The discussion between the panelists highlighted the complexity of antitrust issues related to digital platforms for gig work, including:
- The impact of different factors—such as local demand, workers’ preferences for non-monetary benefits, multihoming, and available alternatives for work—on competition for labor among these platforms;
- Challenges presented by factors such as the diversity of gig workers and the frequent evolution of laws related to gig work in assessing and mitigating the extent of anticompetitive issues in these markets; and
- Important questions at the intersection of antitrust and labor law—including gig workers’ rights and classification.
Overview Of Digital Platforms
Dr. Caminade first defined digital platforms as businesses that use technology to connect two or more different types of users to facilitate the exchange of goods and services. She provided examples of digital platforms involving a wide variety of workers and types of work, such as connecting clients with architects and designers, patients with therapists, and passengers with drivers.
She then listed some characteristics that workers on digital labor platforms share, including that they:
- Can use these platforms flexibly, choosing how much and when to work;
- May work on multiple platforms simultaneously (i.e., multihome);
- Face little to no sign-up costs to use these platforms;
- Inherently assume the risk of demand being low at any given time that they choose to work; and
- Cannot be forced to sell their work on a platform: platforms need to incentivize workers to work at any given time or location.
State of the Law and Public Policy
Ms. Brass followed Dr. Caminade by summarizing potential antitrust concerns related to these types of platforms and the current state of the laws and public policies that address these issues. She explained that under a 1941 Supreme Court ruling, labor disputes between employers and employees are exempt from the interdiction of collusive conduct by the Sherman Act. Whether gig workers are considered employees or independent contractors could be a key determinant in their ability to collectively bargain. However, Ms. Brass highlighted that this distinction has recently been successfully challenged, setting precedent for independent contractors to be granted labor exemptions from collusive conduct prohibited by the Sherman Act.
Ms. Brass also noted that whether gig workers should be considered employees or independent contractors has been heavily litigated. She commented that courts have recently tended to find that it was proper to classify gig workers as independent contractors, but that there have been exceptions depending on the industry or workers involved. Ms. Brass acknowledged the concern that some gig workers may be “misclassified” as independent contractors and therefore may be excluded from protections that they are entitled to as employees, such as minimum wage and overtime compensation. However, Ms. Brass noted that determining the appropriate classification for a group of workers is a highly fact-intensive inquiry that depends on the nature of work performed at a specific place of work and statutory requirements that vary by state, by job, and over time. Additionally, federal standards and the guidelines issued by the National Labor Relations Board have changed over time and may change again.
Ms. Brass then detailed how recent California legislation illustrates the “web of evolving standards” that determine whether gig workers are independent contractors or employees. She explained that in Dynamex Operations West Inc. v. Superior Court (2018), California laid out a new standard under its Industrial Welfare Commission (IWC) Wage Orders called the “ABC Test” to determine whether a worker is an independent contractor or an employee. Under this test, a worker is presumed to be an employee and is only considered an independent contractor if three strict conditions are met. In 2019, California enacted Assembly Bill 5 (“AB 5”) which codified the ABC Test and clarified that it applied to a much broader set of workers while carving out approximately 60 exemptions. A year later, AB 2257 increased the number of exemptions to 109. In November 2020, California voters passed “Prop 22,” the Protect App-Based Drivers and Services Act, which exempted app-based workers from AB 5, and lifted the presumption that they were employees.
Economic Tools to Assess Anticompetitive Concerns in Labor Markets
Dr. Marinescu then provided insight into how economists view these issues and introduced some of the empirical tools used to assess anticompetitive behavior in labor markets. She explained that one measure of competition is the ability of a firm to enforce and sustain a significant wage decrease for its workers and identified two characteristics of labor markets that would affect firms’ ability to decrease wages: labor supply elasticity and network effects of platforms. In a perfectly competitive market, workers’ labor supply for a given firm is infinitely elastic, implying that the smallest decrease in wages paid by a firm would result in all workers leaving. As a result, in equilibrium, workers are paid exactly the amount they contribute to the firm (or their “marginal product”). The less elastic the labor supply, the more likely it is that a firm can pay its workers less than they contribute without losing all its workers. Additionally, Dr. Marinescu explained that when a platform aims to bring together two sides of a market, network effects can amplify antitrust concerns. In the case of labor platforms, network effects imply that both sides of the market benefit from being on a platform featuring lots of workers and customers. Therefore, network effects are inherently advantageous to larger platforms and could introduce competition challenges for smaller or newer platforms.
However, Drs. Marinescu and Caminade noted that the competitive concerns raised by these network effects could be partly mitigated by multihoming. Multihoming results in a higher labor supply elasticity and more competition among platforms for workers. The extent to which multihoming boosts competition depends on how easily workers can multihome.
