Constitutional or Not, Proposition 22 Is Bad for California

Overview

On August 20, 2021, Alameda Superior Court Judge Frank Roesch struck down Proposition 22, which would have classified app-based drivers as independent contractors, not employees.[1] Defining those drivers as independent contractors is problematic for California in two ways. As independent contractors, app-based drivers often underpay their taxes and insurance, which creates knock-on effects for the state’s ability to accurately tax this economic sector and financially burdens the state when they fail to pay. And while this classification policy decision may benefit app-based companies in the short term, their unsustainable business models will eventually collapse. Regardless how Proposition 22 fares on appeal, it remains a poor policy choice.

Analysis

Delivery network companies (DNCs) and transportation network companies (TNCs) operate unsustainable business models

Proposition 22 became a political necessity for five DNCs and TNCs (DoorDash, Instacart, Postmates, Uber, and Lyft) after a judge enjoined them from classifying their drivers as independent contractors, meaning that the companies needed to treat their drivers as employees under Assembly Bill 5.[2] Doing so would require the companies to pay considerably more in employment benefits and taxes, a prospect so dreadful that some threatened to shut down in California if Proposition 22 didn’t pass.[3]

To understand why the DNCs and TNCs feel so threatened by employee classification for their drivers, one need merely glance at their balance sheets. Of the five companies that bankrolled Proposition 22, only one is turning a profit. Postmates (which Uber acquired in 2020) lost $182 million between 2015 and 2017.[4] Uber lost $14.2 billion between 2018 and 2020.[5] Lyft lost $5.2 billion between 2018 and 2020.[6] DoorDash lost $1.3 billion between 2018 and 2020.[7] Only Instacart, which went from losing $300 million in 2019 to making $50 million in 2020, is currently profitable.[8]

Proposition 22 results from this unprofitability: DNCs and TNCs are spending money to shield their inherently money-losing business models from government regulation that would further deteriorate cashflow. Even with Proposition 22 in effect, all five companies raised their rates in early 2021.[9] It is somewhat unclear why that was necessary — the companies claim that it was to pay for the benefits included in Proposition 22, while others argue their prices were already artificially low due to venture capital subsidies[10] — but the reality that faces these companies remains easily understood. Backed by record amounts of venture capital,[11] losing billions of dollars every year, fighting against paying state-mandated employee benefits while forbidding drivers from setting their own prices (as could real entrepreneurs and small business owners), DNCs and TNCs have stranded themselves both economically and politically. By focusing on rapid growth and undervaluing their drivers, they have cultivated business models that they cannot possibly maintain.

App-based drivers struggle to pay taxes

One recent study estimates that between insurance, purchase and leasing payments, tolls, licenses, permits, gas, and maintenance, it costs an uberX driver $8.85 per hour on average to operate a personal vehicle while working through the app.[12] Thus, a driver who earns $18.85 per hour is actually making $10 when these vehicle expenses are subtracted, well below the state’s $13 minimum wage. Then come taxes.

Calculating vehicle operating expenses for app-based drivers is emblematic of the financial obstacle course that self-employment entails, and in no exercise is this difficulty more acute than when paying taxes. The current tax regime is so complex that some experts speculate it may actually turn workers away from the gig economy.[13] For app-based drivers, this complexity likely leads to underreporting or overreporting. According to the Aspen Institute:

The complexities of tax filing for independent workers contribute to low rates of compliance and high costs for both workers and government. Lacking withholding and facing limited reporting requirements, independent workers are vastly less likely than traditional workers to accurately document their earnings for taxes. In fact, the IRS estimated that more than 40 percent of the total tax gap—the difference between taxes owed and taxes paid on time—from 2008 to 2010 was attributable to self-employed individuals underreporting their income. This underreporting equates to more than $190 billion in lost revenue per year. While many independent workers underreport their earnings, others overpay taxes by missing deductions and credits. Almost half of online-platform workers surveyed by the Kogod Tax Policy Center of American University reported being unaware of any deductions, expenses, and credits they were eligible for to lower their tax bills.[14]

Whether app-based drivers underreport (harming the state) or overreport (harming themselves), California is sure to emerge worse off than it would be if these workers were taxed as employees and did not have to self-report. For workers making around (and perhaps well below) minimum wage, hiring an accountant is simply not an option.[15] Without such professional guidance, the result is often faulty reporting.

