WHEN I WAS a youngster, the only time I heard someone refer to a “gig” in the employment context was when
some hip musician spoke about a job. The “gig” was often only a one-night, or otherwise short-term, engagement.
The venerable Grove Dictionary of Music and Musicians describes the term as one “commonly applied to a musical
engagement of one night’s duration only; to undertake such an engagement.” One can find a plethora of
suggestions as to why this type of engagement is called a “gig.” Whatever its origin, the term was slowly co-opted
by hipsters to mean, more generally, a person’s job, or his or her set of predilections or interests (“that’s not my
Apparently picking up on the informal or transitory nature of the type of employment connoted by the term
“gig,” some have come to use the term “gig workers” to describe Uber and Lyft drivers, as well as others who
take on episodic work as they please, typically facilitated by online (that is, Internet-based) applications or other
software. The set of working relationships arising in this fashion is sometimes referred to as the “gig economy” or
“online gig economy.”
The Congressional Research Service (CRS), in a report entitled “What Does the Gig Economy Mean for
Workers?,” describes the gig economy in this way:
The gig economy is the collection of markets that match providers to consumers on a gig (or job) basis in
support of on-demand commerce. In the basic model, gig workers enter into formal agreements with ondemand
companies (e.g., Uber, TaskRabbit) to provide services to the company’s clients. Prospective clients
request services through an Internet-based technological platform or smartphone application that allows
them to search for providers or to specify jobs. Providers (i.e., gig workers) engaged by the ondemand
company provide the requested service and are compensated for the jobs.
The CRS report notes that other terms sometimes used for the “gig economy” are “peer-to-peer markets or
the peer economy, the sharing economy, collaborative economy, matching economy, and talent marketplaces.”
See ERS Summary, available at https://www.fas.org/sgp/crs/misc/R44365.pdf
In explaining the origin of the gig economy, the CRS report cites the importance of technology, as follows:
Technological advancement and the proliferation of the smartphone have reshaped the commercial landscape,
providing consumers new ways to access the retail marketplace. On-demand companies1 are one
such innovation, and underpinning on-demand commerce is the gig economy, the collection of markets that
match service providers to consumers of on-demand services on a gig (or job) basis. Id. at 1.
The CRS report characterizes Uber (providing driver services) and Handy (providing home cleaners and
household services) as “flagship” on-demand companies.
Seth Harris and Alan Krueger, at Cornell University and Princeton University, respectively, have argued that
the traditional distinction between employees and independent contractors is not apt when it comes to the gig
economy. They assert, in a paper for the Hamilton Project entitled “A Proposal for Modernizing Labor Laws for
Twenty-First-Century Work: The Independent Worker,” available at www.hamiltonproject.org, that Congress,
and to the extent necessary, state legislators should pass legislation establishing a new worker classification for
what they call “independent workers.” The prototype for this class of “independent worker” would be the Uber
or Lyft driver. In arguing that these workers fit neither the traditional “employee” nor “independent contractor”
definitions, Harris and Krueger focus heavily on the primary role an “intermediary” plays in the relationship between
the worker and his or her ultimate customer; the ultimate customer in the case of an Uber or Lyft driver
would be the passenger. Harris and Krueger refer to Lyft and Uber, and other companies serving a similar role,
as “intermediaries” since they “are the intermediary between the independent worker and the ultimate customer.”
The CRS report asserts that the business models for the various types of intermediaries vary meaningfully,
Some companies allow providers to set prices or select the jobs that they take on (or both), whereas others
maintain control over price-setting and assignment decisions. Some operate in local markets (e.g., select
cities) while others serve a global client base. Although driver services (e.g., Lyft, Uber) and personal and
household services (e.g., TaskRabbit, Handy) are perhaps best known, the gig economy operates in many
sectors, including business services (e.g., Freelancer, Upwork), delivery services (e.g., Instacart, Postmates),
and medical care (e.g., Heal, Pager). (Footnotes omitted). Id. at 2.
INTERMEDIARIES’ POSITION: WORKERS ARE INDEPENDENT CONTRACTORS • As the CRS
report describes it:
With some exceptions, on-demand companies view providers as independent contractors—not employ-
1 On-demand companies such as Uber (which provides driver services) or TaskRabbit (whose providers perform household errands
and tasks) use Internet- or application-based technological platforms to match consumers to a select pool of potential service providers.
The on-demand aspect of these companies refers to their use of technology to respond to a customer’s immediate need (e.g., for a driver
or delivery) or specific need (e.g., for a particular rental or service provider skillset).
ees—using their platforms to obtain referrals and transact with clients.2 This designation is frequently
made explicit in the formal agreement that establishes the terms of the provider-company relationship.3
In addition, many on-demand companies give providers some (or absolute) ability to select or refuse jobs,
set their hours and level of participation, and control other aspects of their work. In some ways then, the
gig economy can be viewed as an expansion of traditional freelance work (i.e., self-employed workers who
generate income through a series of jobs and projects). Id.
In arguing that gig workers seem to be neither employees nor independent contractors, Harris and Krueger
say the following:
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 indefinite relationship with any employer,
and they do not relinquish control over their work hours or the opportunity for profit 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). Harris & Krueger, at 10.
