Executive summary
Numerous state-level studies show that between 10 and 20 percent of employers misclassify at least one worker as an independent contractor. Independent contractor (IC) misclassification occurs when a worker who should be considered a direct employee of a business—and receive a W-2 form to file with tax returns—is treated as a self-employed, “independent” contractor, and receives a 1099-MISC (miscellaneous income) form instead. The overall numbers have likely increased in recent years as workers in such traditional industries as construction, trucking, and stagecraft have been joined by a growing cadre of “on-demand workers,” who often get their assignments via the Internet (Weber and Silverman 2015). Independent contractors working in the on-demand economy include technical workers, house cleaners, drivers, and scores of others—some of whom are misclassified employees. All independent contractors, in old or new industries, are ineligible for benefits such as the minimum wage, overtime pay, unemployment insurance, and workers’ compensation.
Misclassified workers can now be found in almost every sector of the economy, working for small companies to publicly traded multinational corporations. For example, Atlanta stagehands for concerts produced by Live Nation, a company listed on the New York Stock Exchange that has held shows for such artists as Maroon 5 and Billy Joel, have been misclassified as ICs by a staffing provider (Vail 2015; DePillis 2015). An estimated one-third of construction workers in Southern states such as North Carolina and Texas have been misclassified (Ordonez and Locke 2014a). And roughly 20,000 employees of CrowdFlower Inc., a San Francisco–based startup that breaks down digital jobs such as data entry, are misclassified, alleges a case now moving through the courts (Weber and Silverman 2015).
The costs to tax and social insurance systems and to workers add up. Businesses that misclassify fail to pay mandatory payroll taxes—Social Security and Medicare (FICA) and unemployment insurance (UI)—and workers’ compensation insurance. The independent contractor is made responsible for the full FICA tax (rather than half). The loss of billions of dollars in tax revenue creates a significant financial burden for local, state, and the federal governments, not only due to lost revenue but also because of the added cost of providing social services to uninsured workers. Businesses also are harmed by the practice of worker misclassification. Law-abiding firms that pay their taxes and properly classify their workers as employees face a competitive disadvantage and may feel pressured to cut corners with their workers’ employment status if they wish to remain competitive.
As a rule, companies found to be misclassifying workers and violating tax laws by the Internal Revenue Service usually do not get penalized by federal authorities due to legal constraints on the IRS. Not only are they not fined, they are often allowed to continue misclassifying workers under a tax loophole known as “Safe Harbor,” according to an investigative report by McClatchy (Ordonez and Locke 2014b). When companies are found in violation of state unemployment insurance laws, they are fined and assessed retroactive taxes. The issue, however, is that states have limited audit capacity. Improvements in coordination of state and federal efforts to uncover misclassification have begun in recent years to boost enforcement.
Beyond the dollars in play, there are millions of Americans wrongly left uninsured, without benefits and without job security.
This paper presents state and federal evidence on the magnitude and severity of IC misclassification and recent trends. It looks at related tax issues and other public policy considerations. In addition to more systematic research and existing and planned improvements to the enforcement of labor standards, mixed policy approaches are needed. The scale of the problem will require solutions that go beyond individual worker complaints or court cases, and that include a combination of prevention, information, inspection, and collective worker rights.
Following are some of the major findings of the report:
Misclassification is most common in industries where it is most profitable (such as construction, where workers’ compensation insurance premiums are high), and in industries with scattered worksites where work is performed in isolation. Housecleaning, in-home care, and trucking are industries in which misclassification is particularly common. New “sharing economy” businesses create cause for concern about possible misclassification because it is unclear how “autonomous” these workers really are.
Employers who misclassify avoid paying payroll taxes and workers’ compensation insurance, are not responsible for providing health insurance, and are able to bypass requirements of the Fair Labor Standards Act, as well as the 1986 Immigration Reform and Control Act.
Misclassified workers are ineligible for unemployment insurance, workers’ compensation, minimum wage, and overtime, and are forced to pay the full FICA tax and purchase their own health insurance.
Misclassification undermines worker bargaining power and leaves workers more vulnerable to wage theft.
When workers are misclassified, federal and state governments lose out on revenue from income taxes. Federal and state unemployment insurance, worker compensation, and disability insurance systems are adversely affected.
Employers who play by the rules are disadvantaged by higher labor and administration costs relative to employers who misclassify.
There are a number of policies to address misclassification, including stepped-up enforcement, higher fines, information campaigns, and stronger collective bargaining.
Introduction
Independent contractor (IC) misclassification occurs when a worker who should have received a W-2 tax form to file with tax returns receives a 1099-MISC (miscellaneous income) form instead. Thus, a worker who should be considered a direct employee of a business is treated as self-employed, an “independent” contractor. This practice can constitute fraud; it violates tax and employment laws. When an employee is treated as a self-employed worker, the business employing him or her fails to pay mandatory payroll taxes (Social Security and Medicare [FICA] and unemployment insurance [UI]) and workers’ compensation insurance premiums. The worker is responsible for the full FICA tax (rather than half) and is ineligible for company-based benefits. Unless the worker appeals to administrative agencies and wins, access to unemployment insurance and to coverage by wage and hours laws (e.g., minimum wage and overtime) are denied. While misclassified, a worker has little or no access to labor rights. Independent contractor misclassification is emblematic of changes in employment relationships and of the threats to worker bargaining power. It is not a novel practice, but it seems to have become more common as the frequency of labor standards violations has increased over the past two decades. The way businesses are organized, with lengthening subcontracting chains in particular, has facilitated evasion and violation of labor standards (Weil 2014; Bernhardt et al. 2009). Concurrently, some sectors that provide propitious settings for labor standards violation (e.g., scattered worksites, project-based hiring, subcontracting) have continued to find ways to shed responsibility for their workers. Some sectors, such as construction, low-wage in-person services, and light manufacturing, stand out as settings in which labor standards violations are more common. And settings in which these violations are common are often ones where IC misclassification also occurs.