Recent changes in federal law seek to partially address underreporting by independent contractors, but for California this is unlikely to make a big difference. The 2021 federal stimulus bill contains a provision (which goes into effect on January 1, 2022) aimed at closing a tax loophole that allowed many gig economy companies to avoid sending 1099-K tax forms to workers who earned less than $20,000.[16] The law provides that companies need not report information “with respect to third party network transactions of any participating payee if the amount which would otherwise be reported . . . does not exceed $600.”[17] By lowering the mandatory reporting threshold, the law may succeed in tightening down on underreporting by app-based drivers at the federal level.

Receiving a federal 1099-K form could inspire app-based drivers to pay their California state taxes more accurately, but the change is unlikely to have any drastic effects on the state for two reasons. One is that California only requires residents to file tax returns on gross income exceeding $18,496.[18] This amount is close to the $20,000 threshold, meaning that a relatively small portion of 1099-Ks being issued under the new law will need to be reported in California.[19] The other problem is that 1099-K forms are received annually, while independent contractors are generally required to pay estimated taxes quarterly. Consequently, underreporting by app-based drivers classified as independent contractors in California will almost certainly remain a problem even after the new 1099-K requirement goes into effect, regardless of what may be a considerable improvement at the federal level.

All told, classifying app-based drivers as independent contractors rather than employees severely impedes California’s ability to collect payroll taxes, which means less funding for the state overall. As of June 2020, the state’s Division of Labor Standards Enforcement estimated that worker misclassification “results in an estimated annual loss of $7 billion per year in payroll tax revenue to the state, that otherwise could have supported General Fund programs for public safety, education, and public infrastructure.”[20] That loss amounts to approximately 12.9% of California’s budget deficit in 2020–21.[21] This shows that the current system’s incentives for tax underpayments for DNCs and TNCs has potentially significant negative economic effects on the state budget.

The state will be stuck with the bill for unemployment insurance and workers’ compensation

While certain limited benefits are included in Proposition 22,[22] notably absent is any mention of unemployment insurance or workers’ compensation. If app-based drivers were hired as employees rather than as independent contractors, the companies employing them would be responsible for these costs — which are significant. The UC Berkeley Labor Center recently reported that Lyft and Uber alone would have been responsible for paying $413 million in unemployment insurance between 2014 and 2019 if their app-based drivers were employees.[23] Reuters similarly calculated that each full-time driver would cost these companies $3,140 annually in workers’ compensation insurance.[24] Against those higher annual costs, Proposition 22 as a one-time expenditure of over $200 million by its five biggest funders is a comparatively cheaper business strategy.[25] Indeed, Lyft and Uber have stated that they would drastically reduce the number of drivers working for them if forced to pay these benefits.[26]

As independent contractors under Proposition 22, app-based drivers themselves are responsible for paying these costs. A survey conducted between May and June of this year found that only 10% of app-based drivers were receiving Proposition 22’s healthcare stipend, and 16% (double the national rate) were uninsured.[27] Paying insurance premiums may be particularly difficult for these workers given that they are not entitled to a standard minimum wage, overtime pay, sick days, or paid family leave, and are barred from collective bargaining.[28] Without insurance, the financial burden will shift to the state to care for these workers through various social welfare programs. This means that ultimately other taxpayers will be responsible for their care.

Yet the costs of letting DNCs and TNCs forgo paying unemployment insurance and workers’ compensation are not borne solely by the state itself. One study estimated that law-abiding businesses pay an additional $2.54 billion in workers’ compensation and $831.4 million in unemployment insurance premiums annually due to employers misclassifying their employees as independent contractors.[29] So while the state will wind up subsidizing the bulk of app-based drivers’ unpaid unemployment and workers compensation costs, the financial burdens will also be dispersed across California’s economy.

Underreporting taxes may also translate into lower Social Security benefits.[30] The California Supreme Court implicitly considered such financial constraints when it disapproved systematic worker misclassification in Dynamex Operations W. v. Superior Court, which inspired AB 5:

[T]he minimum employment standards imposed by wage orders are also for the benefit of the public at large, because if the wage orders’ obligations are not fulfilled the public will often be left to assume responsibility for the ill effects to workers and their families resulting from substandard wages or unhealthy and unsafe working conditions.[31]

Until DNCs and TNCs prove that they can maintain sustainable business models, California has no long-term policy incentive to create sweeping exemptions for their entire industry. Delaying their failure only creates the illusion that they are good for California and entrenches them deeper in the state budget quandary.[32]

Imagining the world without Proposition 22

If Proposition 22 does not survive its legal challenges, and DNCs and TNCs are forced to comply with AB 5 once again, these companies will have two viable options to choose from. Either one would improve the companies’ sustainability and benefit California.

