Submitted via https://www.regulations.gov/comment?D=WHD-2020-0007-0001

Amy DeBisschop
Division of Regulations, Legislation, and Interpretation
Wage and Hour Division
U.S. Department of Labor
Room S-3502
200 Constitution Avenue NW
Washington, DC 20210

Re: Comments on Independent Contractor Status Under the Fair Labor Standards Act (RIN 1235-AA34)

Dear Ms. DeBisschop:

The Economic Policy Institute (EPI) is a nonprofit, nonpartisan think tank created in 1986 to include the needs of low- and middle-income workers in economic policy discussions. EPI conducts research and analysis on the economic status of working America, proposes public policies that protect and improve the economic conditions of low- and middle-income workers, and assesses policies with respect to how well they further those goals. EPI submits these comments on the Department of Labor’s (Department/DOL) Notice of Proposed Rulemaking regarding the standard for determining who is a covered employee and who is an independent contractor under the Fair Labor Standards Act (FLSA).1

The Department’s proposed interpretation is contrary to law because it ignores the plain language of the FLSA’s definition of “employ,” which “includes to suffer or permit to work,”2 and ignores U.S. Supreme Court and federal circuit court authority interpreting the Act. The Department is attempting to impermissibly narrow this very broad definition of “employ” by proposing a restrictive interpretation of the long-accepted “economic realities” test.3 This five-part test has always been interpreted by the Supreme Court in its totality, not weighing any one factor more than another. But now, DOL proposes amending it in a fashion that places undue weight on two factors and then narrows those two factors further—individual control over the work and opportunity for profit or loss. As a result, the Department is proposing to constrict the FLSA’s broad coverage in a way that will undermine its statutory intent. This proposal makes it easier for employers to classify workers as independent contractors, and if finalized, it would lead to an increase in the share of the workforce that are independent contractors.

In the proposed rule, the Department egregiously fails to estimate the transfers between employers, workers, and the social insurance system that would occur if this proposal were finalized. The requirements that agencies must follow as a part of the rulemaking process are very clear, and among them is the requirement that agencies must assess all quantifiable costs and benefits “to the fullest extent that these can be usefully estimated.”4 There is no question that DOL could have produced estimates; in what follows, I show that it is straightforward to produce estimates using data researchers routinely use and taking a methodological approach that is in the spirit of estimates the Department of Labor undertakes on a regular basis. One plausible explanation for why DOL left out the required estimate is that any good-faith estimate would have shown this rule would result in a substantial transfer from workers and the social insurance system to employers.

In this comment I will estimate these transfers. The basic structure of this analysis is to estimate (1) the change in the value of a job to a worker if the worker is classified as an independent contractor instead of an employee as a result of this rule, and (2) the change in payments to social insurance funds if a worker is classified as an independent contractor instead of an employee. Multiplying these figures by the estimated number of workers who would shift to independent contractor status if this rule were finalized will yield the aggregate impact of the rule on workers and on social insurance system coffers.

To estimate (1) and (2) above, I need to determine the earnings of workers whose classification would change from payroll employee to independent contractor as a result of this rule. The workers most likely to be affected by this rule are workers in lower-wage occupations in labor-intensive industries, such as delivery workers, transportation workers like taxi drivers and some truckers, logistics workers including warehouse workers, home care workers, housecleaners, construction laborers and carpenters, agricultural workers, janitors, call center workers, and staffing agency workers in lower-paid placements. Using Occupational Employment Statistics data from the Bureau of Labor Statistics where available, the median annual earnings for these jobs ranges from $24,850 for maids and housekeeping cleaners to $48,330 for carpenters. Given that I do not have a way to determine the precise earnings of those workers whose classification would change if the rule were finalized, I will assume that a worker whose status would change as a result of this rule is at the bottom of this range, $24,850, to be extremely conservative.