Several parties are directly affected by IC misclassification. The misclassified worker stands to lose significant social protection and compensation (e.g., loss of overtime pay or nonpayment of wages) as well as representation rights under the National Labor Relations Act. Directly employed workers, even when unionized, find it more difficult to seek and obtain their employer’s compliance with labor standards in settings where misclassification is common, and misclassified workers can be made to work with substandard terms of employment. The unemployment insurance system, federal and state tax collection, and the workers’ compensation system are all adversely affected by IC misclassification. Employers who play by the rules and comply with all employment laws lose when they are underbid by others who have lowered their labor costs by shedding workers and avoiding mandated payroll taxes and compliance with wage and hour laws.
Misclassification results either from ignorance or criminal misconduct; it is unlikely to be acknowledged by employers completing surveys or understood by misclassified workers. Thus no existing national survey reveals the number of businesses that engage in it or the number of workers it affects. Nevertheless, government agencies conduct regular audits, and available evidence indicates that the incidence of IC misclassification is significant in some industrial sectors and seems to be on the rise. Results of state-level studies presented later in this paper indicate that anywhere from 10 to 20 percent of employers misclassify at least one worker. Importantly, IC misclassification is often a business model. For example, the percentage of workers misclassified by Massachusetts employers that engaged in this practice ranged from 25 to 39 percent between 2001 and 2003 (Carré and Wilson 2005).
This paper reviews what employers who misclassify achieve with this practice. It then turns to the salient policy concerns generated by IC misclassification and explores who has a stake in it and why. The paper then presents state and federal evidence to date on the magnitude and severity of IC misclassification and trends in its magnitude and severity in recent years. The next section explores recent patterns of misclassification as well as ambiguous employment arrangements that are challenging existing understandings of what is meant by the term “self-employment.” The final section reviews options for future research and areas where policy development is needed.
Employers who misclassify
Businesses misclassify workers as independent contractors to avoid several employment-related obligations and thereby save on labor and administration costs and gain advantage over competitors. IC misclassification enables these businesses to avoid mandatory payroll taxes: the employers’ half of the Social Security pension contribution and the Medicare tax, which totaled 15.3 percent of gross wages for 2013 and 2014 (Internal Revenue Service 2014).1 In addition, these employers avoid paying both state and federal unemployment insurance taxes because independent contractors are not considered employees and thus not covered by the UI system (unless they appeal and win coverage as employees).
Businesses that misclassify also avoid legally required workers’ compensation insurance and state mandates concerning it, as well as disability insurance.2 Workers’ compensation programs vary across states, so the cost that employers seek to avoid varies as well. Workers’ compensation premiums for individual employers and specific jobs are affected by injury severity and frequency of past claims, and premiums vary across industries and occupations. Partly as a result, some industries have a greater stake in avoiding premiums (and related workplace and disability-related disputes) and have a greater incidence of misclassification than others. A Massachusetts study provided a broad estimate of losses: Over the period 2001–2003, up to $7 million of workers’ compensation premiums were not paid for misclassified construction workers and up to $91 million for misclassified workers across all industries (Carré and Wilson 2004).3 Workers’ compensation insurance premiums are higher for hazardous construction work such as roofing (costs may exceed 50 percent of payroll) and provide a sizable incentive to avoid covering workers.4
Businesses misclassify workers to achieve less direct but possibly significant savings by avoiding having to comply with federal (Fair Labor Standards Act [FLSA]) and state wage and hours laws that mandate minimum wages and overtime premiums. Businesses also misclassify to bypass requirements of the 1986 Immigration Reform and Control Act, which forbids employers to knowingly hire undocumented immigrants and requires them to verify immigration status. By treating such workers as independent contractors, businesses can avoid carrying such workers on their payroll and any resulting penalties.
IC misclassification undermines worker bargaining power, for both workers who are misclassified and the directly employed workers alongside whom they work. As noted, misclassified ICs are not covered by basic labor standards, particularly laws affecting work hours and compensation. It is therefore easier for employers to enforce bargains on work hours and compensation for the self-employed that not only deviate from the workers’ compensation agreement but also result in effective hourly wages below the federal or state minimum and in actual work hours that go beyond 40 in a week, which under the FLSA would require premium pay. It is also easier for employers to renege on a compensation agreement, to pay cash “under the table” (i.e., unreported on a 1099-MISC tax form), or to shortchange workers on agreed compensation. These vulnerabilities of misclassified workers—and the fact that some employers exploit them—have a ripple effect on directly employed wage workers in these workplaces, hemming in their ability to bargain for higher compensation and to resist standards violations by their employers.
Salient policy concerns: Who has a stake?
What are the policy implications of IC misclassification? Who has a stake in policy and collective action to address, reduce, and eliminate it? First, misclassified workers, those most directly concerned, operate as if they were autonomous microentrepreneurs but have neither the autonomy to choose when to work nor for whom, nor the ability to line up alternative sources of income while engaged in a particular arrangement.
Misclassified workers lack access to workers’ compensation insurance at group rates. Ostensibly, individual coverage is possible and often mandated for the self-employed, but the quality of individual plans varies, and state mandates as to what exclusions and disqualifying clauses are allowed also vary. As noted, misclassified individuals pay the self-employment tax (the full Social Security and Medicare tax), which amounts to 15.3 percent of earnings below $118,500. They do not have income replacement during periods of unemployment and do not benefit from minimal labor standards concerning an hourly wage and overtime. Because they are not employees, they do not have access to employer-based benefits such as paid sick time or disability insurance; the lack of access to these benefits is most visible for ICs, who are most likely to perceive the consequences. Misclassified ICs cannot resort to the Department of Labor for redress when their hire agreement is violated, as, for example, due to underpayment of work hours or failure to receive a 1099-MISC tax form accurately reporting their total compensation.