Option 1 is to admit that app-based drivers are employees and accept the responsibility and investment that comes along with that. The limited benefits included in Proposition 22 suggest that these companies already acknowledge this responsibility to some extent. Nothing (other than the DNCs and TNCs) prevents switching to a business model wherein app-based drivers are employees that set their own schedules and receive benefits proportional to work performed.[33] This would require the companies to provide all of the benefits that employees are entitled to in California — not just a select few. Such investment in workers would improve the DNCs’ and TNCs’ long-term business prospects by transforming them into some of the largest employers in the state. Members of those ranks receive considerable respect and are more durable for it.

Option 2 is to lean into the independent contractor theory, and allow drivers to set their own rates and negotiate with customers. Providing drivers with this autonomy would lend substance to (for example) TNCs’ assertion that they are software companies, not transportation companies, and should be regulated accordingly.[34] The ability to negotiate is a hallmark of self-employment and entrepreneurial capitalism. As David Weil, the former Administrator of the Wage and Hour Division of the U.S. Department of Labor, observed: “[c]lassification, really, is about protecting people in what is inevitably an unequal bargaining relationship: employment.”[35] Allowing drivers to bargain with end users would give them the negotiating parity characteristic of independent contractors. Uber briefly experimented with letting drivers set their own multipliers (meaning Uber was still setting the base rate), but quickly abandoned the policy when it turned out that the company could make more money by controlling rates and keeping them uniformly low.[36] Other gig economy companies like TaskRabbit proved that it’s possible to succeed while letting their workers negotiate with clients. True independent contractors should at least have the ability to bargain for their own pay.

Conclusion

Proposition 22 is currently on appeal, and this article takes no position on whether the trial court ruling should be upheld. Regardless how that proceeding resolves, policy ideas like Proposition 22 are not beneficial to California’s economy because they remove more workers from the comprehensive employee regulation scheme. Creating incentives or legal requirements for workers to be classified as independent contractors has several negative individual and macro effects: individuals will be harmed by liability for underpaying their taxes; workers will be exempted from the array of protections developed for their benefit over the past century; and the public treasury will be harmed from lost income and broader economic impacts. The better policy here is to classify app-based drivers as employees with flexible schedules and proportional benefits, or to classify them as independent contractors but allow them to negotiate their own rates. A little innovation here would go a long way.

—o0o—

Harry Libarle is an editor at the Hastings Law Journal.

  1. Castellanos v. State of California (2021) (RG21088725, app. pending) at 11–12. ↑

  2. Uber Technologies, Inc. v. State of California (2020) (CGC-20-584402) at 32–33. ↑

  3. Siddiqui, Uber and Lyft Don’t Want to Make California Drivers Employees, So They’re on the Verge of Shutting Down, Washington Post (Aug. 19, 2020). ↑

  4. Curry, Postmates Revenue and Usage Statistics (2021), Business of Apps. ↑

  5. Iqbal, Uber Revenue and Usage Statistics (2021), Business of Apps. ↑

  6. Iqbal, Lyft Revenue and Usage Statistics (2021), Business of Apps. ↑

  7. Curry, DoorDash Revenue and Usage Statistics (2021), Business of Apps. ↑

  8. Curry, Instacart Revenue and Usage Statistics (2021), Business of Apps. ↑

  9. Sandler, Every Major Gig Company Has Now Raised Prices in California After Proposition 22, Forbes (Feb. 19, 2021). ↑

  10. Ibid. ↑

  11. See, e.g., Sherman, Why Can’t Uber Make Money?, Forbes (Dec. 14, 2017). ↑

  12. Helling, How Much Does It Really Cost to Drive Your Car for Uber and Lyft?, Ridester (Aug. 27, 2021). ↑

  13. Watson, Improving the Federal Tax System for Gig Economy Participants, Tax Foundation (Oct. 18, 2019). ↑

  14. Fitzpayne et al., Tax Simplification for Independent Workers, Aspen Institute (Sept. 2018) at 1–2. ↑

  15. See Reich, Pay, Passengers and Profits: Effects of Employee Status for California TNC Drivers, IRLE Working Paper No. 107-20, (Oct. 2020) at 2 (estimating that most California TNC drivers are paid below minimum wage); Jacobs & Reich, The Effects of Proposition 22 on Driver Earnings: Response to a Lyft-Funded Report by Dr. Christopher Thornberg, Institute for Research on Labor and Employment, U.C. Berkeley Institute for Research on Labor and Employment (Aug. 26, 2020) at 1 (concluding that “should Proposition 22 pass, Uber and Lyft drivers could earn as little as $5.64 an hour, well below California’s minimum wage.”). ↑

