AB 5 and Uber’s Response to the Gig-Economy Reclassification

When Assembly Bill 5 took effect on January 1, 2020, California reset the legal default for millions of workers. The statute codified a worker-friendly test for distinguishing employees from independent contractors, and it placed the burden of proof on the hiring entity rather than the worker. The companies most exposed to that shift — the ride-hailing and delivery platforms that built their economics on contractor labor — did not quietly comply. Instead, Uber announced that it would keep classifying drivers as independent contractors, argued that its core business was technology rather than transportation, and helped bankroll a ballot measure designed to carve its industry out of the law entirely. That sequence raises a question worth examining closely: what does a company owe a statute it has decided, in advance, not to follow?

What AB 5 actually changed

AB 5 did not invent its standard from scratch. It codified the so-called ABC test that the California Supreme Court had adopted two years earlier in Dynamex Operations West, Inc. v. Superior Court (2018) 4 Cal.5th 903, and extended that test beyond the wage-order context in which the court had first applied it. Under the ABC test, a worker is presumed to be an employee unless the hiring entity proves all three of the following: (A) the worker is free from the entity’s control and direction in performing the work; (B) the work falls outside the usual course of the hiring entity’s business; and (C) the worker is customarily engaged in an independently established trade or business of the same nature as the work performed.

The bill, signed by Governor Gavin Newsom on September 18, 2019, was codified at Labor Code section 2750.3.

The ABC test’s burden-shifting structure is what made compliance, not just argument, the live question. Prong B is the one that mattered for the platforms. It asks whether the worker’s labor is part of what the company sells. A bakery that hires an outside plumber clears prong B easily; a bakery that hires bakers does not. For a service that exists to move passengers, the question of whether driving falls “outside the usual course” of the business is not a close call to most observers — which is precisely why the response to AB 5 turned on contesting that premise.

The test AB 5 replaced

Before Dynamex, the governing standard came from S.G. Borello & Sons, Inc. v. Department of Industrial Relations (1989) 48 Cal.3d 341. Borello weighed a long list of factors, the most important being the hiring entity’s right to control the manner and means of the work, alongside secondary considerations such as who supplied the tools, the method of payment, and whether the work was part of the regular business. That multifactor inquiry gave companies room to argue. A platform could point to the flexibility drivers enjoy — setting their own hours, using their own vehicles, declining trips — and frame those features as marks of genuine independence.

The ABC test is harder to game. Flexibility addresses prong A, but a company can lose on prong B regardless of how much autonomy its workers enjoy. By design, the test narrows the room for argument and shifts the burden onto the company. AB 5 thus did more than restate prior law; it changed who has to prove what, and it did so in a way that the most affected industry could not satisfy without rethinking its labor model.

Uber’s response: redefining the business

Rather than reclassify, Uber maintained that drivers remained properly categorized as independent contractors and signaled it would defend that position in litigation and in operations. The central argument was definitional: the company described itself as a technology platform that connects riders with drivers, contending that transportation is the drivers’ business, not Uber’s, so driving falls outside the company’s “usual course” under prong B. That framing is not new — platforms had advanced versions of it for years — but AB 5 raised the stakes by making prong B dispositive.

The more consequential part of the response unfolded outside the courtroom. Uber, together with Lyft, DoorDash, and other delivery platforms, helped fund a 2020 ballot initiative — Proposition 22 — written to exempt app-based ride-hailing and delivery drivers from AB 5 and to substitute a lighter package of platform-defined benefits. The companies committed substantial sums to the campaign, which became one of the most expensive ballot-measure efforts in California history. The strategy was candid: if the legislature would not exempt the industry, the industry would ask voters to do it directly.

What is troubling, and what is ordinary

Some of the reaction to AB 5 was unremarkable. Litigating an unfavorable statute and lobbying to change it are routine, legitimate uses of the legal and political system, and many businesses do both. The more difficult questions concern the conduct that sat between compliance and advocacy. Announcing in advance that the company would not reclassify, while a sweeping new default was already in force, effectively asked workers and regulators to treat the platform’s preferred reading as settled before any court had ruled. And funding a ballot measure to overturn a law one is simultaneously declining to follow blurs the line between contesting a rule and self-exempting from it.

For employment law generally, the episode is a study in how a presumption operates only as well as it is enforced. AB 5 made employee status the default and put the burden on the company. But a default is only a starting point; it still requires agencies to investigate, plaintiffs to sue, and courts to rule. When the most affected firms can absorb litigation costs and finance a statewide campaign, the practical question is not only what the law says but who has the resources to make it bite — a dynamic that also shapes adjacent areas of labor and employment doctrine.

How the dispute looks with hindsight

The fight did not end with AB 5’s effective date. Proposition 22 passed in November 2020, exempting app-based drivers from the ABC test and replacing employee status with a narrower set of contractual guarantees. The measure then drew constitutional challenges of its own, and its validity moved through the California courts over the years that followed before the ride-hailing classification dispute reached the state’s highest court. The throughline is that a single statute did not settle the matter; it opened a multi-front contest across the legislature, the ballot box, and the judiciary, each forum producing its own provisional answer.

For readers tracking how classification rules ripple into other rights, the gig-economy debate connects to broader questions about the boundary between work and personal life — questions this publication has examined in the context of employee privacy while working from home. The common thread is that the label attached to a worker — employee or contractor — cascades into wages, benefits, liability, and protections that workers may not realize turn on a definition fought over in a single prong of a single test.

This publication offers commentary and analysis for general informational purposes and does not provide legal advice; readers with specific questions about worker classification should consult qualified counsel.