Bitcoin, Gold and the Dollar: The Correlation Investors May Be Overestimating
The dollar matters — but far less than the popular narrative suggests
One of the most persistent narratives in global markets is straightforward:
Dollar down = Bitcoin and gold up.
Dollar up = Bitcoin and gold down.
Directionally, there is some truth to this. Both assets are priced in U.S. dollars, both can benefit from concerns over monetary debasement, and gold in particular has historically exhibited an inverse relationship with the dollar.
But the data since 2022 tell a much more interesting story.
From 2022 through August 2026, the average 30-day correlation with the DXY Index has been approximately -0.13 for Bitcoin and -0.41 for gold.
That distinction matters.
A correlation of -1.0 would imply an almost perfect inverse relationship. Zero would imply virtually no linear relationship.
Bitcoin’s -0.13 is therefore extremely weak. Gold’s -0.41 is meaningful, but still nowhere close to suggesting that the dollar dominates gold’s price formation.
Using the squared correlations as a simple explanatory-power intuition:
| Asset | Avg. 30-Day Correlation with DXY | Correlation Squared | Approx. Variation Not Associated with DXY |
|---|---|---|---|
| Bitcoin | -0.13 | 1.7% | 98.3% |
| Gold | -0.41 | 16.8% | 83.2% |
Correlation-squared here is an intuitive approximation based on the average rolling correlation, rather than the R² of a single full-period regression.
The conclusion is striking:
The dollar may influence Bitcoin and gold, but it does not explain most of what they do.
For Bitcoin, the dollar relationship has been particularly weak. For gold, DXY remains important, but more than four-fifths of the variation implied by this simple calculation lies elsewhere.
That makes the next question far more important:
If the dollar isn’t driving these assets, what is?
The 2022–2026 Experience
The annual performance numbers make the disconnect easier to see.
| Year | Bitcoin | Gold | DXY | Dominant Macro Theme |
|---|---|---|---|---|
| 2022 | -64.2% | ~-0.8% | +7.9% | Fed tightening, inflation shock, crypto deleveraging |
| 2023 | +155.5% | +12.7% | -2.1% | Crypto recovery, banking stress, rate expectations |
| 2024 | +121.6% | +26.7% | +4.3% | Bitcoin ETFs, central-bank gold buying |
| 2025 | -6.5% | +63.7% | -7.1% | Gold re-rating, weaker dollar, fiscal/liquidity concerns |
| 2026 YTD* | ~-10% | ~+7% | roughly flat | Fiscal concerns, Treasury yields, geopolitics, changing liquidity |
2026 figures are through late August and vary slightly by pricing source and cutoff.
Bitcoin return data show gains of roughly 155% in 2023 and 122% in 2024 before negative returns in 2025 and 2026 YTD. Gold, meanwhile, rose roughly 13% in 2023, 27% in 2024 and an extraordinary 64% in 2025.
The dollar simply cannot explain those differences.
2024 is perhaps the clearest example.
DXY rose approximately 4.3%.
If the simplistic “strong dollar = weak Bitcoin and gold” framework were sufficient, both assets should have struggled.
Instead:
Bitcoin: +121.6%
Gold: +26.7%
DXY: +4.3%
Bitcoin and gold both appreciated dramatically while the dollar strengthened.
That is not a statistical curiosity.
It tells us something fundamental about market structure:
Asset-specific demand can overwhelm the currency translation effect.
Bitcoin: DXY Is Probably the Wrong Primary Variable
Bitcoin is often described as an anti-dollar asset.
Over very long horizons there is some logic to this argument. Bitcoin has a fixed ultimate supply of 21 million coins, whereas fiat money supply can expand.
But that does not mean Bitcoin trades as a simple inverse DXY instrument.
Research using data through 2024 found Bitcoin’s negative relationship with the dollar to be intermittent rather than structurally stable. Short-term negative coherence appeared periodically but tended not to persist.