The CRS report also seems to identify a “differentness” about gig work. Contrasting gig jobs from the type of
independent contractor work the report characterizes as “traditional freelance jobs,” the report says the following:
However, gig jobs may differ from traditional freelance jobs in a few ways. The established store-front
and brand built by the tech-platform company reduces entry costs for providers and may bring in groups
of workers with different demographic, skill, and career characteristics.4 Because gig workers do not need
to invest in establishing a company and marketing to a consumer base, operating costs may be lower and
allow workers’ participation to be more transitory in the gig market (i.e., they have greater flexibility
around the number of hours worked and scheduling). Id.
2 Some on-demand companies use a hybrid staffing model in which providers (gig workers) are classified as independent contractors,
but workers responsible for company operations are hired as bona fide employees. For example, home-cleaning company Handy
engages its “home service professionals” (the company term for individuals who provide cleaning services) as independent contractors,
but hires workers in management, engineers, finance, marketing and other positions as employees; . . . . Other on-demand companies
hire providers as employees (not independent contractors). These companies include the butler-service company Hello Alfred and the
office-services company Managed by Q. . . . . Finally, some providers who are engaged as independent contractors have disputed this
designation. . . .
3 See, e.g., Handy’s Service Professional Agreement, which states “Handy’s role is limited to offering the technology platform as a
referral tool for Service Requesters and Service Professionals and facilitating payments from Service Requesters to Service
4 Notably, many popular gig economy activities—such as delivery and personal-errand services – are considered to be low-skill, lowpay
In addition, a few commonly held characteristics of the provider-company relationship, when taken together,
set gig work apart from the traditional freelance worker model.
On-Demand Companies Collect a Portion of Job-Earnings. On-demand companies collect commissions
from providers for jobs solicited through the company platform. Commissions often take the form of
a flat percentage rate applied to job-earnings, but some companies employ more sophisticated models.
For example, in 2014, Lyft announced that it would return a portion of its 20% commission on rides as a
bonus to certain high-activity drivers. (footnote omitted)
On-Demand Companies Control the Brand. On-demand companies rely on the jobs brokered
through their platform to generate revenue, and therefore have a clear stake in attracting and retaining
clients. Consequently, and to varying degrees, these companies are selective about who can operate under
their brand. Some on-demand businesses condition provider participation on a background check and
credentials (e.g., some require past work experience, licenses or asset ownership),5 and some reserve the
right to terminate the relationship if the work delivered does not meet company-defined standards of
quality and professionalism.
On-Demand Companies Control the Provider-Client Relationship. Some on-demand companies
discourage or bar providers from accepting work outside the platform from clients who use the company’s
platform. The provider agreement for business-services company Upwork, for example, includes a noncircumvention
clause that prohibits providers from working with any client that identified the provider
through the Upwork site for 24 months.6 Likewise, “service professionals” working with the on-demand
home-services company Handy must agree that they will “not affirmatively solicit [clients] originally referred
through the Handy Platform to book jobs through any means other than the Handy Platform.”
(Footnote omitted.) This is a potentially important difference between gig work and traditional freelance
work, because it curtails the provider’s ability to build a client base or operate outside the platform.7
HOW MANY GIG WORKERS ARE THERE? • In reviewing the literature, it quickly becomes apparent that
it is unknown how many gig workers there are in the U.S. economy. Various sources use proxies for determining
that number, such as through the study of self-employment numbers or what the CRS report refers to as “nontraditional
work arrangements,” but the self-employed and those engaged in nontraditional work arrangements
are clearly very sloppy proxies for gig workers.
The Bureau of Labor Statistics (BLS) collected data during several years in the period 1995 through 2005 on
5 For example, home-services company Handy requires that providers have paid experience in home cleaning and be legally permitted
to work in their location. The delivery-services company Postmates require its operators to own a mode of propelled transport (e.g.,
car, bicycle, scooter) and have valid driver’s license. The City of San Francisco requires that Uber and Lyft drivers register with the
city for a business license. See S.F. Law: Business License crackdown, The San Francisco Chronicle, C1, April 22, 2016.
6 Upwork providers may opt out of the non-circumvention clause by paying a fee ($2,500 minimum).
7 Relatedly, it is unclear whether the provider’s professional reputation can be separated from the on-demand platform used to
transact with clients. That is, the client may view the provider’s performance as being the result of company policy or culture, or
otherwise perceive some value-added from engaging the provider through the on-demand platform. If
so, the client may be less willing to hire the same provider off-platform.
contingent workers (that is, those who did not expect their jobs to last) and workers in alternative employment
arrangements (such as independent contractors, on-call workers, temporary help agency workers, and workers
employed by contract firms). The BLS has announced that it will resume this Contingent Work Supplement as
part of the Current Population Survey in May 2017. Perhaps as part of that initiative data will be collected on gig
Harris and Krueger offer 600,000 as a rough estimate of the number of gig workers in 2015. This would be
about 0.4 percent of the U.S. employment population. Harris and Krueger reached this number based on very
imprecise data, focusing on Uber driver data from 2014 and “Google Trends” data recording the number of
times the studied on-demand company names were searched for in the Google search engine. According to the
CRS report, “McKinsey Global Institute estimates that ‘less than 1%’ of the U.S. working-age population are
contingent workers operating through ‘digital marketplaces for services.’”
