Essays on Cryptocurrencies: Experiments Measuring Attitudes, Uses, and Mechanisms

Sarah Sylvester

Advisor: Kevin A McCabe, PhD, Department of Economics

Committee Members: Daniel Houser, Johanna Mollerstrom

Online Location, Zoom
March 26, 2024, 10:00 AM to 12:00 PM


This dissertation focuses on various aspects of cryptocurrencies, from the individual user’s side to specific mechanism designs. There are three main aspects of cryptocurrency markets that I study, consisting of nonowners, who may be open to using cryptocurrencies in the future; current owners of cryptocurrencies, who use various institutions to achieve profits; and cryptocurrency mechanisms themselves, which provide avenues for users to engage in cryptocurrency transactions. To understand these aspects, I conduct a variety of experiments analyzing individuals’ current perceptions of and willingness to own cryptocurrencies, as well as the mechanisms to trade, hold, and invest in cryptocurrencies.

Chapter One, co-authored with Johanna Mollerstrom, examines how different factors impact an individual's willingness to own cryptocurrencies. We analyze this through an online survey experiment with a diverse population and a laboratory experiment with incentivized measures and a smaller sample. We first find that cryptocurrency owners are typically young males with a high tolerance for risk. Secondly, to understand how a lack of knowledge is a barrier to cryptocurrency use, we provide participants in our survey with varying information treatments concerning cryptocurrencies and evaluate their willingness to own cryptocurrencies after receiving the information. Our results indicate that information conveying the ease of use of cryptocurrencies or the security features of cryptocurrencies is effective at increasing nonowners’ willingness to own cryptocurrencies. Treatments varying the source of the information, however, showed no difference, indicating that participants care more about the content than the source of the information. In our related laboratory experiment, we use the same measures as the survey but conduct an incentivized version, where participants earn bonus rewards in Bitcoin in the form of a paper wallet. We also measure whether receiving a small amount of cryptocurrencies can increase the willingness to own them in the future and find support for this. 

Chapter Two is a continuation of the laboratory experiment co-authored with Johanna Mollerstrom that looks at cryptocurrency use outside of the willingness to own them. In the laboratory setting, we conduct simple trust and risk games to determine how behavior is affected by the introduction of cryptocurrencies. Individual perceptions of cryptocurrencies can influence their adoption but could also impact actions with the currency once they are owned. It is possible that individuals view cryptocurrencies as akin to “dirty money," which can lead to more selfish, unfair practices compared to “clean money." Also possible is that since cryptocurrencies are separate from traditional money, they will avoid negative associations with money such as greed and corruption. We find evidence that participants act differently when interacting with Bitcoin compared to dollars. Participants engage in less risky behavior when playing with Bitcoin and are particularly risk-averse when they receive an initial bonus of cryptocurrencies. This indicates that mechanisms designed for interacting with cryptocurrencies should consider the difference in behavior that individuals have when using cryptocurrency assets instead of money.

Chapter three moves from individual use to the evaluation of a popular cryptocurrency trading mechanism. Co-authored with Kevin McCabe, Aleksander Psurek, and Nalin Bhatt, we conduct agent-based model simulations evaluating the efficiency of Automated Market Makers (AMMs). AMMs are a method of trading cryptocurrencies in a decentralized manner. This work helps us understand the current market and use of AMMs and determines the parameters and conditions that lead to inefficient transactions. Through the simulation experiments, we model a decentralized finance system where AMMs provide trading opportunities to individual agents. We specifically analyze arbitrage agents, who are traders that use token price differences in various markets to make trades, earning profits and theoretically equilibrating the prices of cryptocurrencies across markets as they do so. By nature of their incentives, arbitrageurs make no-risk profits, making it an attractive option for cryptocurrency traders. We provide the first analysis of arbitrage agents using simulations and explore how their behavior can influence an AMM’s prices and volatility. We find that AMMs with lower liquidity depths experience large price fluctuations away from tokens’ external market prices, and that AMMs without active arbitrageurs lead to long-standing price misalignment. These results indicate that liquidity, arbitrage activity, and additional features should be considered when using and creating AMMs.