. Tax withholding
by intermediaries would support the principle of neutrality
between employment status and independent worker status
since most employees benet when their employers withhold
state and federal income and payroll taxes.
18 A Proposal for Modernizing Labor Laws for Twenty-First-Century Work: The “Independent Worker”
In addition, to maintain neutrality with employees, we
propose that intermediaries pay half of independent workers’
contributions toward the Federal Insurance Contributions
Act (FICA) payroll taxes for Social Security and Medicare.
Although in the long run, intermediaries are likely to shi the
ultimate burden of paying for FICA contributions to workers
through fee adjustments, as explained below, the assignment
of legal tax liability can potentially aect tax incidence
in some cases. In addition, tax shiing can take place for
employees as well, so requiring intermediaries to contribute
half of FICA contributions will make it easier for employees
and independent workers to compare their compensation,
because they will be on more-equal footing.
Our principles that guide the creation of an independent
worker status lead us to suggest a more nuanced approach to
some other benets, such as allowing intermediaries to opt into
workers’ compensation insurance. We conclude that it would
not be ecient or feasible to require intermediaries to provide
this class of workers with other protections and benets, such
as overtime protection or unemployment insurance.
WORKERS’ COMPENSATION INSURANCE
State laws typically require that employers provide their
employees with workers’ compensation insurance. Although
this paper is principally focused on federal labor, employment,
and tax laws, workers’ compensation insurance is the oldest
social insurance program in the United States and undeniably
an integral part of America’s employment social compact.
erefore, we consider it a necessary, if challenging, part of
any discussion of independent workers.
Workers’ compensation provides cash compensation and
medical benets to employees who experience workplace
injuries or illnesses. In addition, it provides employees’
survivors with compensation in the event of a fatality. It
is a strict liability system—that is, the employee need not
show that the employer was negligent or otherwise at fault
in order to collect benets. Payments are made based on
a state-established matrix that principally considers the
severity of the employee’s work-related injury or illness and
the employee’s tangible economic losses. For example, an
employee’s permanent total disability would result in greater
compensation than a partial impairment (Burton 2009).
Workers’ compensation itself is the product of a grand
bargain. In principle, employees receive reasonably predictable
compensation for work-related injuries and illnesses without
the cost and complication of proving in court that their
employers failed to protect them from injury or illness. In
return, employers receive immunity from costly lawsuits under
state tort laws that could result in judgments against them that
could be many multiples of the amounts they pay for any single
workers’ compensation claim, or even substantially more.
Because there is no employment relationship with the
intermediaries with which they work, and therefore no immunity
provided by workers’ compensation laws, independent workers
currently are able to use tort law to seek compensation for
injuries or illnesses that result from their work relationships
(for fatalities, survivors could bring the claims), assuming they
are classied as independent contractors. Risk of tort judgments
should create incentives for intermediaries to use reasonable
care in their dealings with independent workers.
11
Texas and Oklahoma allow employers to opt out of their state
workers’ compensation insurance system and employees of
employers that have opted out of the system are able to bring
tort actions against their employers, but some employers have
succeeded in signicantly dulling the incentives of these
states’ tort laws and deprived their employees of fair recoveries
using the limited damages remedies permitted by ERISA
(Grabell and Berkes 1974). Because independent workers are
not employees, ERISA would not be available as a tool to avoid
responsibility to independent workers under state tort laws
(ERISA, 29 U.S.C. 18 § 1003(a)).
Tort laws require, in most cases, that an intermediary
commit some act or omission before being held liable, which
may more accurately reect the nature of the relationship
between intermediaries and independent workers than
workers’ compensation’s no-fault strict liability system.
Independent workers generally do not perform their work
on an intermediary’s premises or use equipment supplied
by an intermediary. In the case of Uber and Ly, drivers use
their own cars. In the case of TaskRabbit and similar apps,
workers use their own tools or other supplies (or perhaps
their customers’) to work in their customers’ homes, yards, or
businesses. In the case of Mechanical Turk and similar apps,
they are typically working in their homes on their personal
computers. us, a legitimate question can be raised whether
an intermediary should be expected to take responsibility for
injuries, illnesses, or fatalities that are more likely than not
beyond its control, particularly absent any proof of an ability
to avoid the injury through reasonable care.
It is possible to imagine circumstances in which an
intermediary’s negligence may cause an injury to an
independent worker. For example, Task Rabbit may have
received complaints from independent workers that a
particular customer has threatened violence against them. If
Task Rabbit were to send another independent worker without
prior experience with the customer to that customer’s home
for an odd job while negligently misrepresenting that there
had been no complaints against that customer, then tort
liability might be possible if the customer were to attack the
independent worker. Similarly, Uber, Ly, or other driving
services might require their drivers to submit information
through their apps while driving in a manner that unduly
The Hamilton Project • Brookings 19
distracts the drivers and leads to accidents and driver injuries.
Tort liability might be possible in these circumstances as well,
and provide an appropriate remedy.
Although the tort system may oen be the best solution
for addressing work-related injuries for intermediaries and
independent workers, it is possible that in some instances
workers’ compensation insurance would oer a more ecient
solution, although workers’ compensation would be rife with
adverse selection concerns if employees can opt into the system.
We therefore propose that intermediaries be permitted to opt
to provide expansive workers’ compensation insurance policies
to the independent workers with which they work without
transforming these relationships into employment. In exchange
for this no-fault insurance coverage, intermediaries would receive
limited liability and protection from tort suits. States would
provide the legal framework within which these policies would
operate, but not operate the systems themselves. States could
require that the policies provide the same level or more protection
to independent workers than their state workers’ compensation
system. While opt-in and voluntary systems of insurance can
create adverse selection and moral hazard problems, experience
with these policies could inform design changes that might
reduce these risks over time, and intermediaries would have
the right to opt out of the system and be subject to tort actions if
adverse selection and moral hazard cause workers’ compensation
insurance to be prohibitively expensive.
WAGE AND HOUR PROTECTIONS AND
UNEMPLOYMENT INSURANCE
As discussed in greater detail above, measuring the working
hours of independent workers in the same manner as the hours
of employees is impossible. is reality of the independent
worker–intermediary relationship makes certain rules that
depend on the measurement of working hours—particularly a
minimum wage for each hour worked and overtime for hours
worked in a week in excess of forty—impossible to properly
administer for independent workers. Since their circumstances
are quite dierent, neutrality does not require the same legal
treatment of independent workers and employees. Accordingly,
similar to independent contractors, independent workers
would not be covered by the FLSA’s requirements of overtime
pay and the minimum wage in our proposal.
12
Rather, our view
is that compensation and benets issues should be the subject
of bargaining between (preferably organized) independent
workers and intermediaries. Moreover, the easy entry and exit
from independent work should provide some protection against
substandard wages and exploitative work hours.
Similarly, because independent workers control when and
whether they will work, the fundamental principles of the
federal–state unemployment insurance system do not apply.
Unemployment insurance benets are generally provided to
employees who lose their jobs through no fault of their own,
and not to those who voluntarily opt out of their jobs or stop
working temporarily by choice. Employers pay a tax on their
payrolls to fund unemployment insurance benets for laid-o
employees, although, ultimately, employer-funded benets of
this sort are oen funded largely out of workers’ wages. Since
independent workers are not employees, and they would not
be eligible for unemployment benets, their earnings would
not be subject to this payroll tax (Woodbury 2009).
Consistent with our discussion of pooling arrangements above,
intermediaries should be permitted to pool resources across
workers and create a private unemployment insurance system,
or a system of individual accounts for independent workers
who stop working.
13
Such a system could come about as a result
of collective bargaining between independent workers and
intermediaries, or it could be established by intermediaries
acting on their own. Organized independent workers may also
seek protections against or compensation from intermediaries
that cease doing business with particular independent workers
for economic or other reasons that lack sucient cause.
Facilitating any or all of these systems would move independent
worker status closer to neutrality with employee status and
improve the eciency of the labor market.