LITIGATION • In a number of lawsuits, gig workers have asserted that they are employees for purposes of the
Fair Labor Standards Act (FLSA), including its minimum wage and overtime compensation requirements. In one
such case, a Lyft driver filed a putative class action complaint asserting that Lyft had a “uniform policy of willfully
misclassifying its drivers as independent contractors when, in fact, each such driver is and/or was an employee of
Lyft.” Frederic v. Lyft, Inc., No. 8:15-cv-01608 (M.D. Fla., filed July 8, 2015). This complaint alleged that Lyft violated
the FLSA as well as a Florida Deceptive and Unfair Trade Practices Act by, among other things, failing to
pay drivers the federal minimum wage, failing to pay overtime compensation for hours worked in excess of 40
hours in a week, and subjecting drivers to citations and criminal charges for operating without a proper permit.
In claiming that Lyft drivers are employees, the complaint asserted that Lyft controls the manner and means
by which drivers must accomplish their work, retains the right to hire and fire drivers in its sole discretion, sets the
rate of pay for drivers (and prohibits drivers from setting their own rate of pay for their services), limits the geographical
locations in which drivers may work and restricts the distance they may transport customers, permits
drivers to work only certain hours each day, and prohibits drivers from hiring other employees to assist them.
The lawsuit was filed in July 2015, but in December 2015 the court was notified that the dispute had been resolved
and the case was dismissed with prejudice.
At least four state wage-related claims have been raised in California, and at least two of these make FLSA
claims as well. Two of the California actions are O’Connor v. Uber Technologies, Inc., No. cv-13-03826-EMC (N.D.
Cal., filed August 8, 2015) and Cotter v. Lyft, Inc., No. 3:13-cv-04065-VC (N.D. Cal., filed July 15, 2015). O’Connor
was filed as a putative class action on behalf of those who have worked as Uber drivers anywhere other than Massachusetts,
initially asserting, among other things, that Uber violated California law by failing to remit to drivers
the entire gratuity paid them by customers, or by causing drivers not to receive tips they would otherwise receive
based on Uber’s communications to customers that the gratuity is included in the price of the car service and
there is, therefore, no need to tip drivers. The initial complaint also asserted that Uber drivers have been misclassified
as independent contractors and that Uber should therefore be required under California labor law to pay
drivers’ business expenses (such as expenses for the drivers’ vehicles, gas, and maintenance). An amended and
substantially expanded complaint asserts that Uber drivers are employees, not independent contractors, because
they have little discretion in their driving work for Uber, and that under either California law or the FLSA (or
both) they are entitled to overtime pay, the opportunity to take rest and meal breaks, expense reimbursement,
minimum wage pay, and accurate and on-time payment of all wages due. (The claims other than those relating to
minimum wage and overtime pay appear to be grounded strictly in California law.)
The federal district court hearing O’Connor certified a class of certain Uber drivers. O’Connor v. Uber Technologies,
Inc., 2015 U.S. Dist. Lexis 116482 (N.D. Cal. Sept. 1, 2015). See also, 311 F.R.D. 547 (N.D. Cal. Dec. 9,
2015), certifying in the same action a subclass of drivers, and holding that certain arbitration agreements and contracts from 2014 and 2015 are unenforceable. In a related case, Del Rio v. Uber Technologies, Inc., No. 3:13-cv-
03826-EMC (N.D. Cal., filed September 21, 2015), claims similar to those raised in O’Connor are made in a putative
class action complaint. As in O’Connor, the court denied Uber’s motion to compel arbitration. Del Rio v. Uber
Technologies, Inc., 2015 U.S. Dist. Lexis 169063 (N.D. Cal. Dec. 16, 2015).
In Cotter v. Lyft, Inc., plaintiffs filed a putative class action challenging “Lyft’s policy of misclassifying its Drivers
as ‘independent contractors’ unprotected by the California Labor Code.” The complaint asserts that the drivers
are in fact employees of Lyft, and objects to Lyft’s policy of taking 20 percent of gratuity payments given by riders
to drivers. The complaint asserts that this is an illegal practice because the California Labor Code prohibits an
employer from taking any amount of a gratuity given to an employee. The complaint also asserts that the failure
to pay drivers minimum wage and provide drivers with wage statements accurately reflecting their hours worked
or rates of pay, as well as the practice of failing to reimburse drivers for mileage costs, are all in violation of the
California Labor Code. The Cotter plaintiffs’ motion for preliminary approval of a proposed class action settlement
was denied. See 2016 WL 1394236 (Apr. 7, 2016). In another California proceeding, Berwick v.
