[Paper]

Workers’ outside options play a central role in determining the transitional costs of labor demand shocks. Using comprehensive administrative data, we examine the worker-level effects of the decline of coal -- a regionally concentrated labor demand shock that reduced coal sector employment by more than 50 percent between 2011 and 2021. We show that coal workers experienced large and persistent earnings losses compared to similar workers with less or no connection to coal. Unlike worker-level analyses of labor demand shocks in more spatially diffuse industries, we find that non-employment is an important margin of adjustment. When employed, coal workers earn substantially lower wages than prior to coal’s decline. Sectoral or regional mobility does little to mitigate these losses, while SSDI receipt increases substantially. Our findings suggest that transitional costs are higher in geographically concentrated industries when skills do not easily transfer across sectors.

Energy inputs per unit of output have steadily trended downward, a pattern often attributed to successful energy efficiency policy. We assess whether this trend reflects targeted improvements in energy use or broader technical progress. To do so, we develop a model of factor-biased technical change and apply it to a half century of establishment-level data from the U.S. Census Bureau. We find that innovations in energy-augmenting productivity account for only about one-third of establishment-level improvements in energy efficiency. Decompositions indicate that aggregate improvements in energy-augmenting productivity largely reflect reallocation of market share across incumbents.

Does investment in distressed regions encourage people to live in areas which limit economic opportunity? We study a four billion dollar grant program that funds affordable, low-emissions housing in regions facing economic and environmental disadvantages. Our empirical approach compares census blocks which are awarded grants to similar blocks which apply and are denied funding. We find investment expands the supply of housing in targeted regions by 41 percent, lowering local rents while not impacting home values. Incumbents in awarded areas become less likely to migrate over time, suggesting these areas become more desirable for local residents. This decline is driven by fewer short-distance moves to similarly-distressed regions rather than reduced long-distance migration to higher-opportunity areas. We conclude by evaluating the extent to which investment in high-density areas reduces migration lower-density regions of the economy.

Do lucrative early-life job opportunities lock in better or worse long-run outcomes? We study this question in the context of the fossil fuel industry, where repeated local booms and busts create sharp variation in labor demand. Using administrative tax records and demographic data from the U.S. Census Bureau, we leverage a shift-share instrument to examine whether exposure to a local boom at the start of workers' careers leads to changes in eventual educational attainment, location choices, and lifetime earnings. We also aim to identify how these impacts vary by population and study the mechanisms, such as skills, place attachment, and industry-specific learning, that propagate them.

[Paper]

[Paper]

This paper provides the first estimates of within-industry heterogeneity in energy and CO₂ productivity for the entire U.S. manufacturing sector. We measure energy and CO₂ productivity as output per dollar energy input or per ton CO₂ emitted. Three findings emerge. First, within narrowly defined industries, heterogeneity in energy and CO₂ productivity across plants is enormous. Second, heterogeneity in energy and CO₂ productivity exceeds heterogeneity in most other productivity measures, like labor or total factor productivity. Third, heterogeneity in energy and CO₂ productivity has important implications for environmental policies targeting industries rather than plants, including technology standards and carbon border adjustments.

[Paper]

The natural environment is central to all aspects of life, but efforts to quantify its influence have been hindered by data availability and measurement constraints. To mitigate some of these challenges, we introduce a new prototype of a microdata infrastructure: the Census Environmental Impacts Frame (EIF). The EIF provides detailed individual-level information on demographics, economic characteristics, and address level histories –- linked to spatially and temporally resolved estimates of environmental conditions for each individual –- for almost every resident in the United States over the past two decades. This linked microdata infrastructure provides a unique platform for advancing our understanding about the distribution of environmental amenities and hazards, when, how, and why exposures have evolved over time, and the consequences of environmental inequality and changing environmental conditions. We describe the construction of the EIF, explore issues of coverage and data quality, document patterns and trends in individual exposure to two correlated but distinct air pollutants as an application of the EIF, and discuss implications and opportunities for future research.

[Paper]

This paper introduces the Gridded Environmental Impacts Frame (Gridded EIF), a novel privacy-protected dataset derived from the U.S. Census Bureau’s confidential Environmental Impacts Frame (EIF) microdata infrastructure. The EIF combines comprehensive administrative records and survey data on the U.S. population with high-resolution geospatial information on environmental conditions. While access to the EIF is restricted due to the confidential nature of the underlying data, the Gridded EIF offers a broader research community the opportunity to glean insights from the data while preserving confidentiality. We describe the data and privacy protection methods, and offer guidance on appropriate usage, presenting practical applications.

[Paper]

The Census Bureau’s Environmental Impacts Frame (EIF) is a microdata infrastructure that combines individual-level information on residence, demographics, and economic characteristics with environmental amenities and hazards from 1999 through the present day. To better understand the long-run consequences and intergenerational effects of exposure to a changing environment, we expand the EIF by extending it backward to 1940. The Historical Environmental Impacts Frame (HEIF) combines the Census Bureau’s historical administrative data, publicly available 1940 address information from the 1940 Decennial Census, and historical environmental data. This paper discusses the creation of the HEIF as well as the unique challenges that arise with using the Census Bureau’s historical administrative data.

[Paper]

The transition to clean energy represents a fundamental and important shift in economic activity. We present new facts about workers in clean and legacy energy sectors between 2005 and 2019 using linked, administrative employer-employee data for all W-2 workers in the United States. We show that both clean and legacy energy establishments hire a disproportionate share of non-Hispanic White and male workers compared to the working population, that workers rarely move from legacy to clean firms, and that, conditional on education, workers do not earn more in clean firms than in legacy firms. The occupational categories of jobs at clean firms differ notably from occupations at legacy firms and, on average, tend to be performed by workers with higher levels of education. Regional overlap in employment opportunities is not sufficient to facilitate worker transitions from legacy to clean firms. Substantially lower earnings outside of the energy sector combined with low mobility between legacy and clean firms suggests that the costs of the clean transition on workers in legacy fossil fuel sectors may be substantial. At the same time workers moving into clean activities from outside of the energy sector experience significant increases in earnings and greater job stability, suggesting that clean jobs are 'good jobs' for those who can access them.

[Paper]

Black households have higher residential energy expenditures than white households in the US. This residential energy expenditure gap persists after controlling for income, household size, home-owner status, and city of residence. It decreased but did not disappear between 2010 and 2017, and it is fairly stable in levels across the income distribution, except at the top. Controlling for home type or vintage does not eliminate the gap, but survey evidence on housing characteristics and available appliances is consistent with the gap being driven at least in part by differences in housing stock and related energy efficiency investments.