UC Berkeley

This dissertation studies behavioral responses of workers and businesses to tax and transfer policies. It examines how these policies affect the incentives of economic agents, whether they foment rent-seeking behavior, and how frictions might limit the set of choices. To address these questions, it exploits plausibly exogenous policy changes using quasi-experimental techniques from labor and public economics as well as large administrative datasets.

The proper taxation of high earners is front and center in the policy debate. While higher progressive income taxes may mechanically reduce income disparities, critics argue that they can backfire by lowering work effort and hours. The first chapter of my dissertation studies intertemporal labor responses of high-wage earners to income tax changes. This chapter is co-authored with Guillermo Cruces and Victoria Castillo. We exploit a large and quasi-randomized income tax holiday in Argentina that exempted a group of high-wage earners from the income tax for 2.5 years and left another comparable group of workers taxed. We provide one of the cleanest evidence to date that the labor supply of high-wage earners responds very little to temporary income tax changes. We report a precise and very small wage-earnings elasticity of 0.02. Responses are larger for more flexible outcomes (overtime hours) and more elastic groups (job switchers and managers). We also find avoidance responses from new entrants who faced no tax if their first monthly wage was below a fixed threshold. Our evidence suggests that low responses might be driven by labor demand constraints (eg. fixed contractual hours) and labor market rigidities, and that employer-employee cooperation is required for wage earners to be able to respond to tax changes.

Most countries also provide some sort of financial aid at the low end of the income distribution (e.g., the EITC in the U.S.). An understudied question in Economics is whether employers capture part of those transfers by lowering wages. The second chapter, co-authored with Santiago Garriga, studies whether the way family allowances (tax credits) are disbursed affects the wage of workers. We exploit an unusual reform in Argentina that was gradually rolled out and shifted the disbursement responsibility from firms to the government, reducing the saliency to employers. Our event study estimates show that employers capture about 10-20 percent of the transfer when they mediate its disbursement. In terms of the mechanisms, the increase in monthly wages after the event is more consistent with a labor demand story rather than pay equity concerns. For example, the effect is explained by new hires rather than incumbent workers. Our evidence suggests that wages do adjust to the way transfers are disbursed, rejecting the null hypothesis that transfers are captured dollar for dollar by workers.

In addition to income taxes, many countries tax small and large businesses under different regimes. The third chapter, co-authored with Santiago Garriga and Jorge Puig, estimates the response of self-employed and firms to two revenue taxes---monotributo and the gross receipts tax---across the revenue distribution. We exploit several revenue-dependent discontinuities (notches) in Argentina that provide incentives to underreport sales. For self-employed workers and small firms, we find sizeable responses that are stronger for higher tax incentives and in sectors with more space for manipulation such as service-based activities. In the case of medium and large firms, bunching is less striking but it suggests that even large firms are able to underreport their gross sales to avoid facing higher tax rates. Firms also seem to find more costly the indirect administrative cost of becoming a collection agent than the direct fiscal cost of the gross receipts tax. Our results also suggest that some entities face substantial adjustment frictions or inattention to tax discontinuities. We cannot rule out, however, that large firms adjust other margins to compensate for the higher tax pressure.