Dr. Marinescu then turned to the empirical methods used to assess labor supply elasticity for gig workers, which varies widely due to the different ways workers use digital platforms. For example, a full-time Uber driver would make different labor supply decisions and tradeoffs from a part-time driver who has another primary job. Dr. Marinescu highlighted empirical studies using data from gig workers that find that their labor supply elasticity is low, allowing platforms to pay workers significantly less than their marginal productivity. She explained that there was a general lack of data tracking workers across platforms and employers, which would provide a better understanding of workers’ alternatives to gig work.
Citing a 2021 Pew Research Center study, Ms. Brass then pointed out that most workers take up gig work for reasons other than the lack alternative job opportunities. The study found that gig work is not a primary source of income for many of the workers engaged in it, and they value gig work for a variety of reasons that may go beyond simply earning extra money. These findings, Ms. Brass argued, imply that non-monetary factors affect labor supply elasticity for gig work and should be considered when evaluating issues of labor market competition in such jobs.
Potential Competitive Implications of Classifying Gig Workers as Independent Contractors
The conversation then turned to potential competitive implications of classifying gig workers as independent contractors, which may limit their ability to collectively bargain with platforms. Collective bargaining can lead to an increase in workers’ wages when firms have monopsony power in the labor market.
However, Ms. Brass pointed out that more than 60% of gig workers in the Pew study considered themselves to be independent contractors. It is also not obvious that classification of workers as independent contractors leads to lower total compensation for workers. Dr. Caminade noted that gig work provides valuable benefits to workers, such as flexibility. Because workers can work when they choose, labor platforms often need to provide additional financial incentives to work at particular locations or times (e.g., surge pricing in ridesharing). She also highlighted that digital labor platforms may increase competition in other ways, including by providing workers with options that make it particularly easy for them to sell their labor. This could lead to increased labor market competition and a shift away from a small number of traditional firms (or a single firm) holding monopsony power.
These business models are attractive to workers, and Ms. Brass referred again to the Pew study which found that there are several non-monetary reasons for workers to value gig work and that not all workers have the same preferences. Dr. Caminade added that gig work offers those who value factors such as flexibility, control over their own schedule and the ability to choose jobs in particular locations or pursue opportunities that may be unavailable through traditional employment. For gig workers, these factors may outweigh some of the benefits from employment.
Practical Implications of Classifying Gig Workers as Employees
The panelists then discussed the practical implications of current policies surrounding gig work. Ms. Brass pointed out two concerns with enforcing the FTC’s policy statement. First, there is no legal presumption at the federal level that gig workers are misclassified, and each state has its own laws regarding classification and the right to collectively bargain. Second, it is unclear whether the FTC’s statutory authority extends to workers. These questions are likely to become highly litigated.
Dr. Caminade noted that classifying workers as employees could have the unintended consequence of limiting options for workers and reducing competition in the labor market. Employment would lead to higher onboarding and fixed costs (e.g., HR and regulatory costs). Firms would respond by offering fewer work opportunities or by shifting away from an on-demand staffing model, reducing the flexibility and types of opportunities available to workers. Also, additional fixed costs could make it harder for new platforms to enter the market and compete for workers, in turn providing fewer opportunities for workers to multihome. Dr. Marinescu countered that there is no strong empirical evidence that employment would decrease because of the costs associated with increased employee benefits, and that the literature on minimum wage has documented that increases in labor costs, such as wages, could increase employment.
Can Unionization Mitigate Antitrust Concerns with Gig Work?
Mr. Hasdoo then asked the panelists about the implications of allowing workers to unionize, and whether unions could be the answer to concerns about potential power imbalance between gig workers and digital labor platforms.
Dr. Marinescu highlighted research documenting that unionization tends to lessen the adverse effects of labor market concentration on wages, outside of the gig work context. The question then follows of whether gig economy labor markets are concentrated. Dr. Caminade warned that concentration in the product market is not necessarily coincident with concentration on the labor market. She illustrated her point using the (hypothetical) example of a carless consumer who can choose only between a small number of platforms when looking for a ride. In that case, the product market could be considered concentrated. However, a concentrated product market does not imply concentration on the labor side: drivers may have more options, such as working for taxi companies or for a delivery service.
Ms. Brass noted that allowing more groups of workers to collectively bargain creates exceptions to competition and a slippery slope towards further exceptions. She warned against the inclination to respond to concerns about potential imbalance in market power by expanding exceptions, stating that changing the rules could allow for collusion, instead of strengthening competition. She also cautioned against making micro-industry level exceptions in rapidly changing technology and labor markets.