AFFORDABLE CARE ACT AND HEALTH INSURANCE
Maintaining neutrality between independent workers and
employees aer the employer mandate that is part of the
ACA takes eect presents an important set of challenges. e
mandate, also known as the Employer Shared Responsibility
Provision of the Aordable Care Act, requires that rms
with y or more full-time equivalent employees oer health
insurance that meets minimum value and aordability
standards for their employees. Employers who do not
oer such coverage to at least 95 percent of their full-time
employees are subject to a penalty. e threshold for full-
time employment under the statute is thirty or more hours
of work per week (26 U.S.C. 43 § 4980H). However, because
work hours are immeasurable for independent workers,
determining eligibility for the mandate and for coverage
under the mandate is problematic. Nonetheless, in our view,
independent workers should benet from the social compact
that supports employer-provided health insurance, and their
intermediaries bear some responsibility under that compact.
If independent workers are treated like independent
contractors with respect to the employer mandate, they would
not be counted toward the 50 full-time employee threshold,
and intermediaries would not be subject to a penalty for
failing to oer independent workers health insurance. As a
result, intermediaries may be viewed as free riding on other
employers who provide health insurance to their independent
workers. For example, an independent worker may have a
20 A Proposal for Modernizing Labor Laws for Twenty-First-Century Work: The “Independent Worker”
traditional employment relationship with an employer on
another job that provides health insurance or a spouse whose
employer provides family coverage. Such free riding would
violate the neutrality principle. Moreover, if independent
workers turn to exchanges to purchase health insurance and
receive tax subsidies, intermediaries will have an advantage
over other employers that would pay a penalty in similar
circumstances.
Consequently, we propose that intermediaries be required
to pay a contribution equal to ve percent of independent
workers’ earnings (net of commissions) to support health
insurance subsidies in the exchange as a solution to the free
rider problem and to support health insurance tax subsidies.
is ve percent gure could be adjusted over time depending
on health insurance costs and earnings growth.
The Hamilton Project • Brookings 21
I
n his landmark book e Fissured Workplace, David Weil
(2014) oers a thorough account of the shi from traditional
bilateral and long-lasting employment relationships
to more-diverse arrangements principally resulting from
corporations outsourcing many of their functions. Whether or
not ssuring of the workplace is the cause, independent workers
are not the only workers who nd themselves in some form of
triangular relationship with customers and other enterprises
in U.S. labor markets. Our view is that the application of our
proposed independent worker category should not be limited to
the online gig economy. In fact, the very nature of law—treating
like cases alike—requires that this new category include any
group of workers who satisfy the denition of independent
workers we oered above. Accordingly, if there are workers in
triangular relationships with intermediaries and customers,
then they should be considered for independent worker status.
We cannot oer a comprehensive list of potential candidates
for independent worker status in this paper, but some
obvious candidates should be discussed and analyzed. In
particular, many traditional taxi drivers (as opposed to Uber
and Ly drivers), temporary stang agency employees,
labor contractors, members who secure jobs through union
hiring halls, outside sales employees, and (perhaps) direct
sales employees occupy the points of triangles with other
economic actors. In some of these cases, under existing law, an
employment relationship is formed. In others, the workers are
classied as independent contractors. Below we evaluate some
specic cases. In a couple of cases, by applying the principles
discussed in this paper, we conclude that workers should be
reclassied (or considered for reclassication) as independent
workers. In fact, we believe the neutrality principle requires
it. In other cases, we conclude that there should be no re-
classication given the nature of the work relationship.
For this latter category of work relationships, we nd that
there are meaningful dierences from the independent
worker-intermediary relationship. As we explain below, in
several cases, the employer exercises more control over the
worker’s work hours, work tasks, and means of performing
the work. As a result, work hours are not immeasurable like
those of independent workers, and the employer’s greater
control contributes to the worker’s greater dependence
upon the employer. In addition, some of these relationships
are expected to last for longer periods than an independent
worker’s relationship with an intermediary. is suggests even
greater worker dependence upon the employer. ese factual
distinctions that are fundamental to the task of classifying
employees, independent contractors, and independent
workers should produce dierent legal results.
We hasten to add that none of these distinctions depend upon
technology, in general, or the use of an Internet-based app, in
particular. If a temporary stang agency or a union hiring hall
used an online app to conduct its business, our conclusions
would not change because the core of their business models
and relationships with workers would not have changed.
TAXI DRIVERS
Nearly 500,000 Americans worked as a taxi or limo driver
as their main or secondary job per month in 2015, according
to a tabulation of the Current Population Survey. For the
tiny percentage of readers who have not encountered them,
taxi drivers transport customers from place to place by car.
Taxi drivers may have any of three relationships with taxi
companies. Owner-operators, in essence, are their own taxi
companies: the driver owns the taxi and bears responsibility
for all aspects of the taxi and her work schedule, including
potentially leasing the taxi to others. ere is no triangle in
this relationship; rather, it is a bilateral relationship between
the owner-operator and the customer. An owner-operator is
an independent business. Independent subletters are owner-
operators without the ownership. ey lease a taxi, but
operate it in whatever manner they see t without direction
or involvement by the lessor. Again, like owner-operators,
the relationship is bilateral, not triangular. us, like owner-
operators, and other small business owners who lease the
premises on or in which they work, the best argument is that
these independent subletters are independent businesses.
e triangular relationship and more complicated classication
task comes with workers who rent or lease taxis for a day or
longer, but who essentially work for the taxi company that
leases the cab. e driver may pay a at rental fee for the use
of the taxi for a specied period or receive a portion of the
day’s fares from the taxi company. Since the lease suggests the
drivers have assumed some or all of the risk of opportunity
or loss, these workers are typically classied as independent
Chapter 5. Are independent workers different from
other third-party players in labor markets?
22 A Proposal for Modernizing Labor Laws for Twenty-First-Century Work: The “Independent Worker”
contractors. However, the rented taxi is branded with the cab
company’s name, telephone number, and (perhaps) Web site
address. Customers who do not hail a taxi on the street (a
practice that is usually limited to a few large cities and many
airports) place an order either by phoning the cab company’s
dispatcher, completing a form on the company’s Web site,
or using an “e-hailing app” when one is available. e cab
company then dispatches a taxi driver to pick up the customer
using a two-way radio system, an in-taxi mobile data terminal,
or some other communications device.
Apart from the technological dierence, this relationship
between drivers and taxi companies closely resembles the
triangular relationship between independent worker-drivers,
ride services such as Uber and Ly, and riders. It also has
some indicia of the independent contractor relationship and
other indicia that it is an employment relationship, just like
independent workers in the online gig economy. ere are
two principal nontechnological dierences that make taxi
drivers more like employees than independent contractors.
First, Uber, Ly, and other online ride services do not
require drivers to rent the services’ cars. Plainly, both taxi
drivers and online ride service drivers invest capital in the
enterprise—cash for taxi drivers, personal cars for the online
ride services’ drivers—and thereby take on some opportunity
for prot or loss. Yet the drivers for the online services look
more like independent contractors in this regard. Drivers for
the online ride services can benet by taking a tax deduction
for depreciation of their vehicles and from the opportunity to
use their vehicles at their discretion for purposes other than
driving the services’ customers. Taxi drivers derive neither
benet. e taxi companies’ greater control over the vehicles’
use and its condition, as well as their ability to depreciate
the vehicles, suggests that the taxi drivers may be more like
employees than the online ride services’ drivers in this regard.
Second, with the exception of rides hailed on the street, taxi
companies appear to have more control over matching customers
and drivers than the online ride services. Taxi companies oen
decide which taxi will pick up each customer, whereas the online
ride services leave this choice to their drivers (or at least the choice
of rst refusal), within some broadly dened rules. It also seems
unlikely that a taxi driver, apart from meal and restroom breaks,
would stop picking up riders during a shi when she must earn
back the investment in the taxi rental. Online ride service drivers
do not have shis. ey oat in and out of working, essentially
at will. is means that taxi companies have greater control over
their riders’ work processes—another factor that suggests taxi
drivers are closer to employees than independent contractors.
In sum, taxi drivers who rent or lease their vehicles bear a close
resemblance to the independent workers that operate in triangular
relationships in the online gig economy. In particular, they share
some indicia of independent contractors as well as some indicia of
employees. e same arguments that suggest that Uber and Ly
drivers should qualify as independent workers in our proposed
legal architecture would apply equally well to many taxi drivers’
work relationships. Furthermore, assigning a similar legal
status to workers in the same relationship with an intermediary,
regardless of the nature of the technology employed, will support
the neutrality principle.
