Throughout the developing world, many water distribution systems are unreliable. As a result, it becomes necessary for each household to store its own water as a hedge against this uncertainty. Since arrivals of water are not synchronized across households, serious distributional inefficiencies arise. We develop a model describing the optimal intertemporal depletion of each household's private water storage when it is uncertain when water will next arrive to replenish supplies. The model is calibrated using survey data from Mexico City, a city where many households store water in sealed rooftop tanks known as tinacos. The calibrated model is used to evaluate the potential welfare gains that would occur if alternative modes of water provision were implemented. We estimate that most of the potential distributional inefficiencies can be eliminated simply by making the frequency of deliveries the same across households which now face haphazard deliveries. This would require neither costly investments in infrastructure nor price increases.

I study the canonical private value auction model for a single good without the quasilinearity restriction. I assume only that bidders are risk averse and the indivisible good for sale is a normal good. I show that removing quasilinearity leads to qualitatively different solutions to the auction design problem. Expected revenue is no longer maximized using standard auctions that allocate the good to the highest bidder. Instead, the auctioneer better exploits bidder preferences by using a mechanism that allocates the good to one of many different bidders, each with strictly positive probability. I introduce a probability demand mechanism that treats probabilities of winning the indivisible good like a divisible good in net supply 1. With enough bidders, it has greater expected revenues than any standard auction, and under complete information, it implements a Pareto efficient allocation.

I study the canonical private value auction model for a single good without the quasilinearity restriction. I assume only that bidders are risk averse and the indivisible good for sale is a normal good. I show that removing quasilinearity leads to qualitatively different solutions to the auction design problem. Expected revenue is no longer maximized using standard auctions that allocate the good to the highest bidder. Instead, the auctioneer better exploits bidder preferences by using a mechanism that allocates the good to one of many different bidders, each with strictly positive probability. I introduce a probability demand mechanism that treats probabilities of winning the indivisible good like a divisible good in net supply 1. With enough bidders, it has greater expected revenues than any standard auction, and under complete information, it implements a Pareto efficient allocation.

I study multiunit auction design when bidders have private values, multiunit demands, and non‐quasilinear preferences. Without quasilinearity, the Vickrey auction loses its desired incentive and efficiency properties. I give conditions under which we can design a mechanism that retains the Vickrey auction's desirable incentive and efficiency properties: (1) individual rationality, (2) dominant strategy incentive compatibility, and (3) Pareto efficiency. I show that there is a mechanism that retains the desired properties of the Vickrey auction if there are two bidders who have single‐dimensional types. I also present an impossibility theorem that shows that there is no mechanism that satisfies Vickrey's desired properties and weak budget balance when bidders have multidimensional types.