Take-home earnings represent only a portion of labor costs to an employer and the value of a job to an employee. In order to estimate the total compensation of a worker whose classification would change as a result of the rule, I use data from the Employer Costs for Employee Compensation (ECEC) program of the Bureau of Labor Statistics. The ECEC breaks down total worker compensation into regular pay (wages and salaries), supplemental pay (overtime, premium pay, shift differentials, and nonproduction bonuses), paid leave, insurance (health, life, long- and short-term disability), retirement benefits, and legally required benefits (Social Security, Medicare, federal and state unemployment insurance, and workers’ compensation). The ECEC has these compensation profiles by occupation and industry, but there obviously isn’t a compensation profile for the set of workers whose classification would change as a result of this rule. To be extremely conservative, I assume that the compensation profile of workers whose status would change as a result of this rule is the same as the compensation profile of workers in the occupation/industry breakdown with the lowest level of total compensation, service occupations in private service-providing industries. The cost per hour worked for various components of total compensation for service occupations in private service-providing industries is given in Table 1.5 Table 1 also calculates the ratio of each of these components to pay (wages, salaries, and supplemental pay), which I will use to estimate various compensation components of workers whose classification status would change as a result of the rule.

Table 1 shows that 2.6% of these workers’ pay is composed of supplemental pay (mostly overtime and other premium pay like holiday pay, but also shift differentials and nonproduction bonuses). Recall that I am conservatively assuming workers whose classification status would change as a result of the rule earn $24,850 annually. That means that $24,202 is earned as regular pay, and $648 was earned as supplemental pay. Table 1 also shows that paid leave benefits are equivalent to 5.5% of pay, which is equal to $1,357 for a worker who earns $24,850. Further, Table 1 shows that insurance (health, life, long- and short-term disability) and retirement benefits are equivalent to 11.2% of pay, which is equal to $2,771 for a worker who earns $24,850.

The employee will have 7.65% deducted for Social Security and Medicare, which is equal to $1,901 for a worker who earns $24,850.6 The employer is also required to pay $1,901 to cover the employer share of Social Security and Medicare. Table 1 also provides the cost to the employer of other legally required benefits—unemployment insurance and workers’ compensation—showing that these costs are equivalent to 1.0% and 2.7% of pay, respectively, which is equal to $259 and $677, respectively, for a worker who earns $24,850.

These figures are in the first column in Table 2. Table 2 shows that the total value to an employee of a job where the pay is $24,850 is estimated to be $27,077, after adding in the value of paid leave, insurance, and retirement benefits, and subtracting off FICA (Social Security and Medicare) taxes.7 The first column in Table 2 also shows that the total payment to social insurance funds (Social Security, Medicare, unemployment insurance, and workers’ compensation) associated with this payroll job is $4,739 (summing the employer contribution to social insurance and the worker share of Social Security and Medicare).

The second column of Table 2 estimates how these quantities would be different if the worker were classified as an independent contractor instead of as an employee. The first question that arises is what pay the worker would receive as an independent contractor. The Department states that that “in a competitive labor market, any reduction in benefits and increase in taxes is likely to be offset by higher base earnings—referred to as an ‘earnings premium,’” which would mean that “in theory, companies would likely have to pay more per hour to independent contractors than to employees because independent contractors generally do not receive employer-provided benefits and have higher tax liabilities,” and “any tax-related transfers from employers to workers are likely to be offset by higher wages employers pay to ensure workers’ take-home pay remains the same.”8 However, the Department goes on to note that “this expected wage premium may not always be observable at a statistically significant level” and in fact, the Department’s own analysis of 2017 Contingent Worker Supplement (CWS) data “did not show a statistically significant difference” between the wages of employees and independent contractors with the same demographic characteristics in the same occupation.

This is not surprising when considering the fact that the theory that businesses will not be able to pay less in total compensation to workers if their status shifts from employee to independent contractor—that their base pay will rise to make up for a reduction in benefits—is based on the assumption of perfectly competitive labor markets. There is broad and growing evidence that perfect competition is rare, and that most labor markets do not function competitively—particularly low-wage labor markets like those under consideration here, in which workers are more likely to lack the power to bargain for higher wages to compensate for their loss of benefits and increase in taxes when they become independent contractors.9,10 Further, very-low-wage employees whose wages are elevated by the minimum wage could easily see their wages drop when, as independent contractors, they no longer legally must be paid the minimum wage.

After review by the Office of Information and Regulatory Affairs (OIRA) at the White House, the following sentence was removed from DOL’s original economic analysis: “The Department anticipates a positive wage effect due to the expected increase in labor force activity, but did not attempt to quantify estimates of changes in earnings.” That removal following OIRA’s review is no surprise, given that a positive wage effect of the proposed rule can’t be supported by the evidence.

