The Combined Effects of Housing and Policy on Energy Poverty

• In Event: Lightning Session: Natural Resource, Energy, and Environmental Policy

Abstract

Energy poverty is a pressing issue of quality of life and social equity that remains deeply consequential in the United States. Three primary factors shape household energy poverty: household characteristics, housing conditions, and the broader external environment. Existing economic research on housing and energy poverty has largely focused on the micro scale, examining how specific attributes such as housing quality and age affect energy poverty outcomes. While such micro-level analyses can distinguish the effects of individual housing characteristics with considerable precision, they tend to overlook neighborhood-level features and the role of relevant policies.

This study centers on housing and investigates the combined influence of housing attributes, neighborhood characteristics, related policies, and administrative capacity on energy poverty, with particular attention to the contextual effects of the areas in which homes are situated. We begin by selecting California and Illinois as our study settings, as both states provide long-panel utility disconnection data at the ZIP code level, enabling a comparative analysis of energy poverty across warm and cold climate regions. We then identify key housing attributes, including quality, age, and tenure status, alongside neighborhood characteristics such as density, infrastructure, and environmental conditions. Next, we employ large language models to extract keywords related to energy poverty from housing policies enacted at the federal, state, and county levels, encompassing both housing quality regulations and energy assistance programs, and construct weighted measures to assess their influence on energy poverty outcomes. Using fixed-effects models and instrumental variable approaches, we conduct a comprehensive econometric analysis of the causal and combined effects of housing attributes, neighborhood characteristics, and housing policies on energy poverty. Finally, through heterogeneity analysis, we examine how the effectiveness of different housing policies varies across housing types and geographic contexts, and assess how local governments' administrative capacity shapes policy implementation outcomes.

Disconnection data are drawn from utility-reported records mandated by state governments in both states, as well as the Utility Disconnections dataset compiled by the Energy Justice Lab at Indiana University, available at the ZIP code level for the period 2020 to 2024. Housing attribute data come from Cotality, neighborhood characteristics from the U.S. Census Bureau, and policy data from various levels of government as well as compilations by the National Low Income Housing Coalition (NLIHC).

This study contributes to the theoretical understanding of housing-related determinants of energy poverty and provides empirical evidence on the integrated effects of housing, neighborhood, and policy factors. On a practical level, the findings can help more accurately identify areas and housing types most in need of targeted intervention, evaluate the effectiveness of energy poverty mitigation measures, support the design of context-sensitive policy combinations at the regional scale, strengthen administrative capacity, and ultimately reduce energy poverty at the local level.

Author

• Zheyu Zhang, University of Maryland, College Park
  Presenting Author