Project details

Responsible White
Fairness White
Algorithm White
Theory-driven White

Automated Reasoning for Economics

The design of fair mechanisms for group decision making is an important and challenging research topic in Economics. Examples for the need of such mechanisms are numerous. What is the right voting rule to elect the members of a municipal committee? What is the right procedure to assign students to schools? What is the right way of dividing limited funds between areas such as education and environment?

In this project, using tools from the field of automated reasoning, we develop algorithms to support economists in their quest to identify good mechanisms for some of these domains.

We do this by encoding the societal demands on such mechanisms as constraints in a logical language, which makes it possible to reduce the economic design task to a systematic search across the huge space of all logically conceivable mechanisms. The main tool we employ in our research are so-called SAT solvers, which are very fast algorithms to determine whether a given set of logical constrains can or cannot be satisfied.

Unfortunately, often the results obtained are negative, showing that it is mathematically impossible to design a mechanism of the desired kind. But even such negative results are important, as they provide clear guidance to economists for where looking for new mechanisms is likely to be most fruitful.

To date, we have been able to contribute new insights in specific areas of policy, such as voting for committees or matching job seekers to companies. We also have helped to further develop the methodology at the core of this emerging area of research, and we have enabled others to make similar contributions by providing advanced training courses.

Papers related to this project

Automating the Analysis of Matching Algorithms