Caveat Venditor : The Conditional Effect of Relationship-Specific Investment on Contractual Behavior

We examine the effect of relationship-specific investment on contract complexity, which has broad implications because complex contracts and vertical integration have similar causal origins. It is usually assumed that transaction cost economics (TCE) implies that relationship-specific investment always has a positive effect on complexity. Using a simple TCE model and the assumption that firms are credit and cash-flow constrained, we predict that the effect can be of opposite sign on opposite sides of a relationship. We also derive hypotheses on omitted-variable bias in OLS estimates of the effect of relationship-specific investments, showing that such biases reduce the probability of finding opposite signs. Our main results are on the empirical side, where we examine these predictions using detailed survey data on a broad sample of agreements made by Romanian firms. Seller relationship-specific investment has a positive effect on contract complexity while buyer investment has a negative effect. This result is obtained from a variety of estimating procedures that address omitted-variable bias in different ways, but not from OLS. The result disappears in the small sample of agreements that involve buyer prepayment, suggesting that the interaction of seller vulnerability and credit constraints drives the results. Hence, financial and legal development are substitutes in solving transactional problems. The broader context of the paper is therefore institutional development. The paper shows that firms invest more in law when courts are of better quality