Skip to content

Law and Politics in Employee Classification


Benjamin Sachs (@bsachs) is the Kestnbaum Professor of Labor and Industry at Harvard Law School.

As has been widely reported, the U.S. Department of Labor issued an “opinion letter” yesterday concluding that an unnamed “virtual marketplace company” does not employ the workers who make the company viable. Instead, the letter finds that these workers are independent contractors. The letter is flawed in multiple ways. As Sharon will explain, deciding a major issue of employment law – maybe the major contemporary issue of employment law – through an informal process that allows one party to present all the facts is decidedly inappropriate. There are also multiple substantive problems: as Charlotte pointed out, the letter considers relevant to the control inquiry the fact that this VMC’s workers can also work for other VMCs. I suppose the fact that Wal-Mart workers can also work for Target suggests that Wal-Mart workers are independent contractors of Wal-Mart. Generalizing, I suppose if low wage workers must rely on multiple jobs to make ends meet this should incline decisionmakers to conclude that those workers are all independent contractors.

But there’s another observation worth making about this letter. The letter concludesthat to determine whether a worker is an employee or an independent contractor under the Fair Labor Standards Act the decsionmaker should use an “economic dependence” test. And to determine economic dependence, the decisionmaker should use a six-factor inquiry cobbled together from Supreme Court case law. That’s a lot of factors. But, in fact, the six factors are just a starting point. Because, in one of the best sentences in the opinion letter, the Department concludes that “[o]ther factors may also be relevant, and the appropriate weight to give to each factor depends on the facts.”

Just say that out loud to yourself.

A legal test with six factors, plus an unidentified list of other factors that “may” be relevant and which is to be applied in an unspecified fact dependent way, is an indeterminate test. By this I mean simply that the test does not – cannot – determine the outcome of any particular case. In fact, this test has become so indeterminate – so chock full of factors and subfactors and “other [unnamed] factors” – that it does not even appear to constrain decisionmakers. The result is that the determination of who is an independent contractor and who is an employee – the primary legal question at the heart of debates over the gig economy and much of the labor market more broadly – is becoming an entirely political question. This opinion letter ought to be understood that way: a political determination, made by the Trump Department of Labor, to treat gig workers as independent contractors. The decision is dressed up as a legal opinion – it walks through an analysis of each factor and cites lots of cases – but this is a political document and a political determination.

If I am right, the obvious next question is whether there’s an alternative. No legal test can function to keep politics out of legal decisionmaking. But some tests are more constraining than others – they function better to ensure that legal outcomes flow from the test rather than from the political orientation of the legal decisionmaker. If we favor this kind of constraint, I think there’s a solid argument for the ABC test over the “economic dependence” or “economic realities” tests.

For one thing, the ABC test has three and only three factors. This, in itself, is an advantage. Moreover, in order to qualify as an independent contractor, all three factors must be satisfied. So there’s no need to weigh factors, and certainly no need to say that “the appropriate weight to give to each factor depends on the facts.” To the contrary, under ABC, if a single factor is not satisfied, the decisionmaker is obligated to determine that the worker is an employee. And, finally, the second two factors in the ABC test – either of which is outcome determinative – are more concrete than any of the factors in the economic realities or economic dependence tests. These second two factors are:

(B) that the worker performs work that is outside the usual course of the hiring entity’s business; and

(C) that the worker is customarily engaged in an independently established trade, occupation, or business of the same nature as the work performed for the hiring entity.

To be sure, there are things to argue about in assessing these factors, including about the nature of the hiring entity’s business. But, as these things go, asking whether (1) someone is doing work that is the kind of work the firm usually does and (2) whether that person has an independent business in the same field as the work she does for firm, are questions with relatively ascertainable answers.

Finally, the choice of which legal test to use is, in itself, a political question. Hence the current debate in California. And, as I have myself written, the ABC test would seem to imply that Uber drivers, for example, are employees. But we can evaluate the merits of a legal test in multiple ways. One is the outcomes it likely will produce in cases that we can predict. Another is whether the test actually functions to constrain legal decisionmakers such that decisions flow from the test (that is to say, from the law) rather than from the policy preferences of the decisionmakers. On the constraint side, the Department of Labor’s VMC opinion letter makes it plain that the multi-factor economic dependence and economic realities tests enable decisionmakers to enact their preferences. ABC is much more promising on this front.

This piece is cross-posted from On Labor.