Gold is often explained with a simple macro rule: higher real yields are a headwind, lower real yields are supportive, and a stronger U.S. dollar usually adds pressure. The framework is useful. The problem begins when it is treated as a mechanical forecast.
Gold pays no coupon or interest. When inflation-adjusted government-bond yields rise, investors can earn a higher real return elsewhere, increasing the opportunity cost of holding a non-yielding asset. Because gold is priced internationally in U.S. dollars, a stronger dollar can also make the metal more expensive for buyers using other currencies.
Those relationships are economically sound, but they do not operate in isolation. Inflation surprises, Federal Reserve expectations, energy prices, geopolitical demand, positioning and liquidity can all change which force dominates on a particular day. Two market windows show why these variables work better as an explanatory framework than as a trading signal.
February 13, 2024: a relatively clean macro-pressure event
The U.S. inflation release on February 13, 2024 is a useful example of the framework working in the expected direction. At 8:30 a.m. ET, the U.S. Bureau of Labor Statistics reported that January CPI had increased 0.3% month on month and 3.1% over the previous 12 months. Economists polled by Reuters had expected annual CPI of 2.9%, so the reading challenged expectations that inflation was cooling quickly enough to support early Federal Reserve rate cuts.
The repricing was visible in inflation-adjusted bond yields. The Federal Reserve's daily 10-year Treasury Inflation-Indexed Security series shows the real yield rising from 1.93% on February 12 to 2.02% on February 13, a nine-basis-point increase in one session.
Gold moved in the opposite direction. Reuters reported that spot gold was down 0.9% at $2,002.29 per ounce at 9:28 a.m. ET. Earlier in the session it had traded at its lowest level since December 13, 2023. That wording matters: the intraday low, rather than the later $2,002.29 quote, is the point that established the period low.
The mechanism was straightforward. A firmer inflation report reduced confidence in early rate cuts. Real yields rose, and gold came under pressure. But even this relatively clean example should not be converted into a formula. A nine-basis-point rise in real yields does not imply that gold must fall by a particular percentage. The useful information was that several markets were repricing the expected path of U.S. monetary policy at the same time.
September 2026: the framework in a noisier market
September 2026 provides a more complicated test. Investors were dealing with inflation concerns, oil above $100 per barrel, sharply higher Treasury yields, geopolitical tension in the Middle East and uncertainty over whether the Federal Reserve would return to rate increases.
The 10-year real yield moved sharply higher: 2.43% on September 8, 2.46% on September 9, 2.55% on September 10 and 2.60% on September 11. It remained 2.60% on September 14. If real yields alone determined gold's daily direction, the metal might have been expected to weaken steadily through the period. It did not.
On September 9, 2026, Reuters reported that spot gold rose 1.4% to $4,414.30 per ounce by 1:54 p.m. EDT. The dollar was near a two-week low, while Brent crude had crossed $100 per barrel amid a widening Middle East conflict. The benchmark 10-year Treasury yield had also climbed to its highest level since November 2023 before easing. Currency weakness and geopolitical demand were therefore offsetting part of the rates headwind.
The inflation data that followed reinforced the pressure on rates. At 8:30 a.m. ET on September 10, the BLS reported that the Producer Price Index for final demand rose 0.4% in August and 5.4% over the previous 12 months. At 8:30 a.m. ET on September 11, the BLS reported that CPI increased 0.4% in August and 3.4% over the previous 12 months.
The nominal Treasury market reflected those concerns. Reuters reported that the 10-year yield traded as high as 4.979% on September 11 before pulling back to about 4.93% after the CPI report. On September 14, the 10-year yield moved above the psychologically important 5% threshold and was last reported at 5.01%.
By September 15, the relationship looked more conventional again. Reuters reported spot gold down 0.1% at $4,293.29 per ounce at 1:45 p.m. EDT as a stronger dollar and elevated Treasury yields weighed on the metal while oil-driven inflation concerns increased expectations of a Federal Reserve rate rise.
Then the short-term balance shifted again on September 16. Spot gold rose 1.3% to $4,347.91 per ounce at 12:08 GMT as the dollar weakened, oil prices eased and U.S. bond yields pulled back ahead of the Federal Reserve decision.
The conclusion is not that real yields stopped mattering. They were clearly part of the pressure surrounding gold throughout the period. What changed from session to session was the balance between real yields, the dollar, energy-driven inflation concerns and geopolitical demand.
Why the dollar deserves separate attention
Real yields measure one part of gold's opportunity cost. The dollar changes how that pressure is transmitted internationally. When the dollar strengthens, gold becomes more expensive in local-currency terms for many buyers outside the United States. When the dollar weakens, that headwind can ease even while real yields remain elevated.
September 2026 demonstrated this clearly. Gold advanced on September 9 while the dollar hovered near a two-week low, weakened on September 15 when the dollar and Treasury yields were both applying pressure, and rebounded on September 16 as those pressures eased. The real-yield signal had not disappeared; another important variable had changed direction.
Inflation can create two competing gold channels
Inflation itself can affect gold through opposing channels. Investors may seek gold as protection against inflation or broader uncertainty. But stronger inflation can also convince markets that policy rates will remain higher, pushing real yields and the dollar upward and making a non-yielding asset less attractive.
The February 2024 CPI event was dominated by the second channel. September 2026 showed both channels operating at once: high energy prices increased inflation concerns and geopolitical uncertainty while also strengthening expectations for tighter monetary policy. That combination helps explain why gold did not move in one smooth direction.
A better way to use the framework
Instead of asking whether one data release is simply bullish or bearish for gold, it is more useful to ask what changed across several markets. Did the data materially change expectations for Federal Reserve policy? Did real yields rise or fall? Did the dollar reinforce that move or offset it? Was there another source of demand, such as geopolitical risk, official-sector demand or positioning, that could overwhelm the rates relationship? Finally, how did gold itself respond?
That last question is especially useful. If gold remains firm while both real yields and the dollar are rising, the divergence itself contains information: another source of demand may be absorbing what would normally be a macro headwind. It still does not tell us where gold must trade next.
Explanation is more useful than certainty
The relationship between gold, real yields and the dollar remains economically important. The February 2024 inflation surprise showed a relatively clean chain: inflation exceeded expectations, real yields rose and gold fell. The September 2026 period showed why the same framework cannot be reduced to a trading formula. Real yields stayed elevated, but gold's day-to-day direction changed as the dollar, bond yields, oil prices and geopolitical conditions changed.
The most defensible conclusion is therefore a modest one. Higher real yields generally create a headwind for gold, and a stronger dollar can reinforce that headwind. But neither variable operates alone. Used properly, the framework helps explain why gold is under pressure - or why it is resisting that pressure. It does not remove uncertainty, and it does not forecast the next price move.