One fundamental question of economics is how consumers respond to price variation in the long run, with applications across a variety of fields. But there is a dearth of causally-identified long-run elasticity estimates, due to challenging empirical conditions. In this paper, I leverage a novel source of exogenous and persistent price variation to estimate the long-run price elasticity of demand in the setting of residential electricity. In this setting, I find that consumers are sixteen times as responsive to prices in the long run compared to the short-run, with elasticity estimates of -2.24 and -0.14 respectively. I explore mechanisms and find that in the long run, consumers respond differently to temperature across price regimes, with these differences accounting for 34 percent of the observed consumption differences. These findings highlight the potential impacts of price-based policies on demand and emphasize the importance of setting prices to reflect social marginal costs.

With the U.S. Environmental Protection Agency, we developed a machine learning model to predict sites where inspections would uncover severe violations of hazardous waste regulations. We estimate that using our model to target inspections will increase the “hit rate” by 46 percent. As is often the case, the model’s data are highly selected (representing about ~2 percent of sites), suggesting that classic selection bias concerns make our estimate’s relevance to the full population unknown. We therefore conducted a national field test of the model’s versus the EPA’s inspection targets; the model’s relative performance was even better, increasing the hit rate by 79 percent.

The U.S. power system faces a 2035 decarbonization target though the exact pathway to the target remains unclear. Policy instruments, like carbon taxes and forcing coal plants to retire through various mechanisms, could help achieve the target. It is critical to understand and compare the performances of these policies as adoption of any such policy could lead to significant costs, different emissions pathways, and political challenges. In this paper, we explore the ramifications of adopting a ‘‘second-best’’ decarbonization policy. Specifically, we assume a particular carbon tax to be the ‘‘optimal’’ policy and compare it to ‘‘suboptimal’’ carbon tax and forced coal retirement policies in terms of emissions and costs. We use a power system dispatch model that co-optimizes unit commitment, energy, and regulation capacity to simulate system evolution over multiple years, including retirements and renewables/storage expansion, under each policy scenario. Our case study highlights the trade-offs between ‘‘optimal’’ and ‘‘suboptimal’’ policies. We find that ‘‘suboptimal’’ carbon taxes could achieve similar emissions results because, counter-intuitively, higher carbon taxes do not always achieve more emission reductions due to the complexity of dispatch and retirements. In contrast, forced coal retirements result in lower costs but higher emissions than the ‘‘optimal’’ policy, with a large range of possible outcomes across different retirement cases.

In electricity markets, generators are rewarded both for providing energy and for enabling grid reliability. The two functions are compensated with two separate payments: energy market payments and ancillary services market payments. We provide evidence of changes in the generation mix in the energy market that are driven by exogenous changes in an ancillary services market. We provide a theoretical framework and quasi-experimental evidence for understanding the mechanism, showing that it results from the multi-product nature of power plants combined with discontinuities in costs. Although research in economics typically focuses solely on the energy market, our results suggest that spillovers between markets are important as well. Furthermore, policy changes relating to grid operations, grid reliability, or climate change could have unintended effects.