Educational content only. This is not financial advice or an investment recommendation. The author is not a registered financial adviser. Trading leveraged instruments carries risk of loss.
In short: Gold pays no interest, so what a bond pays after inflation is the real competition. That number is the real yield, and it is one of the strongest known drivers of the gold price. The headline 10-year yield everyone quotes is not that number — it contains it. Splitting the headline into its two parts is what separates a useful reading from a misleading one, and it takes about two minutes using three free public series.
Why bonds compete with gold at all
Gold pays you nothing. No coupon, no dividend, no interest. Hold an ounce for a year and at the end of that year you have exactly one ounce. Whatever return you get comes entirely from the price moving.
A Treasury bond is the opposite. It pays a coupon on a schedule, backed by the US government.
So the two assets are competing for the same money, and the comparison between them is what economists call opportunity cost — what you give up by choosing one thing over another. When you hold gold, you give up whatever the bond would have paid you.
When bonds pay very little, that sacrifice is small and holding gold is cheap. When bonds pay a lot, the sacrifice is large and holding gold gets expensive.
That is the mechanism, and it runs in one direction: higher bond yields raise the cost of holding gold, which pushes down on the gold price.
Most commentary stops there. Stopping there will give you the wrong answer regularly, because one question has not been asked yet.
The problem with the headline yield
Say the 10-year yield is 4.5%. You lend the government money for ten years and they pay you 4.5% a year.
Here is the question that rarely gets asked: 4.5% of what?
If prices in the economy are rising at 2% a year, your 4.5% is not really 4.5%. Inflation is quietly eating part of it. In terms of what your money can actually buy, you are earning roughly 2.5%.
Now suppose inflation runs at 4.5% instead. The bond still pays 4.5%. But your purchasing power has not grown at all. In real terms you have earned nothing.
Same bond. Same headline number. Completely different outcome.
And this is the point: gold does not compete with the headline yield. It competes with what the bond delivers after inflation. That number is the real yield.
The decomposition
The headline yield — the one on every news ticker — is the nominal yield. It breaks into two parts:
Nominal yield = Real yield + Expected inflation
Three numbers, all three public and free:
| Component | What it is | FRED series |
|---|---|---|
| Nominal | The headline 10-year Treasury yield | DGS10 |
| Expected inflation | What the bond market expects inflation to average over ten years | T10YIE |
| Real yield | What is left after expected inflation is stripped out, from inflation-protected Treasuries | DFII10 |
A detail worth knowing, because it removes any doubt about whether the identity holds: FRED defines T10YIE as DGS10 minus DFII10. Nominal equals real plus breakeven by construction. It is not an approximation — it is how the series is built.
Now here is why the split matters. When you see the nominal yield rise, on its own you have learned almost nothing, because you do not yet know which component moved.
If the nominal rose because the real yield rose, bonds genuinely pay more after inflation. Gold just got more expensive to hold. That is downward pressure on gold.
If the nominal rose because expected inflation rose, while the real yield stayed flat, then bonds are not paying more in real terms. Nothing has changed in the competition between the two assets. The headline moved and the thing that matters did not.
That is why gold sometimes sits still while the yield everybody is watching climbs.
The four cases
| Driven by real yield | Driven by expected inflation | |
|---|---|---|
| Nominal up | Bonds pay more after inflation. Pressure on gold, downward. | Real competition unchanged. Neutral for gold, sometimes supportive. |
| Nominal down | Holding gold gets cheaper. Supportive for gold. | Often misread as good news for gold. Usually is not. |
There is also a shortcut worth committing to memory, and it comes with a strict condition.
You can deduce the direction of the real yield without looking it up — but only when the nominal and the breakeven are moving in opposite directions. If nominal rises while breakeven falls, the real yield is definitely rising, and by more than the headline moved.
If both are moving the same way, you cannot deduce it. It depends on which one moved more, and at that point you have to check the actual number. Applying the shortcut outside its condition is one of the easiest ways to reach a confident wrong conclusion.
