
Two very different paths: weaken the dollar — or change the money itself. Neither is comfortable, and one of them we have already tested.
$40 trillion in debt.
Seriously — how do you get out of that?
We started looking at it. There are a few ways. None of them are particularly comfortable, and the one everyone in crypto likes best is the one our own data has been quietly voting against all year.
That is the total public debt of the United States on 17 September 2026, straight from the Treasury's own ledger; $32.4 trillion of it is held by the public rather than by government trust funds. It crossed $36 trillion at the start of 2025 and $39 trillion at the start of this year. Source: Treasury Debt to the Penny; FRED GFDEBTN.
The number itself is not the problem. The cost of carrying it is. Federal interest payments were running at $1.25 trillion a year in the second quarter of 2026 — up from $1.16 trillion a year earlier and about $350 billion in 2021. Debt is 122.6 % of GDP. The deficit in fiscal 2025 was $1.78 trillion, so the pile grows while the interest on it compounds. Source: BEA via FRED A091RC1Q027SBEA, GFDEGDQ188S, FYFSD.
So the honest question is not "how does the debt go away". It doesn't. The question is: who ends up absorbing the cost? Taxpayers, bondholders, holders of dollars, or everyone through inflation.
It doesn't, in the sense people mean; no large country has repaid a debt of this size. What can change is the real burden, and there are only eight ways for that to fall:
| 1 · Growth | The economy outgrows the debt. Slow, and it needs productivity, not just prices. |
|---|---|
| 2 · Higher taxes | Politically the hardest; every dollar is visible on someone's return. |
| 3 · Spending cuts | Interest, Social Security, Medicare and defence are most of the budget; the rest is small. |
| 4 · Inflation | Nominal GDP rises, the old debt does not. Paid for by everyone holding dollars. |
| 5 · Lower real interest rates | Cheaper to carry — but the market sets long rates, and in 2026 it is setting them high. |
| 6 · Dollar depreciation | Each dollar owed buys less of the world. Paid for by holders of dollar assets, many of them foreign. |
| 7 · Financial repression | Keep rates below inflation for years. Options 4 and 5 done quietly, at bondholders' expense. |
| 8 · Change the money | Move the system to different rails — or a different unit — so the old debt is denominated in something the state controls differently. |
The first three are the honest ones and the slow ones. The two that come up again and again in market conversations are 6–7 and 8: let the currency take the hit, or change what money runs on. We took both seriously enough to write down the mechanism, and then checked each against what 2026 prices have actually done.
What if the U.S. simply allowed — or engineered — a dramatically weaker dollar?
It works on paper because the debt is written in dollars. Make each dollar worth less and the $40 trillion buys less of the economy it is owed against. Governments have done versions of this before; economists call the polite version financial repression: keep interest rates below inflation for years and let the bondholders pay.
What if the problem isn't the debt but the monetary system underneath it? The President has pushed publicly, and repeatedly, for far lower interest rates. The crypto version of the argument goes one step further: move the rails.
To be clear about what that means and doesn't: crypto cannot "replace" the U.S. monetary system today. It is a hypothesis about where the system drifts, not a description of where it is.
If Scenario 01 were under way you would expect to see the fingerprints: inflation expectations rising, real yields falling, the dollar sliding. 2026 shows the opposite. The 10-year real rate — what a lender earns after expected inflation — went from 1.94 % at the start of the year to 2.61 % in September, the 99th percentile of every reading since 2003. Nominal 10-year yields rose from 4.19 % to about 5.0 %, and almost all of that rise was real rate, not inflation expectation. The dollar index is up on the year. Source: FRED DGS10, T10YIE; Yahoo DX-Y.NYB; Apex finding "The debasement thesis".
That is the expensive way to carry debt, not the sneaky way. Right now the United States is paying its creditors a real return of about 2.6 % — the reverse of financial repression. Long Treasuries have lost 6.6 % this year because of it.
Scenario 02 is where the data got interesting, just not in the way its fans expect. Stablecoin supply — dollars living on blockchains — has gone from about $205 billion at the start of 2025 to $312 billion in September 2026, and almost all of it is dollar-pegged. Crypto is not competing with the dollar here. It is becoming a distribution layer for the dollar, and every one of those tokens is backed by, mostly, U.S. Treasury bills. The escape hatch is buying the debt. Source: DefiLlama stablecoin charts.
Apex's 60-day correlation between Bitcoin and gold reached a five-year high this year while its correlation with the Nasdaq fell — Bitcoin started trading more like a monetary asset. That is genuine evidence for the "change the money" story, and we log it as such. But the same Bitcoin is down about 8 % in 2026 while gold is up 2 % and the real rate is up 67 basis points. Over the 120 days we measured in late August its correlation with the real rate was about −0.28: rates up, Bitcoin down. A monetary escape hatch that falls when money gets more expensive is behaving like a long-duration asset, not like an exit. Source: Apex findings "Bitcoin's gold link at a five-year high" and "The debasement thesis".
We tested the debasement thesis — high inflation becomes bullish for crypto — three separate ways during 2026. It failed each time, and it sits in our register as FALLEN with its own invalidation conditions. Most of the year's rise in yields was real rate, not inflation expectation; Bitcoin fell with rising real rates instead of rising against them; and the one criterion that would flip our view — Bitcoin moving with rates while the Nasdaq moves against them — has not been met on any day we have measured.
The catch for Scenario 01 is just as concrete. Apex's dollar pattern (gold up, dollar down on the same day) does show Bitcoin moving with it — about +0.7 % on those days, 69 % of the time, on two years of data. But the effect is gone by the next day, and the oil leg of the pattern vanished entirely on the two-year sample. A dollar slide would be visible in gold and Bitcoin the same afternoon. It would not be a thesis you could position for the night before. Source: Apex finding "The dollar pattern".
| What we know | $40.09 T total debt, $32.4 T held by the public · interest running at $1.25 T a year · debt 122.6 % of GDP · the real 10-year rate at a 20-year high · the dollar index up in 2026 · stablecoin supply up about 50 % in 20 months and almost entirely dollar-pegged · Bitcoin's gold link at a five-year high, and Bitcoin down on the year anyway. |
|---|---|
| What we don't know | Whether policymakers would tolerate a deliberately weaker dollar · whether inflation ends up doing the work indirectly, through a Fed that is pushed to cut into it · whether crypto becomes a competing monetary system or simply the dollar's next layer · whether the current real-rate regime is a peak or a plateau. |
| What we think | The question isn't whether the dollar disappears. It's what the next version of the monetary system looks like — and in 2026 the data says it looks like the dollar, on new rails, paying its creditors more than it has in twenty years. |
Live: fetched from the sources named in each cell when the page is built, at most an hour old.
Apex will watch the evidence. The numbers above are regenerated from their sources every hour; the findings are re-measured on their own schedules and their status is public at /research/.
Every number on this page names its source in the text; every Apex finding links to its evidence card, including the ones that failed. It describes the past and the present under stated conditions and promises nothing about the future. Octavian Apex is information software, not investment advice. All deep dives · Research · Markets · Understanding markets