Dollar Decline and the Rise of Bitcoin

Educational, not financial advice. This article is about macroeconomics and a volatile asset class. It contains no recommendation to buy, sell or hold anything. Bitcoin can lose most of its value in months — it has, twice — and currency forecasting has a poor track record, including mine. Figures verified 13 July 2026; every number here decays. Speak to a licensed adviser before acting, and never put money into bitcoin that you cannot afford to see halve.

Key takeaways

• The dollar's share of disclosed official reserves fell from 72% in 2001 to 58% in 2024 (Federal Reserve). That is about six-tenths of a percentage point a year. It is erosion, and at that pace the dollar stays the majority reserve currency into the late 2030s.
• The Fed's trade-weighted dollar index hit its all-time high in January 2025 and is down 7.2% since. It is still higher than it was in July 2021 or July 2016. A 7% round-trip is a cycle, not a collapse.
• I measured bitcoin's annualised volatility over the trailing year at 43.1%, against 4.3% for the dollar index — ten times. Bitcoin also fell 50% between October 2025 and today. That disqualifies it as a reserve asset now, whatever it becomes later.
• The central banks that are diversifying out of dollars are buying gold, not bitcoin: 863 tonnes in 2025 (World Gold Council). Official bitcoin holdings inside international reserves are, worldwide, roughly nothing.
• The strongest fact in bitcoin's favour is real and I will not pretend otherwise: its issuance rate is now about 0.8% a year and falling toward zero. That is a genuine monetary property. It is just not the same thing as being usable as money today.

The dollar's decline is real, and it is boring

Start with the number that the whole argument rests on, because almost nobody who shouts about the dollar's collapse ever puts it on the page.

The Federal Reserve's own survey of the dollar's international role reports that the dollar made up 58% of disclosed global official foreign exchange reserves in 2024, down from 72% in 2001 (The International Role of the U.S. Dollar — 2025 Edition, 18 July 2025). The euro is second at about 20%. The Chinese renminbi — the supposed successor, the currency of the world's second-largest economy — is at roughly 2%.

Do the arithmetic that the headline writers skip. Fourteen percentage points, over twenty-three years, is 0.61 points a year. Extend that line and the dollar crosses below half of world reserves somewhere in the late 2030s — and it would still be, by a wide margin, the largest reserve currency on earth, because nothing else is even close to 20%.

That is what the data says. It says the dollar is losing ground slowly to a crowd of small currencies, not to a single rival. "De-dollarisation" is happening. It is happening at the speed of continental drift.

The dollar's "collapse" narrative peaked at the dollar's all-time high

Here is the fact I could not find on a single other page about this topic, and it is the one that should make you suspicious of the genre.

The Federal Reserve publishes a Nominal Broad U.S. Dollar Index — the dollar measured against a basket of the currencies of America's actual trading partners, weighted by trade (Federal Reserve H.10; daily series DTWEXBGS on FRED). It is the least gameable measure of the dollar there is.

Its all-time high was 130.04, on 13 January 2025 — the highest the trade-weighted dollar has ever been in the history of the series. The "collapse of the dollar" story was at full volume that month.

Since then it has fallen to 120.69 (2 July 2026). That is a drop of 7.2% from the peak: 7.5% of it during 2025, and then nothing much — the index is up 0.9% so far in 2026.

Read the bottom three rows again. The dollar today is worth more, against the currencies America actually trades with, than it was five years ago and ten years ago. It gave back an unusually strong 2024. That is the entire move.

If someone tells you the dollar is collapsing, ask them for the index level. If they cannot give you one, they are selling something.

The debt is enormous and the trade deficit is not getting worse

Two numbers, because the bearish case deserves its strongest form.

US total public debt outstanding is $39.41 trillion as of 9 July 2026, of which $31.71 trillion is held by the public (US Treasury, Debt to the Penny). That is a real number, it is published daily, and it is going up. Foreign investors hold about $9 trillion of marketable Treasuries — 32% of the total, down from nearly 50% in 2014 (Federal Reserve). That falling foreign share is the single most legitimate worry on the bearish side, and it is the one metric I would actually watch.

The trade deficit, though, does not cooperate with the collapse story. The 2025 goods-and-services deficit was $901.5 billion — down $2.1 billion from $903.5 billion in 2024 (Bureau of Economic Analysis). It shrank. Marginally, but it shrank. A $1.24 trillion goods deficit was partly offset by a $339.5 billion services surplus, which is the part of the American economy that the doom pieces never mention.