Uber Technologies, Inc., California’s Labor Commissioner concluded that an Uber driver was an employee of Uber
for purposes of the California Labor Code. This meant Uber was required to indemnify the employee for “all
that the employee necessarily expends in the discharge of the employee’s duties.” The Labor Commissioner ordered
Uber to reimburse the driver $3,878, of which $3,622 was for mileage at the 2014 IRS mileage rate of 56
cents per mile and $256 was for toll charges. Uber was also assessed interest in the amount of $274. No. 11-
46739-EK (Cal. Lab. Comm. June 3, 2015). The Labor Commissioner in Berwick said there is a presumption of
employment (as opposed to an independent contractor relationship) under California labor law where one retains
the services of another. In describing why Uber failed to rebut this presumption, the court said the following:
By obtaining the clients in need of the service and providing the workers to conduct it, Defendants retained
all necessary control over the operation as a whole. The party seeking to avoid liability has the
burden of proving that persons whose services he has retained are independent contractors rather than
employees. In other words, there is a presumption of employment. (Labor Code § 3357; Borello, [S.G. Borello
& Sons, Inc. v. Dept. of Industrial Relations, 48 Cal. 3d 341 (Cal. 1989)], at pp. 349, 354.)
Ownership of the vehicle used to perform the work may be a much less important factor in industries other
than transportation. Even under the traditional, pre-Borello common law standard, a person making pizza
deliveries was held to be an employee of the pizzeria, notwithstanding the fact that the delivery person
was required to provide his own car and pay for gasoline and insurance. (Toyota Motor Sales v. Superior Court
(1990) 220 Cal.App.3d 864, 876.)
“The modern tendency is to find employment when the work being done is an integral part of the regular
business of the employer, and when the worker, relative to the employer, does not furnish an independent
business or professional service.” (Borello supra, at p. 357.) Plaintiff’s work was integral to Defendants’ business.
Defendants are in business to provide transportation services to passengers. Plaintiff did the actual
transporting of those passengers. Without drivers such as Plaintiff, Defendants’ business would not exist.
Defendants hold themselves out as nothing more than a neutral technological platform, designed simply
to enable drivers and passengers to transact the business of transportation. The reality, however, is that
Defendants are involved in every aspect of their operation. Defendants vet prospective drivers, who must
provide to Defendants their personal banking and residence information, as well as their Social Security
Number. Drivers cannot use Defendants’ application unless they pass Defendants’ background and DMV checks.
Defendants control the tools the drivers use; for example, drivers must register their cars with Defendants,
and none of their cars can be more than ten years old Defendants refer to “industry standards” with respect
to drivers’ cars, however, it is unclear to what industry, other than the “taxi” industry, Defendants
are referring. Defendants monitor the Transportation Drivers’ approval ratings and terminate their access
to the application if the rating falls below a specific level (4.6 stars).
While Defendants permit their drivers to hire people, no one other than Defendants approved and registered
drivers are allowed to use Defendants’ intellectual property. Drivers do not pay Defendants to use
their intellectual property.
The passengers pay Defendants a set price for the trip, and Defendants, in turn, pay their drivers a nonnegotiable
service fee. If a passenger cancels a trip request after the driver has accepted it, and the driver
has appeared at the pick-up location, the driver is not guaranteed a cancellation fee. Defendants alone
have the discretion to negotiate this fee with the passenger. Defendants discourage drivers from accepting
tips because it would be counterproductive to Defendants’ advertising and marketing strategy.
Plaintiff’s car and her labor were her only assets. Plaintiff’s work did not entail any “managerial” skills
that could affect profit or loss. Aside from her car, Plaintiff had no investment in the business. Defendants
provided the iPhone application, which was essential to the work. But for Defendants’ intellectual property,
Plaintiff would not have been able to perform the work.
Uber reportedly has filed an appeal of the Labor Commissioner’s decision with the Superior Court of California
for the County of San Francisco. Berwick v. Uber Technologies, Inc., No. CGC-15-546378 (Sup. Ct. San Francisco
County, filed June 16, 2015). Uber attempted to force the driver to arbitrate her dispute under the terms of
her contract with Uber. But on September 21, 2015, the San Francisco Superior Court Judge hearing the matter
concluded that the contract was contradictory and “substantively unconscionable,” and therefore refused to
compel the driver to arbitrate the dispute.
In addition to the California actions described above, Florida’s Department of Economic Opportunity concluded
that a couple of Uber drivers were independent contractors, not employees, for purposes of Florida unemployment
insurance. The drivers were therefore not entitled to apply for unemployment insurance in Florida.
The Florida agency noted both the California Labor Commissioner’s decision in Berwick and an advisory opinion
from the Commissioner of the Oregon Bureau and Labor and Industries that “[u]nder Oregon law . . . Uber
drivers are employed.” Oregon Bureau of Labor and Industries Advisory Opinion regarding The Employment of
Uber Drivers (October 14, 2015). The Florida Department of Economic Opportunity nonetheless concluded that
the Uber drivers at issue were not employees for the purposes of Florida unemployment insurance.
DOL WAGE AND HOUR DIVISION INTERPRETATIONS • The Department of Labor’s Wage and
Hour Division has over the past many months issued a couple of Administrator’s Interpretations (AIs) that some
interpret as reflecting an attempt by the DOL to expand the notion of who is an employee (as opposed to an independent
contractor) for purposes of the Fair Labor Standards Act (which, recall, imposes the federal minimum
wage and overtime rules). The FLSA standards for determining who is an employee apply also, incidentally, in
determining who is an employee for purposes of the Family and Medical Leave Act.
