The Fed's Bitcoin Experiment: Price Hype Works, But the Ledger Shows a Stalled Base

The Federal Reserve Bank of Cleveland just ran the largest behavioral experiment on Bitcoin investors ever conducted. The result is a cold, quantitative confirmation of a mechanism every trader already knows: price action drives expectations, and expectations drive new capital. But the data hides a brutal inflection point. The US household holding rate is stuck at roughly 12%, despite Bitcoin blowing past $120,000. The marginal new investor is becoming harder to acquire, and the Fed's own numbers expose the asymmetry between the bull case and the actual adoption curve. When an institution like the Cleveland Fed drops a working paper, you don't read it for the poetry. You read it for the code. And the code here is the experimental design. This isn't another survey where you ask people if they like crypto. This is a randomized controlled trial. The researchers took the Nielsen Homescan Panel, a dataset covering tens of thousands of US households, and randomly split participants into different information groups. Some got told about Bitcoin's past returns. Some got told about the S&P 500. Some got nothing. Then they measured the change in expectations and the actual shift in asset allocation. This is the gold standard for causal inference. The economists involved—Olivier Coibion and Yuriy Gorodnichenko—aren't just any researchers. These are heavyweights in the macro expectations field, known for their work on inflation expectations and how households form views about the future. When they run a randomized trial on Bitcoin, it's not a footnote. It's a data point that suggests the Fed is actively mapping the psychology of crypto market participants. Let me get into the core mechanics of what the experiment found. The most important number is the expected return gap. Bitcoin holders in 2025 expect a return of 13.8%. Non-holders expect 4.7%. That's a 9.1 percentage point chasm between the faithful and the skeptics. In 2021, that gap was even wider. Holders expected 22%, non-holders expected 7%. The narrative was hotter, but the base was smaller. Only 3% of households held Bitcoin in 2021. That number surged to 11% in 2022, then plateaued around 12% through the 2025 rally. Here's where the analysis gets cold. The experiment showed that past 12-month returns of 14.3% in Bitcoin were enough to boost allocation intent by roughly 2 percentage points against a control group average of 4.3%. That's not nothing. But it's not a flood. A 2-point bump in allocation intent from a 14% annual return is the sign of a market that is maturing, not an exponential adoption curve. The days of 3% to 11% penetration in a single year are over. The marginal non-holder today isn't someone who needs a nudge. They are someone who needs an education. The paper finds that expectations and perceived risk explain twice as much of the holding decision as demographics. That means the story isn't about age or income, though those matter—40-year-olds are 13 points more likely to hold than 60-year-olds. The real story is narrative. The bullish narrative sets expectations. Expectations set allocation. And allocation sets the price. This is a feedback loop. The Fed just quantified it for the first time. Here is the contrarian angle the market doesn't want to hear. The Fed didn't run this experiment to give you a bull case. They ran it to understand the transmission mechanism. Why does this matter? Because if the Fed is studying how price shocks affect household expectations, it is studying how a future price crash will affect household expectations. This is not a marketing handout. It's a risk assessment. Consider the source of the new capital. The paper notes that most of the additional allocation comes from checking accounts, savings accounts, or cash. Bitcoin isn't cannibalizing the S&P 500. It's pulling money out of zero-yield or low-yield storage. That sounds bullish in a vacuum, but it is actually a sign of vulnerability. The money in checking and savings is sticky money. It's the buffer for real life. When a household moves their safety net into a volatile asset, they are doing so because the expectation of future returns has overwhelmed their perception of risk. The Fed's research shows this is a psychological phenomenon, not a financial one. If expectations reverse—and in a bear market they always reverse—the flow doesn't just stop. It reverses too. Money that flows from the balance sheet flows back to the balance sheet. The paper's own data shows that the holding rate dipped in 2023-2024 before recovering in 2025. The cycle is real. There's another hidden gem in this paper that most commentators will miss. The Fed found that participants with limited knowledge of crypto showed the strongest response to price information. The uninformed are the most reactive to the recent past. This is not a bullish signal for market stability. It means the marginal buyer is often the least informed. They are buying because the price went up, not because they understand the infrastructure. The paper calls this an 'enthusiasm spillover'—those who received information about the S&P 500 were also more likely to hold crypto. This suggests that a broad risk appetite can contaminate asset classes. But the institutional takeaway is clear: if you are a whale, a market maker, or a quant, the play is not to follow the narrative. The play is to front-run the narrative. You don't need to believe the Fed's paper to see the gap. Let's talk about the actual data points for the key levels. The research confirms that the 'wealth effect' of Bitcoin is real but diminishing. At 12% penetration, there is still a huge pool of 88% non-holders. But the study suggests that simply showing a 14% return to an uninformed household only shifts allocation by 2 percentage points. That means the cost of customer acquisition in the market is rising. The next wave of adoption will not be driven by price. It will be driven by knowledge. The barrier is not capital. It is ignorance. Around 40% of non-holders in the panel admit they don't know much about crypto. The experiment showed these individuals are the most reactive to price information. But their reaction is not an intellectual conviction. It is a reflexive response. The more information they have, the more they behave like the market. This is an instability risk. Let me be clear about the broader market structure. Bitcoin is trading above $120,000. The paper is a working paper, not a final publication. It has not been fully peer-reviewed. The Fed authors themselves explicitly state that the views are theirs and not the Federal Reserve Bank of Cleveland or the Federal Reserve System. Do not mistake this for an official institutional endorsement. This is a data point. It is not a policy. But why would the Fed be running a randomized controlled trial on Bitcoin expectations? Because they are building a framework. If expectations are the driving force behind crypto allocation, then expectations are the target for future regulation or communication. The Fed is building a toolkit to understand the psychological impact of price movements. The next step could be warnings, education campaigns, or even tighter regulatory frameworks around leverage and leverage. The report's own data suggests that the 'self-reinforcing' nature of Bitcoin's bull market is a double-edged sword. Price rises create expectations, which create new investors, which push price higher. But in a bear market, the loop works in reverse. Price drops. Expectations collapse. Holders exit. Price drops further. The paper shows this through the timeline of adoption: 3% in 2021, 11% in 2022, a dip in 2023, and then back to 12% in 2025. The market's memory is short. The exit door is just as wide as the entrance door. So what is the takeaway for a strategist? We have a paper from an elite academic team showing that a 14% return can move allocation by 2 points. We have a holding rate that is stagnant. We have a flow of capital from savings accounts. And we have a 40% of the population with no knowledge base. The market is not on a linear path to 50% penetration. It is in a liquidity grab phase. The winners will be the ones who recognize that price is not a product. Expectations are the product. We are in a bull market, and the machine is working exactly as the Fed describes it. But the machine is also dangerous. When the code bleeds, the ledger keeps the truth. The black box of price data is starting to show its constraints. The next leg of the bull market will be built on education and infrastructure, not just a price sticker. In the interim, the smart money is not chasing the market narrative. The smart money is positioning for the volatility the narrative creates. Is the Fed’s study a reason to buy? No. Is it a reason to respect the mechanic? Yes. The real signal here is not 'Bitcoin goes up'. The signal is that the expectation gap is the asset, and the asset is the chase. The execution is everything.