Research

Abstract: Building energy efficiency has been a cornerstone of greenhouse gas mitigation strategies for decades. However, impact evaluations have revealed that energy savings typically fall short of engineering model forecasts that currently guide funding decisions. This creates a resource allocation problem that impedes progress on climate change. Using data from the largest U.S. energy efficiency program, we demonstrate that a data-driven approach to predicting retrofit impacts based on previously realized outcomes is more accurate than the status quo engineering models. Targeting high-return interventions based on these predictions dramatically increases net social benefits, from $0.93 to $1.23 per dollar invested.

Abstract: Free-Floating Car-Sharing (FFCS) services allow users to rent electric vehicles by the minute without restrictions on pick-up or drop-off locations within the service area of the rental company. Beyond enlarging the choice set of mobility options, FFCS may reduce congestion and emissions in cities, depending on the service's usage and substitution patterns. In this paper, we shed light on this by analyzing the universe of FFCS trips conducted through a leading company in Madrid during 2019. We correlate FFCS usage patterns with data on traffic conditions, demographics, and public transit availability across the city. We find complementarities between FFCS and public transport in middle-income areas with scarce public transport options. Moreover, we find that the use of FFCS peaks earlier than overall traffic and is broadly used during the summer months. This suggests that FFCS may have smoothed road traffic in Madrid, contributing to a reduction in overall congestion.

Abstract:  Evaluations of energy efficiency programs reveal that realized savings consistently fall short of projections. We decompose this 'performance wedge' using data from the Illinois Home Weatherization Assistance Program (IHWAP) and a machine learning-based event study research design. We find that bias in engineering models can account for up to 41% of the wedge, primarily from overestimated savings in wall insulation. Heterogeneity in workmanship can also account for a large fraction (43%) of the wedge, while the rebound effect can explain only 6%. We find substantial heterogeneity in energy-related benefits from IHWAP projects, suggesting opportunities for better targeting of investments. 

Abstract: In many parts of the developing world, farmers widely use deliberate fires to burn vegetation and clear land to plant crops. These agricultural fires, however, are known to be associated with health costs due to increased air pollution. We contribute to underpinning the associated health cost estimates by studying the effects of these fires on hypertension risk. Despite being one of the leading causes of mortality globally, there is little direct evidence on how hypertension risk changes with exposure to pollution from agricultural fires. To overcome common data and empirical challenges in this setting, we match blood pressure readings from nearly 784,000 individuals across India with satellite data on 1.2 million agricultural fires, wind direction realizations, and local ambient air pollution. We find that the incidence of hypertension increases by 1.8% for each standard deviation increase in the number of upwind fires observed one day before the blood pressure readings. We find that the impact is stronger among older males, smokers, individuals that were already on blood pressure medication, and individuals belonging to socially marginalized groups. Our estimates suggest that agricultural fires in India lead to hypertension-related additional mortality, associated with USD 9 billion annually in costs.

Abstract:  This paper provides novel estimates of the implicit cost of carbon abatement associated with the COVID-19 crisis. We compare that to the costs from renewable investments that would lead to similar abatement. Focusing on the Spanish economy and its power sector, we combine machine learning and simulation tools to construct a precise counterfactual of market performance in absence of the crisis. Results suggest that power sector CO2 emissions fell by 4.13 Million Tons (about 11.5%) during 2020 due to the pandemic, less than half of the actual year-on-year emissions reductions. Investing in renewables to achieve similar carbon abatement would yield an implicit cost of 60-65 Euro/Ton of CO2. Conversely, the pandemic caused a substantial GDP loss in Spain, relative to the extent of overall carbon abatement. The resulting cost of carbon abatement associated with the pandemic thus exceeded 7 thousand Euro/Ton.

Abstract: We explore the mechanisms driving the effectiveness of a widely-used behavioral intervention that reduces energy consumption by repeatedly mailing social comparison-based home energy reports (HERs) to households. With a randomized controlled trial, we introduce HERs in a college residence, where tenants do not pay energy bills. Our results indicate that HERs induced almost no behavioral changes for heating demand, with precise estimates that allow us to rule out thermostat changes greater than 0.36ᵒF. This suggests that behavioral channels, such as competitiveness, social norms, or moral suasion, may not motivate conservation in the absence of direct monetary incentives. 

Abstract: This paper compares energy-efficient appliance adoption rates across U.S. residential markets. The focus is to explore variation across tenure modes (rented or owner-occupied residences).  Bivariate probits are used to correct for endogenous determination of tenure mode and energy efficiency outcomes. Results suggest that, when compared to renters, homeowners are significantly more likely to have energy-efficient appliances. The mechanisms that could be driving those differences are also investigated. Heterogeneity analyses reveal that rented dwellings are more likely to have efficient appliances when landlords incur utility payments. Adoption rate differences are also shown to be inversely related to energy prices. Those findings are consistent with a problem of asymmetric information in the housing market, typically referred to as the “landlord-tenant problem.” This paper is also the first to assess how tenancy duration influences efficiency investments in this context. Results suggest that investments in rented homes are more likely to occur at later periods of tenancy, when relations between landlords and tenants might be better established.

Abstract: We study the energy pricing distortion that arises when apartment buildings share a common heating bill, resembling a tragedy of the commons. Submetering corrects the distortion, but does it improve welfare? We propose a theoretical framework with shared billing and environmental externalities to derive a sufficient statistic for the welfare effect of submetering. We then analyze the staggered introduction of submetering in 265 apartment buildings in Switzerland. Event study estimates show that submetering reduces annual heating expenses by 9.5 percent, on average. Under benchmark assumptions, this average effect implies a welfare improvement only if the social cost of carbon is 239 CHF/tCO2 or larger. Machine learning estimates reveal substantial heterogeneity in treatment and welfare effects, suggesting a role for targeted interventions in this context.

Abstract: I propose a machine learning-based approach for estimating treatment effect heterogeneity in event studies with staggered adoption. The first step is to use machine learning algorithms to predict counterfactual outcomes in the absence of treatment. A distribution of effects can then be obtained by taking the difference between realized and counterfactual outcomes. From this distribution, it is possible to estimate the average treatment effect on the treated (ATT) or conditional ATT, depending on the researcher's objective. With simulations, I show that the ML estimates are unbiased and more efficient than estimates from conventional approaches. My proposal serves as an alternative to standard two-way fixed effects regressions that are potentially biased in the presence of heterogeneous treatment effects. I conclude with an application to real data from a residential energy efficiency program, revealing substantial heterogeneity of energy savings depending on the types and levels of upgrades performed.