Why Gold Doesn't Always Follow Real Interest Rates

Gold's price is really the outcome of four forces pulling at once, not one clean rule. The textbook version, gold pays no yield, so when real interest rates rise the opportunity cost of holding it rises too and gold should fall, is directionally real. It's also only one input among several that are often moving at the same time, which is why gold spends a lot of its life looking like it's defying the textbook when it's actually just responding to something else in parallel.

Real yields. Start with what "real interest rate" actually means here, since this is where a lot of casual explanations go wrong. The relevant real yield for gold isn't nominal rates minus today's inflation print, it's the market's expected real return, best approximated by the yield on inflation-protected government bonds, which already embeds where investors expect inflation to head, not just where it's been. Gold's relationship with real yields was genuinely imperfect even in the clearest recent test case. Through 2021, inflation surged, but markets were simultaneously pulling forward expectations for Fed tightening, and real yields began repricing up from deeply negative levels as that expectation built. Gold did not respond to the headline inflation number alone, because the market was pricing the expected policy response at the same time, and that response worked against gold even as inflation itself would have argued for it. Real yields are one of gold's most persistent drivers over medium time horizons, but they are not sufficient on their own to explain a given week's move.

Safe-haven demand. A separate channel operates alongside the real-yield story: gold's role as a safe-haven asset during acute risk-off events. During a geopolitical shock or a sudden growth scare, gold can rally even if real yields haven't moved much, on demand for an asset perceived as insulated from the specific crisis at hand. This is a genuinely different mechanism from the real-rate story, even though the two often push in the same direction together.

Official-sector demand. A third force, and one that has grown more important over the past several years, is central bank buying. A number of central banks, particularly in emerging markets, have been accumulating gold reserves as part of a broader push to diversify reserves away from any single currency. This flow doesn't respond to what US real yields are doing week to week, and it has added a persistent source of demand that a simple real-yield model doesn't capture. The precise scale of this buying is reported with some lag and inconsistency across sources, worth treating cautiously rather than citing a single number as settled, but the direction of the effect is well documented.

Acute liquidity stress, the exception box. The most counterintuitive wrinkle shows up during the most acute phase of a genuine market liquidity crunch, like March 2020. In moments like that, gold can fall sharply for a brief window even though every other signal (falling real yields, a major risk-off shock) points the other way. The mechanism is closer to plumbing than fundamentals: when investors across a portfolio need cash fast, alongside real dollar funding stress more broadly, they sell whatever is liquid, gold included, regardless of what gold's own fundamentals are doing. This effect is usually short-lived, and gold typically resumes more standard safe-haven behavior once acute liquidity stress passes, but it explains some genuinely confusing headlines from past crises.

Put together, the honest version of the gold story is that real yields, safe-haven flows, official-sector demand, and acute liquidity conditions are all operating on gold at once, and any one of them can dominate in a given stretch. Treating gold as a simple inverse real-yield trade is a reasonable starting point and a reliable way to be surprised.

Gold is one of the more genuinely multi-causal assets in markets, worth understanding as four forces rather than one rule.