This dissertation represents an effort to advance interdisciplinary research in issues relevant for energy and environmental policy, combining economics with applied engineering and ecology. It includes work that is informed by theoretical and empirical studies, and is conceptually centered in the notion that competitive markets lead to inefficient combinations of risk and yield. In the first two chapters of the dissertation, I study this in the context of wind energy capacity investments, where profit-maximizing developers choose the location and timing of the construction of wind farms. The final chapter of the dissertation is an empirical study on the effects of intensive aquaculture on water pollution.

This dissertation presents a three-part study in modern empirical environmental economics. In these three studies, I focus on five core economic issues—equity, incentives, environmental quality, consumer behavior, and causality—and ask what environmental economics can teach us about three common topics: energy consumption, cannabis legalization, and pesticide application. The first chapter examines how residential natural gas consumers respond to changes in the price of natural gas. With 70 million consumers, residential natural gas has grown to a first-order policy issue. This first chapter provides the first causally identified, microdata-based estimates of residential natural-gas demand elasticities using a panel of 300 million bills in California. To overcome multiple sources of endogeneity, we employ a two-pronged strategy: we interact (1) a spatial discontinuity along the service areas of two major natural-gas utilities with (2) an instrumental-variables strategy using the utilities' differing rules/behaviors for internalizing upstream spot-market prices. We then demonstrate substantial seasonal and income-based heterogeneities underly this elasticity. These heterogeneities suggest unexplored policies that are potentially efficiency-enhancing and pro-poor. The second chapter explores what may be unintended—or unconsidered—results of cannabis legalization. Cannabis legalization advocates often argue that cannabis legalization offers the potential to reduce the private and social costs related to criminalization and incarceration—particularly for marginalized populations. While this assertion is theoretically plausible, it boils down to an empirically testable hypothesis that remains untested: does legalizing a previously illegal substance (cannabis) reduce arrests, citations, and general law-enforcement contact? The second chapter of this dissertation provides the first causal evidence that cannabis legalization need not necessarily reduce criminalization—and under the right circumstances, may in fact increase police incidents/arrests for both cannabis products and non-cannabis drugs. First, I present a theoretical model of police effort and drug consumption that demonstrates the importance of substitution and incentives for this hypothesis. I then empirically show that before legalization, drug-incident trends in Denver, Colorado matched trends in many other US cities. However, following cannabis legalization in Colorado, drug incidents spike sharply in Denver, while trends in comparison cities (unaffected by Colorado's legalization) remain stable. This spike in drug-related police incidents occurs both for cannabis and non-cannabis drugs. Synthetic-control and difference-in-differences empirical designs corroborate the size and significance of this empirical observation, estimating that Colorado's legalization of recreational cannabis nearly doubled police-involved drug incidents in Denver. This chapter's results present important lessons for evaluating the effects and equity of policies ranging legalization to criminal prosecution to policing. Finally, the third chapter investigates the roles pesticides play in local air quality. Many policymakers, public-health advocates, and citizen groups question whether current pesticide regulations properly equate the marginal social costs of pesticide applications to their marginal social benefits—with particular concern for negative health effects stemming from pesticide exposure. Additionally, recent research and policies in public health, epidemiology, and economics emphasize how fine particulate matter (PM2.5) concentrations harm humans through increased mortality, morbidity, mental health issues, and a host of socioeconomic outcomes. This chapter presents the first empirical evidence that aerially applied pesticides increase local PM2.5 concentrations. To causally estimate this effect, I combine the universe of aerial pesticide applications in the five southern counties of California's San Joaquin Valley (1.8M reports) with the U.S. EPA's PM2.5 monitoring network—exploiting spatiotemporal variation in aerial pesticide applications and variation in local wind patterns. I find significant evidence that (upwind) aerial pesticide applications within 1.5km increase local PM2.5 concentrations. The magnitudes of the point estimates suggest that the top decile of aerial applications may sufficiently increase local PM2.5 to warrant concern for human health. Jointly, the three parts of this dissertation aim to carefully administer causally minded econometrics, in conjunction with environmental economic theory, to answer unresolved, policy-relevant questions.