So: a very large debt, a stable-to-slightly-improving trade balance, a slowly declining reserve share, and a currency near the top of its historical range. That is a country with a fiscal problem. It is not a country whose money is failing.

The people actually leaving the dollar are buying gold, and they are not buying bitcoin

This is the part of the argument that settles it, and it is settled by watching what institutions do rather than what commentators say.

If you genuinely believed the dollar was finished, and you ran a central bank, you would be moving reserves into the alternative. So which alternative are they moving into?

Central banks bought 863.3 tonnes of gold on a net basis in 2025 — down 21% from 1,092.4 tonnes in 2024, but still far above the 2010–2021 annual average of 473 tonnes (World Gold Council). Poland alone took 102 tonnes. Kazakhstan took 57. Brazil came back to the market for 43.

Now the bitcoin column. Official bitcoin held inside the international reserves of the world's central banks: as far as any public record shows, none.

The exception proves the point, and you have to read the actual document to see it. In November 2025 the Czech National Bank became the first central bank to buy bitcoin, and it was reported everywhere as a landmark. Here is what the CNB's own press release says: the purchase was USD 1 million, it was made "outside the international reserves," it also contained a dollar stablecoin and a tokenised deposit, its purpose was "to gain practical experience with holding digital assets," and — the sentence that never makes the headline — the CNB is "not planning to include bitcoin or other digital assets in its international reserves in the near future" (Czech National Bank, 13 November 2025).

"The value of bitcoin may fluctuate substantially. No investor should buy bitcoin without being aware of the significant risks involved."

— Aleš Michl, Governor of the Czech National Bank, in the same press release announcing that his central bank had bought bitcoin.

One million dollars. A rounding error, held deliberately outside the reserves, by a bank whose governor used the announcement to warn you about the risk. Set that beside 863 tonnes of gold. That is the honest scoreboard for the "central banks are fleeing to bitcoin" thesis, and it is not close.

I measured the volatility, and it is the whole argument

Everyone says bitcoin is volatile. Almost nobody says how volatile, relative to the thing it is supposed to replace. So I ran it.

Method, so you can reproduce it: I took daily closing prices for BTC-USD from the Coinbase Exchange public API, and daily values of the Fed's broad dollar index (DTWEXBGS) and the euro–dollar rate (DEXUSEU) from FRED. For each, I took the standard deviation of daily log returns over the trailing 365 days to 13 July 2026 and annualised it — √365 for bitcoin, which trades every day, and √252 for the currency series, which do not.

Bitcoin is ten times as volatile as the trade-weighted dollar and about seven times as volatile as the euro–dollar exchange rate. And this was a calm year for bitcoin by its own standards: measured over the full six years of data I pulled, its annualised volatility is 62%.

Then there is the number that no amount of theory survives. Bitcoin closed at $124,720 on 6 October 2025. Today, 13 July 2026, it closed at $62,412. It has lost almost exactly half its value in nine months.

Sit with what that would mean for a reserve manager. A central bank holds reserves so that when its currency is under attack, or an import bill comes due in a crisis, it can sell an asset for a known amount of hard money. The entire job of the asset is to be worth roughly what you thought it was worth on the day you need it. An asset that can halve in nine months does not do that job. It is not a bad reserve asset; it is a category error. It is a position, and reserves are not for taking positions.

This is also why the "bitcoin is a hedge against dollar weakness" claim keeps failing in practice. Over the past nine months the dollar was flat and bitcoin fell 50%. Bitcoin did not trade like an escape from the dollar. It traded like a high-beta risk asset, which is what it is.

The honest case for bitcoin, which is better than its advocates make it

I want to give the other side its strongest argument, because the version you usually hear is worse than the real one.

The real argument is not "number go up." It is issuance. Bitcoin's supply is capped at 21 million coins, and roughly 20.06 million are already in circulation — about 95.5% of the total that will ever exist. Since the April 2024 halving, each block pays 3.125 BTC, blocks come about every ten minutes, so new issuance runs at roughly 164,000 BTC a year. Against 20.06 million outstanding, that is an inflation rate of about 0.8% a year — and it halves again around 2028, and again after that, toward zero.

Point being: bitcoin's monetary expansion rate is now lower than the Federal Reserve's 2% inflation target, and it is set by arithmetic rather than by a committee. That is a genuine property. It is the reason serious people take the asset seriously, and anyone who waves it away has not understood the argument.

But notice what that argument actually establishes. It establishes that bitcoin's supply is predictable. It says nothing whatsoever about its price, which is set by demand, and demand is the thing that fell by half since October. Scarcity is necessary for a store of value. It is nowhere near sufficient. Tulips were scarce.