CME research reaches a similar conclusion. Crypto’s negative correlation with the dollar became relatively noticeable in 2022–23 — at times approaching -0.4 — but weakened substantially toward zero by 2025 and early 2026.
This makes intuitive sense.
Bitcoin’s biggest moves since 2022 have often been generated by Bitcoin-specific events rather than FX events.
The more relevant Bitcoin variables are:
1. Global liquidity
Bitcoin is arguably more sensitive to the availability and price of global liquidity than to DXY itself.
VanEck research, for example, argues that global money supply has considerably more explanatory power for Bitcoin than the dollar alone, estimating that changes in global M2 can explain more than half of Bitcoin’s price variance in its model.
This distinction is critical.
DXY measures the dollar relative to other currencies.
Liquidity measures the amount and availability of money potentially flowing through the global financial system.
Those are not the same thing.
The dollar could remain relatively firm against the euro and yen while global liquidity simultaneously expands.
Under that combination, Bitcoin can rise alongside DXY.
2. Institutional adoption
Bitcoin’s 2023–24 rally cannot be understood without considering the transformation of its investor base.
The arrival and growth of spot Bitcoin ETFs altered the transmission mechanism between traditional financial portfolios and crypto.
Bitcoin increasingly became accessible through the same brokerage, custody and portfolio infrastructure used for conventional securities.
That represents a structural demand shock.
And structural demand does not require DXY to fall.
3. The halving and supply structure
Bitcoin has another characteristic the dollar cannot explain:
its supply schedule is largely predetermined.
The 2024 halving reduced the rate at which new Bitcoin entered circulation.
When new demand encounters structurally constrained incremental supply, price can rise regardless of whether DXY moves 2% higher or lower.
This is one reason treating Bitcoin simply as “digital gold versus the dollar” misses much of its actual market structure.
4. Leverage and positioning
Bitcoin is also heavily influenced by:
- futures open interest
- perpetual funding rates
- options positioning
- liquidations
- stablecoin liquidity
- exchange balances
- ETF flows
- long-term holder behaviour
A leveraged liquidation cascade can move Bitcoin 10–20% without any meaningful movement in DXY.
That alone helps explain why a 30-day Bitcoin/DXY correlation can spend long periods close to zero.
Gold Is More Dollar-Sensitive — But Still Not a Dollar Trade
Gold’s -0.41 average correlation tells a different story.
The dollar clearly matters more for gold than Bitcoin.
This is economically logical.
Gold is globally traded but conventionally quoted in dollars. When the dollar falls, gold becomes cheaper in non-dollar currencies, potentially improving international demand.
But even here, the relationship is far from deterministic.
An implied R² around 17% based on the average correlation means that the overwhelming majority of gold’s variation still cannot be reduced to movements in DXY.
The World Gold Council’s own attribution work demonstrates this complexity.
Its models incorporate not only the dollar but also interest rates, risk variables, momentum and investment demand. In February 2026, for example, the WGC noted that dollar weakness and lower Treasury yields helped gold, but its model still produced a large unexplained positive residual — evidence that additional investment-demand forces were at work.
That is enormously important.
Gold is not one trade.
It is several trades happening simultaneously.
The Five Forces That Matter More for Gold
1. Real interest rates
Gold generates no coupon.
Its opportunity cost therefore changes with the return available from inflation-adjusted bonds.
All else equal:
Real yields falling → supportive for gold
Real yields rising → potentially negative for gold
But even this relationship can break when another source of demand becomes sufficiently powerful.
And that is exactly what the post-2022 market has demonstrated.
2. Central-bank buying
This may be one of the most important structural changes in the gold market.
Central banks do not necessarily buy gold because they expect DXY to decline next week.
Their motivations can include:
- reserve diversification
- geopolitical insurance
- sanctions risk
- reduced reliance on foreign sovereign liabilities
- long-term reserve preservation
This creates demand that can be relatively insensitive to short-term dollar fluctuations.
That helps explain why gold has repeatedly remained resilient even during periods of high U.S. yields and dollar strength.
