Skill-biased remote work and incentives

We document four key trends since the pandemic - a surge in remote work, an increase in performance pay, their joint occurrence, and the skill-biased nature of this complementarity. We develop a firm-worker model that explains this evidence. We show that, under risk aversion, the incentive-compatible performance pay premium falls with worker's skills, as the likelihood of a good performance increases. Hence, the firm uses performance pay if the worker is sufficiently skilled and fixed pay with monitoring, otherwise. The unforeseen pandemic shock forces the firm to adopt remote work and reduces monitoring effectiveness. As a result, the firm relies more on performance pay. Post-pandemic, the firm always sticks to the remote work if the worker is sufficiently skilled. If the worker is too unskilled for performance pay to be cost-effective, the firm sticks to remote work only if remote monitoring is effective. Accordingly, the model predicts that a decline in remote monitoring efficacy could reduce remote work for less-skilled workers only. To test this, we exploit temporal variation in legislation in New York State, using a Difference-in-Differences approach, to estimate the impact of stricter regulations of remote monitoring on the adoption of remote work. Such a test strongly supports our model's predictions. Our findings suggest that pandemic policies and regulations may have played a significant role in shaping the adoption and persistence of remote work and performance pay.