  16. U.S. Treas. Inspector Gen. for Tax Admin., Expansion of the Gig Economy Warrants Focus on Improving Self-Employment Tax Compliance (Feb. 14, 2019) at 31–36. ↑

  17. American Rescue Plan Act of 2021, Pub.L. No. 117–2 (Mar. 11, 2021) 135 Stat 4, § 9674(a). ↑

  18. Cal. Franchise Tax Bd., Residents (accessed Oct. 2, 2021). ↑

  19. It is possible that some app-based drivers who work for multiple gig economy companies will now receive more than one Form 1099-K, which in aggregate could push a driver’s gross income over California’s minimum reporting threshold. Given however that: A) many drivers falling into such a category would previously have been required to file a tax return in California even if they did not receive a Form 1099-K; and B) enough deductions claimed by these drivers would obviate the need to file a return entirely; any such analysis would be highly speculative. ↑

  20. Lon. Dec., Ex. 75, Assem. Com. on Labor & Employment, Background Information of Assem. Bill No. 5 (2018-2019 Reg. Sess.) at 3. ↑

  21. There was a $54.3 billion gap in 2020-21. California State Budget – 2020-2021 at 1. ↑

  22. The benefits include a guarantee of 120% the state’s minimum wage for “engaged time” (the time from when a delivery or rideshare request is accepted to the time when it is complete); a modest healthcare stipend (also tied to a driver’s engaged time); additional compensation of thirty cents per engaged mile driven; and limited occupational accident insurance. Bus. & Prof. Code § 7449(f). ↑

  23. Jacobs & Reich, What Would Uber and Lyft Owe to the State Unemployment Insurance Fund?, U.C. Berkeley Labor Center (May 7, 2020). ↑

  24. Bellon, Uber, Lyft Spend Big in California to Oppose Even Costlier Gig-Worker Law, Reuters (Oct. 5, 2020). ↑

  25. Ballotpedia, California Proposition 22, App-Based Drivers as Contractors and Labor Policies Initiative (2020) (accessed Sept. 6, 2021). Proposition 22 was the most expensive ballot initiative in California history by over $60 million. Ballotpedia, What Were the Most Expensive Ballot Measures in California? (accessed Sept. 6, 2021). ↑

  26. Bellon, Uber, Lyft Spend Big in California to Oppose Even Costlier Gig-Worker Law, Reuters (Oct. 5, 2020). ↑

  27. McCullough, Most California Rideshare Drivers Are Not Receiving Health-Care Benefits Under Proposition 22, PolicyLink (Aug. 19, 2021). ↑

  28. Bus. & Prof. Code § 7465(c)(4). In particular the restrictions on collective advocacy corrode political process by systematically carving out workers. See Dubal, The Drive to Precarity: A Political History of Work, Regulation, & Labor Advocacy in San Francisco’s Taxi & Uber Economies (2017) 38 Berkeley J. Emp. & Lab. L. 73, 135; see also Castellanos at 11. ↑

  29. National Employment Law Project, The On-Demand Economy & State Labor Protections (Jan. 2017) at 3. ↑

  30. Bruckner & Hungerford, Failure to Contribute: An Estimate of the Consequences of Non- and Underpayment of Self-Employment Taxes by Independent Contractors and On-Demand Workers on Social Security (2021) 9 Ind. J. L. & Soc. Equality 52, 54. By one estimate, on-demand platform workers underpaid Social Security contributions by $2.03 billion in 2014. Id. at 77. ↑

  31. Dynamex Operations W. v. Superior Court (2018) at 953. ↑

  32. See Legis. Analyst, California’s Fiscal Outlook (Nov. 18, 2020) (noting that operating deficits are expected to grow in the coming years). ↑

  33. Proposition 22 states that “recent legislation has threatened to take away the flexible work opportunities of hundreds of thousands of Californians, potentially forcing them into set shifts and mandatory hours, taking away their ability to make their own decisions about the jobs they take and the hours they work.” Bus. & Prof. Code § 7449(d). AB 5 does no such thing. In fact, the legislation expressly provides that “[n]othing in this act is intended to diminish the flexibility of employees to work part-time or intermittent schedules or to work for multiple employers.” Assem. Bill No. 5 (2019-2020 Reg. Sess.) § 1(g). ↑

  34. Dubal at 126. ↑

  35. Weil, Lots of Employees Get Misclassified as Contractors. Here’s Why It Matters, Harvard Business Review (July 5, 2017). ↑

  36. Uber, Upcoming Changes to the Driver App (May 18, 2021). ↑