Approved Article With Interlinks

Gold averaged US$2,386/oz across 2024, the highest annual price on record. Crescat Capital says it is heading to $20,000. That is not a typo, and it is not a rounding error: it is a figure roughly eight times the current level, and it deserves a serious answer rather than a reflexive dismissal.

The number is not a headline grab. It is the output of two independent macroeconomic models the firm laid out in its July 2026 and August 2026 research letters, each built on distinct inputs and each pointing at the same destination. One tracks the widening gap between global money supply and the world’s gold stock. The other reads the S&P 500-to-gold ratio against its behaviour in 1929 and the 1970s.

This piece works through both frameworks in full, sets them against what central banks are actually doing right now, and gives an honest account of where the analysis holds up and where it asks you to accept the most.

Central bank accumulation is no longer the marginal story

Before touching any model, start with what is empirically verifiable. Central banks added roughly 1,045 tonnes of gold to official reserves in 2024, according to the World Gold Council’s full-year data published on 5 February 2025. That marked the third straight year above 1,000 tonnes, out of total global demand of a record 4,974 tonnes.

A separate accounting from In Gold We Trust puts the figure at 1,086 tonnes, reflecting a different scope. Either way, the direction is unmistakable, and the pace held through the first half of the year: H1 2024 net buying reached 483 tonnes, running 5% above the previous H1 record.

World Gold Council demand data published in early 2026 confirmed that central bank net purchases remained above the 1,000-tonne threshold for a third consecutive year, reinforcing the structural rather than cyclical character of official-sector accumulation.

The buyers themselves tell you what kind of demand this is.

• Poland: 90 tonnes
• India: 73 tonnes
• China: 44 tonnes

The pattern is geopolitical, not tactical. These are non-Western reserve managers systematically trimming dollar-centric exposure, and they have access to the same data as everyone else. The clearest quantitative signal of regime change comes from the European Central Bank, which found in June 2025 that official-sector buying now accounts for more than 20% of global gold demand, up from roughly 10% across the 2010s.

The ECB’s June 2025 research links the post-2022 surge in official gold buying directly to Russia’s invasion of Ukraine and the associated sanctions, suggesting reserve managers are structurally hedging against currency-based sanctions and seeking assets outside the dollar system.

Central banks have now bought gold on a net basis for 15 consecutive years. That persistence matters for how you read Crescat’s thesis. The monetary reset the firm describes is not a fringe forecast; it is the revealed preference of sovereign reserve managers acting on different political incentives. For a US investor, that reframes gold from an inflation hedge into a geopolitical reserve asset, and it changes how you should think about the durability of the trend.

The scale of sovereign accumulation reflects the broadening of official reserve functions: central banks are no longer treating gold purely as a passive store of value but as an active instrument for sanctions-proofing, collateral, and reserve diversification outside the dollar system.

What the S&P-to-gold ratio actually measures, and what history says

The S&P 500-to-gold ratio is a simple number: the index value divided by the price of gold per ounce. When it falls sharply, it means stocks have badly underperformed gold in real terms over a stretch of time. Two historical episodes give the ratio its analytical weight, and understanding both is what lets you judge Crescat’s target on its merits.

1929 to 1934: deflation, default, and deliberate repricing

In the crash and depression that followed 1929, US equities fell by roughly 80-90% from their peak to the 1932 trough. The official gold price, meanwhile, stayed fixed at US$20.67/oz under the gold standard, so the ratio compressed almost entirely through equity deflation rather than a rising gold price.

Then came the policy move. Roosevelt devalued the dollar in 1934, repricing gold to US$35/oz, a deliberate government revaluation used as a tool for reflation. This is the historical template Crescat points to when it describes a government repricing gold, and it shows the compression can happen even when the gold price is set by policy rather than the market.

1971 to 1980: monetary regime break and the decade-long gold bull

The second episode is the reverse mechanism. After Nixon closed the gold window in 1971, gold was allowed to trade freely and rose from around US$35/oz to above US$800/oz by 1980, during a decade of high inflation and interest rates that climbed from roughly 4% to 20% before inflation was contained.

The 1970s bull market pattern that Crescat references in Model Two was not a smooth trend: gold experienced a 50% drawdown in 1974-1975 before resuming its climb to the 1980 peak above $800, a sequence that matters for investors thinking about how to size and hold exposure across a multi-year move.

At the 1980 peak, the gold-to-S&P 500 ratio reached 7.58. Here is the point that matters for the current thesis: Crescat’s modelled target ratio is 5.25, comfortably below that 1980 extreme. The firm’s model does not require history to repeat at its most severe to produce the $20,000 figure.

Knowing the mechanics lets you evaluate the assumption directly rather than treating $20,000 as an arbitrary figure. The ratio framework supplies the specific mathematical pathway from current conditions to the target, and a 5.25 endpoint is a genuinely conservative version of the 1980 precedent.

The two frameworks Crescat uses to reach $20,000

The two models start from different places, which is what makes their convergence interesting. One begins with the money supply. The other begins with equity valuations. Both arrive at the same destination.

• Model One inputs: global fiat money supply growth of more than 7% annually; above-ground gold stock growth of approximately 1.5% annually; a divergence sustained over decades.
• Model Two inputs: a 50% decline in the S&P 500; a gold-to-S&P 500 ratio of 5.25; dollar devaluation against gold.

