Research

Gasoline Price Shocks and Consumer Spending: Evidence from Household Purchases (Job Market Paper)

presented at EPIcenter Summer Program Expo 2025 and Southern Economic Association (SEA) 2025

I study how two recent, California-specific gasoline price shocks (Fall 2022 and Fall 2023) affected household retail spending using proprietary high-frequency consumer panel data aggregated to the household–week level. I compare California to other large states that did not experience any significant gasoline price increases, estimating a weekly event study and a pooled event-window difference-in-differences. During the event weeks, California households cut retail spending by about 8 percent relative to the comparison states, with similarly sized declines in food and smaller, noisier responses in non-food. These spending reductions are concentrated among lower income households, who display larger declines in retail spending than higher income households. The spending path tracks the price path—falling within weeks of the spike and reverting slowly as prices ease. The results show that short-lived, highly visible fuel price increases can generate sizable reductions in household spending, a channel that is relevant for other salient, bill-based energy cost increases. These spending reductions represent short run welfare losses, especially for lower income households that cut back on basic consumption when gasoline prices rise sharply. The findings are relevant for the design of gasoline taxes, temporary tax holidays, and targeted rebates, and they suggest that retailers can use promotions during high price episodes to help sustain store traffic while easing the burden on consumers.

Did Deregulation De-Electrify Homes? Evaluating the Social Costs of Natural Gas Market Restructuring for the Energy Transition

presented at the EPIcenter Affiliates Program 2024

This paper investigates whether the US natural gas market restructuring in 1992 has increased the residential adoption rate of natural gas for space heating, water heating, and cooking. If the restructuring made natural gas more attractive, this could lead to more homes, especially new builds, preferring natural gas as their energy source. This could lock these homes into using natural gas since they would incur large switching costs to change to electricity. Given the cleaner electricity generation today compared to decades ago, homes using natural gas instead of electricity may contribute more to CO2 emissions, resulting in unintended social costs. I add to the existing literature by analyzing the effects of this restructuring on three separate household appliances and include both old and new homes in my analysis. I find an 8% to 9% increase in the national adoption rate of natural gas for space and water heating in new homes after the restructuring. Heterogeneous regional effects suggest that the result is mainly driven by the Northeast and South census regions. The adoption patterns are similar for space and water heating, but natural gas adoption for cooking did not respond to the restructuring, suggesting that households have specific preferences for cooking appliances.

Drafts and/or slides available upon request.

Decomposing Racial Gaps in Energy Expenditures (with Bobby Harris and Casey Wichman)

We study the drivers of racial and income disparities in United States household energy expenditures. Using 2005–2022 American Community Survey (ACS) microdata linked to local weather, state-level retail electricity and natural gas prices, geography, and Weatherization Assistance Program (WAP) funding, we measure Black–White energy expenditure gaps by income quintile and owner/renter status. There is no unconditional difference in average energy spending, but conditional on household characteristics Black households pay about 240 dollars more per year for energy than comparable White households, and this conditional gap appears for both owners and renters in every income quintile. Kitagawa–Oaxaca–Blinder decompositions show that roughly forty to sixty percent of the homeowner gap, and a smaller share of the renter gap, is explained by observable differences in location, climate, housing, and household composition, with mixed evidence on differential energy efficiency. Decomposition of changes between 2005–2009 and 2018–2022 indicates that falling retail energy prices and gradual shifts in observable characteristics explain most of the decline in the Black–White homeowner gap, while funding by national efficiency programs like Weatherization Assistance Program is too small to plausibly explain the observed decline in the gap.