Given these findings, I assume that the low-earnings workers under consideration here get no wage premium when they shift from being an employee to being an independent contractor, and instead receive the same regular pay that they received when they were an employee. I further assume that this worker will receive no supplemental pay (such as overtime, holiday premium pay, shift differentials, or nonproduction bonuses), or paid leave. Overtime is not legally mandated for workers who are not payroll employees, and it and other kinds of supplemental pay or paid leave are unlikely to be received by an independent contractor. I also assume that a worker who changes status as a result of this proposed rule would receive no insurance or retirement benefits from their contract job.11

An additional cost that people who switch from being payroll employees to independent contractors have to face is greater paperwork costs. For example, the IRS estimates that business taxpayers spend 13 more hours than nonbusiness taxpayers doing their taxes.12 If we conservatively assume that independent contractors spend 30 minutes per week on other (nontax) paperwork costs that they wouldn’t have to spend if they were a payroll employee, that, plus the additional 13 hours spent on taxes, is an additional 39 hours of paperwork per year. This is equivalent to 1.8% of pay, or $445 annually for an independent contractor who earns $24,202 in regular pay annually. Further, independent contractors are likely to need software for doing their bookkeeping and taxes. A commonly used program for things like tracking expenses is FreshBooks, which costs $162 per year for the cheapest option.13 A commonly used tax preparation program for independent contractors, TurboTax Self-Employed, costs $120 plus $50 per state for the cheapest option, whereas an individual who is a payroll employee with a simple tax return can prepare their return using TurboTax at no cost.14 For an independent contractor who earns $24,202, this sums to a new annual paperwork cost of $777.15

Independent contractors are required to pay both the employee and the employer portions of taxes for Social Security and Medicare, which is 15.3% of pay—though they are able to deduct the employer-equivalent portion of the tax, along with business expenses. Taking these factors into account means an independent contractor who earns $24,202 and has $777 in business expenses will pay $3,310 in Social Security and Medicare taxes.16 Furthermore, when a worker is reclassified from an employee to an independent contractor, the former employer no longer pays for unemployment insurance or workers’ compensation. As Table 2 shows, the value of the job to the worker drops from $27,077 when they are an employee to $20,114 when they are an independent contractor, a drop of $6,963, or 25.7%.17 Social insurance funds get $1,429 less, or 30.2%, when the worker’s status changes.

What might be being left out in this analysis? The Department focuses on “flexibility and satisfaction” as important nonpecuniary attribute that workers may trade income to receive.18 However, it is difficult to imagine that there are a meaningful number of workers who would get more satisfaction from doing the same job for substantially less compensation as an independent contractor than for substantially more compensation as a payroll employee. Many workers indeed may value flexibility, but notably, employers are able to provide a huge amount of flexibility to payroll employees if they choose to; the “inherent” tradeoff between flexibility and payroll employment is greatly exaggerated. Workers also highly value other factors, like income stability, which are much less prevalent among independent contractors and are not taken into account here.

How will the share of workers who are payroll employees and the share of workers who are independent contractors change as a result of this rule? To begin to answer that question, we need to know how many independent contractors there currently are. There is a great deal of uncertainty around this number (the Department notes that “there are a variety of estimates of the number of independent contractors and these span a wide range based on methodologies and how the population is defined”).19 The 2017 Contingent Worker Supplement (CWS) estimated that there were 10.6 million workers who are independent contractors in their main job. This number, however, drastically underestimates the total number of independent contractors by not including workers who do independent contracting on the side, in addition to their payroll job. The Department makes a correction for this issue and estimates that there are 18.9 million individuals working as contractors at a given time. For the sake of my calculations, however, I will limit my analysis to the 10.6 million workers the CWS finds are independent contractors in their main job, since workers who do independent contracting as a side job may, in many cases, make far less than the $24,202 I am assuming workers whose status changes as a result of this rule earn. It should be noted that this means I am leaving out many millions of independent contractors and my estimates will, as a result, be extremely conservative in this way, in addition to other ways my estimate is conservative (some mentioned above and more to come below).