And there is the scale problem. At $62,412, the entire bitcoin float is worth about $1.25 trillion — roughly 3% of the $39.4 trillion US federal debt it is imagined to be replacing, and about 14% of the Treasuries that foreigners already hold. Even if every reserve manager on earth decided tomorrow to put a token 5% of reserves into bitcoin, they could not do it without moving the price to somewhere it has never been. The asset is too small for the job, and the act of making it big enough would be the most violent bull market in financial history — which is, of course, exactly why the people who own it want you to believe the job is imminent.

If you want the sceptical case laid out at length, we have argued that bitcoin might be the next big bust; if you want the case for its durability, we weighed whether bitcoin has really earned the "digital gold" label. And if the mechanics are new to you, start with what cryptocurrency actually is.

The awkward twist: crypto's most successful product is a dollar

The framing of this whole debate — crypto versus the dollar — may simply be wrong, and the Federal Reserve has started saying so out loud.

The fastest-growing thing in crypto is not bitcoin. It is the dollar stablecoin: a token whose entire promise is that it is worth one US dollar, typically backed by short-dated US Treasuries. The Fed put the market capitalisation of dollar stablecoins at roughly $220 billion as of April 2025 (Federal Reserve). Fed Governor Christopher Waller, opening the Fed's own conference on the dollar in June 2026, described distributed ledgers and tokenised assets as "creating new channels for global dollar intermediation," and noted that dollar-backed stablecoins form "a new channel linking global liquidity demand directly to U.S. Treasury markets" (Waller, 22 June 2026).

Read that plainly. A shopkeeper in Lagos or Buenos Aires who wants out of his local currency reaches for crypto — and what he actually buys is a token backed by American government debt. Blockchain technology's killer application, so far, has been exporting the dollar to people who previously could not get one, and funnelling their savings into Treasuries while it does so. On the evidence of the last few years, crypto has been a dollar-strengthening technology. That is not what anyone in 2013 expected, and it is the single most under-reported fact in this argument.

What would have to happen for me to be wrong

A position you cannot falsify is not a position, it is a mood. So here is mine, stated as a set of tripwires. If these things happen, the erosion thesis is dead and the collapse camp was right.

1. The reserve-share drop accelerates past about 2 points in a single year. The twenty-three-year trend is 0.61 points a year. A single year at 2+ points is no longer drift — it is a run. The IMF publishes the currency composition of reserves quarterly; that is the series to watch, not the headlines.
2. Bitcoin's trailing-year volatility falls below roughly 15% and stays there for three years. At 43% it is a trade. At 15% a treasurer can defend holding it in a board meeting. Nothing in bitcoin's history has been within a mile of that, but it is a clean, checkable line and I have drawn it deliberately.
3. A G20 central bank puts more than 1% of its actual international reserves into bitcoin — and publishes it. Not a $1 million lab experiment held outside the reserves. The Czech National Bank says it will run its test for two years and publish the results; that report, when it lands, is the most informative document in this entire debate.
4. The foreign share of marketable Treasuries falls below about 20%. It has already gone from nearly 50% in 2014 to 32% in early 2025. Below 20% and the world is genuinely refusing to fund the United States.
5. A failed Treasury auction, or an actual default on a coupon payment. This is the only one that would happen fast. Everything else is measured in decades.

None of those five has happened. Four of them are not close. Until at least two of them do, "collapse" is a word being used to sell you something, and the correct description of what is occurring is: the dollar is slowly getting less dominant, and remains overwhelmingly dominant.

So what do you actually do about it

Less than the internet suggests, and none of it urgent.

If your income, your rent and your groceries are all priced in one currency, then your real exposure is to domestic inflation, not to the dollar's reserve share. A Kenyan reader worried about the shilling and an American reader worried about the dollar have almost nothing in common except the vocabulary. Currency-debasement content flattens that distinction on purpose, because a single scary story sells to both.

The dull, defensible response to a slowly weakening currency is to own assets that are not that currency: equities of firms that earn abroad, some real assets, and — if you must — a position in bitcoin sized so that a 50% fall is annoying rather than structural. Because a 50% fall is not a hypothetical. It is what happened, this year, while you were reading articles about how bitcoin protects you.

The one thing I would not do is what the collapse genre is designed to make you do, which is treat a 0.6-point-a-year drift as an emergency and restructure your savings around it. Slow problems reward patience. They punish panic, and the people selling the panic have already priced that in.

