Corporate Tax Incentives Retraction: Evidence from Puerto Rico

• In Event: Economic Development Incentives and Local Employment: New Evidence on Subsidies, Spillovers, and Phaseouts

Abstract

Background: Place-based corporate tax incentives (i.e. federal Opportunity Zones, state enterprise zones, local abatements) are a staple of U.S. regional development policy. A large literature documents their effects at implementation, but remarkably little is known about what happens when they are withdrawn: whether the economic activity they attracted persists once the subsidy ends, or whether repeal leaves places worse off. This literature gap is increasingly consequential. Policymakers routinely confront the choice between extending costly incentives and letting them expire, yet they do so without credible estimates of the long-run consequences of retraction. Objective: This paper fills that gap by asking whether corporate tax subsidies produce labor market benefits once the policy ends. Methods: We study the phaseout (1996) and full elimination (2006) of Internal Revenue Code Section 936, a federal tax credit that effectively exempted U.S. multinationals from taxation on income earned in Puerto Rico. For nearly four decades, Section 936 anchored the island's industrial policy and sustained manufacturing as its largest private-sector employer; its repeal represents one of the largest retractions of a place-based tax incentive in U.S. history. Our primary data are a state-level panel drawn from the Quarterly Census of Employment and Wages (QCEW). We estimate the effects of phaseout and elimination using the synthetic difference-in-differences (SDiD) estimator of Arkhangelsky et al. (2021), constructing a weighted donor pool of U.S. states to recover the counterfactual trajectory of outcomes in Puerto Rico absent repeal. Results: The 1996 phaseout generated an islandwide reduction of approximately 9.2 percent in manufacturing employment by 2000 relative to the synthetic counterfactual, and the 2006 elimination deepened these losses to roughly 28 percent by 2019. Aggregate employment and the working-age population declined in parallel, indicating that labor market losses were accompanied by a broader demographic contraction rather than reallocation across sectors. Conclusions and implications: These findings challenge the implicit premise that place-based corporate tax incentives seed self-sustaining local economies. Nearly four decades of federal subsidy did not leave behind a manufacturing base capable of surviving its removal; instead, the repeal triggered cumulative labor market and demographic decline. For policymakers weighing the design, extension, or sunset of similar incentives at the federal, state, and territorial level, our results highlight the importance of implementing transition strategies when large place-based programs expire.

Author

• Dayanara M Diaz Vargas, Northeastern University
  Presenting Author