The 3-Month — 10-Year Spread
The 3-Month — 10-Year Spread
The Federal Reserve measures recession risk not with the popular 2-10 spread, but with the 3-month–10-year spread. This measure has technical advantages and is what policymakers watch closest.
Key takeaways
- The 3-month–10-year spread (3M10Y) inverts alongside the 2-10 spread but is the Fed's official recession indicator
- 3-month T-bills are the shortest liquid risk-free asset; their yield reflects immediate Fed policy with minimal duration risk
- The 3M10Y spread avoids some distortions present in the 2-10 spread, particularly during periods of Fed balance-sheet normalization
- A negative 3M10Y spread is rarer and historically even more ominous than a 2-10 inversion
- The Sahm Rule and other Fed recession-forecasting models use the 3M10Y spread as their anchor
Why the Fed Prefers 3-Month
The Federal Reserve's preferred measure of the yield curve is not the 2-10 spread that dominates financial headlines, but the 3-month–10-year (3M10Y) spread. This choice reflects technical considerations about which measures are least distorted by Fed policy and most predictive of recessions.
The 3-month T-bill is the shortest liquid Treasury maturity. It is essentially the Fed's policy rate plus a tiny premium for counterparty risk. A 3-month bill yielding 5.0% when the Fed Funds rate is 4.75–5.00% is almost a direct read on the current Fed stance. There is little room for expectations about future Fed policy or term premiums to muddy the signal.
The 10-year Treasury, by contrast, embeds expectations about the Fed's rate path over a decade, expectations about long-term real growth and inflation, and a term premium for bearing duration risk. But this is precisely why the 3M10Y spread is informative: it measures the gap between "what the Fed is doing today" (3-month bill) and "what the market expects over the long run" (10-year Treasury). When that gap is negative, it means the Fed has tightened harder than the market expects to be necessary, a sign that Fed policy is about to constrain growth.
Technical Advantages Over 2-10
The 2-10 spread is intuitive and widely cited, but it has technical weaknesses. The 2-year Treasury is a medium-duration asset (not so short that Fed policy dominates, not so long that term premiums dominate). During periods of sharp Fed tightening, the 2-year can move faster than warranted by fundamentals, creating a flattened or inverted 2-10 spread that is more about near-term Fed policy shock than about true recession expectations.
By using the 3-month bill instead, the Fed zeroes in on the policy stance itself. An inverted 3M10Y spread says: "Even though the Fed is tightening now, the market believes the long-term sustainable yield is lower." This is a more fundamental signal than "the 2-year and 10-year have crossed," which could reflect temporary valuation imbalances.
The Fed's balance sheet also matters. During QE, the Fed holds billions of long-term Treasuries, suppressing the 10-year yield artificially. This can flatten or invert the 2-10 spread even if the economy is not facing recession. The 3M10Y spread is less susceptible to this distortion because the Fed's holdings of 3-month bills are minimal; the 3-month rate is set by true supply and demand in the shortest maturity.
The Sahm Rule
One of the most powerful recession-forecasting models is the Sahm Rule, developed by economist Claudia Sahm at the Federal Reserve. The rule is simple: a recession is likely if the 3-month moving average of unemployment rises 0.5 percentage points above the minimum over the last 12 months.
The Sahm Rule uses unemployment data, but it is mechanistically related to the 3M10Y spread. Both measure the same fundamental dynamic: Fed tightening and growth expectations. When the Fed raises rates to fight inflation and the labor market weakens (unemployment rises), both signals fire simultaneously.
The key insight is that the 3M10Y spread inverts before the Sahm Rule fires. This is why the 3M10Y is used in Federal Reserve recession-forecasting models (like those published by the New York Fed): it provides advance warning compared to backward-looking labor data.
Historical Accuracy
The 3M10Y spread has the same perfect track record as the 2-10 spread: every U.S. recession since 1955 has been preceded by a 3M10Y inversion.
However, the 3M10Y spread is less likely to show false positives or technical distortions than the 2-10. There were brief moments in the 2010–2012 period when the 2-10 spread was inverted or near-zero due to Fed QE driving down long yields, but the 3M10Y spread remained positive, correctly signaling no imminent recession.
Similarly, the 2019 inversion in the 2-10 was mirrored in the 3M10Y, but the inversion was less severe and briefer in the 3M10Y, which (correctly) suggested the recession risk was lower than some feared.
The Depth and Duration of 3M10Y Inversions
Like the 2-10 spread, the depth and duration of a 3M10Y inversion correlate with recession severity. A shallow, brief inversion is less ominous; a deep, sustained inversion signals severe trouble ahead.
