The seemingly collusive 95% rule in public procurement: the case of local governments in Japan

Purpose This paper aims to examine a unique form of coordinated behavior in public procurement: the 95% rule. Although it has been recognized by some experts and researchers, the extant literature has not systematically analyzed this phenomenon. Design/methodology/approach This study identifies and quantifies the impact of the 95% rule, showing a sharp drop in the winning probability at 95% of the reservation price; bidders deliberately avoid exceeding this threshold to evade suspicions of collusion. Using regression discontinuity analysis, the authors examine over 13,000 bids for approximately 2,500 pavement contracts awarded by Ehime Prefecture, Japan, from 2014 to 2022. Findings The results show that winning bids accumulate just below the 95% threshold in certain market conditions while losing bids exceed this level consistently. If the winner and loser are predetermined, they only need to set their bid prices based on this threshold. This simple pattern effectively sustains collusive agreements, such as bid rotation and regional division. Social implications The coordinated behavior among firms identified in this paper is easy to understand and may help both experts and the public strengthen bid-rigging oversight. Originality/value This study contributes to the literature on Japanese procurement by elucidating the structural mechanism underlying bidding patterns. The 95% rule also provides a unifying explanation for various observed facts – such as the significant gap between winning and runner-up bids and the stickiness of certain high bid price levels – that were separately identified in previous studies based on Japanese data.