3. Fiscal credibility
There is an important difference between:
a weak dollar
and
concern about the dollar-based financial system.
Gold can benefit from the second without immediately requiring the first.
Rising government debt, large fiscal deficits, growing interest expenses and concerns about sovereign balance sheets can increase demand for assets carrying no counterparty liability.
Gold is unique in this respect.
A Treasury security is somebody else’s liability.
A bank deposit is a bank’s liability.
Gold bullion is not.
That distinction becomes increasingly relevant when markets begin questioning fiscal sustainability.
4. Geopolitical fragmentation
Gold’s role as a reserve asset has become more valuable in an increasingly fragmented geopolitical system.
The freezing of Russian foreign-exchange reserves after the invasion of Ukraine changed the reserve-management debate.
Foreign reserves are only as politically neutral as the institutions holding them.
Physical gold has no issuing government.
For reserve managers concerned about sanctions exposure, that characteristic has strategic value.
The World Gold Council has explicitly highlighted greater caution among central banks and investors toward U.S. exposure following the weaponisation of financial sanctions.
5. ETF and investor flows
At the margin, prices are determined not by narratives but by flows.
When institutional investors significantly increase gold allocations, those flows can dominate modest movements in DXY.
The same principle applies to Bitcoin.
This leads to one of the most useful frameworks for understanding both assets.
Think in Terms of “Marginal Buyer,” Not Just Correlation
Ask:
Who is setting the next price?
For Bitcoin, the marginal buyer might be:
- an ETF investor
- a macro hedge fund
- a corporate treasury
- a leveraged crypto trader
- a long-term institutional allocator
For gold, it could be:
- a central bank
- an ETF investor
- a Chinese household
- a macro fund
- a reserve manager
- a geopolitical hedger
The identity and motivation of that marginal buyer can matter far more than a one-point move in DXY.
Bitcoin and Gold Are Also Not the Same Trade
Another mistake is putting Bitcoin and gold together simply because both are sometimes described as alternatives to fiat currency.
Their correlation with each other has itself been surprisingly weak.
CME analysis found that rolling 12-month correlations between crypto and gold never exceeded roughly +0.41 even during the pandemic liquidity boom, and since 2024 correlations have fallen toward zero.
So there are really three separate relationships:
| Relationship | Structural Strength | Interpretation |
|---|---|---|
| Bitcoin vs DXY | Weak | Dollar is a secondary driver |
| Gold vs DXY | Moderate inverse | Dollar matters, but fundamentals dominate much of the variation |
| Bitcoin vs Gold | Weak / unstable | They should not automatically be treated as the same macro trade |
That has significant portfolio implications.
Owning Bitcoin and gold together is not necessarily doubling the same bet.
The return engines are different.
The Most Important Difference: Monetary Asset vs Liquidity Asset
A useful conceptual distinction is:
Gold behaves increasingly like a monetary-reserve asset.
Its long-term demand is influenced by:
central banks → real yields → fiscal credibility → geopolitics → reserve diversification.
Bitcoin behaves more like a scarce global liquidity asset.
Its price is influenced by:
global liquidity → adoption → ETF flows → supply scarcity → leverage → network confidence.
Both can benefit from monetary debasement.
But the path through which that debasement reaches their prices is very different.
Correlation Is Regime-Dependent
Perhaps the deepest lesson from the data is that correlations themselves are not permanent.
Consider four hypothetical regimes:
| Macro Regime | DXY | Liquidity | Gold Bias | Bitcoin Bias |
|---|---|---|---|---|
| Fed tightening / liquidity contraction | ↑ | ↓ | Negative/Neutral | Strongly Negative |
| Dollar weakness / liquidity expansion | ↓ | ↑ | Positive | Strongly Positive |
| Fiscal stress / high inflation | Mixed | Mixed | Strongly Positive | Potentially Positive |
| Risk-off deleveraging shock | ↑ | ↓ sharply | Mixed/Positive | Strongly Negative |
This explains why simple correlation analysis often fails at turning points.
