Ethereum Investors’ Bias Fuels Market Volatility Study Finds
A new study published in *Finance Sciences* (Polish: *Nauki o Finansach*) by researcher Jia-Ying Lyu from National Sun Yat-sen University in Taiwan examines how a well-known behavioral bias—the *disposition. . .
A new study published in *Finance Sciences* (Polish: *Nauki o Finansach*) by researcher Jia-Ying Lyu from National Sun Yat-sen University in Taiwan examines how a well-known behavioral bias—the *disposition effect*—influences Ethereum investors’ trading behavior between 2020 and 2024. The disposition effect refers to the tendency of investors to sell assets that have gained value more readily than those that have lost value, even when holding onto the latter might be the better long-term strategy. Using publicly available blockchain data and exchange transaction records, Lyu reconstructed the cost basis of over 1. 2 million Ethereum addresses on a daily basis. By applying statistical models, the study found clear evidence that Ethereum investors are about 12–18% more likely to sell tokens when they are in profit compared to when they are at a loss. This pattern persisted across different market conditions and wallet types. The research also reveals that this bias intensifies during periods of high price volatility. For example, during the sharp price swings of early 2022 and the Terra-LUNA collapse, investors became even more prone to realizing gains and cutting losses quickly. This behavior may contribute to increased market fragility during stress periods, as rapid selling can amplify price declines. However, the study identifies an important exception: major Ethereum protocol upgrades—such as the Merge (2022), Shapella (2023), and Dencun (2024)—appear to weaken the disposition effect. Around these events, investors showed slightly less tendency to realize gains prematurely, possibly due to increased attention, optimism about future value, or reduced uncertainty following successful upgrades. From a commercial perspective, the findings carry several implications. For cryptocurrency exchanges, understanding investor behavior can inform the design of user interfaces, educational tools, and risk warnings that discourage impulsive selling. Platforms could introduce features like “loss aversion alerts” or “hold reminders” during volatile periods to help users resist emotional trading. Regulators may also take note, as the disposition effect can lead to suboptimal investment outcomes and contribute to market instability. The study suggests that behavioral biases are measurable even in decentralized markets, reinforcing the need for investor protection measures that go beyond traditional financial systems. For DeFi protocols and dApps, the research highlights the importance of clear communication around upgrades and network changes. If protocol events can temporarily reduce biased behavior, well-timed announcements and transparency could be leveraged to improve investor confidence and long-term holding patterns. Finally, the use of on-chain data to study investor behavior represents a breakthrough in financial research. Unlike traditional finance, where data is often siloed within brokerages, blockchain transparency allows for real-time, verifiable analysis—opening new avenues for academic and commercial research in crypto markets. The study underscores that behavioral finance principles apply not only to stocks and bonds but also to decentralized digital assets, and that market participants who account for these biases may gain a competitive edge in product design, risk management, and user engagement. Read all about it