Real yields, breakevens, gold and tech, and the term premium
Lesson 15 · about 11 min
The nominal 10-year yield is made of two pieces: what the market expects inflation to be, and what it demands on top of that in real terms. Splitting them apart explains why gold sometimes rallies when yields rise, why unprofitable tech got crushed in 2022, and why the same 10bp move on the 10-year can mean two different things.
TIPS and the decomposition
Treasury Inflation-Protected Securities (TIPS) pay a coupon on a principal that is adjusted for CPI. Their yield is therefore a real yield: the return above inflation. Comparing a TIPS yield to a nominal Treasury of the same maturity gives the breakeven inflation rate, the level of inflation at which the two would return the same.
nominal yield ≈ real yield + breakeven inflation
| Component | Where to find it | What moves it |
|---|---|---|
| Nominal 10-year yield | Everywhere | Everything below |
| 10-year real yield (TIPS) | FRED series DFII10; Treasury daily real yield curve | Fed policy expectations, growth expectations, term premium |
| 10-year breakeven | FRED series T10YIE, or nominal minus TIPS | Inflation expectations, energy prices, inflation risk premium |
The two components can move in opposite directions. On an oil spike, breakevens rise and real yields may fall (growth worry), leaving the nominal roughly unchanged. On a hawkish Fed surprise, real yields rise and breakevens fall (the Fed is expected to crush inflation), and the nominal may barely move. Reading only the nominal misses both stories.
The 5-year, 5-year forward breakeven (the expected inflation rate for the five years starting five years from now) is the measure central banks cite when they talk about whether "expectations are anchored." A sustained move in it, up or down, is the kind of thing that changes the reaction function.
Gold
Gold pays nothing, so holding it costs you whatever a safe real return would have paid. That opportunity cost is the real yield. When real yields rise, gold's relative attractiveness falls; when real yields fall, especially into negative territory as in 2020, gold benefits.
| Real yield move | Breakeven move | Usual gold reaction |
|---|---|---|
| Up | Flat | Down |
| Down | Flat | Up |
| Flat | Up | Up (inflation hedge without the opportunity cost) |
| Up | Up, by more | Mixed to up |
| Up sharply on hawkish Fed | Down | Down sharply |
The correlation between gold and real yields has been strong for long stretches and then broken for others. In 2023 and 2024 gold rose to records while real yields stayed high, which analysts attributed to central bank buying and geopolitical demand. Treat the real-yield link as the default explanation, not a law, and check the recent relationship before you trade it.
Tech and long-duration equities
The same opportunity-cost logic applies to stocks whose value is mostly in distant earnings. A rise in real yields raises the discount rate applied to those distant earnings without any offsetting boost to near-term cash flows. This is why NQ underperforms ES on days when real yields jump, and why the 2022 bear market was concentrated in long-duration growth stocks while energy and value held up.
A rise in breakevens, by contrast, is less bad for equities, because higher expected inflation also implies higher nominal earnings. The cleanest single macro variable for "how are growth stocks going to feel today" is the 10-year real yield.
Key idea: Nominal yield = real yield + breakeven inflation. Gold and long-duration tech trade off the real yield; the breakeven tells you what the market thinks about inflation itself.
The term premium, in one paragraph
The term premium is the extra yield investors demand for holding a long bond instead of rolling short bonds, as compensation for the risk that rates move against them over the holding period. It cannot be observed directly; it is estimated by models (the New York Fed's ACM model is the one most often cited, and FRED carries its estimates), and different models disagree by tens of basis points. It was estimated to be negative for much of the 2010s, when demand for safe long bonds was intense, and it rose in 2023-2025 as Treasury supply grew and inflation uncertainty persisted. When yields rise with no change in the expected Fed path and no change in breakevens, the residual is usually labelled term premium, and that is the bear-steepener case from the previous lesson: bad for long-duration assets, and not something the Fed can fix by cutting.
Putting it together on a release day
After a data print, look at three things in this order:
- The 2-year: how did the Fed path change?
- The 10-year real yield: how did the discount rate for long-duration assets change?
- The 10-year breakeven: how did the inflation outlook change?
A hot CPI that raises the 2-year and the real yield while breakevens are flat is a "Fed will tighten" read: bad for NQ, bad for gold. A hot CPI that raises breakevens while real yields fall is a "Fed is behind" read: mixed for NQ, good for gold. Same headline, opposite gold trade.
Try it: On FRED, chart DFII10 (10-year real yield) and T10YIE (10-year breakeven) together with the gold price for the last two years. Find two periods where gold followed real yields inversely and one where it did not. Write one line about what else was happening in the exception.
Recap
- Nominal yield ≈ real yield (from TIPS) + breakeven inflation; the two components often move in opposite directions.
- Gold trades primarily off the real yield, as its opportunity cost, though the link has broken for stretches.
- Long-duration tech is sensitive to real yields for the same reason; breakeven rises are less damaging to stocks.
- The 5y5y forward breakeven is the anchoring measure central banks cite.
- The term premium is the model-estimated compensation for holding long bonds; when it rises, long yields rise without any change in the Fed path.
See it drawn
Original diagrams for the ideas on this page. Illustrative, not real market data.