Skip to main content

Frontiers in Mechanism Design: Methodology and Applications

Periodic Reporting for period 2 - Frontiers in Design (Frontiers in Mechanism Design: Methodology and Applications)

Reporting period: 2018-03-01 to 2019-08-31

Mechanism design is viewed as the engineering side of economics. The goal is to design institutions–contracts, markets (`mechanisms') –that have desirable outcomes, for example allocate private goods or coordinate the use of pollution permits efficiently, or achieve adequate redistribution of income. The search for the best institution is modelled as the search for the non-cooperative game whose equilibrium outcome function (the game’s outcome function is essentially a social choice function mapping characteristics to economic outcomes) is best given an objective. The game can capture a dynamic relationship that lasts indefinitely. In this process the designer is selecting the `rules' of the game--the institution--and the choice is final and committed forever.

The standard mechanism design paradigm relies on two key fundamental assumptions: 1. The designer of the institution—the `principal'—does not have any privileged information. 2. The designer chooses the mechanism and commits to it once and for all. These assumptions often fail in today’s “big data” world: Firms (online retailers, insurance companies, banks) do have privileged—and often certifiable —information that may affect the contractual terms that they propose. Also, they interact repeatedly with the same agents, and, as they learn about them, they attempt to change the terms by making personalized offers. Finally, a mechanism often interacts with outside markets—e.g. a government insurance program interacts with private insurance markets. The performed research provides methods and foundations to design optimal mechanisms for precisely those highly relevant situations. In the process applications of the methods are developed in the design of online marketplaces when consumer privacy is at stake and in the design of tests (such as emission tests) when firms may have incentives to manipulate test results.
The research output so far includes the following conceptual innovations:

The PI has developed a `revelation principle’ to study contract design when the principal lacks commitment and is free to propose any mechanism at the beginning of each period. This fundamental tool will allow researchers to address practical questions when designers have less than full commitment to future contractual arrangements: These include the design of social insurance programs, monetary policy and contracts that govern international trade.

The PI has made a further methodological advancement in the area of mechanism design with limited commitment. She provides a formal connection between contract design under imperfect commitment and a game with a time inconsistent player. This formalisation captures the key tension that lack of commitment brings and can be used to derive the costs of lack of commitment. Such calculations should be take into account when public officials are tempted to change the rules.

The research performed thus far includes the formulation of a framework to study test-design in the presence of manipulations and cheating and the derivation of optimal tests that make cheating unprofitable.

Finally, the PI was developed an analytical framework to study the design of selling protocols when firms that superior information vis a vis their customers. This framework allows one to assess whether or not sellers can exploit their informational advantage with the risk of harming consumers. The earlier works had suggested that if sellers can provide evidence of their product (give a sample to consumers or show a certificate of quality) equilibrium forces will favour information revelation and thus information will unravel. The new research suggests that this insight may be an artefact of restrictive assumptions on what procedures sellers can use. The new research shows that result is overturned if firms can use more elaborate contractual arrangements as many do in practice (think for instance tech companies, telecommunication companies, insurance companies etc) and in some cases sellers can indeed use their informational advantage to exploit their customers.
All aforementioned innovations are beyond the state of the art of mechanism design theory. Given that mechanism design theory is a central area of modern economic theory that has not only made a big impact of how institutions and contracts are designed, but has also had a big impact on computer science and operations research, we expect and hope the results of this research to have spill overs to those disciplines.