Governments often delegate responsibility for regulatory enforcement to third-party firms that sell compliance-related services to regulated buyers, who may be firms or consumers. This paper uses the context of passenger vehicle emissions testing required by the Clean Air Act to empirically illustrate the welfare tradeoffs caused by between-firm competition under this market-based enforcement regime. In theory, competition may lead to more choices and lower compliance prices for buyers, increasing their surplus. On the other hand, it may also lead sellers to lower enforcement stringency in order to attract customers, generating externalities from incomplete enforcement. 

 I employ a dataset comprising millions of emissions tests that allows me to directly observe, for a majority of tests, whether a firm cheated to allow a customer to pass. In this sample, 0.74\% of test results are falsified, and cheating cars emit 50-70\% more air pollution than observably similar non-cheating vehicles. Two reduced-form empirical strategies indicate that firms cheat more in response to increases in competition in the testing market, but do not lower prices. In ongoing work, I compliment this evidence with a structural model of consumers' choices over firms and cheating vs fair tests that takes into account preferences for prices, cheating, firm characteristics, and driving distances to obtain a test. This model will allow me to quantify the benefit of restricting competition in terms of reduced pollution, and weigh it against the costs in terms of changes in test prices and driving distances. 

Dockless, shared e-bikes and scooters are a rapidly-growing segment of the local transportation market despite being taxed at higher rates per mile of travel than many conventional transportation modes, including motor vehicles. This paper asks what the optimal tax rate for shared bikes and scooters would be when accounting for environmental externalities and redistribution. We derive a sufficient-statistics model that describes how optimal tax rates depend on both extensive-margin (number of trips) and intensive-margin (duration of trips) demand responses to price changes, as well as the income of bike and scooter users and net externalities relative to other travel modes. We estimate the parameters of this model using trip-level data from a large bike and scooter company and find that in several U.S. cities, use of their vehicles is concentrated among low-income individuals. The model accordingly suggests that policymakers in these cities might want to subsidize shared bike and scooter use even if it generates only modest positive externalities by displacing trips by motor vehicles.