During a traditional risk-off event, Bitcoin may behave like a high-beta liquidity asset and fall.
Gold may initially fall because investors need dollars, but later rally as safe-haven demand emerges.
The same DXY move can therefore produce completely different responses.
A More Useful Dashboard for the Next 12 Months
Rather than asking only:
“Where is DXY going?”
investors should probably monitor a broader set of variables.
| Bitcoin | Gold |
|---|---|
| Global M2/liquidity | U.S. real yields |
| ETF inflows/outflows | Central-bank purchases |
| Stablecoin supply | Gold ETF flows |
| Institutional adoption | Fiscal deficits/debt |
| Halving/supply dynamics | Geopolitical risk |
| Futures leverage | Physical Asian demand |
| Long-term holder supply | Reserve diversification |
| Regulation | Inflation expectations |
| Risk appetite | DXY |
Notice where DXY appears.
It remains relevant — but it is only one variable.
The Counterintuitive Scenario Investors Should Watch
The most interesting possibility over the next year may be a world where:
the dollar does not collapse, yet Bitcoin and gold still rise.
Why?
Because their structural demand functions do not require a dollar bear market.
Gold could appreciate because central banks continue diversifying reserves, fiscal concerns intensify and investment demand increases.
Bitcoin could appreciate because global liquidity expands, institutional adoption grows and available supply becomes increasingly constrained.
In other words:
Dollar weakness would be a tailwind. It is not a prerequisite.
The opposite is equally important.
A falling dollar does not guarantee rising Bitcoin.
If DXY declines because the U.S. economy enters a severe downturn accompanied by falling risk appetite, deleveraging and contracting liquidity, Bitcoin could initially fall despite dollar weakness.
Similarly, gold can rise alongside a strengthening dollar during severe geopolitical or financial stress.
The 2024 Lesson
2024 deserves special emphasis because it effectively falsified the simplistic dollar framework.
The dollar rose.
Gold rose strongly.
Bitcoin more than doubled.
Three supposedly contradictory market movements occurred simultaneously.
The explanation is straightforward once we stop treating DXY as the master variable.
Bitcoin experienced powerful structural and institutional demand.
Gold experienced powerful monetary, institutional and central-bank demand.
Those forces were stronger than the negative effect generated by dollar appreciation.
Fundamentals overwhelmed correlation.
What the Correlations Really Tell Us
The -0.13 Bitcoin/DXY correlation does not tell us that the dollar is irrelevant.
It tells us that the dollar alone has been a poor forecasting model for Bitcoin.
Similarly, gold’s -0.41 correlation confirms that dollar movements matter, but they are nowhere near sufficient to explain gold’s behaviour.
This distinction is essential.
Correlation identifies association.
It does not identify the complete economic mechanism.
And correlation itself changes when the underlying macro regime changes.
Investment Conclusion
The popular framework says:
Buy Bitcoin and gold because the dollar will weaken.
The evidence suggests a more sophisticated thesis:
Own Bitcoin and gold only if you believe in their respective fundamental demand structures — and treat dollar weakness as an additional potential catalyst rather than the core investment thesis.
Since 2022, Bitcoin’s average 30-day DXY correlation of approximately -0.13 suggests an exceptionally weak dollar relationship.
Gold’s -0.41 relationship is considerably stronger, yet even there the dollar explains only a minority of price variation under a simple correlation-squared interpretation.
The implication for the next year is important.
For Bitcoin, watch global liquidity, institutional flows, leverage, adoption and supply.
For gold, watch real yields, central-bank buying, fiscal credibility, geopolitics and investment flows.
Watch DXY for both.
But don’t make it the whole story.
The bigger insight
Bitcoin and gold may both be alternatives to traditional monetary assets, but neither is simply an inverse-dollar trade.
The dollar can determine the wind.
Fundamentals increasingly determine the destination.
Data sources: CME Group, World Gold Council, Federal Reserve/FRED.
