The Federal Reserve Bank of Cleveland just released a working paper that should terrify anyone who believes Bitcoin adoption is driven by rational assessment of utility. The code does not lie, but it often omits. This study, however, is not about code. It is about the raw, unfiltered psychology of retail investors, and the findings are as cold as they are damning. The central mechanism is simple: show a potential investor a positive price chart, and their probability of buying increases by 2.5 percentage points. That is the entire engine of new demand. Not scaling solutions. Not institutional custody. Not regulatory clarity. A number on a screen.
I have spent the last decade auditing smart contracts and tracing on-chain flows. I have seen how narratives are built and how they collapse. But this study, based on a randomized controlled trial using the Nielsen Homescan Panel, strips away all the technical veneer. It isolates the variable that matters most in a market devoid of fundamentals: expectation. And the conclusion is that Bitcoin's growth is not a story of technological triumph, but a textbook case of behavioral response to a price signal. Zero trust is not a policy; it is a geometry. And the geometry of this market is a feedback loop where price creates expectation, and expectation creates buyers.
The Experiment: A Controlled Burn
The study, authored by Olivier Coibion and Yuriy Gorodnichenko, two heavyweight macroeconomists, did not ask people how they felt about crypto. They ran a randomized controlled trial. They split thousands of households into groups and fed them different information. One group saw the 12-month return for Bitcoin (14.3%). Another saw S&P 500 returns. A control group saw nothing. Then they measured the change in self-reported probabilities of holding crypto over the next 12 months.
This is the gold standard of causal inference. It is not a survey asking for opinions; it is a lab experiment measuring reactions. The results are stark. The group exposed to the high Bitcoin return increased their holding probability by roughly 2 percentage points (from a base of 4.3%). The S&P 500 information had a spillover effect, also increasing crypto interest, but the Bitcoin signal was the primary driver. The researchers found that the effect was strongest among participants who admitted they knew little about crypto. The less you know, the more a single data point moves you.
This aligns with my own experience auditing projects during the 2021 NFT bull run. The most vulnerable users were not the degens; they were the newcomers who saw a profile picture sell for 100 ETH and assumed the technology was the reason. The price was the argument. The code was irrelevant. Compiling the truth from fragmented logs is impossible when the logs are empty.
The 12% Ceiling: The Marginal Cost of New Believers
Here is where the data gets uncomfortable for the maximalists. The study tracks ownership rates over time. In 2021, roughly 3% of households held crypto. By 2022, that number had exploded to 11%. By mid-2023, it peaked around 12%. In 2025, with Bitcoin trading above $120,000, the ownership rate is... 12%. It did not move.
Think about that. The price went from $60,000 to $120,000, and the adoption rate flatlined. This is the tell. The 'wealth effect' is real, but it is hitting a wall. The study shows that the gap between holders and non-holders' expected returns is narrowing (13.8% vs 4.7% in 2025, down from 22% vs 7% in 2021). This suggests that the pool of people who can be convinced by a price chart is shrinking. The marginal new investor is harder to find, or they are already in.
I saw this exact pattern with the EigenLayer restaking model. The narrative was 'shared security' and 'yield amplification.' The code was complex, but the incentive was simple: high APY. Early adopters piled in. But when the risk of slashing became apparent, the inflow stopped. The price of the token dropped, and the narrative shifted from innovation to risk management. The technology did not change; the expectation did. Security is the absence of assumptions, and the market assumes price trends are permanent.
The Silent Variable: The Uninformed Majority
The study reveals a crucial demographic detail: approximately 40% of non-holders say they know 'not much' or 'nothing' about crypto. This is the battleground. The study shows that these low-knowledge participants are the most responsive to price information. They are not evaluating the technical merits of Bitcoin versus Ethereum. They are reacting to the number. This is not adoption; it is speculation. It is a lottery ticket purchase.
