← The collection

Bond Convexity.

Shift the yield of a fixed-coupon bond and compare exact repricing with two approximations.

Interactive experimentintuitiveField note ·
INTERACTIVE EXPERIMENT / M18

Bond Convexity

Shift the yield of a fixed-coupon bond and compare exact repricing with two approximations.

TEST THE MECHANISM

A fictional annual-pay bond has face value 100, coupon 5 and baseline yield 5%. Compare exact repricing with first- and second-order approximations.

220
-33
50.0105.0160.0
Price response to a parallel yield change · Scale 50.0 to 160.0 · Yield change −3 to +3 percentage pointsExact priceLinear durationWith convexity
Exact shifted price
92.64
Linear approximation
92.28
Second-order approximation
92.65
Convexity measure
75.00

Convexity captures curvature, not a guaranteed positive return. This option-free fixed-cash-flow example omits credit risk, callable features, market conventions and nonparallel yield-curve shifts.

This annual-pay option-free bond omits default, embedded options, day-count conventions and nonparallel yield changes.

THE SHORT VERSION

Bond Convexity, explained.

Bond convexity measures the curvature of the price–yield relationship, supplementing duration’s first-order sensitivity.

01 / THE MECHANISM

Why it happens

A face-100 annual bond pays coupons of five. Exact discounted cash flows are compared with a linear duration approximation and a quadratic approximation using the second derivative at a 5% baseline yield.

Bond convexity measures the curvature of the price–yield relationship, supplementing duration’s first-order sensitivity.

Read the result

Compare exact price and both approximations at the selected yield shift. A convexity measure is a sensitivity, not a return forecast.

02 / FOLLOW IT THROUGH

A worked example

Reprice a fixed-coupon bond

  1. Start with the face-one-hundred bond paying five annually at a five-percent yield.

  2. Move the yield by one percentage point and compare linear and quadratic estimates with exact discounting.

  3. Try a larger move or longer maturity to inspect where a local approximation becomes less accurate.

OPTIONAL DEEPER DETAILGo deeper: inside the model

Inside this model

A face-100 annual bond pays coupons of five. Exact discounted cash flows are compared with a linear duration approximation and a quadratic approximation using the second derivative at a 5% baseline yield.

03 / BEYOND THE EXPERIMENT

Where this idea is useful

A practical use

When a yield move is large, inspect approximation error instead of extending a local linear sensitivity without qualification.

CHECK YOUR INTUITION

A common misconception

THE TEMPTING CONCLUSION

“Duration exactly predicts every yield-driven price change.”

THE MORE USEFUL DISTINCTION

Duration is a first derivative evaluated locally. Convexity adds curvature, and even the second-order approximation can diverge for larger changes.

What this explanation leaves out

  • This annual-pay option-free bond omits default, embedded options, day-count conventions and nonparallel yield changes.
ONE MORE QUESTION

Do all bonds have this same positive curvature?

No. Embedded options and changing cash flows can alter the relationship; this board uses an option-free fixed-payment bond.

TAKE THE IDEA WITH YOU

How large a move can your local sensitivity reasonably approximate?

Associated thinkers

Associated withFrank Fabozzi ↗

Associations marked provisional are awaiting source review.

Further reading

Explore the original research or the teaching reference behind this experiment.