The Federal Reserve Bank of Cleveland published a working paper. The image is innocent: Bitcoin at $120,000, holding rate at 12%. The metadata confesses: the price-expectation engine is sputtering.
Context: The Experiment
This is not a survey. It is a randomized controlled trial (RCT) embedded in the Nielsen Homescan Panel—a dataset tracking tens of thousands of U.S. households. Researchers randomly assigned participants to receive historical return information for Bitcoin, the S&P 500, or GameStop. Then they measured changes in expected returns and asset allocation decisions. The experimental design is the gold standard for causal inference. In my years of auditing smart contracts and tracking on-chain liquidity, I have learned that correlation is a trap. This study is a rare attempt to isolate cause from noise.
Core: The Evidence Chain
The study’s three key findings form a coherent chain:
- Price information shifts expectations asymmetrically. Participants who saw Bitcoin’s past 12-month return (14.3%) raised their expected future return by approximately 2.5 percentage points. The effect was strongest among those who previously knew little about crypto. Tracing the ghost in the machine: the marginal investor is the least informed.
- Holding rate is sticky at 12%. The share of U.S. households holding Bitcoin jumped from 3% in 2021 to 11% in 2022, then stabilized around 12% through 2025—even as price tripled from $40k to $120k. The velocity of new adoption is collapsing. In 2020, I built a Python script to track liquidity decay in DeFi pools. The same pattern appears here: the marginal utility of price appreciation as a marketing tool is declining.
- The money comes from savings, not speculation. The study shows that most of the new allocation is drawn from checking accounts, savings accounts, or cash. This is not a rotation out of stocks or gold. Bitcoin is expanding the total risk pool, but at a glacial pace. Yields decay, but the logic remains immutable.
Contrarian: The Limits of the Wealth Effect
Many market participants will interpret this study as bullish: “Bitcoin’s price rise attracts new investors.” That is true, but the magnitude is underwhelming. The treatment group increased Bitcoin allocation by only 2.5 percentage points—from 4.3% to 6.8% of a hypothetical portfolio. That is a small shift for a 14.3% return signal. The effect is not zero, but it is far from the euphoric “FOMO” narrative.
More importantly, the study cannot predict how this effect scales. Does a 30% return produce twice the new demand? Or does the marginal response diminish further? The authors note that “the study cannot determine whether each Bitcoin price increase generates the same level of new demand.” That is a polite way of saying the relationship is nonlinear and likely saturating.
Forensic architecture reveals the architect: the Federal Reserve is not endorsing Bitcoin. It is building a data-driven framework to understand how speculative assets embed into household balance sheets. The working paper status, with its disclaimer that it does not represent Fed policy, is a deliberate signal. The study is a reconnaissance mission, not a policy brief.
Takeaway: The Next Signal
The next week’s signal is not the price. It is the holding rate. If the 12% threshold remains unbroken as price consolidates or rises, the bull thesis shifts from “new money enters” to “existing holders refuse to sell.” That is a different macro regime—one that favors liquidity providers and options sellers over directional traders. The metadata never forgets.
