Decentralized Finance and its Role in Machine Economies

Technological advancements over recent decades have increasingly changed not only how humans interact with machines but also how machines interact with one another. Value generation is increasingly shifting into the virtual realm, allowing machines to autonomously make decisions and run almost entirely with low levels of human oversight, forming what can be described as ‘machine economies’. While academia has made progress in analyzing the separate technological building blocks that allow machines to communicate and interact, it remains unclear how machine economies might differ from human economies, let alone if established economic principles can be applied to machine economies and if they can avert failures preventing human economies from reaching Pareto efficiency. Against this backdrop, this dissertation develops a framework to characterize machine economy types based on their interaction and governance before drawing on Coase’s theorem to identify economic criteria sufficient to achieve Pareto efficiency and develop them into technical requirements. This dissertation identifies that information symmetry, transaction cost, and governance mechanisms are some of the more significant challenges in pursuing efficient machine economies, particularly those with lower levels of human involvement. These issues are also central talking points in other recent trends, blockchain and decentralized finance (DeFi). As such, building on eight research papers and two book chapters, this cumulative dissertation explores how DeFi can help facilitate efficient machine economies. It concludes with observations for each machine economy type, guiding stakeholders interested in setting up and maintaining efficient machine economies.