COMBINING LAB EXPERIMENTS AND INDUSTRY DATA IN TRANSACTION COST-ANALYSIS - THE CASE OF COMPETITION AS A SAFEGUARD

Extant empirical work in transaction cost analysis relies almost exclu sively on cross-sectional survey data on firm behavior to test predict ions about vertical integration, long-run contracts, alliances, and th e like. Such designs shed useful light on the causal mechanisms suppos edly responsible for the effects, but do not develop the details. Many of these designs also offer scarce insight into the profit implicatio ns inherent in the theory. Cognizant of these gaps, we combine a labor atory experiment and cross-sectional industry practice data to offer t he first empirical evidence of ''invited'' competition as a safeguard for buyers' specific investments. Specifically, our lab data offer dir ect evidence of the suppression of opportunism. As the theory predicts , competition from a licensee holds down price hikes that exploit lock ed-in buyers in follow-on time periods. The lab data also show that co mpeting with a licensee becomes more attractive to the original monopo list as potential buyers need to make larger supplier-specific expendi tures. The reverse is true when the focal product provides buyers with larger economic value. Interestingly, these effects show some systema tic deviation from the subgame perfect predictions of the game-theoret ic models. Our industry data show that firms behave according to the p rediction that products requiring greater levels of supplier-specific investments are more likely to be licensed. We close with a discussion of the economics of safeguards and the methodological implications of the studies.