The first of these Administrator’s Interpretations, AI 2015-1, was issued on July 15, 2015. It says “most workers
are employees” under the FLSA’s broad definitions. In determining whether an individual is an employee, AI 2015-1 applies a multi-factor “economic realities” test. The relevant factors under this test include:
The extent to which the work performed is an integral part of the employer’s business;
The worker’s opportunity for profit or loss depending on his or her managerial skill;
The extent of the relative investments of the employer and the worker;
Whether the work performed requires special skills and initiative;
The permanency of the relationship; and
The degree of control exercised or retained by the employer.
For each of these factors, the AI offers an example to help illustrate how the factor is applied, as follows:
Is the Work an Integral Part of the Employer’s Business?
Example: For a construction company that frames residential homes, carpenters are integral to the employer’s
business because the company is in business to frame homes, and carpentry is an integral part of
providing that service.
In contrast, the same construction company may contract with a software developer to create software
that, among other things, assists the company in tracking its bids, scheduling projects and crews, and
tracking material orders. The software developer is performing work that is not integral to the construction
company’s business, which is indicative of an independent contractor.
Does the Worker’s Managerial Skill Affect the Worker’s Opportunity for Profit or Loss?
Example: A worker provides cleaning services for corporate clients. The worker performs assignments
only as determined by a cleaning company; he does not independently schedule assignments, solicit additional
work from other clients, advertise his services, or endeavor to reduce costs. The worker regularly
agrees to work additional hours at any time in order to earn more. In this scenario, the worker does not
exercise managerial skill that affects his profit or loss. Rather, his earnings may fluctuate based on the
work available and his willingness to work more. This lack of managerial skill is indicative of an employment
relationship between the worker and the cleaning company.
In contrast, a worker provides cleaning services for corporate clients, produces advertising, negotiates contracts,
decides which jobs to perform and when to perform them, decides to hire helpers to assist with the
work, and recruits new clients. This worker exercises managerial skill that affects his opportunity for profit
and loss, which is indicative of an independent contractor.
Is the Relationship between the Worker and the Employer Permanent or Indefinite?
Example: An editor has worked for an established publishing house for several years. Her edits are completed
in accordance with the publishing house’s specifications, using its software. She only edits books
provided by the publishing house. This scenario indicates a permanence to the relationship between the editor
and the publishing house that is indicative of an employment relationship.
Another editor has worked intermittently with fifteen different publishing houses over the past several
years. She markets her services to numerous publishing houses. She negotiates rates for each editing job
and turns down work for any reason, including because she is too busy with other editing jobs. This lack of
permanence with one publishing house is indicative of an independent contractor relationship.
What is the Nature and Degree of the Employer’s Control?
Example: A registered nurse who provides skilled nursing care in nursing homes is listed with Beta
Nurse Registry in order to be matched with clients. The registry interviewed the nurse prior to her joining
the registry, and also required the nurse to undergo a multi-day training presented by Beta. Beta sends the
nurse a listing each week with potential clients and requires the nurse to fill out a form with Beta prior to
contacting any clients. Beta also requires that the nurse adhere to a certain wage range and the nurse
cannot provide care during any weekend hours. The nurse must inform Beta if she is hired by a client and
must contact Beta if she will miss scheduled work with any client. In this scenario, the degree of control exercised
by the registry is indicative of an employment relationship.
Another registered nurse who provides skilled nursing care in nursing homes is listed with Jones Nurse
Registry in order to be matched with clients. The registry sends the nurse a listing each week with potential
clients. The nurse is free to call as many or as few potential clients as she wishes and to work for as
many or as few as she wishes; the nurse also negotiates her own wage rate and schedule with the client. In
this scenario, the degree of control exercised by the registry is not indicative of an employment relationship.
More recently, in AI 2016-1, the DOL’s Wage and Hour Division established standards for determining
whether there is “joint employment” for purposes of the FLSA. When two or more employers jointly employ an
employee, the employee’s hours worked for all of the joint employers during the work week are aggregated and
considered as one employment, including for purposes of calculating whether overtime pay is due. Some read the
recent Administrator’s Interpretation as signaling a new aggressiveness in finding the existence of joint employment.
The new AI on joint employment (2016-1) describes both what it calls “horizontal joint employment”—
where an employee has an important relationship with two or more employers that are associated or related with
respect to the employee—and “vertical joint employment.” With respect to “vertical joint employment,” the AI
has the following to say:
Vertical joint employment exists where the employee has an employment relationship with one employer (typically
a staffing agency, subcontractor, labor provider, or other intermediary employer) and the economic realities
show that he or she is economically dependent on, and thus employed by, another entity involved in the work.
This other employer, who typically contracts with the intermediary employer to receive the benefit of the employee’s
labor, would be the potential joint employer. Where there is potential vertical joint employment, the
analysis focuses on the economic realities of the working relationship between the employee and the potential
The AI focuses particular attention on the construction, agricultural, janitorial, warehouse and logistics, staffing,
and hospitality industries.
NLRB’S EMPLOYEE ANALYSIS • The National Labor Relations Board has also, in the view of some, expanded
its notion of who is an employee (as opposed to an independent contractor) within the meaning of the
National Labor Relations Act. This arguable expansion of who constitutes an employee began with the NLRB’s
decision in FedEx Home Delivery, 361 NLRB No. 55 (2014). There the NLRB “restated and refined” its independent
contractor test under the NLRA.