Because they are classified as self-employed, misclassified workers are not covered under federal and state wage and hour laws, most notably the minimum wage and overtime premium mandates. In practice, they are exposed to risks of underpayment (i.e., working more hours than compensated for). Recent cases have exposed IC misclassification involving wage fraud. For example, in 2014 the California labor commissioner’s office examined more than 300 claims for wage theft related to misclassification, a “dramatic rise from 2011,” according to news accounts. In all, more than 500 complaints were filed in 2012 and 2013 (Lopez 2014). Misclassified ICs also forgo access to employer-based benefits, most notably health insurance where provided—a major disadvantage relative to wage employees, particularly prior to implementation of the Affordable Care Act.
Second, for coworkers of misclassified ICs, misclassification is a threat to bargaining power over wages and working conditions. Not only is the possibility of being replaced by a lower-cost worker a disciplinary tool, but the possibility that misclassification offers employers to skirt labor standards altogether has a chilling effect on a worker’s ability to raise concerns about, or bargain for, improvements in either the terms of employment (pay, work hours, etc.) or working conditions. In union environments, if misclassification occurs it may undermine compliance with the terms of a bargaining agreement.
Third, employers who play by the rules, payroll their workers, and comply with tax requirements and labor standards are disadvantaged by higher labor and administration costs relative to employers that misclassify. This is the case overall, across all industries, but constitutes a particularly striking disadvantage in industries where contract bidding is constant, where contracts are generally awarded to the lowest bidder, and where a notable share of employers engage in the practice. For example, the labor cost differential is estimated to range from 20 percent to 40 percent of payroll when employers do not pay the unemployment insurance tax, workers’ compensation premiums, the employer share of Social Security, and pension or medical insurance (Belman and Block 2009, 11). When the workers perform dangerous work, the labor cost savings from avoiding workers’ compensation insurance premiums can, by themselves, approach 50 percent of wages (CPWR 2010).
Playing by the rules is hard to do when competitors misclassify workers
Prepared by EPI
In February 2015, after hearing testimony from truck drivers and stagehands, the Georgia Senate Insurance and Labor Committee voted unanimously to study misclassified workers in the state and to find ways to enhance job protections for workers in this category. Along with the hardships faced by misclassified workers, much of the discussion in a public hearing by committee members focused on how companies that wrongly misclassify employees as ICs gain an unfair advantage over their competitors by escaping payroll taxes. The Georgia Senate’s President Pro Tempore, David Shafer, told the committee, “I’m persuaded this failure to enforce existing laws is unfair. It’s unfair to workers who are working very hard to support themselves and their families, and it’s unfair to businesses who obey the law and are then put at a disadvantage.” (Jones 2015a)
It has been estimated that more than a third of construction workers in Southern states such as North Carolina and Texas have been wrongly classified as independent contractors. Annual losses amount to about $400 million in Florida, $467 million in North Carolina, and $1.2 billion in Texas (Ordonez and Locke 2014b).
When and where misclassification spreads
It appears that IC misclassification spreads under one or both of the following conditions: (1) where it is most profitable to commit fraud (i.e., where avoiding mandated costs results in significant savings, such as in construction, where workers’ compensation insurance premiums are high, and in other sectors where injuries are common), and (2) where the business organization is such that the practice is easy to conceal and employer responsibility hard to pin down. The latter is the case in long subcontracting chains (another characteristic of construction), where the work is project-based, where companies rise and fall, and where there is a high rate of turnover of employers (or “bosses”).
IC misclassification also occurs in work settings where monitoring is more difficult because worksites are small, are scattered, and employ few workers. It occurs where workers operate in isolation, such as with housecleaning and in-home care, where the worker is dispatched by an agency but is treated by the agency as a self-employed worker.
Misclassification also occurs in settings where the work is performed individually but within the context of a large business organization, which entails layers of subcontracting. Trucking deregulation and restructuring over the past 20 years have resulted in heavy use of independent contractors for driving, many of whom are misclassified employees (see the section “Recent patterns and trends to watch”). In fact, several courts have determined that arrangements in which drivers lease trucks and are subject to a number of behavioral requirements and mandated costs under the lease agreement constitute employment and that employees were misclassified as ICs.5
Several components of the social protection system also are adversely affected by IC misclassification, most directly the unemployment insurance and the workers’ compensation insurance systems. The federal unemployment insurance system and state systems established to finance UI benefits are impacted by misclassification fraud through the loss of UI tax revenue. If a worker files for UI benefits, challenges his or her IC status, and is found to have been improperly classified by the employer as self-employed, the worker is eligible for benefits— while no tax was collected from the employer. Fines and retroactive taxes have to be collected from the employer (except in cases where the employer can receive exemption through establishing good-faith ignorance).
State worker compensation and disability insurance systems are adversely affected in similar ways. Misclassified ICs who are injured on the job and are found to be misclassified may, in many states, receive benefits from a public “uninsured employer fund” set up to compensate employees of businesses who fail to purchase insurance.6 Workers’ compensation insurance benefits are also paid to injured workers whose employer has retroactively enrolled them in the company’s insurance policy, according to insurance industry fraud units. This practice entails enrolling a worker as of the Monday of the week of injury.7 Waiting until an employee is injured before paying insurance premiums makes the system more expensive for employers that comply with the law.
Federal and state income tax systems face significant revenue losses in income tax revenues and therefore have a stake in the eradication of IC misclassification. A 1984 IRS study found that 15 percent of employers engaged in misclassification and misclassified 3.4 million workers as independent contractors, resulting in an estimated loss of $1.6 billion in FICA tax, unemployment tax, and income tax combined (or $3.5 billion, as of 2014, adjusted for inflation) (Bickley 2011, 10).
According to the IRS, independent contractors, and the self-employed more generally, tend to underreport income. A 2007 IRS report indicates that in 2001, 57 percent of income for “non-farm proprietors” went unreported (report cited in Hurst, Li, and Pugsley 2010). A recent study of income underreporting in household surveys spanning several national data sets and a range of years (1980–2003, but with variation across datasets) finds that the self-employed (misclassified or not), as a group, underreport 30 percent of their income (Hurst, Li, and Pugsley 2010).
