Tacit Collusion and Voluntary Disclosure: Theory and Evidence from the U.S. Automotive Industry

Abstract

This study examines how industry peers share information when they are engaged in tacit collusion. We develop a model of firms' information sharing and production decisions and use it to establish that firms engaged in tacit collusion are more likely to share information when current market demand is low and when the firms' decision horizon is long. Using 31 years of monthly production forecast data shared among the Big Three U.S. automobile manufacturers, we find empirical evidence that is generally consistent with the predictions of the model. The frequency, horizon and accuracy of the shared production forecasts decrease when the expected demand increases, suggesting less information sharing when firms have greater incentives to compete aggressively to capture greater current demand rather than to tacitly agree to restrict production. The production forecast frequency and horizon also decrease when the firms focus more on short-term profit, consistent with less information sharing when firms place less weight on the future benefit from tacit collusion.