Economics and computation (EconCS), also sometimes called algorithmic game theory (AGT), is an interdisciplinary field consisting of economists and computer scientists.
Economists have developed a rich and deep theory of individual behavior. Their models shed light on the design of optimal rules governing economic interaction under incentive constraints. Computer scientists, in turn, have developed a theory of computational thinking, driven by algorithms and complexity. This group works at the intersection of these two fields. Economics informs what outcomes algorithms can compute in the presence of self-motivated individuals. Computer science offers a theory of complexity and approximation, thereby both imposing realistic constraints on the space of solutions and relaxing the metric by which we judge them. Together, these interdisciplinary researchers can design and build markets and platforms with robust and socially-desirable properties.
Our theories revolve around pricing, matching, information, learning, and networks. We explore pricing policies and auction designs that efficiently compute approximately optimal outcomes in equilibrium even in the presence of constraints on the complexity of the solution. We study matching policies in dynamic or stochastic environments, both with and without the convenient tool of money. We study markets for information, developing tools for a market maker to solicit beliefs from individuals in order to make predictions about the world. We ask how individuals learn, and how a centralized platform might coordinate the exploration and exploitation of these individuals to minimize the regret induced in the process. And we think about the social networks in which all these individuals interact, and how the network structure impacts outcomes.
Our theories shed light on many of the traditional economic applications, including spectrum auctions, organ transplantation, school choice, and the like. But of particular interest to our group are the new online markets and meeting places, often designed and built at computer science companies like ours. In retail markets, consumers purchase household products from the comfort of their own homes at online retailers like Amazon, giving rise to questions about optimal pricing policies. In labor markets, drivers are hired through apps like Uber and Lyft; unskilled laborers can be hired in TaskRabbit; and short-term skilled labor contracts are available through platforms like Up Work and Mechanical Turk, giving rise to questions about matching policies. Markets for reviews, like Yelp, give rise to questions about social learning. And social media platforms like Facebook and Twitter give rise to questions about the impact of network structure.
Together, we are building a combined theory of economics and computer science to guide the development of tech-enabled modern markets and platforms.