Underreporting can be higher if the self-employed report cash income. A U.S. GAO report cites earlier IRS reports that found that self-employed workers operating formally underreport 32 percent of their business income but that “informal suppliers” (self-employed reporting cash income) do not report 81 percent of their income (U.S. GAO 1997, 3; U.S. GAO 1996, 1994). Thus, when misclassified ICs (and other independent contractors) are paid in cash, the odds are quite high that little of their income is reported for income tax purposes. Underreporting may have worsened over time. A 1974 IRS report found that independent contractors (misclassified or not) did not report 26 percent of their income. (U.S. Treasury Department 2001, 7).
If recent evidence is indicative, the impact of this pattern of underreporting cash income is exacerbated by the practice of combining IC misclassification with payment in cash. This practice renders monitoring and audits more difficult. For example, a 2012 Massachusetts government recap of misclassification enforcement noted that “the Attorney General’s Office and the JTF [Joint Task Force] received wage and hour complaints against multiple subcontractors working at sites operated by Pulte Homes. The workers alleged that they were not paid for overtime and not paid for all hours worked. Workers were often paid in cash and by checks that bounced.” As part of enforcement efforts, penalties were issued to five employers (subcontractors) for “failure to pay wages and overtime, failure to issue suitable pay stubs, failure to furnish records for inspection, and misclassification of workers as independent contractors” (Attorney General of Massachusetts 2012). Cash payments also make it easy to shortchange workers on pay that is due.
Other systems also are affected. For example, when employers treat employees as ICs and fail to pay FICA taxes on their behalf, the financing of Medicare and Social Security can be undermined.
IC misclassification impacts rights under the National Labor Relations Act. Because misclassified workers are treated as self-employed, their right to organize for purposes of collective bargaining with an identifiable employer is not protected. If organized as a group, they risk charges of antitrust violation (Kennedy 2005).
IC misclassification also impacts the work of the Equal Employment Opportunity Commission and the Pension Benefit Guarantee Corporation (U.S. GAO 2009, 7).
Safe Harbor and labor violations
Usually, when an employer is confronted by the IRS for wrongly misclassifying workers, the IRS has few, if any, ways to seek redress and lost tax revenue. As a result of what is known as “The Safe Harbor Rule,” or more technically, Section 530 of the Revenue Act of 1978, companies can continue to misclassify workers as ICs for tax purposes even if it has been demonstrated that they are really employees. The same rule prohibits the IRS from seeking back taxes or in any way ordering a change of status for the worker.
‘Safe Harbor’ rules shelter misclassifiers
Prepared by EPI
To gain Safe Harbor, all an employer has to do is to show a “reasonable basis” for using ICs, such as it has always structured its workforce this way or that it is a widespread industry practice. Section 530 was supposed to be on the books for only one year to allow Congress to update the law and for companies to reclassify workers and come into compliance during that period. But the rule has been on the books for 37 years, costing estimated billions annually in lost revenue.
Sen. Sherrod Brown (D-Ohio) and Rep. Jim McDermott (D-Wash.) have both introduced bills to close the Safe Harbor loophole (Office of Sherrod Brown 2010), and President Obama has proposed closing the loophole in his 2012, 2013, 2014, and 2015 budget proposals. The U.S. Treasury estimates that eliminating Safe Harbor would generate $9 billion in tax revenue over 10 years. (Ordonez and Locke 2014b)
Magnitude, severity, and trends
Because independent contractor misclassification is fraud, its magnitude is not easily documented. Employers do not report workers they misclassify in national surveys, and misclassified workers who may not be aware of their status (until they file a claim for unemployment benefits or workers’ compensation insurance) also do not consistently report their independent contractor status. Therefore, estimates of misclassification that are currently available rely on administrative data sources, either audit data from enforcement agencies or IRS one-time studies of tax records. A common source of information is the audit data from state unemployment insurance agencies; failure to pay unemployment insurance tax is usually the result of a worker being misclassified.
State studies yield consistent findings that IC misclassification has been on the rise since at least the late 1990s, and that it is worse in industries where workers’ compensation insurance costs are comparatively high and rising (construction being a prime example) and where it is easy to misclassify workers.
Several dimensions of misclassification matter:
Its prevalence: the frequency with which it occurs, as a percentage of employers. Prevalence highlights the relative differences in propensity to misclassify across employers and sectors.
Its severity: the percentage of workers misclassified among the total number of workers employed by misclassifying employers. Severity gives further detail about employer behavior and the degree of impact on their workforce.
Its extent: the percentage of workers misclassified among all workers. Extent enables us to estimate the impact of misclassification on the workforce as a whole; it casts the phenomenon in the context of both the labor market as a whole and the impact on, for example, unemployment insurance and income tax systems.
As of 2012 an estimated 21 states have released studies of misclassification based on unemployment insurance tax audit data; the policy interest is such that we expect this number to increase. Existing studies report that, across most states, the prevalence of misclassification ranges from 11 percent to 30 percent depending on the method used by state unemployment insurance agencies to select companies for audits.8 Audit methods vary across states in the extent to which they target employers for audit: They can base the audits on specific criteria (e.g., a record of prior violation), or use a random sample of employers within industries prone to misclassification, or a mix of both methods (see de Silva et al. 2000 and Carré and Wilson 2005 for reviews of methods used). The composition of the group of employers audited has implications for the prevalence found. States that rely heavily or exclusively on targeted audits report substantially higher prevalence (as high as 62 percent), as can be expected. Nevertheless, there is remarkable consistency in the prevalence of misclassification found across states using similar audit methods.