TEMPORARY STAFFING AGENCIES
Agencies such as Kelly Services and Manpower provide
temporary workers to client employers to perform work
ordinarily undertaken by the client’s employees, typically in
the client’s workplace. One common compensation scheme
involves the temporary stang agency receiving a percentage
of the temporary worker’s hourly wage for every hour worked.
ere is little question that the client employer forms an
employment relationship with the worker because it controls
almost all aspects of the employee’s work and, at least for the
duration of the relationship, the employee does not and cannot
work for anyone else during the hours committed to the client
employer. For this period, the employee is economically
dependent on the client employer and, to the extent it shares
decision-making with its client, with the temporary stang
agency. ese important characteristics of this relationship
distinguish it from the independent worker’s relationship with
the intermediary.
A question may arise, however, about whether the temporary
stang agency is a “joint employer” of the temporary
worker.
14
In some cases, for example under the FLSA, the
answer is usually “yes.” But this relationship also diers
from the independent worker–intermediary relationship.
e temporary stang agency ordinarily conducts a skills
assessment when it begins working with a temporary worker.
Based on this skills assessment, the agency decides the clients
and jobs to which the temporary worker will be referred. It
exercises signicant control over the employment relationship
in this way. Presumably, the worker may refuse some number of
assignments, but the desire to sustain a continuing relationship
with the temporary stang agency would limit these choices.
e temporary stang agency may also retain some ability
to hire and re the employee, or to transfer her to a new job.
It almost always pays the employee and takes responsibility
for tax withholding, payroll taxes, and workers’ compensation
premiums. Furthermore, as with other employees, there
is an expectation that the temporary worker will sustain
a relationship with the agency beyond the completion of a
particular assignment. In all of these ways, this relationship
looks like a traditional employment relationship rather than
the independent worker–intermediary relationship, and it is
usually treated as an employment relationship.
LABOR CONTRACTORS
Labor contractors operate in a manner that is somewhat
similar to temporary stang agencies, but they play a larger
The Hamilton Project • Brookings 23
role in managing the temporary workers in the workplace.
ey nd the workers and provide their labor to the client
employer to ll a temporary need, but the labor contractors
directly supervise the work of the employees on behalf of the
client employer, usually in the client’s workplace. In this way,
the labor contractor’s control over the employee is greater
than that exercised by the temporary stang agency, and an
employment relationship is more likely, as a result.
UNION HIRING HALLS
In certain industries such as construction and maritime, some
employers nd new employees through hiring halls established
and managed by the unions with which the employers have a
collective bargaining relationship. e employer identies a job
opening, contacts the hiring hall in search of a union member
to ll the job, and the union runs a form of competition among
its members—oen based, in part, on objective factors like
seniority in the union, industry, and/or occupation—to ll the
job. e successful union member becomes an employee of the
employer for the duration of the job.
15
Unlike with independent workers, that relationship with the
ultimate customer—the employer—is not eeting. It can last
months or years. Unlike with temporary workers, the union
is not a joint employer. It has no role in the employment
relationship aer running the employment competition.
e hiring occurs according to criteria determined by the
employer, not the union, with certain boundaries established
through collective bargaining. Furthermore, the union is
an instrumentality of the collective of its members. It has
no independent interest or prot motive. In fact, it receives
no payment for running the hiring competition. Rather,
its payment comes in the form of regular dues from its
members—a payment for collective bargaining and other
services rendered—and per capita contributions from the
employer to a trust fund established to nance the hiring hall.
OUTSIDE SALES EMPLOYEES AND DIRECT SALES
WORKERS
Outside sales employees may be the least like independent
workers of these examples. ey typically receive commissions
for selling an individual company’s goods within a territory or
to a list of customers and prospective customers provided by
that company. Some may receive a small base salary that may
or may not be charged against the commissions.
16
Critically,
these sales transactions are conducted at the customer’s
place of business, not at the seller’s facility. Yet the principal
dierence between sales employees and independent workers
is that the former sells goods and the latter sells services. is
is more than mere dierence in form: In addition to exercising
some control over territories and target customers, the
company purveying the goods makes every decision about the
goods, and oen controls aspects of the sales system, including
marketing. e goods purveyor is ultimately responsible to
the customer for the quality and performance of the goods.
Future sales will depend in large part on the quality of the
product, not on the quality of the salesperson.
e neutrality principle requires us to consider whether direct
sales workers should be treated like outside sales employees
given the similarity in their functions, or whether they should
be independent workers. Direct sales workers also aliate
with a company and sell its products. However, there is oen
a meaningful dierence: direct sales workers typically have
greater freedom than outside sales employees to make decisions
about how, when, and to whom they will sell the company’s
products. For example, they generally have the ability to set
their own work hours and days without any direction from the
purveyor of the goods they sell. Direct sales typically involves
face-to-face discussions (or the use of personal social media
channels like Facebook) and product demonstrations that may
be targeted to friends, neighbors, family, and others. Direct
sales workers usually earn commissions, but they receive no
salary. Like independent businesses, some direct sellers have
the opportunity to recruit additional workers to sell their
products and, in return, may receive a portion of the new
seller’s commissions from the company. For these reasons,
under existing law, they are generally treated as independent
contractors rather than as employees.
Although they may not be employees, the question of whether
direct sales workers should be classied as independent
workers turns principally on the answers to three questions.
First, what is the extent of the product-purveying company’s
involvement with the direct sales worker’s customers? If
the company plays a role in identifying potential customers
and connecting them to the direct sales worker in a manner
that resembles the eorts of intermediaries in the online gig
economy, then the direct sales workers may be independent
workers. In other words, the workers, companies, and
customers may have the kind of triangular relationship that
characterizes the independent worker status. Second, does
the company or the worker bear the risk and opportunity
for prot and loss? Independent workers primarily bear
the risk associated with the amount of time and eort they
invest in providing direct customer service. eir additional
investments, like the use of a personal car or tools, are limited.
But if direct sales workers are required to invest in a sizable
inventory of products that they may or may not be able to sell,
whether for themselves or for the sales teams they assemble,
then they are operating more like independent businesses
and probably should be classied as independent contractors
rather than as independent workers. ird, does the purveyor
of goods exert other forms of control over the workers, such
as requiring uniforms? If so, this would militate in favor of a
determination of independent worker status.
24 A Proposal for Modernizing Labor Laws for Twenty-First-Century Work: The “Independent Worker”
I
n the face of new and emerging work relationships in the
online gig economy, this proposal aims to improve worker
classication in three signicant ways: reducing legal
uncertainty, enhancing economic eciency, and strengthening
the social compact.
REDUCING LEGAL UNCERTAINTY
One of the goals of our proposal is to reduce legal uncertainty
associated with the determination of employee, independent
contractor, and independent worker status. In the short run,
a change in law may introduce some additional uncertainty
as new statutory, regulatory, and judicial rules are interpreted
in the context of real-world factual circumstances and
information about the changed legal rule spreads. It is
not uncommon, for example, for the amount of litigation
associated with a particular statute or regulation to increase
temporarily aer an amendment is adopted because the
amendment disrupts received understandings of the meaning
and application of the law. Congress could also increase
uncertainty if new laws are laced with ambiguous language.
On the other hand, a third legal category governing the
treatment of workers will also reduce uncertainty in both
the short and long terms. Legal rules dening “independent
workers” can and should more closely reect the actual
experience of workers in that category than the current
denitions of “employee” and “independent contractor.” As
a result, employers, workers, lawyers, regulators, and judges
seeking to apply this new denition to the facts of a particular
case may nd reaching a conclusion about how the law applies
to these workers both easier and less uncertain. An apt
metaphor is a large tent that is suspended between two poles
positioned at distant ends. With only the two poles, the middle
of the tent will ap sloppily in any reasonably strong wind. But
the introduction of a third pole to hold up the middle of the
tent will reduce the apping and give more shape to the tent,
even if the tent is not perfectly taut.
One way in which legal uncertainty could be reduced would
be to establish a default condition, such as a strong rebuttable
presumption that all workers are employees. Employers could
seek to rebut the presumption by demonstrating that the
workers satisfy all elements of independent contractor status,
which would have to be clearly articulated in a new legal rule.