A worked example: 2021 and 2022
Two consecutive years show both sides of this, and the contrast is unusually clean.
2021
| 4 Jan 2021 | 31 Dec 2021 | Change | |
|---|---|---|---|
| Nominal 10y | 0.93% | 1.52% | +59 bp |
Breakeven (T10YIE) | 2.01% | 2.56% | +55 bp |
Real yield (DFII10) | −1.08% | −1.04% | +4 bp |
The headline yield rose 59 basis points over the year. On the simple story, that is steady pressure on gold for twelve months.
Decompose it and the picture changes completely. Expected inflation rose 55 basis points over the same period. The real yield moved four basis points, and stayed deeply negative the whole way.
So almost the entire move in the headline came from the inflation side. In real terms, bonds were paying no more at the end of the year than at the start.
2022
| 3 Jan 2022 | 30 Dec 2022 | Change | |
|---|---|---|---|
| Nominal 10y | 1.63% | 3.88% | +225 bp |
Breakeven (T10YIE) | 2.60% | 2.30% | −30 bp |
Real yield (DFII10) | −0.97% | 1.58% | +255 bp |
A much larger move in the headline — and this time expected inflation actually fell. Which means the real yield did all the work and then some: a 255 basis point swing, from deeply negative to clearly positive.
Notice the shape of it: nominal up, breakeven down. Opposite directions. This is precisely the case where the shortcut applies — the real yield had to be rising, and by more than the headline.
One more detail. The 10-year real yield ended 2022 at its highest level in more than a decade, peaking that year at 1.74% on 3 November. It went higher still the following year — 2.49% on 19 October 2023 — which is worth knowing before treating any single reading as an extreme.
Same headline direction in both years. Very different readings. The decomposition is the only thing that tells them apart.
How to check it yourself
You can reproduce all of the above for free in a couple of minutes.
Go to FRED, run by the Federal Reserve Bank of St. Louis, and pull three series:
DFII10— 10-year real yieldT10YIE— 10-year breakeven inflation rateDGS10— 10-year nominal yield
Put them on one chart and you can see at a glance which component is driving any move in the headline.
One practical detail that catches people out: these series publish with a lag. Friday’s values typically appear the following Monday. If you are looking at the chart over a weekend, the last data point visible is not the most recent trading day. Worth knowing before drawing a conclusion from a chart that is missing its final bar.
Common errors
Reading the nominal yield as if it were the real yield. The single most common mistake, and the reason the same headline move gets opposite interpretations from different commentators.
Applying the deduction shortcut when both series move the same way. The shortcut has a condition. Outside that condition it produces confident nonsense.
Treating the relationship as mechanical. It is a strong relationship, not an equation. More on this below.
Drawing conclusions from a chart with a missing final bar. See the publication lag above.
What this does not tell you
This matters more than it might sound.
Real yields are one of the most important drivers of gold. They are not the only one, and the relationship is not deterministic. Central bank buying, currency moves, geopolitical stress and physical demand all push on the price as well. There have been extended stretches where the observed relationship between real yields and gold weakened considerably.
So this is not a formula that tells you where gold goes next, and it should not be used as one. It is a framework for understanding what a move in bond yields actually means — which is a different thing, and a more durable one.
And when the framework and the price appear to disagree, the most likely explanation is not that the market is broken or being manipulated. It is that something in the picture is not being accounted for yet: a publication lag, a component nobody is looking at, a driver from outside this framework. An incomplete model is far more common than a rigged one.
Sources
- Federal Reserve Bank of St. Louis, FRED: DFII10, T10YIE, DGS10. Yield figures retrieved 27 September 2026.
- Board of Governors of the Federal Reserve System, H.15 Selected Interest Rates.
- US Department of the Treasury, Daily Treasury Par Yield Curve Rates and Daily Treasury Par Real Yield Curve Rates.