Is Bitcoin a safe haven or a risky bet? People have argued about that for fifteen years without settling it. The reason they cannot settle it is that they are asking about the wrong thing. It is not a property of the asset. It depends on what has run out.
The treatise made this argument in 2026 on a thin record. This piece marks it against the evidence, including one recent episode that does not fit, and updates the scoring table accordingly. Conditional, falsifiable, not investment advice.
The question almost always gets put the wrong way round. Is crypto a safe haven or a risk asset? Is Bitcoin digital gold or a leveraged bet on cheap money?
Both camps line up their episodes. Both are sometimes right. And the argument never ends, because it assumes the answer is a fact about the asset.
It is not. Crypto has two opposite jobs. Which one it is doing on any given day is decided by the thing that decides everything in this framework: what has actually run out.
The two flights
The first flight runs toward the dollar. In a panic, when everyone scrambles for cash at once, the scarcest thing on earth is dollars. For an unanswerable reason: debts are written in dollars. A margin call, a maturing bond, a mortgage payment. Each one is an obligation you can settle only one way.
So anything sellable gets sold to raise dollars. Crypto is the easiest thing in any portfolio to sell, because it trades on weekends and never closes. It goes down first and hardest. That is not the hedge failing. That is the hedge being the most liquid item in a fire sale.
The second flight runs away from the dollar. This happens when what is in doubt is not the supply of dollars but the worth of them. A central bank bent to political will. A deficit paid for with the printing press. Here what has gone scarce is not dollars but trust in dollars.
And in that world people want something a government cannot print. Crypto goes up.
The same asset, the same chart, means opposite things depending on which shortage is on top. That is not a hedge working or failing. That is one instrument reading two completely different diseases.
Scoring it against the record
Episode
Which scarcity was binding
What crypto did
Verdict
March 2020
Dollars. A textbook dash for cash as the pandemic hit and every leveraged position sought liquidity at once.
Fell harder than equities, exactly when the haven property was being tested for the first time at scale.
Fits
Mid 2020 to 2021
Trust. The rescue was read as debasement, with the money stock rising at rates without postwar precedent.
Violent rally on the debasement narrative. The same asset, the opposite direction, within months.
Fits
2022
Dollars. Rate rises forced a system-wide deleveraging.
Crypto fell hardest of any major asset class, and TerraUSD, a claim on dollars that was not backed by dollars, failed outright.
Fits
March 2023
Trust, narrowly. Not the dollar itself but the banks holding it, as three lenders failed in a week.
Bitcoin rose roughly a third over the week of Silicon Valley Bank’s failure, while USDC broke its peg on reserves stranded at that same bank.
Fits
October 2025
Neither, on the framework’s own gauge. A tariff headline, not a liquidity event.
A $19 billion liquidation cascade, Bitcoin down roughly 14 percent in a day. Behaved as leveraged beta, not as a hedge against anything.
Partly
2025 as a whole
Trust should have been the story. Deficits wide, the central bank under open political pressure, the debasement narrative loud.
Bitcoin’s correlation with software equities reached +0.64 by the fourth quarter, the tightest since 2022. The market did not price it as a monetary hedge.
Does not fit
The one that does not fit
The second flight predicts that when the dollar’s credibility comes into question, money runs to something outside the dollar system.
Through 2025 the conditions were about as good as they have been in decades. Enormous deficit. Central bank under open political pressure. Debasement discussed everywhere. And crypto did not do what the theory says it should have done.
An honest framework has to sit with that rather than explain it away. Three readings are available, and they do not exclude one another.
One. There is a threshold and it was not crossed. Worrying about debasement is not debasement. A currency being criticised in opinion columns is still the thing everybody prices in, and until it visibly stops being that, nobody has a practical reason to leave.
Two. The owners changed. The asset that rallied on debasement fear in 2021 was held mostly by people who believed a story about money. The asset held in 2026 sits in exchange-traded funds, on institutional balance sheets, inside portfolios measured against stock market benchmarks. Those holders sell when stocks fall, because that is what their mandate says. The asset did not change. Its owners did.