In companies that violate the law, misclassification is severe. For example, in Massachusetts, from 2001 to 2003, 13 to 19 percent of employers overall misclassified at least one worker but, among these employers, 25 to 39 percent of the workforce was misclassified. In construction, the situation was worse: 14 to 24 percent of employers misclassified, and among these, 40 to 48 percent of workers were misclassified (Carré and Wilson 2005). An analysis of Michigan audit data for 2003–2004 found that 30 percent of employers misclassified and that 24 percent of workers in these employers’ workforces were misclassified (Belman and Block 2008, 14). The Michigan study used audits from construction, trucking, and security guard industries and from 10 percent of the remainder of employers.
The extent of misclassification—the share of a state’s workforce that is misclassified—is a measure that appears low (i.e., a preponderance of workers reported “on the books” are classified properly). Yet the implications for tax systems of even a small percentage of unreported wages are great. And so are the damaging consequences for individual workers.
For example, with 8.4 percent of Michigan workers misclassified during 2003–2004, conservatively estimated UI revenue losses amounted to $16.8 million. State income tax revenue losses were estimated to amount to $19.5 million if workers underreported 30 percent of their income, or up to $32.5 million if they underreported 50 percent of their income (Belman and Block 2008, 14–15). For 2001–2003, Massachusetts UI tax losses related to IC misclassification were estimated to range from $12.6 million to $35.1 million. State income tax losses over the same period were estimated to range from $152 million if 50 percent of misclassified worker income was unreported in tax filings, to $91.5 million if 30 percent of worker income was unreported (Carré and Wilson 2005, 16).
Caveats about state audit data
State audit data are the best source available to state regulators for understanding the magnitude and, importantly, the trend in the prevalence of misclassification. Some state studies indicated steady growth of prevalence from the 1990s to the early 2000—for example, Massachusetts.9 More recent studies (e.g., Belman and Block 2008 on Michigan, Kelsey et al. 2006 for Illinois) indicate no decline, and if anything, increases in the prevalence of misclassification.
Nevertheless, the choice of audit methods by states is driven by the need for efficiency in finding fraud and recovery of tax revenue losses rather than by research needs per se. The logic for this approach is undeniable: Searching for fraud in places it is unlikely to occur is an inappropriate approach to enforcement and a waste of resources. Nevertheless, as a consequence, understanding the importance of misclassification is hampered by the lack of nationwide random sample surveys to assess misclassification and its extent across all industries—an ambitious undertaking.
Another facet of misclassification is that state audit data focus on unreported or underreported wages as the strongest indicators of IC misclassification, which they are. In some cases, and in particular industries, employers also underreport wages for workers who are wage employees. The practice impacts tax collection (both employment tax and payroll tax), but the workers affected by the practice do not lose access to UI, workers’ compensation, or employer-based benefits. A small sample of audits in Michigan found that, in construction, 38 percent of those with underreported wages were misclassified, while in trucking all workers with underreported wages were misclassified. The bulk of dollar amounts that are unreported are paid out to workers misclassified as independent contractors (Belman and Block 2008, 9).
Perhaps the most significant limitation of state audit data is that they do not and cannot point to cases where wages are paid entirely “under the table,” and unreported altogether. This practice is invisible in audit information and requires extensive investigation; some misclassified IC earnings, and the prevalence of misclassification (and related violations), are underassessed. It is possible that a practice of undeclared work and cash payments could be uncovered as part of a UI tax audit, but the practice is, on the whole, not easily documented.
What is known at the federal level
Aside from research planned by the IRS’s National Research Program, the last comprehensive federal estimate of independent contractor misclassification was generated for tax year 1984. For that year, the study found that 15 percent of employers nationwide and across industrial sectors engaged in misclassification of a total of 3.4 million workers. At the time, the estimated tax losses encompassed $1.6 billion in uncollected Social Security taxes, Medicare taxes, federal unemployment taxes, and federal income taxes (Treasury Inspector General for Tax Administration 2013; U.S. GAO 1989).
Methods using tax records have been developed for estimating IC misclassification at the national level. A study prepared by the IRS for the Government Accountability Office rested on the combined use of business and individual tax information. This approach entailed matching “1099 information returns” filed by businesses on behalf of their independent contractors with individual income tax returns for the workers concerned. This method enabled analysts to apply criteria such as deriving all or most of one’s income from a single business payer (a strong indicator of misclassification) and thus to estimate the percent of workers misclassified. The study found that very stringent criteria (e.g., at least $10,000 of income all from a single business payer) pointed to misclassification that, in turn, was confirmed, through an IRS audit, in virtually all cases (U.S. GAO 1989).
Subsequent federal reports have called for increasing attention to misclassification in light of reports of increasing incidence at the state level and high profile cases of egregious misclassification practices. As a result, an IRS study currently underway will review taxpayer records and 1099-MISC tax forms issued by enterprises that have also filed employer’s quarterly federal tax returns in order to identify “questionable worker classification issues” (Treasury Inspector General 2013, 1).10
Recent patterns and trends to watch
Recent indications are that independent contractor misclassification has not abated and, if anything, might have increased during the period of protracted unemployment and underemployment following the 2007–2008 financial crisis. Worker bargaining power in the job market was weakened, enabling some employers to exact favorable terms of employment such as IC misclassification with greater ease.
In addition to these general trends, specific industries, and some companies within those, have surfaced as using IC misclassification systematically.
Pervasiveness in trucking
Court cases and public demonstrations have called attention to pervasive misclassification in the trucking industry. Since trucking deregulation in 1980,11 the industry has made heavy use of self-employed drivers, or “independent contractors.” Large transport companies sold their trucks to the drivers, then contracted with them on a per-load basis (Bensman 2009).
