Incentive and Sampling Effects in Procurement of Innovation

In this paper, we study the procurement of an innovation in which firms exert effort and create innovations, where the quality of innovation is stochastic. Both the effort level and the quality of innovation are unverifiable, and the procurer cannot extract up-front payment from the firms. Given the uncertainty of quality realization, there is a trade-off regarding the number of participating firms in the procurement process: If many firms participate in the process, they may be discouraged from expending their initial investment because each of them has a small chance of winning (we call this incentive effect). At the same time, as the number of participants increases, the procurer has a growing chance of getting a higher quality because of the randomness of the quality realization (sampling effect). Therefore, the procurer faces a nontrivial problem of how many firms to invite in the procurement process. We consider two prominent contest mechanisms, a first-price auction and a fixed-prize tournament. We show that if the randomness is large enough, it is optimal for the buyer to invite as many firms as possible in both mechanisms, and the fixed-prize tournament outperforms the first-price auction. In the limit at which the randomness vanishes, inviting only two firms is optimal in both mechanisms, and the firstprice auction outperforms the fixed-prize tournament. Under the first-price auction, we show that any equilibrium converges to an equilibrium as the randomness diminishes and provide a characterization of the limit equilibrium. We also provide a constructive example of a mixed-strategy equilibrium with two firms when the randomness is moderate.