Gold and Real Interest Rates: The Opportunity-Cost Logic, and When It Breaks Down

The Core Logic

Gold pays no interest, no dividend, and no coupon. Holding it means giving up whatever return you could otherwise have earned on a comparably safe asset. That forgone return is gold’s opportunity cost, and the cleanest measure of it is the real interest rate: the yield on a safe bond after subtracting expected inflation.

In practice, analysts typically use the yield on Treasury Inflation-Protected Securities (TIPS) as a proxy for the real interest rate, because TIPS yields already strip out the market’s inflation expectations by construction. When TIPS yields fall, the cost of forgoing interest to hold gold falls with them, so gold becomes relatively more attractive. When TIPS yields rise, the opposite logic applies. This is the single most frequently cited explanation for gold price moves in professional commentary, and for good reason: in recent data the correlation between gold and TIPS yields runs around -0.9, which is a strong relationship by the standards of cross-asset correlations.

Why the Relationship Is Usually Strong

A few features of gold make this logic more reliable than it might be for other commodities. Gold has essentially no consumption dynamic that dominates its price the way, say, industrial demand dominates copper. It is held primarily as a store of value and a monetary asset, which means its price responds more cleanly to the pure discount-rate logic that real yields represent. There is also no credit risk or maturity mismatch to complicate the comparison: unlike corporate bonds or equities, gold’s only real “yield” competitor is a risk-free real return, which is exactly what TIPS approximate.

The mechanism runs in both directions and interacts with the other forces in our four-dimensional framework. Falling real rates typically coincide with a weaker dollar, since lower US real yields make dollar-denominated assets less attractive to international capital relative to other currencies, and a weaker dollar independently supports gold prices for buyers using other currencies. The real-rate channel and the dollar channel therefore tend to reinforce each other rather than operate separately, which is part of why real rates have historically had such a large apparent effect on gold.

The Relationship Through History

The real-rate logic is not a recent theoretical construct; it shows up clearly across several historical episodes, and looking at them helps calibrate how much weight the relationship deserves.

The clearest historical illustration is the early 1980s. Federal Reserve Chair Paul Volcker’s disinflation campaign pushed nominal interest rates to levels that, once adjusted for the rapidly falling inflation rate that followed, produced sharply positive real yields for the first time in years. That shift coincided with the end of gold’s 1970s bull market and the start of a two-decade decline, a pattern covered in more detail in our history of gold price cycles. The 2000s told the opposite story: a weakening dollar and, particularly after the 2008 financial crisis, unconventional monetary easing that pushed real yields toward zero and eventually negative, coincided with gold’s rise from the low hundreds of dollars an ounce to a peak near $1,921 in September 2011. And the 2013 “taper tantrum,” when the Fed signaled an eventual end to quantitative easing well before actually raising rates, produced a rapid rise in real yields and a corresponding sharp drop in gold, a fast, clean illustration of the mechanism operating in real time rather than over a multi-year horizon.

Each of these episodes supports the same underlying logic, even though the surrounding macro details differed considerably from one period to the next, which is part of why the real-rate framework has held up as a useful lens across very different monetary regimes.

When the Relationship Breaks Down

A correlation of -0.9 is strong, not perfect, and the periods when it weakens are informative in their own right.

Central bank buying can dominate. As documented in our research on central bank gold demand, official-sector purchases are largely insensitive to real rates. When that demand is large enough relative to the rest of the market, it can push gold higher even during periods when real rates are rising or holding steady, muting the correlation over that stretch.

Geopolitical shocks operate on a different timescale. Real rates move gradually, tracking monetary policy expectations that build over weeks or months. A war, a sanctions announcement, or a sovereign-credit event can move gold sharply within days, in a way that has nothing to do with the interest-rate backdrop at that moment. In the short window around such an event, the real-rate relationship can appear to break down completely, even though it typically reasserts itself once the event passes.

Extreme monetary regimes complicate the picture. During the zero-lower-bound and negative-rate era in parts of Europe and Japan after 2009, the relationship between nominal policy rates and gold weakened because policy rates were pinned near zero for extended periods regardless of what was happening to inflation expectations, which is precisely the situation real yields are supposed to capture, but extended near-zero regimes reduce the variation in the input, thereby reducing what a correlation coefficient calculated over a period like that can tell you.

Currency-specific dynamics can offset the dollar-based version of the relationship. A US-based real-rate analysis captures the US dollar price of gold well, but an investor holding gold priced in a different currency is also exposed to that currency’s own real-rate and exchange-rate dynamics, which do not always move in the same direction as US real rates.

A Practical Way to Use This

We would suggest treating the real-rate relationship as the single most useful starting hypothesis for gold, not as a mechanical trading signal. Before attributing a gold price move to any other cause, real rates are the first thing worth checking, because historically they have explained more of the variance in gold prices over multi-month periods than any other single factor we track. But the exceptions above are not edge cases to be dismissed; they are the periods that matter most for understanding gold, because they are exactly when gold behaves differently from what a real-rates-only model would predict, which is when the other three dimensions of our framework, geopolitics, supply and demand, and technicals, are doing more of the work.

As of January 2026, TIPS yields have fallen alongside the Federal Reserve’s rate-cutting cycle that began in September 2025, and that decline lines up with gold’s advance from roughly $3,200 per ounce in mid-2025 to above $4,600 by January 2026. The real-rate story is consistent with a meaningful share of that move, but it is not the whole story: sustained central bank buying and elevated geopolitical tension were present over the same period and, in our reading, both contributed as well.

The Limits Worth Repeating

No correlation coefficient, however strong, is a forecasting tool on its own, and real rates are no exception. A -0.9 correlation describes a historical relationship over a specific period; it does not guarantee the relationship holds with the same strength going forward, and it says nothing about magnitude or timing for any individual future rate move. Readers should treat the real-rate framework as a way of organizing analysis, consistent with how we describe the broader four-dimensional framework, rather than as a basis for a specific price prediction.