Among drivers, those who operate with little autonomy and yet are classified as self-employed have become more common, and their problems have become visible in recent years. According to one study, based on 10 surveys of drivers at seven ports, 82 percent of workers in the part of the industry that hauls containers from ports to warehouses are misclassified as independent contractors (Smith, Bensman, and Marvy 2012; Smith, Marvy, and Zerolnick 2014).12
A study of truckers at the ports of New York and New Jersey provided indications of the ways in which drivers who were being classified as independent contractors were operating with limited autonomy from the transport company (Bensman 2014). Drivers were prohibited from making deliveries for other companies and, in effect, could work for only one (which is indicative of company control); many were assisted in leasing their trucks by the trucking companies, which took possession of the leases. The trucking companies obtained insurance for the drivers and billed them for it, then took that amount out of their weekly pay (Bensman 2014).13
In the Los Angeles–Long Beach port area, only 10 percent of port truck drivers are estimated to be directly employed by companies, with all others considered to be owner-operators. A contractor arrangement would theoretically leave drivers free to work for multiple companies, set their work schedule, and hire other drivers. Drivers report that, in practice, companies exert control over work, such as enforcing environmental compliance for trucks and leasing trucks for drivers (Lopez 2014). In some cases, port truckers have challenged their status in court—which is how details of contracts have come to light. For example, practices by transport companies operating in the port of Los Angeles have been called into question by individual drivers, sometimes with the support of the Teamsters union. Three companies, Green Fleet Systems of Long Beach, Pacific 9 Transportation in Carson, and Total Transportation Services Inc. in Rancho Dominguez, have been challenged by workers in court under the FLSA and California law for misclassifying them as ICs as well as violating minimum-wage laws—and some cases were won.14 Other similar cases have also been won.15
In early 2015, additional attention was brought to bear on cases that involve “stage hands and other backstage workers,” involved in occupations related to supporting concert performances and other shows. The entertainment industry has historically made use of workers who operate through the International Alliance of Theatrical Stage Employees (IATSE), which dispatches workers to sites to be payrolled directly by the venue or promoter. Recently, Live Nation, described by the Washington Post as one of the country’s largest concert promoters, has drawn attention from unions and regulators about its hiring practices, particularly in the South, as indicated by news coverage of cases in Georgia. It hires the services of stage hands, ultimately, through a staffing company, which in turn dispatches workers to sites but classifies them as “independent contractors.” Concerns have been raised regarding several aspects of the employment relationship: the lower pay that the staffing company, Crew One, provides (about half the union rate); the fact that ICs would have been on the venue payroll under earlier practices; and the fact that, according to critics, Crew One’s activities risk flooding what has been a “craft-like” market, which is characterized by an organized, skilled labor supply, with a broader range of workers, some of whom may not be certified to perform the tasks (e.g., lighting or carpentry) for which Crew One is dispatching them. The stage hands have been recognized as Crew One employees by the National Labor Relations Board (NLRB), a determination Crew One has challenged in court, and have voted to join the union. Crew One has so far refused to bargain. The practice of combining subcontracting and staffing services has drawn attention from the IATSE as well as from regulators. Cases of IC misclassification in both entertainment stage hand work and trucking have come to the attention of legislators in Georgia, prompting calls for investigation and regulation (DePillis 2015; Jones 2015b).
Tied (in)dependent contractors: Recent court decisions regarding FedEx Ground and other transport
Within the transportation industry, FedEx has taken the lead by implementing a business model for ground delivery (FedEx Ground) that, for years, has relied almost exclusively on drivers treated as independent contractors, sometimes owner-operators (i.e., drivers who own their delivery truck). Under the operating agreement with the company, drivers are responsible for leasing or owning their truck, truck repairs, and other expenses. They are treated as self-employed with respect to taxation and employment law. FedEx refers to these workers as “independent businesses.”
The ambiguity created by the fact that drivers operate on their own has given FedEx significant leeway in running the ground operation and reducing costs (by not paying payroll taxes or offering benefits). According to numerous accounts, FedEx Ground mandates that its drivers purchase and use FedEx uniforms and dictates standards of operation and service delivery. It also controls the routes and route changes to a great degree and, as a result, sets total hours of work for drivers. The specificity of these requirements, and workers’ concern about working conditions and compensation, have given rise to significant litigation, in both individual and class action lawsuits. Because of these cases, details of implementation of the operating agreement have been made clearer, and the difficulties workers face have been highlighted.
As far back as the early 2000s, drivers initiated court cases to claim employee status and coverage by labor standards. As of late 2010, most decisions at the state level and in district courts vindicated FedEx Ground. What was at stake in these cases was how courts used and interpreted multifactor tests for dependent worker status based on dimensions of control over the pacing and scheduling of the work or whether they found that drivers could in theory avail themselves of the opportunity for entrepreneurship beyond their sole client, FedEx Ground. In a December 2010 decision, a federal district court judge in Indiana, presiding over dozens of IC misclassification cases in a “multi-district litigation,” granted summary judgment in favor of FedEx Ground in 42 IC misclassification lawsuits brought by drivers in 27 states, including California and Oregon (Reibstein 2014).
However, in the last year there have been significant changes in how courts view the FedEx Ground drivers. In August 2014, the U.S. Court of Appeals for the 9th Circuit Court reversed the December 2010 Indiana district court decision with respect to FedEx Ground operations in California and Oregon and determined the workers to be employees under state law, and thus covered by federal and state wage and hour laws. Also, in September 2014, the Kansas Supreme Court ruled that FedEx Ground “contractors” are employees, not independent contractors. Appeals from FedEx Ground are expected.
These recent court cases provide significant information on the detailed wording of the operating agreements between FedEx Ground and its drivers, as well as on the practical effects of such agreements. The FedEx Operating Agreement permits a driver to delegate to other drivers, take on additional routes, or sell his route to a third party. However, the appeals court noted that FedEx Ground may refuse to let a driver take on additional routes or sell his route to a third party, and FedEx Ground’s senior managers have the authority to reject proposed replacement drivers based on failure to meet company standards such as grooming requirements (Reibstein 2014). These requirements reduce the autonomy and control of drivers and weaken the case for their being self-employed. Similarly, judges in recent decisions have considered that drivers with these contracts should be allowed to work for other companies in order to be considered self-employed. Some courts have found that the workload at FedEx Ground has made it impossible for drivers to have other jobs at other companies (Schultz 2014). These recent decisions have moved beyond the “letter” of the operating agreements and, instead, have focused on the worker’s ability, in practice, to exercise autonomy.
