What's in a Name? the Effect of Changing Definitions of “employee” on Worker Outcomes

Abstract: This paper provides causal evidence on how the definition of employment – whether a worker can legally be designated as an employee or an independent contractor – affects labor market outcomes. We exploit the random assignment of U.S. Circuit Court Judges to specific cases to obtain exogenous variation in employment definitions over time, across states, and across occupations. More “employee” definitions (relative to "contractor" definitions) increase wages for workers in non-manual jobs by 10%, with no effect for workers in manual jobs. We also find evidence that wages increase more for workers in states without right-to-work laws, suggesting there is variation in effect by preexisting legal rules. We also find that more “employee” definitions increases unionization rates by 2 percentage points, again with larger effects in states without right-to-work laws. There is no effect on either aggregate employment or contracting rates, suggesting that the wage gains are not driven by compositional changes in the workforce. These results highlight the importance of legal definitions of employment in the operation of the labor market.