Recently in Japan, it has been a subject of intensive policy discourse how local governments could protect the exhausted construction industry while maintaining the efficiency of the procurement system. In the past decade, local governments repeatedly raised the lower bid limit provision until thousands of tie bids resulted in random selection of contractors. However, there are not enough theoretical arguments supporting the logic that raising lower limits contributes to the long-term growth of the industry. In this paper, we first build a benchmark model, in which we regard the bidding system as asymmetric first price auction with a ceiling price and a preannounced lower limit. Then we analyze how a raised lower limit contributes to bidders' expected payoff under Bayesian Nash equilibrium. The result implies that the existing “improvement” of the system is not best suited for the Government's aim to realize the long-term growth of the industry. It rather decreases the competitiveness of high techniques, leading to a “homogenized” market where companies with different private information have no distinct advantage in the auction. In order to evaluate the lower limit bidding system, we introduced “Random Subsidy Mechanism”, which gives subsidy to a randomly chosen bidder. The comparison indicates that raising lower limits is even worse and superior bidders prefer random subsidy. Finally, we extend the benchmark model to (1) asymmetric auction and (2) without disclosing the exact lower bid limit. (1) suggests that construction companies with low techniques tend to bid the exact lower limit more aggressively, and (2) shows that hiding the exact lower limit results in the same equilibrium as the benchmark model with exact lower limit disclosed. This argues against the optimistic stand that the efficiency of procurement bidding system can be recovered simply by hiding the lower limits.
View full abstract