Although, as noted above, some countries have established a
default rule of this sort, careful deliberation and debate would
be required to determine if a default rule is appropriate in the
context of the U.S. labor market.
ENHANCING ECONOMIC EFFICIENCY
In an ideal labor market with no frictions and perfect
information, the cost of many of the benets that employers are
legally required to provide to employees would be ultimately
borne by employees themselves in the form of lower wages. In
other words, the cost of benets—like taxes—can be shied
from one party to another. Indeed, in the case of mandated
benets, the likelihood of shiing costs from employers to
employees is even higher than it is for taxes since employees
directly value the benets that they receive, which leads to an
outward shi in labor supply (Summers 1989).
Of course, the labor market oen is characterized by frictions
and imperfect information. is is particularly likely to be
the case in traditional employment relationships, where the
employment relationship is expected to endure and employees
and employers make investments in the relationship. Moreover,
individual employees typically face bargaining disadvantages
compared with employers. In this situation, the assignment of
which party is initially required to pay for benets can aect
the party that ultimately bears the cost of the benets. e
default can matter in a bargaining model.
Research has found that 80 percent or more of employers’
costs of providing employee benets, such as health insurance
or workers’ compensation insurance, is ultimately borne by
employees in the form of lower wages (Gruber 1994; Gruber
and Krueger 1991). In addition, the lion’s share of payroll taxes
are likely to be shied from employers to employees because
labor supply is more inelastic than labor demand.
ese observations suggest that most of the economic impact
of requiring intermediaries to provide certain benets or
pay for certain payroll taxes will ultimately be oset in the
form of lower net fees collected by independent workers and
higher commissions taken by intermediaries. For example, if
an intermediary is required to pay for half of its independent
workers’ Social Security contributions, whereas before
independent contractors paid for both halves themselves, fee
Chapter 6. Economic Analysis of the Proposal
The Hamilton Project • Brookings 25
schedules that remunerate independent workers will likely
eventually be adjusted to reduce the independent workers’
compensation by the amount of the intermediary’s Social
Security contributions.
is analysis changes, however, when the independent worker
status enables intermediaries to provide benets in a more
ecient manner than would be the case under independent
contractor status. For example, if intermediaries can provide
life insurance benets more cheaply than workers could
purchase them on their own, then a surplus is created that
enables both sides to benet.
In the standard Coasian explanation for why rms employ
workers as opposed to contracting with external parties to
provide services, transaction costs make it more ecient
for rms to directly employ and supervise workers than to
specify and monitor all of the contingencies required for their
services in a contract. Firms thus nd it more ecient to use
hierarchies, directives, and internal structures to ensure that
the desired work is performed. e emerging online apps have
the potential to greatly reduce the transaction costs associated
with contracting and monitoring the provision of certain
services. us, the Coasian explanation for the growth of
online intermediaries is that new computer and information
technology enables a more ecient means for companies
to contract with third parties (i.e., technology lowers the
transaction costs that induce companies to hire employees
rather than to contract work out).
e Coasian analysis overlooks the role of rent sharing,
morale, and internal labor markets within rms. Because
employee morale is critical for productivity, and because
morale is aected by employees’ perceptions of fairness, rms
oen nd that they must share some rents with workers in
order to maintain high morale, quality, and productivity
(Blanchower, Oswald, and Sanfey 1996). Rent sharing is more
likely to occur in less-competitive industries that have product
market rents to share, in highly unionized industries, and in
highly capital-intensive industries. If networking technology
leads to more disintermediation of traditional employment, it
could have the eect of reducing rent sharing while it raises
productivity. is is less likely in industries where most of the
work is already conducted by independent contractors, such as
taxi services, or in sectors where product markets are highly
competitive and employees are nonunionized.
e standard Coasian analysis assumes a perfectly competitive
and ecient labor market with no transaction costs. However,
in actual labor markets workers and rms oen implicitly or
explicitly (in the case of unionized workplaces) bargain over
wages and face signicant frictions that create transaction
costs. Although the legal assignment of responsibility for
paying taxes or funding benets is irrelevant in a competitive
market because the ultimate incidence would be shied
between the parties based on their relative elasticities of supply
and demand, if a work relationship is marked by bargaining
power, then switching the party responsible for paying for
taxes or benets can have consequences for incidence. e
legal assignment of responsibilities, for example, is likely to
aect the default position in bilateral bargaining settings, and
thus to inuence the ultimate outcomes.
STRENGTHEN THE SOCIAL COMPACT
Over the course of the 20th Century, a social compact developed
between employees and employers in the U.S. that protected
employees from dangerous working conditions, provided a
minimum level of economic security, and dened norms of
fairness. e social compact has served workers, employers,
and society well. is social compact is jeopardized by the
misclassication of employees into independent contractor
status. It is also challenged by emerging forms of work that do
not t neatly into the employee-employer relationship.
Establishing a new legal classication for independent workers
would help to strengthen the social compact. In particular,
components of the social compact that are appropriate for
their working relationship, such as Civil Rights protection
and the right to collectively bargain, would be extended to
independent workers. is would have the immediate eect
of bringing more workers under the umbrella of important
components of the social compact. In addition, adhering
to the neutrality principle would help maintain the social
compact for traditional employees by reducing the incentive
for employers to reclassify workers as independent contractors.
26 A Proposal for Modernizing Labor Laws for Twenty-First-Century Work: The “Independent Worker”
T
he online gig economy represents a small but rapidly
growing segment of the workforce, especially in the
ride-sharing and food-delivery sectors. is new
and emerging sector has the potential to provide many new
opportunities for workers and customers but raises serious
challenges to the administration of existing employment,
labor, and tax law. In particular, the workers who utilize
intermediaries to identify customers to deliver services, such
as car rides, do not t neatly into existing legal categories of
independent contractors and employees. We have sought
to cra a new employment status that we call “independent
workers,” to ll this void and improve the eciency and
fairness of the labor market, and reduce legal uncertainty.
Many workers in the “oine economy” who are currently
classied as independent contractors, such as taxi drivers,
would also t into this new category.
Independent workers would receive some protections and
benets of employees, such as the right to organize and the
requirement that intermediaries contribute half of Social
Security and Medicare payroll taxes, but not others, such as
time-and-a-half for overtime hours. Most importantly, we
think that reforms along the lines that we propose would help
to protect and extend the hard-earned social compact that has
protected workers and improved living standards over the
past century, reduce uncertainty, and enhance the ecient
operation of the labor market.
Chapter 7. Conclusion
The Hamilton Project • Brookings 27
Chapter 8. Appendix
APPENDIX TABLE
Description of Several Prominent Online Intermediary Companies
Description/Business Model Date Formed Size
Agent
Anything
• Similar to TaskRabbit, but only college
students perform tasks (must verify
through .edu email address).
• Client posts a “mission” and a price; an
interested agent accepts the mission and
becomes responsible for completing it.
• Also has option to facilitate bidding among
agents. This model appears to be a hybrid
between TaskRabbit’s new model (where the
client is presented with options to choose among)
and old model (where the Taskers bid for tasks).
• June 2010:
founded
N/A
Axiom • Provides “tech-enabled legal services”
including data analytics of contracts, due
diligence, adjusting to regulatory changes,
etc. Axiom also provides secondments
(temporary in-house counsel).
• Attorneys and staff are paid an annual
salary, but only for the months actually
worked/staffed on project.
• 1999: founded • 1,500 employees in 11 offices.
• Recently signed a $73M contract
to handle a trade agreement.
• Almost $200M in sales as of summer 2015.
• $28M in 2013 from 1 investor.
Caviar • Similar to GrubHub, but specializes
in delivering to restaurants that
ordinarily do not deliver their food.
• Couriers are compensated per-
delivery. They are interviewed and
background checked but not trained.
• Takes 25% commission per delivery.
• March 2013:
launched
• Caviar was acquired in August 2014 by
“Square” (portable iPhone device for credit card
payments). Caviar received only Square stock
in the transaction but was valued at $90M.
• By August 2014, had raised
$15M in venture funding.
ChaCha • A Web site and app launched in 2006. Users
get answers to questions—essentially, a search
engine where answers are generated by
humans (“Guides”) rather than by an algorithm.
• Guides are paid a few cents for each
question they answer. ChaCha has some
rules about who can become a Guide—e.g.,
must have access to highspeed internet,
must complete training and orientation,
and pass a “Readiness Test.”
