Research

Work in Progress

Adjustment Frictions and the Cost of Environmental Regulatory Uncertainty (Job Market Paper)

When regulators are appointed by elected government officials, political power shifts create regulatory uncertainty through regulator turnover. Regulator turnover can change how rules are enforced, generating fluctuations in enforcement intensity for regulated facilities. Moreover, adjusting pollution levels in response to new enforcement regimes often involves adjustment costs. This paper quantifies the welfare cost of such regulatory fluctuations in California’s water quality enforcement, emphasizing the roles of adjustment costs and political inefficiency—that is, inefficiencies stemming from politically driven variation in enforcement intensity. Using linked data on enforcement, compliance, and regulator membership, I show that facility violations gradually adjust following regulator turnover, consistent with adjustment costs. Estimating a structural model of facility pollution decisions, I find substantial adjustment costs: past pollution influences abatement 3.5 times more than current fines, reflecting slow adjustments in abatement behavior. I estimate welfare losses equal to nearly one-third of total fines—28% from adjustment costs and 72% from political inefficiency—relative to a stable enforcement regime. While these adjustment costs raise costs for facilities, they also stabilize pollution outcomes by dampening responses to regulatory swings. However, uncertainty about future regimes weakens this stabilizing effect and amplifies adjustment costs. I further show that institutional stability improves welfare: doubling regulator term lengths to eight years increases welfare by approximately half.

Measurement Errors in Weather Data —with Derek Lemoine, Wint Thu

Accurately estimating the economic impacts of weather is increasingly important, yet challenging due to inherent measurement error in weather data. Beyond the classic attenuation bias, endogenous entry and exit of weather stations can further distort these estimates. This concern is particularly salient as recent studies increasingly rely on panel fixed effects regressions, where the influence of measurement error can be amplified. To address this issue, we propose an instrumental variable approach that uses weather measures constructed to keep measurement errors constant over time and across space. We build such data and apply it to a set of canonical studies on the effects of weather. Our results show that correcting for this bias alters the estimated effects, often changing the magnitude and in some cases, even the sign of the effects.

Do Firms Avoid Pollution? Water Pollution and Beverage Manufacturers’ Location

I study a novel channel through which environmental pollution imposes a cost on firms: the quality of raw material inputs. Input water quality greatly determines the quality of beverages and thus, water pollution can affect beverage manufacturers’ profits and where they operate. Specifically, I investigate how water pollution impacts the entry decisions of beverage manufacturers. I use annual allotments of Clean Water State Revolving Fund to a state as an instrumental variable for water quality in downstream, adjacent counties. I find that past allotments increase downstream water quality and that a 10% increase in mean dissolved oxygen concentrations leads to an increase in the number of net entry of beverage manufacturers by 2.36 firms.