It should be noted that the Department emphasizes the difference between workers whose classifications would change to independent contractor status as a result of the rule and workers who are newly hired as independent contractors. Given the high degree of churn in the labor market—particularly in the low-wage labor market—the distinction between newly hired independent contractors and workers whose status changes is not a relevant distinction, and the Department should drop this emphasis—and should instead focus on the change in the share of the share of the workforce that is made up of independent contractors. Throughout this comment I have, for simplicity, referred to workers shifting from being payroll employees to independent contractors, but that should be understood to represent an increase in the share of the workforce that is made up of independent contractors. It is further worth noting that the Department provided no evidence that this rule would increase overall employment levels, merely citing a McKinsey Global Institute study that finds that 15% of those not working are interested in becoming an independent contractor, but providing no evidence that this rule would likely lead to those workers joining the workforce.20 In fact, by reducing the income of lower-income workers who have a high marginal propensity to consume, this rule would reduce aggregate demand in the economy and likely decrease the overall number of jobs.

How much will independent contracting increase as a result of this rule? The DOL proposal would potentially allow companies to legally argue that workers who are now misclassified as independent contractors or who are working “off the books” would be legitimately classified as independent contractors under the narrow terms of the proposal. As such, one approach would be to use the percentage of workers misclassified or working off the books under current law to estimate the number of workers who could be reclassified as independent contractors under the proposed rule. However, due to severe data constraints, estimates of the share of workers who are misclassified as independent contractors or working off the books are limited. A recent paper estimates that between 12.4% and 20.5% of workers in the construction industry are either misclassified as independent contractors or working off the books.21 Conservatively assuming that the bottom of this range applies more broadly to the lowest-paid quartile of the U.S. labor market, that is 4.9 million low wage workers who may be affected by this rule.22 Of course, these are workers who are already not getting the benefit of being a payroll employee, so the cost impacts described above would not apply without a substantial increase in enforcement. However, this exercise does provide a broad sense of the potential scope of workers affected. Further, even these workers lose something of value under this rule given the current enforcement regime, namely the legal right to the wages and benefits they would receive if they were properly classified. I do not attempt to quantify this effect.

To be exceedingly conservative, I will simply assume that there would be an increase as a result of this rule of 5% in the number of workers who are independent contractors in their main job.23 This translates into an increase of just 530,000 workers who are independent contractors at their main job, given the CWS estimate of 10.6 million workers who are independent contractors in their main job. Multiplying that by my conservative estimate that these workers would lose $6,963 per year yields an aggregate loss to workers of at least $3.7 billion annually. This loss to workers is composed of at least $400 million in new annual paperwork costs, and a transfer to employers of at least $3.3 billion in the form of reduced compensation. Further, social insurance funds would lose at least $750 million annually in the form of reduced employer contributions, meaning this rule also results in a transfer of at least $750 million annually from social insurance funds to employers.24 Table 3 provides the final estimates.

It is important to note that these estimates are lower bounds for many reasons. In particular, I assumed the workers whose status would change are very low-wage workers with a very low level of benefits (meaning that to the extent the status of workers with higher wages or greater benefits changes, I have underestimated the impacts), I only took into account workers who are independent contractors in their main job (meaning that I have left out the impact of any status shift for workers who would be independent contractors for supplemental income, not their primary job), and I assumed that there would only be a 5% increase in the number of workers who are independent contractors in their main job as a result of this rule (meaning that to the extent the status of more workers changes, I have underestimated the impact). Taking into account these factors, the true impact could be many times my estimates. For example, if the true number of workers affected is equivalent to just half of the 4.9 million workers currently estimated above to be misclassified or working off the books, workers would lose on the order of $17 billion annually as a result of this rule.

EPI strongly opposes the Department of Labor’s proposed rulemaking regarding independent contractor status under the Fair Labor Standards Act, because this rule will cost workers billions of dollars annually and will cost the social insurance system hundreds of millions of dollars annually. Further, due to things like occupational segregation by race, discrimination, and other labor market disparities rooted in structural racism, Black and Latinx workers are more likely to work in the occupations affected by this rule. As a result, this rule would exacerbate existing racial disparities. It must not be finalized.

Thank you for the opportunity to submit comments. Please do not hesitate to contact me at hshierholz@epi.org if you have questions.

Sincerely,

Heidi Shierholz
Senior Economist and Director of Policy
Economic Policy Institute
Washington, DC

Endnotes

1. Independent Contractor Status Under the Fair Labor Standards Act, 85 Fed. Reg. 60600–60639 (September 25, 2020).

2. 29 U.S.C. §203(g)

3. See U.S. v. Silk, 331 U.S. 704 (1947).

4. Maeve P. Carey, Cost-Benefit and Other Analysis Requirements in the Rulemaking Process, Congressional Research Service, December 9, 2014.

