Public Procurement and Reputation. An Agent-Based Model

This paper uses an agent-based computational model to investigate whether and how considering the firm's reputation in the public procurement selection process affects the expected final contract cost. We take account of different sets of simulations and a range of model parameters (such as firm skills, level of opportunistic rebate, relative weights of reputation and rebate) and propose a reputation index based on the cost overruns recorded by winning firms at the conclusion of their contracts. We show that this index allows the awarding authority to (i) select the most efficient and the least opportunistic firms, and (ii) to exclude firms that engage in frequent opportunistic behavior whose reputation has declined. Our results suggest that reputation matters, and we derive some implications for policy.