• 2006: founded • In 2006, ChaCha had over $100M in
funding and was generating excitement,
however by 2015 it had laid off most of its
employees and appears to be declining
rapidly in value and in market-share.
28 A Proposal for Modernizing Labor Laws for Twenty-First-Century Work: The “Independent Worker”
Description/Business Model Date Formed Size
Clickworker
GmbH
• Similar to Mechanical Turk, but
specifically for standardized tasks that
cannot be done by a computer.
• “Clickworkers” provide a variety of services,
including taking surveys, digitalization,
translation, creating content, and doing
Web research. Clients come to Clickworker
with a project, Clickworker breaks down the
project into smaller tasks, doles out the work
to individual clickworkers, and conducts
quality control over the final product.
• Clickworkers use their own computers and
choose their own hours. Clickworker tests and
trains the individual clickerworkers to determine
which worker is suitable for what tasks.
• 2005: founded as
humangrid GmbH
• 2011: relaunched
as online
marketplace
• Reportedly has worked with 700,000
freelancers in 136 countries.
• $14.2M in funding as of July 2013.
Curb • Similar to Uber, but draws on a pool of
licensed taxi drivers. Standard taxi rates
apply; the benefit to the rider is tracking
the ride through an app similar to Uber. It
appears the pay structure is the same as in an
ordinary taxi setting: the app merely facilitates/
modernizes the process of hailing a cab.
• 2007: founded as
RideCharge, an
online platform to
book and expense
ground travel for
business travelers
• 2009: becomes
Taxi Magic, a
mobile app for
ordering taxis
on demand
• August 2014:
rebrands as Curb
• Available in 60 cities.
• Draws on 35K cars from 90 cab companies.
• $10.7M in funding as of August 2014.
Eden
McCallum
• Management consulting firm that relies on
a stable of independent consultants who
are selected to teams to work on projects.
Consultants can decline any project.
• 2000: founded • Over 500 independent consultants.
• Over 1,500 projects completed.
• Revenue unclear, but Harvard Business Review
reported it was a “$40M firm” in 2012.
Fiverr • Sellers offer services ranging from
graphic design to translation.
• The default price is a fixed $5 fee, although
the fixed price can be set higher.
• Fiverr profits by taking a processing fee from
the buyer and taking a cut of the fee paid
to the seller. Sellers collect $4 for a $5 “gig”
completed. Buyer pays a $.50 processing fee
on purchases $10 and under. For purchases
over $10, the processing fee is 5%.
• As sellers complete orders and maintain
low cancellation rates and 4+ star rating,
they can move up “levels” which offer
more perks—especially greater exposure
on the Web site and ability to have
more gigs listed at the same time.
• February 2010:
launched
• Ranked among 100 most popular Web sites in
the US (and top 150 in the world) since 2013.
• Raised $30M in Series C funding in August
2014, bringing total funding to $50M.
• Claims over 300M gigs completed since 2010.
The Hamilton Project • Brookings 29
Description/Business Model Date Formed Size
Gengo • Crowdsourced language translation
services headquartered in Tokyo.
• Translators must pass a proprietary proficiency
exam. Translation quality is monitored.
• Businesses post translation work to the
platform. Translators self-assess the difficulty
and timeline. If interested, the translator
selects the job and begins to work.
• Standard rate charged is $.06/word;
standard rate paid is $.03/word. Higher
rates apply for more difficult work.
• 2008: founded
as myGengo
• 2012: rebranded
as Gengo
• 10,000+ registered translators.
• As of May 2015, had raised $24.2M
in 6 rounds from 23 investors.
• 2013 Series B investment of $12M.
Grubhub • A large, public company that delivers food
from local restaurants. It has begun developing
its own delivery service that will function like
Curb’s delivery service. It previously relied
on the restaurants’ own delivery teams.
• 2004: founded
• 2013: merges
with Seamless
• 700 US cities and London.
• Approx. 174K order placed daily.
• Trading on NYSE at $27.34/share on 9/14/2015.
Handy • Allows individuals to hire home
cleaners, plumbers, or handymen.
• Individuals select a day/time and project,
and Handy selects and dispatches a
“professional” to assist. Individuals cannot
review or select the professional.
• Professionals are background
checked and insured.
• Handy takes 20 percent of booking price.
• 2012: founded • $60.7M in 5 rounds of funding from 9 investors.
• Available in 37 cities across
US, UK, and Canada.
• By June 2015 claims to have
completed 1 million bookings.
Hourly Nerd • Businesses hire freelance consultants (current
MBA students and graduates) on an hourly
basis. Consultants choose their own fees
and create their own profiles. Businesses
submit a project and outline their needs.
• HourlyNerd’s algorithm generates
appropriate “experts” who then submit bids
for the businesses to choose among.
• HourlyNerd restricts the “nerds” to those who
have graduated from certain selective schools.
• Consultants are also provided
with a “proprietary toolkit” as well
as formatting templates.
• Takes a 14.5 percent commission.
• February 2013:
launched
• 10,000 consultants.
• Raised $7.8M in Series B (in February 2015).
Instacart • Same-day shopping and delivery from stores
like Whole Foods, Costco, and Petco.
• In summer 2015 Instacart divided the shopping
and delivery role. Shoppers have the option
to switch to employee status, but drivers
may not. Most shoppers have switched.
• For orders over $35, Instacart charges $3.99
for orders delivered in 2 hours and $5.99
for deliveries within the hour. For orders
under $35 (but over $10), two hour deliveries
are $7.99 and hour deliveries are $9.99.
During “Busy Periods” these base prices
rise. Instacart also offers a membership that
is $99/year, but free 2-hour deliveries.
• On top of delivery fees, the customer pays
for the groceries, which are priced based on
agreements between Instacart and the retailer.
• July 2012: founded • 7K shoppers in 16 US cities.
• $274.8M in 5 rounds of investing.
30 A Proposal for Modernizing Labor Laws for Twenty-First-Century Work: The “Independent Worker”
Description/Business Model Date Formed Size
Lyft • Peer-to-peer ridesharing app similar to Uber.
• Takes a 20 percent commission.
• Summer 2012:
launched as a
short-distance
ride share offshoot
of Zimride that
connected riders
and drivers for
long distances
to split costs (not
a paid service)
• $1B in venture funding.
• Valued at $2.5B.
• Operates in 65 U.S. cities.
Mechanical
Turk
• Part of Amazon.
• Crowdsourcing online marketplace
connecting individuals (“Turkers”) and
businesses (“Requesters”) to do work that
computers cannot (e.g., writing product
descriptions, selecting the “best” picture).
• Turkers browse available jobs and complete
them at a payment rate set by the employer.
• Unknown: founded
for Amazon’s
internal use
• 2005: launched
to the public
• N/A
Medicast • On-demand “house call” doctor visits. Patients
can choose from three subscription levels:
Plus ($39/mo for two visits spread over 12
months), Premium ($75/mo for 4 visits over 12
months); On Demand (one-time visit for $249
with an added night/weekend fee of $100).
• Doctors are paid according to the number
of patients seen per day. They are provided
an iPad with Medicast resources on it.
• 2013: founded • 1.94M in 2 rounds of funding.
• Currently operating in Miami/South
Florida and LA/Orange County.
Red Beacon • Peer-to-peer task outsourcing.
• Similar to Task Rabbit, but
specifically for household services
(e.g., repairing leaky faucet).
• Model: (1) a client describes the size and
scope of the project, (2) Red Beacon sends
a list of qualified repair people, (3) the client
selects up to five to interview, (4) the selected
repair people contact the client to discuss
the project and name a price. The client then
selects one to perform the task based on
reviews, description of expertise, and price.
• 2008: founded
• 2012: purchased
by Home Depot
• Home Depot, Inc. purchased for
an undisclosed amount.
Samasource • Similar to Mechanical Turk: provides
business services to companies.
Samasource breaks down large-scale
projects and sources it out to workers.
• Major difference is that it is a 501(c)
(3)—its central aim is helping
workers in developing nations.
• Samasource trains workers in
basic computer skills.
• -Samasource takes a “small
cut” of each transaction.
• June 2008:
founded
• As of March 2015 has had 6,527 workers.
