Yield Curve - The Term Structure of Interest Rates

By EC Assets Research Team, Macro Strategy · Published 19 May 2026 · Updated 15 October 2026

Yield Curve: The yield curve plots interest rates across maturities for instruments of equivalent credit quality (typically Treasuries). Its shape encodes expectations about growth, inflation, and monetary policy.

What the Yield Curve Shows

The yield curve plots the yields of government bonds against their maturities, from bills of a few weeks to bonds of 30 years. The US Treasury publishes a daily par yield curve, and most macro analysis starts from its current shape and recent movement.

Three features matter. The level reflects the general cost of money. The slope, long yields minus short yields, reflects expectations for policy rates and the compensation for holding longer bonds. The curvature, the middle of the curve against the ends, reflects where in time the market expects policy to change.

How It Works

Under the expectations view, a long yield is approximately the average of the short-term rates expected over the bond's life, plus a term premium for bearing interest-rate and inflation risk. The curve therefore contains implied forward rates. From yields y_m and y_n for maturities m and n, treated as zero-coupon rates on a zero-coupon bond basis:

$$f_{m,n} = \left(\frac{(1+y_n)^n}{(1+y_m)^m}\right)^{\frac{1}{n-m}} - 1$$

The front end is anchored by the policy rate; the long end reflects expected inflation, expected growth and the term premium. Term premia cannot be observed directly and are estimated with models such as that of Adrian, Crump and Moench (2013); central bank bond purchases and flight-to-quality episodes have at times compressed them.

A normal curve slopes upward. A flat curve signals that markets expect little change in rates. An inverted curve, with short yields above long ones, usually signals expected rate cuts. Estrella and Hardouvelis (1991) showed that the slope predicts real economic activity, and Estrella and Mishkin (1998) found the 10-year minus 3-month spread among the best single predictors of US recessions.

Worked Example

The US Treasury par curve on 14 October 2026:

Source: US Treasury Daily Par Yield Curve Rates.

The curve slopes upward: the 2s10s spread is 5.19 - 4.82 = 0.37 percentage points and the 3m10s spread 5.19 - 4.28 = 0.91. Treating the par yields as zero rates, the one-year rate expected one year ahead is 1.0482² / 1.0453 - 1 ≈ 5.11 percent, and the five-year rate five years ahead about 5.38 percent. Curvature is often measured with a 2s5s10s butterfly, twice the 5-year yield minus the 2-year and 10-year yields: 2 × 5.00 - 4.82 - 5.19 = -0.01, so the middle of the curve sits almost exactly on a straight line between its neighbours. Those forwards include term premium, so they overstate the policy path the market actually expects by an unknown amount.

The history since 1990 puts inversions in context. Grouping months in which the average 10-year minus 3-month spread was negative:

Computed from US Treasury daily data; recession start dates are NBER business cycle peaks.

When It Applies (and Limitations)

A signal, not a clock. Lead times have ranged from 8 to 16 months in this sample, too wide to time markets on.

False and lucky signals. The longest and deepest inversion in the sample, from 2022 to 2025, was not followed by a recession in that period. The 2019 inversion preceded a recession caused by a pandemic, which no curve could forecast.

Term premium distortions. When the term premium is compressed, for example by central bank bond buying, long yields fall and the curve flattens without any change in growth expectations, weakening the signal.

Which spread matters. The 2s10s and 3m10s spreads can disagree. In 2019 the 3m10s inverted for several months while the 2s10s barely did, and in 2022 the 2s10s inverted four months before the 3m10s.

Other markets differ. German, UK and Japanese government curves follow the same logic but carry their own inflation, credit and policy premia, so a slope that signals stress in one market may be normal in another.

Par versus zero curves. The published curve uses par yields. Exact forward rates and discount factors require a zero curve bootstrapped from them, which differs noticeably when the curve is steep.

Why It Matters for Institutional Investors

Duration and curve positioning. The curve determines the yield earned for extending duration. Steepener and flattener trades, sized to be neutral in DV01, express views on the slope without taking a view on the level.

Regime identification. Slope and level feed most macro regime frameworks, and changes in slope have often coincided with turning points in growth and policy.

Cross-asset links. Bank margins, discount rates for long-duration equities and real assets, and the attractiveness of carry trades all depend on the curve. Rates volatility, measured by the MOVE index, shows how uncertain the market is about where the curve goes next.

Liability valuation. Pension and insurance liabilities are discounted off the curve, so its level and shape directly change the funding position of liability-driven investors.

References

1. Estrella, A., & Hardouvelis, G. A. (1991). The Term Structure as a Predictor of Real Economic Activity. Journal of Finance, 46(2), 555-576.
2. Estrella, A., & Mishkin, F. S. (1998). Predicting U.S. Recessions: Financial Variables as Leading Indicators. Review of Economics and Statistics, 80(1), 45-61.
3. Adrian, T., Crump, R. K., & Moench, E. (2013). Pricing the Term Structure with Linear Regressions. Journal of Financial Economics, 110(1), 110-138.
4. U.S. Department of the Treasury. (2026). Daily Treasury Par Yield Curve Rates. U.S. Department of the Treasury.
5. National Bureau of Economic Research. (2023). US Business Cycle Expansions and Contractions. NBER.