FedEx was followed by the NLRB’s application of its refined test in Porter Drywall, 362 NLRB No. 6 (2015) and
Sisters’ Camelot, 363 NLRB No. 13 (2015). Notably, even prior to FedEx’s expansion of the NLRB’s view of who
constitutes an employee, the Eleventh Circuit reversed an NLRB conclusion that stage hands were employees
rather than independent contractors. Crew One Productions v. NLRB, 812 F.3d 945 (11th Cir. 2016), denying enforcement
to 362 NLRB No. 8 (2015) and 10-RC-124620 (August 21, 2014, unpublished order in representation
What Does This Mean for Employee Benefit Plans?
It is unclear to what degree, if any, the DOL Wage and Hour Division’s more aggressive position (in its AIs)
on who constitutes an employee, and court decisions concerning who is an employee for purposes of the FLSA,
NLRA, or state labor codes, will affect determinations of who is an employee for purposes of retirement plans,
health insurance, or other employee benefit programs. On the one hand, when an individual asserts that he or
she is entitled to coverage under the terms of an ERISA-covered plan, the “employee question” is merely what the
plan means when it uses the term “employee.” Often employee benefit plans say “employees” are eligible for coverage,
but do not explain what the term “employee” means. Some plans, particularly some qualified retirement
plans, are more sophisticated, saying they cover only those individuals who are characterized as employees on the
company’s payroll system. In some sense, this effectively defines the term “employee” to mean those individuals
the company takes the position are employees. (This is an oversimplification, since a plan typically does not incorporate
the “characterized as an employee on the employer’s payroll” rule into the Plan’s definition of “employee,”
but instead simply uses the standard in describing which persons are eligible for coverage under the
plan). Many plans do not include even this refinement, but this type of language became more common following
Microsoft’s agreement in late 2000 to settle class action litigation concerning the eligibility of certain workers for
company-provided employee benefits. For rulings in the Microsoft litigation preceding Microsoft’s agreement to
settle, see Vizcaino v. Microsoft, 97 F.3d 1187 (9th Cir. 1996) (“Vizcaino I”), cert denied, 522 U.S. 1088 (1998) (“Vizcaino
II”); Vizcaino v. United States Dist. Ct. (In re Vizcaino), 173 F.3d 713 (9th Cir. 1999), amended by 184 F.3d 1070
(9th Cir. 1999), cert denied, 528 U.S. 1105 (2000).
If a plan purports to cover “employees,” but does not define the term, one reasonable argument might be that
the term has the meaning the Supreme Court attributed to it under ERISA in Nationwide Mut. Ins. Co. v. Darden,
503 U.S. 318, 323-24 (1992). There, the Supreme Court adopted a common law test for determining employee
status under ERISA, and in doing so referred favorably to the Restatement (Second) of Agency § 220 (1958),
which sets forth a 10 factor test. The Supreme Court did not, however, anoint the 10 factor Restatement (Second)
test as the sole means for determining status as an employee (it also cited its 12 factor test set forth in Community
for Creative Non-Violence v. Reid, 490 U.S. 730, 751-52 (1989) and the IRS’ 20 factor test set forth in Revenue
Ruling 87-41). Of note, since Darden was decided, the American Law Institute promulgated, adopted and published
a Restatement (Third) of Agency (2006) that does not include a section setting forth factors differentiating
an employee from an independent contractor, instead reserving that topic to the Restatement of Employment
Complicating matters, to the extent the Darden formulation of who is an employee under ERISA differs from
the IRS’ 20 factor test set forth in Revenue Ruling 87-41, 1987-1 C.B. 296 (if it differs at all), for qualified retirement
plan purposes it is possible the proper reading of a plan provision referring to an “employee,” and the
proper reading of the term “employee” as used in the Tax Code’s qualified retirement plan rules, is the Revenue
Darden is of some age, being over 25 years old. It will therefore be interesting to observe whether courts, in
interpreting employee benefit plan provisions or in determining who is an employee within the meaning of
ERISA or the employee benefit provisions of the Tax Code, will be influenced by more recent court or DOL determinations
of who is an employee for purposes of the FLSA, NLRA, or state labor laws. Perhaps notable in this
regard is the modest judicial trend toward borrowing “successor employer” liability notions from non-ERISA labor
law when assessing whether purchasers of business assets take on (usually inadvertently) pension-related liabilities
of the seller. See, e.g., Upholsterers’ Int’l Union Pension Fund v. Artistic Furniture, 920 F.2d 1323 (7th Cir. 1990); Chicago
Truck Drivers Union Pension Fund v. Tasemkin, Inc., 59 F.3d 48 (7th Cir. 1995); Einhorn v. M.L. Ruberton Constr. Co.,
632 F.3d 89 (3d. Cir. 2011); Tsareff v. Manweb Servs., Inc., 794 F.3d 841 (7th Cir. 2015); Resilient Floor Covering Pension
Fund v. Michael’s Floor Covering, Inc., 801 F.3d 1079 (9th Cir. 2015).