• Raised $1.5M in 12 rounds from 8 Investors.
The Hamilton Project • Brookings 31
Description/Business Model Date Formed Size
Shyp • Individuals and businesses pay Shyp
to package and send any item.
• “Shyp Heroes” have been background
checked and “extensively” trained at Shyp
Academy. They are not allowed to accept tips.
• Shyp appears to be moving away entirely
from peer-to-peer or contract models
and toward an employee model.
• March 2014:
launched
• Operates in 5 cities.
• As of September 2015 it has raised $62.2M.
Sidecar • Similar to Uber/Lyft but drivers can set
their own prices. Riders can also screen
drivers by selecting a ceiling for how much
they are willing to pay for a ride. This
includes electing to share the ride with a
stranger headed in the same direction.
• Sidecar also offers a delivery service, which
is now its primary focus (according to their
CEO in a statement in August 2015).
• Takes a 25 percent fee from each transaction.
• January 2012:
launched
• Operates in 8 U.S. cities.
• $35M in venture funding.
Skillshare • Peer-to-peer courses/classes.
• “Teachers” create video content for the
Web site, teaching skills from a variety of
areas (e.g., how to use InDesign, how to
build a Web site, how to do calligraphy).
”Members” subscribe to view the video
classes for a monthly fee. Teachers earn
money through a royalties pool and also can
earn bonuses for recruiting other teachers.
• Skillshare provides teachers with materials
to help them get started on creating content.
Skillshare also establishes “class guidelines”
that teachers must follow (including
resolution quality, minimum duration, and
level of creativity/educational nature).
• Skillshare takes 50 percent of the
Premium Membership revenue.
• November 2010:
founded
• April 2011: site
went live
• As of March 2014 it was valued at $20M.
• By March 2014 it had raised
$10.8M in venture funding.
Task Rabbit • Online and mobile marketplace to
outsource small jobs and tasks.
• Users name a task and a price and then
Task Rabbit assigns a Tasker to the job.
• Task Rabbit used to be a bidding-model
marketplace until its July 2014 reboot.
Now, Taskers are assigned instead of
bidding, they must wear a uniform, and
tasks are paid on an hourly basis.
• Task Rabbit takes about 20
percent commission.
• 2008: founded • 2M users.
• 50,000 Taskers.
• $134M valuation.
Thumbtack • Online marketplace for services (e.g., wedding
officiating, personal training, wall painting).
• Thumbtack vets the professionals
(checking licenses, etc.).
• Users describe what service they need
performed. If interested, professionals can
pay $3-25 to send a quote to the potential
customer. The user then evaluates the quotes
and selects a professional to complete the task.
• 2009: founded • 150,000 professionals available.
• 5 million projects/year.
• August 2014: raised $100M through Series
D round of venture capital funding.
32 A Proposal for Modernizing Labor Laws for Twenty-First-Century Work: The “Independent Worker”
Description/Business Model Date Formed Size
Uber • Peer-to-peer ridesharing app.
• App determines the price for each ride.
• Typically takes 20 percent commission
and collects fee for insurance.
• 2009: launched
black car service
• 2012: launched
UberX (taxi-like
service with
regular cars)
• $7B in venture funding.
• Valued at $50B.
• Available in 58 countries and 300 cities.
Upwork • Connect businesses with freelance workers
(e.g., computer programming, translating,
legal work). Similar to Mechanical Turk.
• A business posts a job onto the platform and
interested freelancers apply. The business
interviews the interested applicants and
decides. Business decides if payment
is made per hour or per project.
• Upwork takes 10 percent commission.
• 2003: founded as
Elance-oDesk
• 2015: rebranded
as Upwork
• 4 million registered clients.
• 9 million registered freelancers.
• $1B in work done per year.
Washio • “Uber for laundry.” Delivers
laundry and dry cleaning.
• Customers place an order on the app,
specifying a time window for pickup and
drop off. “Ninjas” collect the clothes, which
are dry cleaned or washed, dried, and
folded. Customers can have clothes picked
up within 30 minutes with WashioNow.
• Ninjas must have a driver’s license, smart
phone, and a car made after 2000.
• 2013: founded • $13M in venture funding.
The Hamilton Project • Brookings 33
Authors
Professor Alan B. Krueger
Princeton University
Professor Seth D. Harris
Cornell University
Alan B. Krueger is the Bendheim Professor of Economics and
Public Aairs at Princeton University. He served as Chairman
of President Barack Obama’s Council of Economic Advisers
from November 2011 to August 2013, and was a member of
the President’s Cabinet. In 2009-10, he served as Assistant
Secretary for Economic Policy and Chief Economist of the
U.S. Department of the Treasury and in 1994-95 he served
as Chief Economist of the U.S. Department of Labor. He is
the founding Director of the Princeton University Survey
Research Center. He wrote for the New York Times Economic
Scene column and Economix blog from 2000 to 2009. Krueger
was named a Sloan Fellow in Economics in 1992 and an
NBER Olin Fellow in 1989-90. He was editor of the Journal
of Economic Perspectives from 1996 to 2002. He was elected
a Fellow of the Econometric Society in 1996, a fellow of the
Society of Labor Economists in 2005, and a member of the
Executive Committee of the American Economic Association
in 2004. Professor Krueger was awarded the Kershaw Prize
by the Association for Public Policy and Management in 1997
for the most signicant contributions to public policy research
by someone under age 40, elected a fellow of the American
Academy of Arts & Sciences in 2002, and awarded the IZA
Prize in Labor Economics in 2006. He earned a BS degree
with honors from Cornell University in 1983 and a PhD in
Economics from Harvard University in 1987.
Seth D. Harris is a Distinguished Scholar at Cornell University’s
School of Industrial & Labor Relations. He served four and one-
half years as the Deputy US Secretary of Labor and six months
as Acting US Secretary of Labor and a member of President
Barack Obama’s Cabinet. Beginning in January 2007, he chaired
Obama for America’s Labor, Employment and Workplace Policy
Committee, and later founded the campaign’s Disability Policy
Committee. He also advised then-Senator Obama on issues arising
in the Senate Health, Education, Labor and Pensions Committee.
In August 2008, he joined the Obama-Biden transition planning
committee’s Agency Review Working Group. Aer Election Day
2008, he oversaw the Obama-Biden transition team’s eorts in
the Labor, Education and Transportation departments and 12
other agencies. From 1997 to 2000, he served as Counselor to
Secretary of Labor Alexis Herman. From 1993 to 1997, he served
as Acting Assistant Secretary of Labor for Policy, Deputy Assistant
Secretary of Labor for Policy, and Special Assistant to Secretary
Robert of Labor Reich. Prior to re-joining the Labor Department
in 2009, he served as a professor of law at New York Law School
and director of its Labor and Employment Law programs, as well
as a visiting professor of law at Seton Hall Law School. He served
as a law clerk to Judge William Canby of the US Court of Appeals
for the Ninth Circuit and to Judge Gene Carter of the US District
Court for the District of Maine. He has published extensively on
workplace and employment issues, from both legal and economic
perspectives. He earned a BS degree with honors from Cornell
University in 1983 and a JD cum laude from New York University
School of Law in 1990 where he was the Editor-in-Chief of the
Review of Law and Social Change.
Acknowledgments
e authors thank David Cho and Carolyn Wald for excellent
research assistance. Diane Whitmore Schanzenbach and
Larry Katz provided helpful comments on an earlier dra.
Disclosure
Seth Harris has no economic relationship with any company
operating in the online gig economy. In the interest of full
disclosure, Alan Krueger acknowledges that he has coauthored
a study commissioned by Uber in the past, although he has no
ongoing relationship with the company or any other company
operating in the online gig economy.
34 A Proposal for Modernizing Labor Laws for Twenty-First-Century Work: The “Independent Worker”
Endnotes
1. Extensive litigation and administrative decision making is already under
way involving Uber and other intermediaries. See, e.g., O’Connor v. Uber
Techs., 2015 U.S. Dist. LEXIS 116482, 80 Cal. Comp. Cases 852 (N.D. Cal.
2015); Levin v. Caviar Inc., Case No. 15-1285 (N. D. Cal. 2015); Singer v.
Postmates, Case No. 14-1284 (N.D. Cal 2015.)