The 2022–2023 inversion in the 3M10Y was dramatic. The spread fell from +150 bps in December 2021 to -50 bps or lower by mid-2023, and stayed inverted or near-zero for an extended period. This deep, sustained inversion was flagged as a red alert by Fed officials and forecasters.
By contrast, the 2019 inversion was shallow, and the 3M10Y recovered quickly, which helped explain why the 2020 recession, while sharp, was also brief (a shutdown recession, not a traditional cyclical downturn).
3M10Y and Fed Forward Guidance
The 3M10Y spread is particularly useful for interpreting Fed forward guidance. When the Fed says "we expect to keep rates high for longer," a positive 3M10Y spread that remains steep is consistent with that message. When the Fed starts signaling rate cuts, the 3M10Y spread typically begins to flatten in advance.
In 2024, as inflation moderated and the Fed began to consider rate cuts (without yet implementing them), the 3M10Y spread began to flatten from its 2023 lows, signaling that the market believed peak Fed tightness had passed. This flattening was a signal that recession risk was diminishing from the peak levels of 2023.
The Relationship Between 3M10Y and 2-10
The 3M10Y and 2-10 spreads are highly correlated but not identical. In most periods, they move together: when the curve flattens, both spreads narrow. When the curve steepens, both widen.
However, when the Fed is in a tightening cycle, the 3M10Y spread narrows faster than the 2-10, because the 3-month rate rises faster than the 2-year rate (the Fed raises the 3-month rate directly, and the 2-year price-in some expectation of eventual cuts). The 2-10 can stay relatively stable or even widen when the 3M10Y is collapsing.
This property makes the 3M10Y spread a "purer" measure of Fed tightening impact, which is why the Fed values it.
Practical Use in Forecasting Models
The Federal Reserve's researchers publish recession probability estimates based on the 3M10Y spread (and other spreads). The formulas are proprietary, but they essentially encode the historical relationship: given the current level of the 3M10Y spread, what is the probability of a recession in the next 12 months?
In mid-2022, as the 3M10Y spread turned negative, these models spiked recession probabilities to 30–40%. By mid-2023, as the spread stabilized near zero and short-term data showed economic resilience, the models revised down recession probabilities to 10–20%. By late 2024, as the spread recovered to positive territory, recession probabilities fell further.
These probability estimates are not perfect — they cannot predict the timing of a recession or whether it will materialize at all. But they provide a quantitative way to track changing recession risk over time.
Global Comparisons
Other developed economies' yield curves can be monitored using similar logic, though the 3-month rate equivalents vary by country. In the UK, the 3-month SONIA (Sterling Overnight Index Average) spread to the 10-year gilt serves a similar role. In the eurozone, the 3-month EURIBOR to 10-year German Bund spread is watched.
These international spreads tend to invert alongside or near the U.S. spread, suggesting a global cyclical dynamic. A worldwide 3M10Y inversion is a signal that global recession risk is elevated.
Limitations and Context-Dependence
The 3M10Y spread, like all economic indicators, requires interpretation. A zero or modestly negative spread is not an automatic recession call; it is a recession warning. Context matters:
- Inflation trends: If inflation is moderating, the Fed may be nearing the end of its tightening cycle, and an inverted 3M10Y spread may mean "near the end of tightening," not "recession imminent."
- Credit conditions: If financial conditions are stable (credit spreads tight, equity volatility low), a 3M10Y inversion may be a false signal.
- Fed forward guidance: If the Fed is explicitly committed to keeping rates high, the market may take it at face value and maintain a positive 3M10Y spread. But if the Fed's actions contradict guidance, the spread will reflect the true expectations.
The signal is powerful, but it is one input into a multi-factor recession analysis, not a deterministic rule.
Monitoring the Spread in Real Time
For investors who want to track the 3M10Y spread in real time, the best sources are:
- Federal Reserve Economic Data (FRED): The St. Louis Fed publishes the spread daily.
- Financial news outlets: Bloomberg, Reuters, and CNBC track both the 2-10 and 3M10Y spreads.
- Treasury market data: Any bond trader can calculate the spread from published 3-month T-bill and 10-year Treasury yields.
As of early 2024, the 3M10Y spread had recovered to modestly positive territory (around +50 to +100 bps), suggesting recession risk was diminished from 2023 highs.
Next
The 3M10Y spread is the Fed's preferred gauge of recession risk, but what does it look like when an inversion actually arrives? The next two articles examine two pivotal historical episodes: the 2022–2023 inversion (the deepest in 40 years) and the 2008 inversion (which preceded the Great Recession). These case studies show how the spread functions in real economic circumstances.