Federal Reserve Research: Cryptocurrency investors adhere to their beliefs and adjust strategies wit

Cleveland Fed Research: The behavior of cryptocurrency investors depends more on beliefs than demographic characteristics

A recent working paper released by researchers at the Federal Reserve Bank of Cleveland pointed out that many abnormal behaviors in cryptocurrencies may stem from the way people form beliefs about digital assets, rather than standard demographic or financial factors. Through analysis of household survey data and random information experiments, the author found that expectations of future cryptocurrency returns are closely related to who holds cryptocurrency, and understanding the recent performance of Bitcoin can significantly change people's planned configurations and actual purchasing behaviors.

This item is titled "Do you understand cryptocurrencies? Research in "Cryptocurrencies in Home Finance" also reveals a mechanism that may explain the continued volatility of cryptocurrencies: when prices rise, they may strengthen bullish expectations and attract more buyers, thereby further amplifying price fluctuations.

Core found that

Return expectations explain cryptocurrency holdings more than demographic characteristics. The paper found that expected returns and perceived risk explain differences in holdings more than age, income or gender.

Many non-holders lack a clear belief in the return of cryptocurrencies. In the 2021 Fed survey, most non-holders said they did not know what the expected return would be in the coming year.

Information about recent Bitcoin performance can change behavior. In a randomized trial in 2025, households were shown Bitcoin's return over the past 12 months, and their expected cryptocurrency allocations and subsequent purchases increased.

Cryptocurrency gains may be regarded as "gambling income" rather than lasting wealth. The study linked the increase in Bitcoin prices to increased spending on durable goods, but found no continued growth in consumption.

Beliefs, not background, determine who buys cryptocurrencies

Researchers-Michael Weber, Bernardo Kandia, Olivier Koiben and Yuri Gorodchenko-used duplicate survey data covering as many as 25000 U.S. households in each round. Their core finding was that people's perceptions of cryptocurrency returns were far superior to observable characteristics in explaining holding behavior.

According to the paper, expected returns are particularly influential: for every 1 percentage point increase in an individual's expected return on cryptocurrencies, the probability of holding cryptocurrencies increases by 0.8 percentage points. When return expectations are combined with risk expectations, the explanatory power is further improved-exceeding the influence of factors such as age, income, and gender.

This makes cryptocurrencies an anomaly compared to traditional asset classes such as stocks, bonds and gold. For these markets, demographic and financial characteristics often have a greater impact on holding patterns than differences in return expectations. The paper concluded that participation in the cryptocurrency market depends more on "what do you believe in cryptocurrency" than on "who you are."

Cognitive gaps may exacerbate cycles of fluctuations

Research also highlights the uneven distribution of knowledge and beliefs among populations. In the 2021 Federal Reserve survey cited by the author, 87% of non-holders of cryptocurrencies said they did not know what the expected return would be in the coming year. Among cryptocurrency holders, the proportion remains high-54% say they do not know the expected return.

Studies describe significant differences among people willing to make predictions. Cryptocurrency holders expect an average return of 22% over the next year, compared with just 7% for non-holders. Holders also tend to think that cryptocurrencies are less risky than non-holders think.

The authors link this belief gap to a potential self-reinforcing market dynamic. If rising prices reinforce optimistic expectations, those expectations attract new players, whose buying behavior in turn supports higher prices. In the paper's words: "Positive returns attract new players, driving prices up further."

It is worth noting that this mechanism does not require underlying asset fundamentals to be the only driver of price changes. Instead, it suggests that divergence and learning-that is, how investors update beliefs based on past results-can be a major source of volatility.

Experimental evidence: Demonstrating Bitcoin's past performance changes configuration and purchasing behavior

For market participants, the most actionable part of the paper comes from a random information experiment. In 2025, researchers randomly assigned families to different groups and showed them information on how Bitcoin compares to other topics, including stocks, game stations and inflation. Participants who saw Bitcoin's return over the past 12 months increased their expected cryptocurrency portfolio allocation by approximately 2 percentage points.

The authors report that this represents an increase of approximately 47% relative to the expected configuration of 4.3% in the control group. They also found that subsequent actual cryptocurrency purchases increased by approximately 2.5 percentage points.

The key is that this effect is concentrated among people who say they don't own cryptocurrencies due to a lack of sufficient information. Those who already believe that cryptocurrencies are a bad investment have not shown the same reaction to information processing.

For readers, this experiment means that retail demand may not be driven solely by price headlines or broad narratives. It may also depend on what people are directed to focus on-specifically, whether they are provided with near-term performance data that can reshape future return expectations.

What happens to consumption when cryptocurrency wealth grows?

In addition to trading behavior, the paper also examined whether cryptocurrency gains translate into broader spending patterns. The authors report on the spillover effects of changes in cryptocurrency prices on household consumption, especially consumer durables. When the price of Bitcoin doubled, a household whose entire financial assets were allocated to cryptocurrency was more likely to purchase consumer durables by 1.4 percentage points.

They described it as an increase of approximately 7% relative to the probability of unconditional purchases. However, the effect did not persist on ordinary spending, a pattern that elicited a sharper interpretation: cryptocurrency gains seem more like "gambling income" or lottery bonuses than stable, confidence-enhancing long-term wealth growth.

This difference is important because it suggests that the impact of cryptocurrencies may be intermittent. Even if price surges boost certain consumer categories in the short term, they may not reshape households 'long-term financial behavior in the same way as more stable forms of wealth.

Implications for investors and builders

For investors and builders, the core warning of the study, while disturbing, is practical: If participation is belief driven and information sensitive, then volatility may remain structurally high. Readers should pay attention to how new retail entrants interpret Bitcoin's recent performance, and whether changes in public information delivery or feedback information access channels will amplify the feedback loop described in the paper.