This dissertation consists of three essays in Energy and Environmental Economics. In the first chapter, an empirical analysis is employed to examine the relationship between energy consumption and economic output in Canada. Using provincial-level Canadian records, this chapter shows that there is a long run equilibrium relationship and a bi-directional Granger causality between energy consumption and economic growth in Canada. These finding have important implications for public policy because they show that constraints on energy consumption may impact future economic growth. In the second chapter, event study methodology and Canadian stock market data are used to assess the impact of seven recent event/announcements regarding the pipelines approval process on the equity returns of energy-related firms. This chapter shows that there is no market reaction (on average) to any of the news events, which implies two possible scenarios: either the market fully anticipated the events and they did not contain any significant new information or these events did not change investor's expectation regarding future profitability and cash flow of Canadian energy firms. In the third chapter, we use a prediction market mechanism to examine the possibility of using derivatives trading as a means of generating objective forecasts of future climate change and the value of marginal damages. This chapter shows that such a market can yield unbiased estimates of the true future climate state. Also, we find that the level of consensus about climate science strongly influences the efficiency with which market uses available information.

Over the past decades, two things have become increasingly apparent: first, climate change and associated environmental impacts are pressing issues, and second, despite this growing threat, existing policies still fall far short. The goal of my research, and what I hope for the field more broadly, is to achieve effective, efficient, and equitable policy. My dissertation research covers a wide range of topics, focusing on three different areas of energy and environmental economics: methane emissions from oil and gas production; flooding on agricultural land; and energy utility regulatory rates of return. The common thread is using applied economic tools and answering policy-relevant questions with data and analysis. Often, the data that are available are far from the ideal dataset, or the policies that are on the table are far from the first best. Here, my coauthors and I adopt the "economist as plumber" mindset, using the tools that are available to address the challenges at hand (Duflo 2017).In my first chapter, my coauthor Wenfeng Qiu and I study emissions of methane, a powerful greenhouse gas, from oil and gas wells in the US. These emissions contribute significantly to climate change--they are approximately as large as the emissions of all fuel burned in the western US electricity grid. Methane emissions are rarely priced and lightly regulated--in part because they are hard to measure--leading to a large climate externality. However, measurement technology is improving, with remote sensing and other techniques opening the door for policy innovation. We present a theoretical model of emissions abatement at the well level and a range of feasible policy options, then use data constructed from cross-sectional scientific studies to estimate abatement costs. We simulate audit policies under realistic constraints, varying the information the regulator uses in choosing wells to audit. These policies become more effective when they can target on well covariates, detect leaks remotely, and charge higher fees for leaks. We estimate that a policy that audits 1% of wells with uniform probability achieves less than 1% of the gains of the infeasible first best. Using the same number of audits targeted on remotely sensed emissions data achieves gains of 30-60% of the first best. These results demonstrate that, because leaks are rare events, targeting is essential for achieving welfare gains and emissions reductions. Auditing a small fraction of wells can have a large impact when properly targeted. Our approach highlights the value of information in designing policy, centering the regulatory innovation that is possible when additional information becomes available.My second chapter is coauthored with Oliver Browne, Alyssa Neidhart, and Dave Sunding. We study high-frequency flood risk on agricultural land. Floods destroy crops and lower the value of agricultural land. Economic theory implies that the hedonic discount on the value of a parcel of flood-prone land should scale with the expected probability flooding. Most empirical studies of the impact of flood risk on property values in the United States focus on the relatively small risk posed by the 100-year or 500-year floodplains, as reported in maps produced by the FEMA. These studies consequently find a relatively small corresponding discount in property values. However, a significant amount of agricultural bottom-land lies in floodplains that flood more frequently. We estimate the hedonic discounts on with agricultural land that floods at these higher frequencies along the Missouri River. As flood risk increases, the value of flood-prone land decreases, with a hedonic discount ranging from close to zero in the 500-year floodplain to approximately 17% in the 10-year floodplain. To illustrate the importance of characterizing these higher frequency flood risks, we consider a climate change scenario, where properties that already face some flood risk are expected to flood more frequently. My third chapter, coauthored with Stephen Jarvis, examines the regulated rate of return on equity utility companies are allowed to collect from their customers. Utilities recover their capital costs through regulator-approved rates of return on debt and equity. The US costs of risky and risk-free capital have fallen dramatically in the past 40 years, but these utility rates of return have not. We estimate the gap between what utilities are paid now, and what they would have been paid if their rate of return had followed capital markets, using a comprehensive database of utility rate cases dating back to the 1980s. We estimate that the current average return on equity is 0.5-4 percentage points higher than historical relationships would suggest, and consumers pay an average of $2-8 billion per year more than they would otherwise. We then revisit the effect posited by Averch and Johnson (1962), estimating the consequences of this incentive to own more capital: a 1 percentage point increase in the return on equity increases new capital investment by about 5% in our preferred estimate.