Bond convexity in Bond Markets: How to Compare Related Measures

Bond convexity in Bond Markets: How to Compare Related Measures

Convexity describes the curvature of the relationship between a bond’s price and its yield. Duration supplies the first-order estimate: when yield changes, price moves in the opposite direction by roughly duration times the yield change. Convexity adds the second-order adjustment that becomes more important as the yield move grows. For a conventional option-free bond, the price-yield curve is usually positively convex: a yield decline raises price by somewhat more than duration alone predicts, while an equal yield increase lowers price by somewhat less.

A common approximation expresses the percentage price change as: −modified duration × change in yield + ½ × convexity × (change in yield squared). The squared term is positive for positive convexity regardless of whether yields rise or fall. The approximation is useful, not exact. Its accuracy depends on the size of the rate change, the bond’s cash-flow structure, and whether the instrument contains options or other features that cause cash flows to change with interest rates.

Convexity versus duration: slope and curvature

Duration and convexity are complements, not substitutes. Modified duration measures the local slope of the price-yield curve and is commonly used to estimate sensitivity to a small parallel yield change. Convexity measures how that slope itself changes as yield changes. Two bonds can have the same modified duration but different convexity, so their estimated price changes may diverge materially when yields move far enough.

Macaulay duration is related but answers a different question. It is the present-value-weighted average time to receipt of promised cash flows, expressed in years. Modified duration is derived from Macaulay duration under a specified compounding convention and translates that timing concept into yield sensitivity. Convexity is likewise calculated from the timing and present values of cash flows, but gives greater importance to cash flows farther in the future. Comparing a bond’s Macaulay duration with another bond’s convexity therefore mixes a cash-flow timing measure with a curvature measure; comparing modified duration and convexity alongside one another is generally more informative for price-risk analysis.

Longer maturities, lower coupons, and more distant principal repayment often increase both duration and convexity for otherwise comparable fixed-rate bonds. That does not mean the measures are interchangeable. A zero-coupon Treasury, for example, has a single distant cash flow and can have substantial duration and convexity. A coupon bond returns part of its value earlier, changing both its sensitivity and the curvature of its price response.

Comparing convexity with DV01, yield-curve exposure, and dollar measures

DV01, also called dollar value of a basis point, converts first-order rate sensitivity into currency terms. It estimates the change in market value from a one-basis-point yield move, ordinarily using duration and the position’s market value. Convexity does not replace DV01: DV01 is most useful for small moves and position sizing, while convexity helps adjust the estimate for larger moves. A portfolio can have a low net DV01 yet retain meaningful convexity exposure, especially when long and short positions offset first-order sensitivity but not curvature.

Dollar convexity is the currency-value counterpart of percentage convexity. It shows how the dollar price response changes as yields change and is more directly comparable across positions of different sizes. Percentage convexity can help compare securities on a normalized basis; dollar convexity helps assess the effect on a particular holding or portfolio. Any comparison should use consistent yield conventions, pricing dates, clean or dirty price treatment, and assumptions about the rate shock.

Neither DV01 nor a single bond-level convexity figure fully captures exposure to a changing yield curve. Treasury yields differ by maturity, and curve movements need not be parallel. Key-rate durations measure sensitivity to selected points on the curve, such as shorter, intermediate, or longer maturities. Key-rate convexities can extend that analysis by showing curvature around those points. For portfolios containing Treasuries, corporate bonds, and municipal bonds, a parallel-shift duration-and-convexity calculation is a useful simplification, but it can miss curve steepening, flattening, and changes in spreads.

Positive and negative convexity are not interchangeable risk labels

Option-free fixed-cash-flow bonds usually display positive convexity. Embedded options can alter that pattern. Callable corporate or municipal bonds may exhibit negative convexity when rates fall because declining yields can make redemption more likely, limiting the bondholder’s price upside and shortening expected cash flows. Mortgage-related instruments can show similar behavior when refinancing incentives change expected principal payments. In these cases, a duration estimate calculated at one yield level may change substantially after rates move.

Effective duration and effective convexity are designed for securities whose expected cash flows vary with interest rates. Rather than assuming fixed promised payments, they reprice the instrument under specified yield or curve shocks and infer sensitivity from the resulting prices. These measures should not be compared mechanically with analytical duration and convexity for a plain fixed-coupon Treasury unless the shock design, option model, and cash-flow assumptions are understood. Model inputs, including volatility and prepayment or call behavior, can influence the reported result.

Spread duration is another nearby but distinct measure. It estimates sensitivity to a change in credit or option-adjusted spread, holding the relevant benchmark curve framework constant. Convexity with respect to yield does not by itself measure how a corporate bond responds to deteriorating credit conditions or how a municipal bond responds to changes in its relative valuation. A credit portfolio may therefore require interest-rate duration and convexity alongside spread measures.

A disciplined comparison for portfolios and liabilities

For savers and fixed-income students, the practical distinction is that duration estimates the immediate rate sensitivity and convexity refines that estimate when moves are larger. For pension trustees and portfolio managers, the comparison also needs to reflect the liability horizon, instrument options, curve exposure, and the units in which risk is being managed. Borrowers face the mirror image: the lender’s bond-price sensitivity corresponds to changes in the value of a fixed-rate obligation, but refinancing rights and call provisions can change the economic outcome.

A sound comparison starts by matching like with like: use modified duration with yield convexity for a fixed-cash-flow bond; use effective duration with effective convexity for an option-affected bond; use DV01 and dollar convexity for portfolio-value effects; and use key-rate measures when nonparallel curve changes matter. Then test the approximation against full repricing over plausible rate and spread scenarios. Convexity is valuable because it reveals what duration leaves out, but it is not a complete description of interest-rate risk.