We model normal-quadratic social learning with agents who observe a summary statistic over past actions, rather than complete action histories. Because an agent with a summary statistic cannot correct for the fact that earlier actions influenced later ones, even a small presence of old actions...

Bidders in procurement auctions often face avoidable fixed costs. This can make bidding decisions complex and risky, and market outcomes volatile. If bidders deviate from risk neutral best responses, either due to faulty optimization or risk attitudes, then equilibrium predictions can perform...

We endogenize product differentiation in a model of sequential search with random firm-consumer match value à la Wolinsky (1986) and Anderson and Renault (1999). We focus on a product design choice by which a firm can control the dispersion of consumer valuations for its product; we interpret...

We study the incentives to acquire information from exclusive news sources versus information from popular sources in a CARA-normal asset market. Each trader is able to observe one of a finite number of news sources. Clustering on the most precise source can happen for two reasons. One is...

In a variety of settings, some payoff-relevant item spreads along a network of connected individuals. In some cases, the item will benefit those who receive it (for example, a music download, a stock tip, news about a new research funding source, etc.) while in other cases the impact may be...

We test the no-trade theorem in a laboratory financial market where subjects can trade an asset whose value is unknown. Subjects receive clues on the asset value and then set a bid and an ask at which they are willing to buy or to sell from the other participants. In treatments with no gains...