Three, and this is the one this framework finds most interesting. In July 2025 the GENIUS Act created federal rules for stablecoins. That market has since grown past $321 billion, and roughly 99 percent of it is denominated in dollars and backed by Treasuries.
So somebody worried about the dollar now has a new option. Something that feels like an exit, is regulated, is liquid, pays interest through its reserves, and is made entirely of dollars.
The state did not close the exit. It built a licensed one that leads back inside, and the outflow took it.
That is a mechanism the treatise did not see coming, and exactly the sort this framework should be able to name once it appears. The treatise argued that stablecoins do not escape the dollar, because a claim on dollars is a bet that dollars stay available. That is still true, and still why they are fragile in a real panic. But it turns out to be beside the point during a slow debasement, where what people want is not escape but somewhere to stand that pays them to wait.
Where the gauge points now
The framework’s own gauge gives a clear reading, and it explains why this period looks nothing like 2020 or 2023. The Credit Counterfeit Index sat at minus 1.31 in the spring of 2026, well below its line and deeply negative. Money is loose. Nothing has run out.
Which means neither flight is running. No dash for cash, so crypto is not being dumped. No visible break in the dollar, so nothing is fleeing toward it either. In that world crypto is simply a high-beta technology asset. Which is exactly how it has behaved.
This is an awkward place for a theory to sit, because it means the framework currently predicts the least interesting outcome available. A claim that only becomes testable in a crisis spends most of its life unable to be proven wrong.
The answer is not to go looking for a crisis. It is to say so plainly, and to write down in advance what each flight would look like, so nobody gets to reinterpret it afterwards.
−1.31
The gauge in spring 2026. Far below its line. Neither flight is running.
+0.64
How closely Bitcoin tracked tech stocks in late 2025. The tightest since the 2022 crash.
$321bn
The size of the stablecoin market, an all-time high, and roughly 99 percent of it dollars.
$19bn
Wiped out in one October 2025 cascade on a tariff headline. Borrowed money, not a view about the dollar.
The falsifiable claims
Claim
Trigger
Falsified if
Horizon
Crypto sells off first in a genuine dash for cash
Any episode where the Credit Counterfeit Index breaches 1.0 and holds for a quarter.
Crypto outperforms equities during a confirmed breach quarter, or falls less than the S&P 500 peak to trough.
next breach
Crypto catches an outflow from a visible loss of monetary trust
A disorderly fall in the dollar, a failed Treasury auction, or an overt monetisation of deficits.
Crypto falls or is flat through such an episode while gold rises, which would indicate the flight found a different destination.
by 2030
Stablecoins absorb what would have been the flight
Stablecoin market capitalisation against Bitcoin’s share of total crypto value during any period of dollar-credibility stress.
Stablecoin supply contracts during a credibility episode while Bitcoin rallies, the opposite of the absorption thesis.
by 2030
Stablecoins break rather than protect in a liquidity panic
Peg stability of the two largest tokens during a breach quarter.
Both majors hold peg through a confirmed dash for cash with no redemption gating or support intervention.
next breach
The institutional-ownership reading is right
Bitcoin’s correlation with equities across the next stress episode.
Correlation with equities falls back below 0.2 in stress without any change in the ownership base.
by 2029
The claim here is narrower than the one people usually argue about. Not that crypto is a hedge. Not that it is not. But that the question has no answer in the abstract, because the asset sits inside the ordering of scarcities rather than outside it. Ask which shortage is biting and the behaviour follows. Ask whether it is digital gold and you will be arguing forever.
And the genuinely new thing in the record is not that crypto failed a test. It is that the second flight now has somewhere else to go. Somewhere licensed by statute, denominated in the very currency being fled, and holding Treasuries against what it owes.
If that holds, then the treatise’s original claim survives in a strange form. Stablecoins really do not escape the dollar. They just turned out to be a very comfortable place to stand while worrying about it.