Research

This paper introduces a novel city-level panel of residential electricity prices constructed from the Federal Power Commission's Typical Electric Bills publication series. This dataset represents the most granular and comprehensive historical record of US residential electricity prices yet assembled. Using a spatial difference-in-differences design that matches each municipal utility to its nearest investor-owned neighbor within the same state, I exploit the plausibly exogenous cost shocks of the 1973 OPEC oil embargo and 1979 Iranian Revolution to identify the causal effect of utility ownership structure on the pass-through of input cost increases to residential customers. I find that municipal utilities charged significantly less than matched investor-owned utilities in the years following both shocks, on the order of 49 cents per 500 kWh bill in 1970 dollars during the OPEC embargo period and 166 cents during the Iranian Revolution period. A novel decomposition of nominal and real price differentials further reveals that the public-private gap reflects two distinct mechanisms: municipal utilities absorbed a greater share of real oil cost increases, and also passed through general price inflation to residential customers at a lower rate than investor-owned utilities did relative to the cross-utility average. Together, these findings suggest that utility ownership structure has economically meaningful consequences for the distribution of energy cost burdens across households during periods of cost volatility, with implications for contemporary debates about the privatization and regulation of essential network industries.

As climate change accelerates, energy affordability has become an increasingly urgent challenge, particularly for low-income households. The Low-Income Home Energy Assistance Program (LIHEAP), the United States' primary mechanism for delivering energy assistance to low-income households, increasingly faces constrained capacity and an uncertain future. This paper investigates how state-level variations in LIHEAP implementation affects program participation. Drawing on LIHEAP State and Territory Plans, we examine variation across three key dimensions: (1) timing and duration of assistance relative to shifting seasonal energy needs; (2) the type of administering state agency and the degree of coordination across the social safety net; and (3) outreach strategies. Using 2016–2022 linear regression estimates, we find that states where LIHEAP is administered by community service agencies are associated with higher participation rates, while administration by welfare agencies is associated with higher total participation. LIHEAP programs that coordinate application processes with other social programs are associated with reaching a larger share of eligible households, whereas passive outreach methods show limited effectiveness. Cooling degree days are not statistically associated with participation, suggesting a lag between evolving climate conditions and program adaptation. Finally, relying on 2024 data, we observe descriptive evidence of misalignment between state weather trends and funding periods. Together, these findings highlight the importance of program design and implementation in shaping access to energy assistance. We offer state-specific insights to improve LIHEAP's responsiveness and enhance the program's capacity to support low-income American households amid a changing climate.

Low-income weatherization programs provide home energy efficiency upgrades to reduce utility costs, improve indoor comfort, and address the health and safety needs of income-constrained households. This review examines seventeen retrospective, outcome-focused evaluations of weatherization programs conducted over the past four decades to assess how program evaluation methods have evolved, where critical gaps remain, and how future studies can improve. This review highlights key developments in evaluation design and causal inference methods, from early pre-post comparisons to recent use of randomized encouragement designs, matching techniques, and staggered difference-in-differences estimators. Second, it evaluates the potential of new digital and physical tools such as smart meters, sensors, and other measurement and verification technologies to improve data collection for more timely and precise evaluation. Third, it identifies growing programmatic needs that evaluations must address, including persistent barriers to participation and the expansion of weatherization services to homes requiring pre-weatherization repairs. Across prior studies, we find that many evaluations remain limited by small or unrepresentative samples, selection-biased control groups, and reliance on self-reported or inconsistent data. No evaluations to date have incorporated community-based participatory research (CBPR) approaches, limiting their responsiveness to local needs. The review concludes by offering recommendations for strengthening future evaluations, including strategies for incorporating CBPR, improving data access, and leveraging advanced econometric tools.

A robust literature has emerged that describes the correlates and consequences of high energy burdens. However, limitations and gaps exist, particularly with respect to units and scales of analysis. We compile a unique combination of aggregate and survey data for Georgia, including a stratified on-line survey of 1800 adults and a co-created survey of 28 income- and/or energy-stressed Black households. The results indicate a systemic ~30% under-estimation of Georgia's average energy burden when only aggregate data are used. We characterize the vulnerabilities, mediators, and consequences that commonly accompany high energy burdens, particularly when compounded by multiple vulnerabilities. Black, low-income female-headed households with children pay a particularly high percent of their income on energy. Energy burden vulnerabilities also correlate with low levels of adoption of heat pumps, LED lighting, smart thermostats, and solar systems – reinforcing evidence that participation in the clean energy transition is uneven. Multiple energy burden vulnerabilities also increase the potential for severe consequences such as home eviction, homelessness, the inability to pay child support, and job termination. Overall, we illustrate the value of multiple scales of analysis to estimate energy burdens, understand the role of mitigating measures, characterize consequences, and design effective policies.

Thin film stress is frequently controlled through adjustments applied to the processing parameters used during film deposition. In this work, we explore how the use of solutes with different intrinsic growth properties influences the residual growth stress development for a common solvent Cu film. The findings demonstrated that the addition of a high atomic mobility solute, Ag, or a low atomic mobility solute, V, results in both alloy films undergoing grain refinement that scaled with increases in the solute content. This grain refinement was associated with solute segregation and was more pronounced in the Cu(Ag) system. The grain size reduction was also associated with an increase in the tensile stresses observed in both alloy sets. These findings indicate that solutes can be used to control the grain size under the same deposition conditions, as well as alter the stress evolution of a growing thin film.