This dissertation studies how agents and firms interplay under different market institutions. In the first two chapters, co-authored with Rodrigo Carril and Michael Walker, we study the implications of policies oriented to intensify competition for procurement contracts. Conceptually, opening contracts up to bids by more participants leads to lower acquisition costs. However, expanding the set of bidders hinders buyers' control over the quality of prospective contractors, potentially exacerbating adverse selection on non-contractible quality dimensions. We study this trade-off in the context of procurement by the U.S. Department of Defense. Our empirical strategy leverages regulation that mandates agencies to publicize contract opportunities whose value is expected to exceed a certain threshold. We find that advertising contract solicitations increases competition and leads to a different pool of selected vendors who, on average, offer lower prices. However, it also worsens post-award performance, resulting in more cost overruns and delays. This negative effect on post-award performance is driven by goods and services that are relatively complex, highlighting the role of contract incompleteness. To further study the scope of this tension, we develop and estimate a model in which the buyer chooses the extent of competition, and the invited sellers decide on auction participation and bidding. We estimate sellers' cost and ex-post quality distributions, as well as buyers' preference parameters over contract outcomes. Simulating equilibrium conditions under counterfactual settings, we benchmark the current regulation design with complexity-tailored publicity requirements, and find that adjustments to publicity requirements could provide savings of 2 percent of spending, or $104 million annually.

One of the main roles of the government involves enacting and enforcing regulation aimed at curbing the undesired behavior of agents. In the third chapter, co-authored with Mushfiq Mobarak, we study the consequences of deploying enforcement activities over illegal activities in local markets in Chile. The paper grapples with a key real-world feature that is that regulated agents adapt to circumvent enforcement. We present and test a model of enforcement with learning and adaptation by auditing vendors selling illegal fish in Chile in a randomized controlled trial and tracking them daily using mystery shoppers. Leveraging experimental variation on the frequency and predictability of enforcement, we can test the model's predictions and find that conducting audits on a predictable schedule and (counter-intuitively) at high frequency is less effective, as agents learn to take advantage of loopholes. A consumer information campaign proves to be almost as cost-effective and curbing illegal sales and obviates the need for complex monitoring and policing. The Chilean government subsequently chooses to scale up this campaign.