5. For service workers in service industries, legally required benefits are not disaggregated into subcomponents (Social Security and Medicare, unemployment insurance, and workers’ compensation). This disaggregation exists for service workers in all industries, however, so I simply apply the share each of these subcomponents constitutes of legally required benefits for services workers in all industries to the total value of legally required benefits for service workers in service industries, to estimate the subcomponents for service workers in services industries.

6. The 7.65% used here is slightly smaller than the 8.3% found in the ECEC data presented in Table 1. I use 7.65% because it is the official tax rate.

7. I should note that “total value” is defined as a relatively near-term concept here, in the sense that paying Social Security taxes would likely increase a worker’s Social Security benefits in retirement, so those taxes are not a long-run negative.

8. 85 Fed. Reg. at 60626-60627, 60628

9. Alan Manning, Monopsony in Motion: Imperfect Competition in Labor Markets (Princeton, N.J.: Princeton University Press, 2003); Anna Sokolova and Todd Sorensen, Monopsony in Labor Markets: A Meta-Analysis, Washington Center for Equitable Growth, February 2020; Arindrajit Dube, Jeff Jacobs, Suresh Naidu, and Siddharth Suri, “Monopsony in Online Labor Markets,” American Economic Review: Insights 2, no. 1 (March 2020): 33-46, https://www.aeaweb.org/articles?id=10.1257/aeri.20180150.

10. It is worth noting that if the labor market were to get to full employment and stay there for an extended period, the economic leverage that this would provide workers might mean that even these workers could recoup some of the value of lost benefits and increased taxes in the form of higher wages. However, the instances of full employment in the labor market are the exception, not the norm.

11. An individual with total household earnings of $24,850 and a household size of three or more (for example, one individual with two children) would likely be eligible for Medicaid. I do not attempt to quantify this effect but it should be noted that to the extent workers who change status as a result of this rule are able to take up Medicaid, the transfer to employers related to health care that would result from this rule would be transfers from the social insurance system to employers, not from the worker to employers.

12. See page 101 in Internal Revenue Service, 1040 and 1040-SB Instructions: Including the instructions for Schedules 1 through 3, Cat. No. 24811V, January 2020.

13. FreshBooks, “FreshBooks Pricing, Lite, Plus and Premium Packages ” (web page), accessed October 26, 2020.

14. TurboTax (website), accessed on October 26, 2020.

15. $777 = $445 + $162 + $120 + $50.

16. $3,310 = ($24,202 – $777)*(1-0.0765)*0.153.

17. It is worth noting that this is a similar effect to what other studies find, for example, a paper titled “Pay, Passengers and Profits: Effects of Employee Status for California TNC Drivers” by Michael Reich of the University of California, Berkeley finds that employee status for Uber and Lyft drivers would increase total driver compensation by about 30 percent.

18. 85 Fed. Reg. at 60628.

19. 85 Fed. Reg. at 60623.

20. 85 Fed. Reg. at 60627.

21. Russell Ormiston, Dale Belman, and Mark Erlich, An Empirical Methodology to Estimate the Incidence and Costs of Payroll Fraud in the Construction Industry, Institute for Construction Economics Research, January 2020.

22. Data from the Current Population Survey from the Bureau of Labor Statistics find that there were 157.5 million workers in the US in 2019; 4.9 million = 157.5 million * .25 * .124.

23. A 5% increase is a conservative assumption given the Department is proposing to amend the five-part economic realities test—which has always been interpreted by the Supreme Court in its totality, not weighing any one factor more than another—in a way that will place undue weight on two factors and then narrows those two factors further, making it more likely that workers will be classified as independent contractors and as a result likely leading to a substantial increase in the number of independent contractors.

24. Some might argue that social insurance funds wouldn’t be hurt by not having employers pay into unemployment insurance and workers’ compensation because independent contractors aren’t eligible for those benefits. However, independent contractors are currently eligible for unemployment insurance benefits through Pandemic Unemployment Assistance, and it is reasonable to assume that this will occur in future recessions, as well. Further, low-paid independent contractors who lose their contracts and are without work, or get hurt on the job, would be likely to need to depend on safety net programs to survive, so the social insurance system as a whole would still be depleted.