Mechanism Design in the real world

Mechanism design theory is highly abstract and theoretical. Nevertheless, the basic intuition behind the mechanism design theory is quite simple: the theory tries to provide us with the tools to find an optimal mechanism to allocate scarce resources efficiently. When agents have private information, the final outcome of a transaction between economic agents may not be efficient for the economy as a whole. An example of this is the provision of a public good, such as street lighting. If market agents have private information about how much they value this public good, they have an incentive to pretend to be relatively uninterested in order to reduce their own share of the provision cost (the so-called free-rider problem). In this case there may be no provision of the public good at all because the required revenues cannot be raised.

Mechanism design tries to overcome these breakdowns of the market mechanism. The art of mechanism design theory is therefore to make the revelation of private information incentive compatible, which means that market agents are strictly better off if they reveal their true preferences. In case of street lighting such a mechanism could look as follows: each person should be asked to report his own willingness to pay for the project and the project will be undertaken if and only if the aggregate willingness to pay exceeds the cost of the project. If the project is undertaken those individuals that are pivotal, i.e. can influence the overall decision, pay the difference between everyone else’s reported willingness to pay and their contributions to the project. It can be shown that truth-telling is optimal for agents in this framework.

A simple example can illustrate the so-called Groves-Clark mechanism. Suppose there are three individuals, A, B and C deciding whether they want to finance street lighting. The total cost of the project is EUR 3000 and the individuals are willing to pay EUR 500, EUR 500 and EUR 2500 respectively. It is hence efficient to introduce the project. The social planner (e.g. the government) might ask everyone to contribute EUR 1000 to the project. If the individuals report their true willingness to pay, only C is pivotal, that is, with C in the process the project goes ahead, with C out of the process the project does not go ahead. C hence pays a tax of EUR 1000. Individuals A or B have no incentive to understate their willingness to pay (in order to prevent the introduction of street lighting in this case, A would have to report a willingness to pay of EUR -500. Individual A would then be pivotal and pay a tax of EUR 1000). Individual C also has the incentive to report his true willingness to pay as over- or understating the true willingness to pay leaves the tax that he needs to pay unchanged.

Mechanism design hence examines how one can extract the private information of market agents, which often prevents the efficient functioning of markets and institutions. The influence of mechanism design theory can be seen in the structure of auctions, such as the UK government’s sale of four 3G mobile phone licenses in 2000. The auction employed an Anglo-Dutch design in which the price rose until all but five bidders had quit and the last five bidders then made ‘‘best and final’’ sealed bids with the winner paying the price of the fourth-highest winning bid. It can be shown that such an auction promotes efficiency, in terms of inducing bidders to reveal their true valuations.

e-Bay is another example, where basic mechanism design theory has been used to design a second-price sealed-bid auction, in which the highest bidder wins but pays the price of the second highest bidder. The second-price sealed bid auction is efficient in that the winner is the bidder with the highest valuation and it induces buyers to reveal their true valuations. The theory of mechanism design has also been used by e-Bay to implement a quality rating system, which allows honest sellers and buyers to build up a good reputation and which tries to prevent the breakdown of transactions due to uncertainty about the quality of the good that is being sold.

-Nobel Economics Prize 2007 – Research is relevant to practice


Related;
A theory of an intelligently guided invisible hand wins the Nobel prize

The Road to a System that Works (Without Shooting People)