This creates a structural vulnerability. The 88% of non-holders are not a wall of future demand; they are a reservoir of potential panic sellers. The same mechanism that drives them in on the way up (a rising price) will drive them out on the way down. The study even notes that a significant portion of the new investments are coming from checking and savings accounts—money that is usually idle. This means Bitcoin is not just competing for risk capital; it is cannibalizing the safety cushion of households. When that cushion is threatened, the exit door will be narrow.
During the Axie Infinity Ronin bridge hack, I warned about the weak validator thresholds. The community downplayed it because the price was high. When the $625 million was drained, the price collapsed, and the users who had poured their savings into Axie's SLP token were left with nothing. The mechanism was identical: price-driven adoption, followed by a systemic shock. The Fed's data confirms that this pattern is not an anomaly; it is the core operating system of the retail crypto market.
The Contrarian View: What the Bulls Got Right
Despite my cold reading of this data, the bulls have one point that is empirically valid: the effect is real. The study proves that high returns do attract new investors. The 2-percentage-point increase is statistically significant (p=0.017). It is not a rounding error. It is a confirmation that the 'get rich' narrative works. In a sideways market, this is the only narrative that matters.
Furthermore, the spillover effect from the S&P 500 information suggests that overall risk appetite is rising. Crypto is not stealing from stocks; it is expanding the total pool of risk capital. This is bullish for the long-term infrastructure, even if it is bearish for the individual latecomer. The study also shows that expectations are anchored. Despite the price crash in 2022, the ownership rate did not fall back to 3%. It held at 12%. This indicates a 'ratchet effect'—once people are in, they tend to stay, even if their expectations are not met. This is a powerful stabilizing force.
However, this stability is a double-edged sword. It means that the next bull run will not be driven by new entrants hitting a 20% adoption rate. It will be driven by the 12% holders doubling down. This is a more fragile dynamic. It is not about expanding the base; it is about extracting more from the existing base. This is where the risk of systemic failure lies. If the 12% are over-leveraged and their expectations are not met, the correction could be brutal. The code does not lie, but it often omits the leverage.
The Regulatory Signal
The fact that the Cleveland Fed is studying this is not an accident. It is a signal. They are not asking if Bitcoin is a security; they are asking how it behaves. This is the first step toward a policy framework based on empirical evidence, not ideology. The study's conclusion that expectations are the primary driver of demand will likely be used to justify investor protection rules. If the Fed determines that the 'uninformed' are being lured by price charts, they may implement measures that require more prominent risk disclosures on exchanges or even restrict marketing based on past performance.
From my perspective, this is overdue. The audit reports I write are technical, but the vulnerabilities are often human. We audit the code for reentrancy, but we should also audit the incentive structures that lure in the uninformed. The Fed is doing that audit now. The finding that most new money comes from savings accounts is a red flag for systemic risk. When the price drops, the Fed will not blame the code; they will blame the market structure. And they will have this paper to cite.
The Takeaway
The takeaway is not that Bitcoin is a fraud. The takeaway is that the current adoption model is a self-limiting prophecy. We are at 12% ownership, and the price is at $120,000. The next doubling of the price will not bring us to 24% ownership. It will likely bring us to 13% or 14%. The pool of uninformed investors is finite. The market is reaching a point of saturation where the marginal dollar is chasing the same 12% of holders.
As a security auditor, I look for the point of failure. The failure here is not in the blockchain; it is in the expectation model. The Fed has handed us the evidence. The market is driven by the uninformed, funded by savings accounts, and activated by a single number on a screen. This is not a foundation for a new financial system. It is a structure built on a price chart, and charts are notorious for reversing.
Zero trust is not a policy; it is a geometry. And the geometry of this market is a pyramid that has just stopped growing. The question is not whether the price will go up again. The question is who will be left holding the savings account money when the chart finally breaks. The code does not lie, but it often omits the exit strategy. The Fed just documented the entry strategy. Read the paper. It is the most honest thing I have seen from a central bank in years.