Collusion * in auctions can be divided into two types:

• Type 1: Collusion that involves the central auctioneer directly.
• Type 2: Collusion that is coordinated by some or all of the participants outside the central auctioneer.

The risk of Type 2 collusion is most imminent in open auctions. When the bidding is organized as an open process, it is easier to sustain a cartel or to signal market partition during the bidding. In the auctioning of the 3G licenses in Germany, for example, two of the main competitors Mannesman and T-mobil managed to signal a market partition during the auction to avoid competing. Although, the open price formation is useful to facilitate price discovery *, concerns about collusion and the protection of the participants’ private information are sufficiently important in many cases to oblige the use of partly or entirely sealed bids instead. In fact, OECD recommends sealed bids to prevent collusion in public procurement.

With partly or entirely sealed bids, a trusted third party is necessary. This creates a risk of Type 1 collusion. To illustrate with two cases, Siemens AG gained by getting confidential information about competitors’ bids in a public procurement in Singapore and Hochtief placed an optimal winning bid for the construction of a new airport in Berlin based on confidential information. Although collusion may be hard to prove in court, there are many cases and an attempt to estimate the costs has been done by the World Bank. In 2000 the estimated cost of bribing in procurement auctions was 12 % of the turnover and in 2004 the cost of corruption in public procurement alone was estimated at $200 billion. Also, the widely used proxy-bidding is a source of Type 1 collusion. In proxy bidding, a bidder places a maximal price and allows the “computer” to automatically overbid others until the maximal bid is reached. Knowing such a maximal bid can be valuable to the seller and others that gain from a high sales price.

Partisia can avoid collusion of Type 1:

With Partisia software solutions, the sealed bids are not available to any individuals or institutions. Therefore, this information cannot be misused to manipulate the price. The sealed bids can only be opened if one manages to collude with multiple persons (key custodians *) across selected institutions. The number of key custodians required to unseal the bids can be customized. Hereby, collusion of Type 1 can effectively be avoided. See here for more about the technology >>

Partisia can limit collusion of Type 2:

It is harder to avoiding collusion of Type 2. While sealed bidding makes it difficult, Partisia software solution limits collusion of Type 2 even further. With encrypted bids no one can tell an empty bid from a carefully placed bid. Hereby, the software solution allows “pseudo participation” which makes it possible to hide a low real participation. E.g. in a procurement situation where a buyer faces the same class of sellers in successive auctions, pseudo participation may easily be implemented by a rule of compulsory participation among the class of sellers in order to qualify for future auctions. The higher participation significantly hamper collusion of Type 2.

References

• Klemperer, P. 2004, Auctions: Theory and Practice, Princeton University Press.
• OECD, 2006, Competition in Bidding Markets, Policy Roundtables
• OECD, october 2008, Fighting Cartels In Public Procurement, Policy Brief
• Six Questions on the Cost of Corruption with World Bank Institute Global Governance Director Daniel Kaufmann
• Wolfstetter, Elmar G. and Lengwiler, Yvan, Auctions and Corruption (May 2000). CESifo Working Paper Series No. 401