Bitcoin, Game Theory, and Mechanism Design: A Research Agenda

The security of Bitcoin relies crucially on the stability of its consensus protocol. While the system has worked well in practice so far, it doesn’t yet work in theory. We don’t know if the system will continue to be stable in the long run, facing a dwindling minting reward and a potential block size increase. What knowledge we do have is a result of fortuitous observations rather than a complete model of miner behavior allowing systematic analysis. Besides, the theory has often predicted mining strategies that haven’t been observed in practice, for reasons that aren’t yet clear. Attempting to fill these gaps leads to many interesting research directions in game theory and economics.

An entirely different set of research directions arises from the fact that cryptocurrencies offer a platform for mechanism design. This platform has powerful new capabilities: for example, creating money out of thin air is a tool available to the mechanism designer. The decentralized model also presents new constraints: there is no trusted party (such as an auctioneer) with private memory and private communication channels to other parties. This is a strange new world for algorithmic mechanism design, and there is much work to be done both in terms of new abstractions and specific applications.

All scheduled dates:


No Upcoming activities yet