What Breakeven Inflation Rates Tell You That CPI Doesn't

A breakeven rate is the gap between the yield on a regular Treasury bond and the yield on an inflation-protected Treasury bond, TIPS, of the same maturity: nominal yield minus real yield. It's often described in shorthand as the market's forecast of average inflation over that period, but that shorthand overstates what the number actually is. A breakeven is a measure of inflation compensation, not a pure forecast, and it bundles together three things that don't always move for the same reason: the market's actual expected inflation, an inflation risk premium investors demand for bearing the uncertainty around that expectation, and technical or liquidity effects specific to the comparatively less liquid TIPS market. Federal Reserve research has decomposed breakevens along these lines precisely because collapsing all three into "expected inflation" leads to real misreadings.

That distinction matters for how to use a breakeven move in practice. CPI and PCE describe inflation that has already happened. Breakevens are one way the bond market digests and reacts to that news, not a separate, superior forecast that supersedes it, and the more useful way to think about the relationship is sequential rather than hierarchical: a CPI print is a data shock, and how nominal and real Treasury yields move in response is one observable way to see how the bond market is absorbing that shock. If a hot CPI print doesn't move breakevens much, that's a signal worth taking seriously, but it isn't necessarily the market calling the print pure noise, since a stable breakeven can also reflect an offsetting move between the expected-inflation component and the risk-premium or liquidity component rather than a clean statement that nothing changed.

Breakevens come in different maturities, and the shape of a move matters as much as its size. A jump concentrated in short-term breakevens, the next one to two years, tends to reflect near-term factors, a specific price shock working through the system, and is generally read as more likely to fade. A move in longer-term breakevens, five years and beyond, or in the 5-year, 5-year-forward measure specifically, forward inflation compensation for the six-through-ten-year window rather than a blend that includes the next five years, is read as a more meaningful signal about whether the market thinks the inflation regime itself may be shifting, distinct from a spot 5-year breakeven that still carries a lot of near-term noise. That measure is still inflation compensation rather than a pure expectation, the same expected-inflation-plus-risk-premium-plus-technical-effects bundle described above, just applied to a later window, so a move in it should be read with the same caution as any other breakeven. Central bank credibility ties into this distinction: a central bank widely seen as committed to its target tends to see long-term breakevens stay anchored through noisy near-term data, while a central bank whose commitment is in question can see longer-term breakevens start moving on comparatively modest current surprises.

Inflation swaps are a separate, related market-based measure priced through a different mechanism, and comparing swap-implied inflation to a breakeven can help flag when something specific to the TIPS market, rather than a genuine shift in inflation views, is driving a move. Swaps carry their own liquidity and risk-premium quirks though, so a gap between the two is a prompt to look closer, not a clean answer on its own about which measure is right.

What confirms it. Treat a breakeven move as context for a CPI or PCE surprise, not a standalone hierarchy-topping signal: check the nominal yield and real yield separately to see which one actually moved, compare the short-maturity breakeven to the 5-year, 5-year-forward measure to separate near-term noise from a genuine regime question, and be more cautious reading a sharp breakeven move during a period of broader market stress, when TIPS liquidity itself is more likely to be part of the story.