We examine how social preferences affect behavior and surplus in relational contracts. Experimental subjects participate in a contracting environment similar to Brown, Falk, and Fehr [Brown, M., Falk, A. & Fehr, E., “Relational Contracts and the Nature of Market Interactions, Econometrica, 72 (2004):747-780] and in social preference experiments adapted from Charness and Rabin [Charness, G. & Rabin, M. “Understanding Social Preferences with Simple Tests.” The Quarterly Journal of Economics 117(2002): 817-869]. Subjects’ behavior during the Charness and Rabin experiment is a significant predictor of behavior and outcomes observed during the subsequent multi-period, finite-horizon, relational- contracting environment, which features market power, unenforceable performance, reputation formation and endogenous matching of trading partners. Compared to subjects who respond to the Charness-Rabin games in a fashion consistent with purely self-interested, competitive or reciprocal social preferences, buyers and sellers with alternative social preference structures engage in contracts with substantially higher quality and price, which leads to greater surplus for both parties. A key difference is that self-interested, competitive and reciprocal buyers respond to early-period shirking by extending subsequent offers that are less generous to the seller, while buyers with other social preferences extend subsequent offers that are more generous. Reciprocal and competitive sellers and, to a lesser extent, self-interested sellers, deliver sub-contractual levels of quality more often, which substantially lowers buyer and total welfare. We conclude that intentional or ‘cold’ measures of social preferences have considerable predictive power in dynamic, interactive (or ‘hot’) economic settings.
