Why Submit an Offer Nobody Else Can See? Removing the Bidding War

In an open ascending auction, bidders learn from each other as the price rises, since a rival continuing to bid reveals that they value the item at least that much, and the winner pays just above what the second most eager bidder was prepared to pay. In a sealed format nobody learns anything, so each bidder must decide in advance what to offer knowing only their own valuation and their guess about the others. That removes the information sharing and removes the escalation, and it changes the optimal strategy fundamentally, since bidding one's full valuation guarantees winning nothing worth having while bidding too little risks losing to somebody who valued it less.

It appears wherever open bidding is impractical or undesirable:

• •Government procurement, where transparency and fairness are required
• •Construction tendering, where proposals are complex and not merely a price
• •Property sales in several countries, notably Scotland
• •Sales of mineral and spectrum rights, in some designs
• •Any situation where bidders must not know each other's identity
• •Cases where assembling everybody at one time is impossible

Deciding what to write is genuinely difficult and the theory of it earned a Nobel Prize. Offering exactly what the item is worth to you means that winning produces no gain at all, so every bidder shades their offer downwards, and how far to shade depends on how many rivals there are and on what they are likely to offer. More bidders means shading less, since the chance of being outbid rises. A variant in which the winner pays the second highest offer rather than their own removes the problem entirely, because bidding one's true valuation becomes the best strategy regardless of what anybody else does, which is an elegant result and is why that variant is used in several electronic marketplaces.

The format depends entirely on nobody seeing the offers before the deadline, and procurement rules go to considerable lengths to guarantee that. Offers arrive sealed or encrypted and are held unopened by somebody with no role in the decision. Opening happens at a stated time in the presence of witnesses, and in public procurement frequently in front of the bidders themselves. Late offers are refused absolutely, since accepting one would let a bidder learn the others first. Electronic systems timestamp submissions cryptographically and cannot decrypt anything before the deadline. Collusion between bidders defeats the whole arrangement and is a criminal offence in most countries, with detection relying on statistical patterns in bidding across many contracts.

Where the item has an objective value that nobody knows precisely, such as an oil lease or a company, the winner faces a specific trap. Every bidder estimates the value and those estimates scatter around the truth, and the bidder who wins is the one whose estimate was highest, which means the winner is systematically the person who overestimated most. Winning is therefore evidence of having been too optimistic, and bidders who do not account for it lose money consistently, which was documented in oil lease bidding in the 1970s and gave the phenomenon its name. Correcting for it means shading offers downwards by an amount that increases with the number of bidders, which is the opposite of the intuitive response.

Bidders submit one offer blind, so nobody learns from anybody and there is no escalation, which changes strategy completely. Offering one's full valuation guarantees no gain, so everybody shades downwards by an amount depending on the number of rivals. Paying the second highest offer instead removes that problem. Where value is objective but unknown, the winner is systematically whoever overestimated most.