Abstract: Feeding America, an organisation responsible for feeding 130,000 Americans every day, distributes food among a nationwide network of food banks. Their allocation mechanism, known as the `Choice System', uses auctions and a virtual currency to give food banks choice over the food they receive. This paper examines the consequences of enabling this choice. I apply a dynamic auction model to food bank bidding data, estimating the distribution of food banks' heterogeneous and time-varying needs. The central challenge is that I do not observe food banks’ inventories --- a key determinant of bidding behaviour. I overcome this difficulty using variation in food banks' winnings (observed shifters of these unobserved states) to identify the model, which I then estimate using a Gibbs Sampler. I then compare welfare under the Choice System to Feeding America’s previous allocation mechanism which gave food banks very limited choice. I estimate that the Choice System increased welfare by the equivalent of a 32.7% increase in the quantity of food allocated. Most of this gain arises because food is allocated in batches, rather than sequentially.

Cairncross Prize (RES/SES 2023 conference), Finalist Young Economist Essay Award (EARIE 2024 conference)

Note: This subsumes my Job Market Paper "Choice, Welfare, and Market Design: An Empirical Investigation of Feeding America's Choice System". 

Abstract: In this paper I develop an empirical model of bidding in repeated rounds of simultaneous first-price auctions. The model is motivated by the fact that auctions rarely take place in isolation; they are often repeated over time, and multiple heterogeneous lots are regularly auctioned simultaneously. Incorrect modelling of bidders as myopic or as having additive preferences over lots can lead to inaccurate counterfactuals and welfare conclusions. I prove non-parametric identification of primitives in this model, and introduce a computationally feasible procedure to estimate this type of game. I then apply my model to data on Michigan Department of Transportation highway procurement auctions. I investigate the extent of cost-synergies across lots and use counterfactual simulations to compare equilibrium efficiency when contracts are auctioned sequentially rather than simultaneously.