Research

Bureaucratic Politics in Targeted Trade Adjustment (Working Paper) (Link)

Recognizing that trade can displace domestic workers and firms, governments have adopted a number of social insurance schemes. In the United States, eligibility for these benefits is made dependent on evidence that an individual worker was harmed by international trade. In this paper, I argue that the bureaucrats who examine and interpret this evidence have career incentives to distort decisions toward the partisan affiliation of the president. To identify this effect, I leverage variation in the 42 years of the Trade Adjustment Assistance (TAA) individual petition-level data (81,453 petitions) and career paths of the Office of Trade Adjustment Assistance (OTAA) investigators. I find evidence that the OTAA investigators are less likely to certify the TAA petitions, and more likely to delay the TAA investigations during the Republican presidencies. These partisan effects are detected when the bureaucrat is likely to face the largest career pressure, prior to obtaining a permanent appointment. Their partisan behavior evaporates after three years when appointments are no longer in question. These findings demonstrate how efforts to target compensation following trade liberalization can systematically politicize policy implementation.

Importing Threat: The Electoral Logic of Economic Relief (Working Paper, co-authored with Robert Gulotty) (Link)

Commercial policies are often only efficient insofar as those harmed can be compensated. In practice, compensatory measures fall far short of distributive harm. We rationalize the paucity of compensation as a strategic effort on the part of elected officials to withhold information about effects of their policy initiatives. We develop a formal model in which citizens must infer the effects of a policy initiative as well as the politician’s commitments from the choice to offer compensation. We find that committed policymakers under-provide compensation to avoid electoral backlash. Using microdata from the US Trade Adjustment Assistance program, we replicate a research design that uses exogenous allocation of petitions for assistance across bureaucrats for causal identification. We find that a ten percent increase in the TAA certification rate decreases support for Democratic candidates by 1.75% in areas hard hit by import-competition. This electoral effect incentivizes pro-distributive politicians to under-provide economic assistance.