2. Truck drivers oer an instructive example. Compare North American Van
Lines, Inc. v. N.L.R.B. [National Labor Relations Board], 869 F.2d 596 (D.C.
Cir. 1989) (under the National Labor Relations Act [NLRA], truck drivers
are independent contractors) with Aetna Freight Lines, Inc. v. N.L.R.B.,
520 F.2d 928, (6th Cir. 1975) (under NLRA, truck drivers are employees).
Compare also Redwine v. Refrigerated Transport Co., 84 S.E.2d 478 (Ga.
Ct. App. 1954) (under state unemployment insurance law, truck drivers
are employees) and Rozran v. Durkin, 381 Ill. 97, 45 N.E.2d 180 (Ill. 1942)
(under state unemployment insurance law, truck drivers are employees);
with Nat’l Trailer Convoy, Inc. v. Undercoer, 137 SE2d 328 (Ga. Ct.
App. 1964) (under state unemployment insurance law, truck drivers are
independent contractors) and Hammond v. Dep’t of Empl.,480 P2d 912
(Idaho 1971) (under state unemployment insurance law, truck drivers are
independent contractors).
3. ere is a subtle but important distinction between a company like Apple
and Ly in this regard. Apple does not manufacture iPhones, which are
integral to its business, but instead contracts out their manufacture to
Foxconn and other suppliers. Apple is not an intermediary that hires
independent workers to provide personal services to third-party customers.
Rather, it enters into a bilateral relationship with Foxconn in which Apple
buys what Foxconn produces according to Apple’s specications. Foxconn
does not interact with Apple’s customers. Apple contracts with Foxconn to
produce a good that Apple and others sell to customers.
4. See, e.g., United States v. Federation of Physicians and Dentists, Inc., CA 98-
475 JJF (D. Del. Oct. 22, 2001) (consent decree), http://www.justice.gov/atr/
case-document/proposed-nal-judgment-116; United States v. Federation
of Certied Surgeons and Specialists, Inc., 64 Fed. Reg. 5831 (Dep’t
Justice 1999) (consent decree); Federal Trade Commission v. College of
Physicians and Surgeons of Puerto Rico, https://www.c.gov/enforcement/
cases-proceedings/9710011/college-physicians-surgeons-puerto-rico-
centralmed-inc-fajardo. See generally Kennedy 2014 at 155–60; Brewbaker
2000.
5. Title VII of the Civil Rights Act, 42 U.S.C. § 2000e (1964) (prohibiting
employment discrimination on the basis of sex, race, color, national origin,
and religion); Age Discrimination in Employment Act, 29 U.S.C. §§ 621–
634 (1967) (prohibiting age discrimination against anyone at least forty
years of age); Americans with Disabilities Act, 42 U.S.C. §§ 101–336 (1990)
(prohibiting discrimination against people with disabilities); Equal Pay Act,
29 U.S.C. § 206 (1963) (prohibiting employment discrimination between
employees on the basis of sex by paying unequal wages for equal work).
6. Exec. Order No. 11,246, 41 C.F.R. 60-1.1 (1978) (prohibiting U.S.
government contractors from employment discrimination on the basis
of race, color, religion, sex, or national origin); Exec. Order No. 13,672,
41 C.F.R. 60-4.9 (2014) (prohibiting U.S. government contractors from
discriminating on the basis of sexual orientation and gender identity);
Rehabilitation Act, 29 U.S.C. § 503 (1973) (prohibiting discrimination
on the basis of disability in programs conducted by federal agencies,
programs receiving federal funding, in federal employment, and by federal
contractors); Vietnam Era Veterans’ Readjustment Assistance Act, 38
U.S.C. § 4212 (1974) (prohibiting discrimination by employers and federal
contractors against disabled veterans and veterans who served active duty
during war).
7. Some state laws protect a long list of groups. See, e.g., Minn. Stat. § 363A.17
(providing that it is an unfair discriminatory practice for a business “to
intentionally refuse to do business with, to refuse to contract with, or to
discriminate in the basic terms, conditions, or performance of the contract
because of a person’s race, national origin, color, sex, sexual orientation,
or disability, unless the alleged refusal or discrimination is because of a
legitimate business purpose”).
8. For example, Section 1981 has been applied not only to discrimination
against African-Americans and white Americans, but also against Latinos,
Jews, and Arabs. See, e.g., St. Francis Coll. v. Al-Khazraji, 481 U.S. 604
(1987); Shaare Tela Congregation v. Cobb, 481 U.S. 615 (1987); see also
Pourghoraishi v. Flying J, 449 F.3d 751 (7th Cir. 2006) (collecting cases).
9. It is worth noting, however, that the dierence in forum may disadvantage
low-wage workers who may be unable to aord private counsel to bring a
claim in federal court. Complaints can be led with the EEOC without the
assistance of counsel, although the EEOC is not an adjudicative body and
may not be able to generate a resolution of the complaint.
10. Ironically, Milton Friedman was a key contributor in the group at the
U.S. Treasury Tax Research Department that helped develop the idea of
withholding taxes (Taylor 2014).
11. We do not seek to address the agency law question of whether an
intermediary should be held liable for the negligent or intentionally
tortious acts of the independent workers with which it does business. is
is not fundamentally an employment question and it would not be resolved
by federal laws. Rather, it is an issue for state courts and legislatures.
12. To the contrary, most aspects of the FLSA’s protections against exploitative
child labor do not require measuring work hours. For this reason, these
same protections should apply in the world of independent workers to
guard against any opportunity for this new form of work relationship to be
used for the exploitation of children.
13. Senator Mark Warner (D-VA) has been on the vanguard in considering
alternative models for providing independent workers with social safety
net benets. See, for example, Warner (2015).
14. e NLRB recently reconsidered and signicantly expanded its denition
of joint employment, for example. See See Browning-Ferris Indus,. Case
32-RC-109684 (NLRB Aug. 27, 2015) (Decision on Review and Direction).
15. ese jobs oen have xed terms or are associated with the completion of
a particular task, like constructing a building or sailing a cargo ship from
one port to another.
16. Perhaps because of the prevalence of commissions in outside sales,
Congress exempted outside sales employees from the FLSA’s minimum
wage and overtime protections (29 U.S.C. § 213(a)(1)). e fact that
Congress felt the need to exempt these workers from the FLSA’s protections
strongly suggests that it had concluded these workers otherwise would be
treated as employees.
The Hamilton Project • Brookings 35
Blanchower, David G., Andrew J. Oswald, and Peter Sanfey. 1996.
“Wages, Prots, and Rent-Sharing.” Quarterly Journal of
Economics 111: 277–51.
Brewbaker, William S. 2000. “Physician Unions and the Future of
Competition in the Health Care Sector.” 33 U.C. Davis Law
Review (545).
Browning-Ferris Indus. 2015. National Labor Relations Board. Case
32-RC-109684
Brudney, James J. 2005. “Neutrality Agreements and Card Check
Recognition: Prospects for Changing Labor Relations
Paradigms.” Iowa Law Review 90: (819).
Bureau of Labor Statistics (BLS). 2015, January 28. “e Economics
Daily.” U.S. Department of Labor, Washington, DC.
Burton Jr., John F. 2009. “Workers’ Compensation.” In Labor and
Employment Law and Economics, edited by Kenneth G.
Dau-Schimdt, Seth D. Harris, and Orly Lobel, 235–74.
Cheltenham, UK: Edward Elgar Publishing.
Estlund, Cindy. 2002. “e Ossication of American Labor Law.”
Columbia Law Review 102 (1527).
Gandhi, Natwar M. 1994. “Improving Independent Contractor
Compliance with Tax Laws.” U.S. Government
Accountability Oce, Washington, DC.
Grabell, Michael, and Howard Berkes. 1974. “Inside Corporate
America’s Campaign to Ditch Workers’ Comp.” ProPublica.
Gruber, Jonathan. 1994. “e Incidence of Mandated Maternity
Benets.” American Economic Review 83 (3): 622–41.
Gruber, Jonathan, and Alan Krueger. 1991. “e Incidence of
Mandated Employer-Provided Insurance: Lessons from
Workers’ Compensation Insurance.” Tax Policy and the
Economy 5: 111–43.
Hall, Johnathan, and Alan Krueger. 2015. “An Analysis of the Labor
Market for Uber’s Driver Partners in the United States.”
