Department of Economics

Document Type



This dissertation consists of two chapters on the economics of judicial decision-making. In chapter 2, using the data from U.S. immigration courts and the Global Terrorism Dataset, I investigate the effect of terrorist attacks on asylum decisions. I find that terrorist attacks within the courts' states reduce the likelihood of granting asylums by 1.9 percent. This effect is driven by Islamic, Right-wing, and Lone-wolf terrorist attacks, and the magnitude of the effect is largest in the case of Islamic terrorist attacks (3.6 %) and smallest in the case of Lone-wolf attacks (1.9 %). Left-wing terrorist attacks have no impact on asylum decisions. Applicants from Latin America and China are the groups bearing the burden of the effects, and there is no evidence indicating that applicants from Muslim majority countries are adversely affected. Analyzing the news coverage of incidents reveals a spill-over of effects on all immigration judges in the U.S. when national news channels cover the incidents. On the other hand, incidents covered by national news do not have a differential impact on the judges within the state of terrorist attacks, which suggests that information dissemination is the channel for spill-overs.

Chapter 3 examines the effect of the appellate board’s error correction on asylum decisions by focusing on instances in which the board remands a case back to the judges (for a new round of hearings) due to an erroneous decision in the first round. Using the dates that judges have hearings for cases with error corrections and the date they render a decision for other cases, I show that having hearings for cases remanded in favor of prosecutors (DHS’s attorney) reduced the judges’ leniency towards immigrants by 2.8 percent. To understand the underlying channel through which the asylum decisions are affected, I focus on the appeal rates of the losing party. This analysis reveals that granted (rejected) asylum applications that are decided during exposure to error corrections are 25 % (3.5 %) less likely to be appealed by prosecutors (immigrants), which suggests that increased accuracy in the decisions is a potential mechanism through which receiving error corrections affect asylum decisions.



Committee Chair

Mocan, H Naci