What Are the Benefit Options for Gig Workers if They Are Not Employees?
If a gig worker is an independent contractor, rather than an employee, his or her options for accessing retirement,
health insurance, and the other benefits of the type employees sometimes enjoy would seem to be the set of
options available to self-employed persons. With respect to retirement benefits, self-employed persons may, of
course, maintain a qualified retirement plan, including through the adoption of a prototype or volume submitter
plan document. This might, for example, be a 401(k) plan.
A self-employed person may prefer to adopt a Simplified Employee Pension (SEP), under which he or she can
contribute as much as 25 percent of his or her net earnings from self-employment (not including contributions to
the plan), up to $53,000 (for 2016). A gig worker could do so by completing a Form 5305-SEP or an IRSapproved
prototype SEP plan offered by a financial institution or plan administration company. The gig worker
would also need to open a SEP-IRA through a bank or other financial institution.
A self-employed person might instead adopt a SIMPLE IRA plan, in which he or she could put all of his or
her net earnings from self-employment, up to $12,500 (for 2016), plus an additional $3,000 if the gig worker is
age 50 or older, as well as either a two percent fixed contribution or a three percent matching contribution. A
self-employed gig worker could do this by completing an IRS Form 5305-SIMPLE (for use with a designated financial
institution), or a Form 5304-SIMPLE (not for use with a designated financial institution), or could adopt a
prototype SIMPLE IRA plan offered by a financial institution or plan administration company. The gig worker
would also need to open a SIMPLE IRA through a bank or other financial institution.
And, of course, gig workers could just contribute to an IRA, Roth or otherwise, or establish a myRA.
With respect to health coverage, a gig worker might wish to purchase coverage through the ACA marketplace.
And a gig worker having high deductible health plan coverage may want to contribute to a health savings
account (HSA). With respect to other welfare benefits, such as long-term disability coverage, a gig worker would
presumably need to resort to the individual insurance market, except to the extent he or she has access to coverage
through one or more associations to which the gig worker belongs.
Could an “Intermediary” Facilitate Benefit Coverage?
Harris and Krueger, in arguing for a new “independent worker” classification, advocate for legislation that
would give independent workers some of what they lose by not being employees. Harris and Krueger argue, for
example, for requiring Uber and other “intermediaries” to contribute half of FICA taxes (as an employer would),
and to withhold income and normal employment taxes. They also argue for modifying the labor laws to allow
independent work organizations to bargain over compensation and the terms and conditions of workers’ contractual
relationships with intermediaries, to expand employment-based anti-discrimination and other civil rights
protections to benefit independent workers, and to allow intermediaries to opt to provide the equivalent of workers’
compensation insurance policies and establish a private unemployment insurance system for independent
Harris and Krueger further propose that because there is no adverse consequence under the employer health
mandate of IRC Section 4980H for failing to offer coverage to non-employees, intermediaries should be required
to pay five percent of an independent workers’ earnings (net of commissions) to support health insurance subsi
dies in the Exchange (aka the Marketplace). In part because of the difficulty in measuring working hours, particularly
where, say, an Uber or Lyft driver has waiting time in which he or she is free to anything he or she wants,
and in part because it may be difficult to allocate hours to particular intermediaries where a driver works for
more than one, Harris and Krueger do not propose that wage and hour protections apply to independent workers.
Others besides Harris and Krueger have suggested changes to the way in which health insurance, workers’
compensation, and other common perquisites of full-time employment are structured, to better fit an economy
having a gig worker population. Arun Sundararajan, a professor at New York University, argued in a Financial
Times op-ed piece in June 2015 for decoupling health coverage, workers’ compensation, and perhaps other benefits
from full-time employment status. Nick Hanauer and David Rolf, in an article entitled Shared Security, Shared
Growth, appearing the Summer 2015 issue of Democracy Journal, argued for establishing a “Shared Security System,”
under which American workers would accrue basic employment benefits independent of full-time employment
Some, including Harris and Krueger, are optimistic about the salutary effects of pooling independent workers
for the purpose of purchasing insurance benefits. They argue that if intermediaries were to pool their independent
workers in arranging for the purchase of insurance, this could help minimize insurers’ adverse selection concerns.
Further, “prices almost certainly would be significantly lower for independent workers and intermediaries
than for individual purchases, and both services and products could be greater in quality, quantity, or both.”
Harris and Krueger have in mind, for example, the pooling of independent workers to purchase health insurance,
disability insurance, and retirement products (as well as automobile insurance, banking and savings products,
liability insurance, tax preparation assistance, and other financial services).
The author’s experience with the less than warm reception insurers offer so-called “association plans” raises
at least some question as to whether Harris and Krueger’s optimism concerning pooling is fully warranted. In
general, insurance companies are skittish about offering health, disability, or life insurance coverage to members
of associations, at least compared to their eagerness to insure employees of employers. The view seems to be that
members of an association constitute a less cohesive and homogeneous group, and therefore a less predictable
group in terms of risk, than a population of employees of an employer in some well-established and well-defined
Uber and other intermediaries could, today, make an argument for the legality of offering to their gig workers
health and welfare insurance products through an insured arrangement. An arrangement covering non-employee
gig workers would either not be subject to ERISA (because the arrangement would cover no employees), or, if
coupled with coverage for employees of the intermediary itself, would be a multiple employer welfare arrangement
(a MEWA). A possible constraint on offering an insured MEWA would come not under ERISA but from
whether a carrier could, under state insurance law, offer coverage to the group absent the formation of an “association”
of gig workers to which the group insurance policy might be issued. Even if, as Harris and Krueger have
proposed, an intermediary were protected from any inference that its independent workers were employees as a
consequence of offering such coverage, the creation of such an arrangement would involve some headache, including
some concerns novel enough that the effort might not be worth the candle from the intermediary’s perspective.