Model One: the fiat supply imbalance

The first framework rests on the gap between how fast money is created and how slowly gold accumulates. Crescat assumes global fiat money supply is expanding at more than 7% a year, against above-ground gold stock growth of roughly 1.5% annually, a rate that has held for about a century. If that divergence were fully priced into gold, the implied outcome is an 80% loss in the dollar’s purchasing power against the metal.

The 1971 monetary shift, when Nixon closed the gold window and severed the dollar’s last formal link to gold, remains the foundational reference point for understanding why fiat supply can expand without a hard ceiling while the above-ground gold stock grows at roughly 1.5% annually.

The honest tension sits with the money-supply input. The IMF’s most recent published figure puts global broad money growth at 3.1% as of June 2024, up from 0.7% a year earlier and described as broadly in line with pre-pandemic levels. That is materially below Crescat’s 7%+ assumption, and it is the single most important empirical question in Model One. If broad money is genuinely growing at the lower rate, the revaluation mathematics need either a longer timeline or a fresh catalyst, such as a coordinated fiscal and monetary response to a new crisis, to accelerate money creation.

Model Two: the equity correction pathway

The second framework runs through equities. Crescat models a 50% decline in the S&P 500 combined with a dollar devaluation against gold, producing a gold-to-S&P 500 ratio of 5.25. At that equity level, the ratio implies a gold price of approximately $20,000/oz, and it does so at a ratio below the 1980 peak of 7.58.

The firm’s projected timeline is 3-7 years from the July 2026 publication, with a central estimate of roughly four years. Understanding the specific inputs gives you a way to track progress: if global broad money growth reaccelerates, or if equity markets correct sharply, the pathway to the target shortens and becomes more credible. Neither model needs to be accepted whole to be useful as a monitoring lens.

Where the thesis is strongest and where it asks you to accept the most

The counterarguments deserve genuine analytical weight, not a token disclaimer. Smart institutional investors disagree with Crescat on specific, falsifiable grounds, and knowing those grounds lets you calibrate the thesis rather than swallow or reject it whole.

• Fed credibility: post-2021 rate hikes have pushed real policy rates positive and slowed broad money growth toward pre-pandemic norms.
• Dollar reserve durability: entrenched network effects keep the dollar dominant in trade invoicing and reserves, and gold buying remains small against total global reserves.
• Supply response: high prices historically pull in more mining investment and recycling, expanding effective supply over time, a dynamic Crescat’s models do not explicitly capture.
• Alternative hedges: inflation-linked bonds, real assets, and commodity baskets protect against inflation without requiring a systemic fiat collapse.

The IMF’s 3.1% broad money figure for June 2024 is the clearest empirical challenge to Crescat’s Model One. It suggests monetary conditions have normalised rather than continuing the excessive expansion the model assumes.

The scale argument is worth sitting with. Central-bank buying above 1,000 tonnes a year is large in tonnage terms, but it remains modest against the total stock of global financial assets, which limits how quickly official demand alone could reprice gold to extreme levels. Mainstream sell-side targets, meanwhile, have generally sat in the low-to-mid $2,000s per ounce range, treating gold as a tactical hedge rather than the cornerstone of an imminent reset.

None of this disproves Crescat’s thesis. What it does is identify the conditions under which the $20,000 target does not arrive: sustained Fed credibility, dollar network resilience, and a functioning supply response. Each is a variable to monitor, because if any weakens materially, so does the counterargument against Crescat. For sizing gold or gold-equity exposure, knowing the specific falsifiers of the view is more useful than knowing the target price alone.

What the $20,000 thesis actually requires you to believe

Pull the analytical arc together and the thesis resolves into three conditional requirements. Global money supply growth would need to sustain or accelerate beyond the IMF’s confirmed 3.1% rate. Equity markets would need a meaningful correction, modelled at 50%. And central banks would need to keep accumulating gold, signalling continued erosion of dollar reserve confidence.

You do not need all three to find the framework valuable. The two-model structure means partial versions of the thesis still support meaningful gold allocations well short of $20,000, which is what makes it usable rather than binary.

Three variables function as real-time indicators of thesis progress:

1. Global broad money growth, measured against the IMF’s 3.1% baseline.
2. Equity market valuations relative to historical norms.
3. Official-sector gold reserve changes, as reported by the WGC and IMF.

Crescat frames the backdrop as a prisoner’s dilemma: any nation able to print currency and buy gold has an incentive to move before others, and the US holds the greatest capacity to do exactly that. The 15-year net buying streak is the variable most clearly already in evidence, on a 3-7 year projected timeline with a central estimate near four years.

Gold’s sovereign collateral role has expanded beyond reserve diversification in the post-2022 sanctions environment, with central banks increasingly treating physical gold as an asset that functions outside correspondent banking infrastructure and cannot be frozen by counterparty action.

You do not have to accept $20,000 as a forecast to see that the directional case for gold rests on more independent sovereign and institutional behaviour than at any point since the 1970s. Treated as a checklist rather than a prediction, the Crescat frameworks give you a structured way to calibrate exposure as conditions evolve, instead of reacting to price moves after the fact.

This article is for informational purposes only and should not be considered financial advice. Investors should conduct their own research and consult with financial professionals before making investment decisions.

Past performance does not guarantee future results. Financial projections are subject to market conditions and various risk factors. The model outputs discussed here remain speculative and subject to change based on market developments.