FedEx Ground drivers incurred significant financial liability as self-employed “tied” contractors according to lawyers in the California and Oregon cases. By “tied” contractors, it is meant they are dependent on their customer (here FedEx Ground) for accessing the customers and for their revenue from truck driving. Between 2000 and 2007, FedEx Ground drivers in California were paid $85,000 to $90,000 per route, according to their lawyer. After expenses, their lawyer noted, “These drivers [were] left with $45,000, maybe $50,000, no health insurance, no retirement for working 55 plus hours a week.” The lawyer noted that these earnings corresponded to “a UPS driver wage prior to the benefit package” (quoted in Sherman 2014).
These recent court decisions are likely to have ramifications beyond FedEx Ground for company practices in the package delivery and heavy freight shipping industries. Most immediately, some observers expect FedEx Ground to shift to subcontracting to companies that will, in turn, put drivers on their payroll (Sherman 2014). The possible impact on compensation remains to be determined.
Beyond coverage under employment standards, putative independent contractors have also appealed for coverage under the National Labor Relations Act in order to receive protection from retaliation during organizing and to access collective bargaining mechanisms. And in September 2014, the NLRB rejected FedEx Ground’s claims that its drivers are independent contractors, finding that they are in fact employees and that FedEx had violated the law by refusing to bargain with a group of them (Eidelson 2014).
Unreported wages and other violations
Beyond IC misclassification and underreported wages, an important concern is the growing practice of unreported wages, that is, employment that remains fully invisible to tax authorities and is discovered only as part of a tax audit, workplace investigation (by, for example, a labor standards agency), or worker complaint. Macro evidence is not readily available, as noted earlier, but state enforcement agencies call attention to this practice as a growing concern, one prompting the need for further collaboration with the U.S. Department of Labor.16 The practice of unreported wages usually indicates that other employment regulations are prone to violation as well. The Memoranda of Understanding between the U.S. Department of Labor and state employment and labor agencies, the adoption of which has been spreading, aim first to address egregious IC misclassification and likely will help locate cases of undeclared or unreported wage payments (see the section “Research needs and options for improving policy and enforcement”).
Web-mediated employment
A recent trend to watch is that of Web-mediated employment, or “tasking.” The most visible version of tasking is orchestrated by TaskRabbit, a Web-based company that has lined up seekers of services with “bidders” (workers) for specific tasks since 2008. The company considers the workers (who numbered about 30,000 in 19 cities in 2014), who do short-term tasks for customers, to be self-employed. Over time, with growth, TaskRabbit has had to contend with the ambiguities of coordinating service, providing some task standardization, and instituting quality control (e.g., providing guidelines on behavior and setting expectations) (TaskRabbit 2015), while trying to maintain an arm’s-length relationship with the workers, describing them as microentrepreneurs or contractors. In 2014, the company instituted an hourly wage floor of $11.20 and provided affiliated workers access to discounts (for items such as health care, tools, and cell phones) (Dwoskin 2014). It also standardized the range of jobs to four broad categories: handyman work, home cleaning, moving help, and personal assistant services. Some have argued that these changes, along with vetting workers, correspond to a shift from a platform where individuals negotiate to an agency-like model or service-on-demand model (Lim 2014). These practices illustrate the complexities of models that straddle the boundary between wage employment and self-employment.
Other companies using Web-based platforms to organize matches between workers and customers are likely to face similar conundrums over time (Weber and Silverman 2015). Most recently, Uber and Lyft—Web-based companies that line up self-employed drivers with customers—have faced court challenges to their classifying drivers as self-employed, because aspects of the work (e.g., standards for service delivery) indicate lack of worker control and independence.17
“At first people like the flexibility of being a contractor, but at the end of the day most people don’t have the luxury to bring in half a paycheck,” Kate Donovan, CEO of Zirtual, the startup that provides remote personal assistants, recently told the Wall Street Journal. Zirtual initially used ICs but later switched to an employee model. Zirtual is an exception. (Weber and Silverman 2015)
Challenges in implementing the definition of employee and self-employed
A number of the cases reviewed in the previous section represent situations where workers unequivocally should have been treated as wage employees. Other cases seem to dwell in a grayer area, where a “dependent contractor” status might be an appropriate description. Some coverage under social protections and labor and employment law could be extended to these workers in a manner that recognizes the dual (or boundary-spanning) nature of their work. This is the subject of discussion in international venues such as the International Labor Organization but is a notion that has, so far, been absent in U.S. mainstream policy discussions (Weber 2015).
For now, workers treated as independent contractors—and employers—face a mosaic of criteria for determining employment status; criteria vary across statutes and, in some cases, across states. This situation facilitates honest mistakes on the part of businesses, but also fraud. It also makes it difficult for workers to know their rights. The IRS criteria for independent contractor status are publicized; other areas of regulation such as health and safety or workers’ compensation may use a slightly different definition. This occurs because the criteria for “employee” are established to suit the implementation of particular regulations. Court decisions significantly drive the updating and implementation of these criteria.
The various criteria for determining employee status are grouped into three main tests (National Employment Law Project 2000):
Common Law Control Test / IRS 20-Factor Test
Economic Reality / Economic Dependence Test
“Suffer or Permit to Work” Test
The Common Law Control Test/IRS 20-Factor Test is used for employee status under FICA, the Federal Unemployment Tax Act (FUTA), Employee Retirement Income Security Act (ERISA), most state workers’ compensation laws, and the National Labor Relations Act.18 It is the narrowest test used to find coverage under labor and employment laws; if employment status is found under this test, a worker is automatically covered under the other two tests (National Employment Law Project 2000).