And although the Tax Code Section 414 regulations make practical the maintenance of a multiple employer
qualified retirement plan, the Department of Labor launched an unfortunate attack on multiple employer plans
in Advisory Opinion 2012-04A. There, the DOL set needlessly difficult standards for an arrangement to be considered
a single multiple employer arrangement, as opposed to a constellation of single employer plans. This
DOL antagonism toward multiple employer plans arguably, though, should not present an impediment to a plan
covering only a set of self-employed individuals. That is because ERISA does not apply to a plan covering no
employees. 29 C.F.R. § 2510.3-3(b). Even if coverage were combined with that for employees of an intermediary,
treating the arrangement as a set of plans rather than as a single multiple employer plan might not much matter
since the “plans” for the self-employed individuals should not be subject to ERISA.
The existing retirement plan options available to self-employed individuals (noted earlier) are not terrible.
There may, therefore, be little motivation for an intermediary to establish for its gig workers a multiple employer
retirement plan. The greater value in “pooling” the purchasing power of gig workers is likely to be with respect to
health and welfare products. Even if the pooling of gig workers’ purchasing power were to offer to those workers
real economic benefits, the favorable tax treatment accorded employees for many employee benefit programs
would remain unavailable (assuming gig workers are not employees).
CONCLUSION • The author will not be holding his breath waiting for Congress to modify the legal standard
for determining what it means to be an employee, such as by creating for “gig workers” some third category of
worker status (that is, other than “employee” or “independent contractor”). The real battle, it seems, will be in
determining whether, and if so when, gig workers are employees for purposes of ERISA and the federal Tax
Though the author does not expect imminent statutory changes to the standard for determining who is an
employee for purposes of ERISA or the federal Tax Code, there has been some activity relating to the treatment
of gig workers under state and local law. For example, the Seattle, Washington City Council reportedly voted
unanimously in December 2015 to give drivers for companies like Uber and Lyft the right to collectively bargain
over wages and working conditions. Reportedly, the U.S. Chamber of Commerce filed suit against the City of
Seattle on March 3, 2016, challenging the ordinance. California’s AB 1727, introduced January 28, 2016, stated:
“This bill would establish for eligible groups of independent contractors the right to organize and negotiate with
hosting platforms, and would declare the activity of such a group to be a labor dispute for purposes of injunctive
relief. The bill would require a hosting platform to meet and negotiate with a group on specified subjects.” AB
1727 was withdrawn by its sponsor, Lorena Gonzales, on April 21, 2016. And then on April 22, The San Francisco
Chronicle reported three gig economy stories. The first, appearing on the first page, reported that Uber settled
its dispute with drivers over their status—the drivers were to remain independent contractors in California
and Massachusetts, but with the right to appeal when they are “deactivated” (barred from using the app). See
Drivers’ lawsuits settled by Uber, The San Francisco Chronicle, A1, April 22, 2016,
second, Uber, Lyft drivers face more hurdles, encompassed two stories—PUC: Tweaking some key rules (reporting that the
state Public Utilities Commission voted to accept rules legalizing fare splitting, display of company logos, and
mandatory inspections of drivers’ cars by licensed repair shops); and S.F. Law: Business license crackdown (reporting
that Uber and Lyft drivers register with the City of San Francisco for a business license). The San Francisco
Chronicle, April 22, 2016, C1. Available at http://www.pressreader.com/usa/san-francisco-chronicle-lateedition/
20160422/282617441929603. The third, Pennsylvania fines Uber $11 million, reported that Pennsylvania
regulators voted to fine the ride-hailing company for operating for six months “without the required approval.”
The story also reported that Uber was fined $7 million in January by California regulators for “failing to provide
sufficient information.” The San Francisco Chronicle, April 22, 2016, C3. Available at
http://www.sfchronicle.com/news/us/article/Pennsylvania-fines-Uber-11M-for-operating-7287441.php. By the
time this article is printed there will undoubtedly be more developments about the employment status of Uber
and Lyft drivers—as well as all other gig workers.
As a parting thought, there may be some modest trend toward intermediaries in the gig economy choosing to
treat gig workers as employees, including for employee benefit plan purposes. Honor, a home health care company
that has been described as based on a gig-economy model, announced on January 19, 2016, that its workers
“will become employees, eligible for benefits—and stock options.” Other gig-economy companies that have purportedly
made the switch from classifying workers as independent contractors to employees are Eden, an office
management startup, Instacart, a grocery-delivery provider, Luxe, an on-demand valet service, Shyp, a courier
and shipping startup, and Munchery and Sprig, meal-delivery services. See