Favoritism in Scoring Auctions Based on Russian Procurement Data

In this paper, we study scoring auctions – a type of auction where not only price but also non-price criteria (quality, qualification, etc.) matter to determine whether there was private collusion between the procurer and the participant. The buyer can exaggerate the quality and qualification scores of the favoured bidder, thus making them win the auction. Quality and qualification scores are intangible and, therefore, cannot be compared across different auctions. Hence, we transform the scoring rule to normalized quality minus price so that it would have a monetary equivalent. We apply statistical methods, as suggested by Huang (2019), to analyze Russian procurement data on more than 20 000 scoring auctions that took place between 2014 and 2018. We find that 20.2% of the auctions are suspicious of having a collusion between the winner and the procurer. The knowledge gained allows the government to deeply screen suspicious auctions and prevent unfair resource distribution, which would lead to economic development.