FAQ

Is the US dollar going to collapse?

On the evidence, no — and the word is doing a lot of work. The dollar's reserve share has fallen from 72% to 58% over twenty-three years, and the Fed's trade-weighted index is currently 7% below its January 2025 record while still sitting above where it was five and ten years ago. That is a slow decline in relative dominance. A collapse is what happened to sterling after 1945: a fast, disorderly loss of the currency's role. Nothing in the current data resembles it.

Can bitcoin actually protect me from a falling dollar?

It has not so far. Since October 2025 the dollar index has been broadly flat and bitcoin has fallen about 50%. Bitcoin has historically traded like a high-risk asset — it tends to fall when investors are frightened, which is precisely when a hedge is supposed to work. Its 43% annualised volatility is roughly ten times the dollar's, so a bitcoin position introduces far more currency risk than it removes.

Why do central banks buy gold instead of bitcoin?

Because gold does the job and bitcoin currently does not. Reserve assets need to be deep, liquid, widely accepted between official institutions, and stable enough that the value you can sell for is close to the value you booked. Gold has a several-thousand-year track record and central banks bought 863 tonnes of it in 2025. Bitcoin has a sixteen-year track record that includes several drawdowns of more than 50%, including one in progress.

Does bitcoin's 21-million cap make it a good inflation hedge?

The supply cap is real — issuance is now around 0.8% a year and heading to zero — and it is the strongest thing about the asset. But scarcity constrains supply, not price. Bitcoin has repeatedly fallen sharply during periods when inflation was rising, which is the opposite of what an inflation hedge is meant to do. Treat the cap as an interesting monetary property, not as a promise about returns.

Is the renminbi replacing the dollar?

Not on any timeline worth planning around. The renminbi is about 2% of disclosed official reserves against the dollar's 58% and the euro's 20%. China maintains capital controls, which is fundamentally incompatible with being a reserve currency — reserve managers must be able to get their money out on demand. Until that changes, the renminbi is a trade-settlement currency, not a reserve one.

Does the $39 trillion federal debt mean the dollar is doomed?

It means the United States has a fiscal problem, which is a different sentence. The debt is real and rising. But a currency fails when holders stop accepting it, and foreigners still hold roughly $9 trillion of Treasuries and over $1 trillion in physical dollar banknotes. The metric that would signal genuine trouble is the foreign share of Treasuries — down from nearly 50% in 2014 to 32% — continuing to fall. Watch that, not the headline debt figure, which has been "unsustainable" in the press for forty years.

How this was researched

Every figure above comes from a primary source: the Federal Reserve, the US Treasury, the Bureau of Economic Analysis, the Czech National Bank, and the World Gold Council. The volatility, drawdown, issuance-rate and market-capitalisation figures are my own calculations from public daily price data, with the method stated in the article so you can rerun them. Where I could not source a claim, I cut it. Figures verified 13 July 2026 and they will go stale; the linked sources will not.

By Muslih Abdiker Ali — founder and editor of TheSchicht, published by Northfast Limited (Kenya). I write about money, markets and the technology underneath them. This piece was drafted with AI assistance; the sourcing, the calculations and the position taken are mine, and I am responsible for every number in it. Corrections and challenges are welcome.

Sources

• Federal Reserve, The International Role of the U.S. Dollar — 2025 Edition (FEDS Notes, 18 July 2025) — reserve shares (58% in 2024 vs 72% in 2001), foreign Treasury holdings, dollar banknotes abroad.
• US Department of the Treasury, Debt to the Penny — total public debt outstanding, 9 July 2026.
• Bureau of Economic Analysis, U.S. International Trade in Goods and Services, December and Annual 2025 — the 2025 trade deficit.
• Federal Reserve, H.10 Foreign Exchange Rates — about the dollar indexes, and the daily series Nominal Broad U.S. Dollar Index (DTWEXBGS) via FRED.
• U.S. / Euro Foreign Exchange Rate (DEXUSEU), FRED — euro volatility comparison.
• World Gold Council, Gold Demand Trends Full Year 2025 — Central Banks — 863.3 tonnes of net official purchases.
• Czech National Bank, The CNB creates a test portfolio of digital assets (13 November 2025) — the $1m portfolio, held outside international reserves.
• Federal Reserve Governor Christopher J. Waller, Welcoming remarks, Fifth Conference on the International Roles of the U.S. Dollar (22 June 2026) — on stablecoins as a new channel for dollar intermediation.
• Coinbase Exchange public API, BTC-USD daily candles — the price series behind the volatility and drawdown calculations.