Working Papers
"Natural Disaster, Federal Assistance, and Local Public Finance" (Job Market Paper)
Abstract: Federal disaster aid is the primary funding source for local recovery, yet little is known about how it affects local governments after a disaster. Using a fuzzy regression discontinuity design with novel administrative disaster-damage data from six states, I estimate the effect of FEMA's Public Assistance program on municipal public finances. Public Assistance raises per-capita intergovernmental revenue by about 6.5 percent, largely through the state rather than the federal government. The transfer expands the budget rather than substituting for local effort: total expenditure rises by about 16 percent while own-source revenue does not move. The transfer arrives in the year of the disaster, while the spending it finances follows at a lag.
"High Temperatures and Crime: Evidence from China" (with Chenxi Tang)
Abstract: We study the effect of high temperatures on crime in China using over 4.4 million criminal judgments from 2014-2020 in a fixed effects model. A one-degree Celsius increase raises daily crime by 0.6 percent. Crime rises with moderate heat but flattens at extreme heat, indicating a non-linear relationship. A cumulative effect model shows limited temporal displacement, with 94 percent of the effect occurring contemporaneously. These findings provide new evidence from China on the short-run social costs of rising temperatures.
"The Impact of Earthquakes on the Marriage Market: Evidence from the Yogyakarta Earthquake"
Abstract: Natural disasters leave individuals and households especially vulnerable, yet little is known about how high-mortality earthquakes reshape marital behavior. Using the Indonesia Family Life Survey, I combine propensity score matching with a difference-in-differences design to estimate the effect of the Yogyakarta earthquake on marriage timing and marriage payments. The earthquake lowered the average age at marriage by 1.55 years: 1.34 years for men and 1.54 years for women. It raised the likelihood of early marriage by 10 percent, with a larger increase in rural areas (14 percent) than in urban areas (9.5 percent). I find no significant effect on bride price payments.
Work in Progress
"Internal Migration and Earthquake Reconstruction Plans" (with Ke Gao)
"Internet Development and Child Marriage in Africa"