The Economic Reality / Economic Dependence Test can be used to determine whether a worker is an employee or an independent contractor for purposes of the FLSA, Agricultural Worker Protection Act (AWPA), Equal Pay Act (EPA), and the Family and Medical Leave Act (FMLA). The factors aim to determine whether and on whom the worker is economically dependent for his or her earnings and working conditions.
The broadest test for finding an employment relationship is the “Suffer or Permit to Work” Test. This may be used only to find coverage under the FLSA, AWPA, EPA, and the FMLA. It may be used to find an employment relationship where one would not be found under the other two, more restrictive tests.
In particular, the implementation and enforcement of the unemployment insurance tax (and UI eligibility) varies across states. A study conducted for the Department of Labor found that, as of 2014, 14 states plus the District of Columbia used the Common Law Control Test,19 22 states used the ABC test,20 10 states used their own test developed from the Common Law Control Test,21 and four states used the IRS 20-Factor Test22 (de Silva et al. 2000, 20–22, A7).
Research needs and options for improving policy and enforcement
Going forward, protecting workers as well as tax and insurance systems from misclassification will require research, enhanced enforcement, and other policy changes. National studies based on tax records could derive an updated estimate of misclassification. Targeted studies of patterns of cash payment and nonpayment of wages that are facilitated by IC misclassification would document an apparent growing problem. Documenting worker experiences with misclassification would shed light on its negative impacts. Issues created by the emergence of Web-mediated employment require exploration.
On the policy front, recent increases in federal and state enforcement and information activities have begun to publicize the problem and to affect practices. Policy options to consider include experimenting with means of enforcing labor standards other than individual worker complaints, or audits, because misclassification is more frequent where other labor laws are broken. Options also include changes to the social protection systems as well as revisiting federal and state statutes and their implementation of the employee/self-employed dichotomy in light of current employer practices and worker needs.
Research
Methods used by the IRS in earlier studies (U.S. GAO 1989) continue to be the tested methods for deriving a national estimate of misclassification and of its tax loss implications. Matching IRS employer “1099 Information returns” issued by employers with workers’ income tax returns (e.g., by singling out those individual returns with at least $10,000, in 1988 dollars, from a single business payer) has proven to be a good indicator of misclassification. As noted earlier, the results of the GAO 1989 study (thanks to the use of IRS follow-up audits) confirmed that most taxpayers reporting at least $10,000 from a single business were misclassified as independent contractors.
The criterion for evidence of possible “economic dependence” used in this early, comprehensive study was stringent according to the 1989 GAO report and, some would argue, might overlook other situations that are indicative of misclassification. Another option to consider—one that could be implemented in future studies—entails considering individual tax returns where the main customer business (the 1099 issuer) paid less than $10,000 to an individual or is responsible for less than 100 percent of a worker’s earnings in a year (U.S. GAO 1989, 6). A pattern of being responsible for 70 percent of a worker’s earnings in a given year may be an indicator of misclassification. Considering the situations documented in recent court cases in industries, such as trucking, with heavy use of independent contractors, implementing this criterion in tax studies (and comparing the results to the use of the narrow criterion) would yield important and useful information on how to estimate national impacts of IC misclassification. As a rule, testing the sensitivity of the incidence and extent of misclassification, and resulting tax losses, with alternative criteria for identifying problematic tax returns would prove fruitful.
Beyond national studies based on tax records, targeted research on specific industries is also possible, and desirable, even if the estimates derived cannot easily be extrapolated to the economy as a whole. It is clear that IC misclassification—like other labor standards violations—clusters in particular industry subsectors. Construction and trucking have received attention because the practice is widespread in these sectors and because the consequences are dire for employees without workers’ compensation coverage in these high-injury sectors.23 Other sectors would benefit from targeted sector-specific studies—for example, the home health care industry, the domestic help staffing industry, and other subsectors providing in-person services to households.
As a whole, and across sectors, targeted studies are useful for capturing instances of cash payment of wages, underreporting of wages by businesses/employers on 1099 forms, and under- or nonpayment of compensation. As has been demonstrated in a study of other labor standards violations, targeting a subset of industries susceptible to violation and tracking worker experience, patterns of wage payment, schedules, and working conditions exposes how pervasive labor standards violations can be in some sectors (see three-city study in selected industries by Bernhardt, McGrath, and DeFilippis 2007; Bernhardt et al. 2009). Studies can start with workers and track evidence back to their employers to document both dimensions of the phenomenon—that is, first, misclassification as self-employed, and second, cash payment and under- or non-payment of compensation.
Ultimately, identifying patterns of cash payment and nonpayment that are facilitated by IC misclassification may turn out to be as important, if not more important, than documenting misclassification that is made apparent from examining reported income (tax returns).
Additional research on the worker experience with misclassification—about the experiences and constraints that lead workers to accept misclassification (when they knowingly do so)—would be useful. In policy discussions, some have argued that workers usually collude in order to have higher earnings. Identifying under what conditions workers engage in collusion—what rationing of total work hours takes place, what substandard wage is offered, what pressures are exerted—would provide useful information for considerations of means of enforcement.
Also, it is time to conduct research on workers and employers operating in the gray zone between wage employment and self-employment, where court cases and audits have not, so far, raised issues of misclassification but, nevertheless, where important questions about worker access to labor standards and social protection arise (in employment that includes, for example, certain kinds of freelancing in desktop publishing and communications, and various arrangements for contract labor, for programming, paraprofessional, and technical health-related fields).24
Enforcement and policy options
The prevention of IC misclassification and enforcement of existing federal and state statutes have received greater attention over the past few years than in the previous two decades. Efforts have taken place at the federal level, at the federal and state levels through collaboration (in departments of labor and sometimes in departments of revenue), and at the state level. The Internal Revenue Service and the U.S. Department of Labor have had a Memorandum of Understanding in place since 2011 to fac