Three Essays in Energy Economics

Abstract

This dissertation is composed of three essays that explore uncertainty, investment, and emissions regulation in the oil and gas industry. Flaring releases carbon dioxide, methane, and other toxic air pollutants that are harmful to the environment and human health. Natural gas extracted from the ground cannot be sold in the markets unless there is sufficient capacity to process the gas. In the first chapter, I theoretically derive a subsidy to offset the flaring damages that stem from insufficient processing capacity. The subsidy depends on the economic relationship between capacity and flaring, which I quantify using an instrumental variable model and new data from North Dakota. I find that processing capacity bottlenecks are indeed an important driver of flaring. I estimate that each gas processing plant requires an ex-ante capacity subsidy of $396 per thousand cubic feet to ensure that flaring emissions remain at the socially optimal levels. Back of the envelope calculations suggest that the subsidy would have reduced flaring by $1.4 billion cubic feet across the state, offsetting roughly $5 million in flaring damages per year. In the second chapter, I link novel data on well-level methane emissions to natural gas-producing wells in Texas, and examine the effect of information disclosure on emission reduction. I analyze changes in natural gas production following the detection of methane leaks by remote-sensing flights and the subsequent disclosure of this leakage information. Since production reported to state authorities must account for both lost and reused gas, any repairs made to leaking equipment after a flyover would likely result in an increase in reported production. My findings reveal that natural gas production tends to decrease in the immediate months following the detection of methane emissions above a detectable threshold of 5 kg/hr. However, once the operators responsible for the emitting equipment are notified, production increases compared to the period just before the leak was detected. This effect is particularly pronounced among wells operated by large publicly traded firms. These results suggest that the disclosure of leak information significantly influences production decisions within the natural gas industry. Two main mechanisms likely drive this response: first, the disclosure provides firms with valuable information about costly-to-detect methane leaks, enabling them to address stochastic equipment failures; second, firms may be aware of existing leaks but choose to delay repairs due to high costs. Nevertheless, the public release of leakage data may compel firms to act more swiftly due to anticipated scrutiny from regulators and the public. In the third chapter, I examine how uncertainty in the crude oil options market influences oil drilling decisions, given the irreversible nature of such investments. Using risk neutral ex-ante probabilities of the outcomes of 13 OPEC (the Organization of the Petroleum Exporting Countries) meetings recovered from crude oil options, I conduct a stacked event study analysis to determine how drilling changes as uncertainty resolves leading up to the events. I find that at maximal uncertainty, firms delay their drilling until the event occurs. When event outcomes are more surprising, firms drill more as uncertainty resolves leading up to the event. When event outcomes are largely expected, the effect of uncertainty on drilling is relatively small. These findings are consistent with the theoretical literature on investment under uncertainty which suggests that as uncertainty increases, the incentive to delay investment grows, widening the gap between the expected benefits and the costs necessary to justify the investment decision.