Industrial Relations Section Working Paper No. 587,
Princeton University.
Hudnall, David. 2015. “Former KC Uber Driver Organizing
National Strike Against the Rideshare Company.” e
Pitch, September 30.
Kennedy, Elizabeth. 2014. “Freedom from Independence: Collective
Bargaining for Dependent Contractors.” Berkeley Journal of
Employment and Labor Law 26 (144).
Kleiner, Morris. 2001. “Intensity of Management Resistance:
Understanding the Decline of Unionization in the Private
Sector.” Journal of Labor Research 22 (3): 519–40.
Lui, Eric. 2013. “Viewpoint: e Decline of Unions is Your Problem
Too.” Time, January 29.
McKinsey Global Institute. 2015, June. “Connecting Talent with
Opportunity in the Digital Age.”
Organisation for Economic Co-operation and Development
(OECD). “Non Regular Employment, Job Security and the
Labour Market Divide.” http://www.oecd.org/els/emp/
Emo2014-Annex-chapter4.pdf
Rothschild, Michael, and Joseph Stiglitz. 1976. “Equilibrium
in Competitive insurance Markets: An Essay on the
Economics of Imperfect Information.” Quarterly Journal of
Economics 90 (4): 629–49.
Summers, Lawrence. 1989. “Some Simple Economics of Mandated
Benets.” American Economic Review Papers and
Proceedings 79 (2): 177–83.
Taylor, Timothy. 2014. “How Milton Friedman Helped Invent Income
Tax Withholding.” Conversable Economist, April 12.
Warner, Mark. 2015. “Asking Tough Questions about the Gig
Economy.” Washington Post, June 18.
Weil, David. 2014. e Fissured Workplace: Why Work Became So
Bad for So Many and What Can Be Done to Improve It.
Cambridge, MA: Harvard University Press.
Weiler, Paul C. 2009. Governing the Workplace: e Future of Labor
and Employment Law. Boston, MA: Harvard University
Press.
Woodbury, Stephen A. 2009. “Unemployment.” In Labor and
Employment Law and Economics, edited by Kenneth G.
Dau-Schimdt, Seth D. Harris, and Orly Lobel, 480–516.
Cheltenham, UK: Edward Elgar Publishing.
References
36 A Proposal for Modernizing Labor Laws for Twenty-First-Century Work: The “Independent Worker”
The Hamilton Project • Brookings 3
ADVISORY COUNCIL
GEORGE A. AKERLOF
University Professor
Georgetown University
ROGER C. ALTMAN
Founder & Executive Chairman
Evercore
KAREN ANDERSON
Principal
KLA Strategies
ALAN S. BLINDER
Gordon S. Rentschler Memorial Professor of
Economics & Public Affairs
Princeton University
JONATHAN COSLET
Senior Partner &
Chief Investment Officer
TPG Capital, L.P.
ROBERT CUMBY
Professor of Economics
Georgetown University
STEVEN A. DENNING
Chairman
General Atlantic
JOHN DEUTCH
Institute Professor
Massachusetts Institute of Technology
CHRISTOPHER EDLEY, JR.
The Honorable William H. Orrick, Jr.
Distinguished Professor; Faculty Director,
Chief Justice Earl Warren Institute on
Law & Social Policy
Boalt School of Law
University of California, Berkeley
BLAIR W. EFFRON
Partner
Centerview Partners LLC
DOUG ELMENDORF
Former Director
Congressional Budget Office
JUDY FEDER
Professor & Former Dean
McCourt School of Public Policy
Georgetown University
ROLAND FRYER
Henry Lee Professor of Economics
Harvard University
MARK T. GALLOGLY
Cofounder & Managing Principal
Centerbridge Partners
TED GAYER
Vice President &
Director of Economic Studies
The Brookings Institution
TIMOTHY GEITHNER
President
Warburg Pincus
RICHARD GEPHARDT
President & Chief Executive Officer
Gephardt Government Affairs
ROBERT GREENSTEIN
Founder & President
Center on Budget and Policy Priorities
MICHAEL GREENSTONE
The Milton Friedman Professor in Economics
Director, Energy Policy Institute at Chicago
University Of Chicago
GLENN H. HUTCHINS
Co-Founder
Silver Lake
JAMES JOHNSON
Chairman
Johnson Capital Partners
LAWRENCE F. KATZ
Elisabeth Allison Professor of Economics
Harvard University
MELISSA S. KEARNEY
Senior Fellow, The Brookings Institution
Professor of Economics
University of Maryland
LILI LYNTON
Founding Partner
Boulud Restaurant Group
MARK MCKINNON
Former Advisor to George W. Bush
Co-Founder, No Labels
ERIC MINDICH
Chief Executive Officer & Founder
Eton Park Capital Management
SUZANNE NORA JOHNSON
Former Vice Chairman
Goldman Sachs Group, Inc.
PETER ORSZAG
Vice Chairman of Corporate and
Investment Banking
Citigroup, Inc.
RICHARD PERRY
Managing Partner &
Chief Executive Officer
Perry Capital
MEEGHAN PRUNTY
Senior Advisor
The Hamilton Project
ROBERT D. REISCHAUER
Distinguished Institute Fellow
& President Emeritus
Urban Institute
ALICE M. RIVLIN
Senior Fellow, The Brookings Institution
Professor of Public Policy
Georgetown University
DAVID M. RUBENSTEIN
Co-Founder &
Co-Chief Executive Officer
The Carlyle Group
ROBERT E. RUBIN
Co-Chair, Council on Foreign Relations
Former U.S. Treasury Secretary
LESLIE B. SAMUELS
Senior Counsel
Cleary Gottlieb Steen & Hamilton LLP
SHERYL SANDBERG
Chief Operating Officer
Facebook
RALPH L. SCHLOSSTEIN
President & Chief Executive Officer
Evercore
ERIC SCHMIDT
Executive Chairman
Alphabet Inc.
ERIC SCHWARTZ
76 West Holdings
THOMAS F. STEYER
Business Leader, Philanthropist &
Clean Energy Advocate
LAWRENCE SUMMERS
Charles W. Eliot University Professor
Harvard University
PETER THIEL
Entrepreneur, Investor & Philanthropist
LAURA D’ANDREA TYSON
Professor of Business Administration
and Economics; Director, Institute for
Business & Social Impact
Berkeley-Haas School of Business
DIANE WHITMORE SCHANZENBACH
Director
4 Informing Students about Their College Options: A Proposal for Broadening the Expanding College Opportunities Project
WWW.HAMILTONPROJECT.ORG
WWW.HAMILTONPROJECT.ORG
1775 Massachusetts Ave., NW
Washington, DC 20036
(202) 797-6484
Printed on recycled paper.
Highlights
Seth Harris of Cornell University and Alan Krueger of Princeton University propose the creation of
a new legal category of workers, to be called “independent workers,” to address the current legal
uncertainty regarding whether workers in the online gig economy should receive employment
and tax benefits and protections. Their proposal would allow independent workers to gain access
to collective bargaining, various forms of insurance, civil rights protections, employer-provided
benefits, and tax withholding.
The Proposal
Create a New Classification for Independent Workers.
Congress and, where necessary, state
legislatures would pass legislation to establish a new classification for independent workers. In
doing so, Congress and state legislatures would consider three guiding principles in the new
worker classification system to recognize that: work hours are difficult or impossible to measure,
businesses should not organize themselves to fit their workers into one status over another,
and workers and businesses should maximize the joint benefits of their relationship. The new
classification would encompass both new types of work, such as jobs in the online gig economy,
and more-established forms, such as taxi driving.
Assign Benefits and Protections to Independent Workers.
Congress would assign new benefits
and protections to independent workers, following the proposed guiding principles. Benefits such
as tax withholding and various forms of insurance would be available to independent workers
without businesses facing full employment classification, while benefits tied to hours such as
minimum wage and overtime pay would be excluded.
Benefits
This proposal would address the uncertainty that workers and businesses face in the current
legal environment regarding a range of legal protections and benefits that employees receive.
Harris and Krueger argue that the proposal would increase efficiency in the labor market,
enhance worker protections, encourage innovation, and decrease costly legal battles by
addressing a key deficiency in current employment law.
