Why do Eurodollar futures (and now SOFR futures) need a convexity adjustment when used for swap pricing?
I'm building a swap curve using futures rates in my FRM prep. My textbook mentions that futures rates are biased upward compared to forward rates and you need a 'convexity adjustment.' Can someone explain intuitively why this bias exists and how to fix it?
The convexity adjustment is one of the trickiest concepts in fixed-income derivatives. Here's the intuition and the math.
Why Futures Rates != Forward Rates:
Futures contracts are marked to market daily, meaning gains and losses are settled in cash every day. Forward rate agreements (FRAs) settle only at maturity. This daily settlement creates a systematic bias:
- When rates rise, futures positions lose money, and you reinvest those losses at higher rates
- When rates fall, futures positions gain money, but you reinvest gains at lower rates
This asymmetry hurts the long position — you get cash when reinvestment rates are bad and lose cash when rates are good. To compensate, futures rates must be higher than the equivalent forward rate.
The Adjustment Formula:
Forward Rate = Futures Rate - (1/2) x sigma^2 x T1 x T2
Where:
- sigma = volatility of the short rate
- T1 = time to futures expiration
- T2 = time to end of the futures underlying period
Worked Example:
Suppose a 2-year SOFR futures contract implies a rate of 4.50%, rate volatility is 1.2% (0.012), T1 = 2.0, T2 = 2.25.
Convexity adjustment = 0.5 x (0.012)^2 x 2.0 x 2.25 = 0.5 x 0.000144 x 4.5 = 0.000324 = 3.24 bps
Forward rate = 4.50% - 0.0324% = 4.4676%
Key Points for FRM:
- The adjustment grows with the square of maturity — negligible for short-dated contracts, significant for 5+ year maturities
- Higher volatility increases the adjustment
- SOFR futures replaced Eurodollar futures post-LIBOR transition, but the convexity issue is identical
- Ignoring this adjustment overestimates swap fixed rates
Master this topic with our FRM Part I practice problems in the question bank.
Master Part I with our FRM Course
64 lessons · 120+ hours· Expert instruction
Related Questions
Why is DV01 so much smaller than dollar duration if both are supposed to measure rate risk?
When should I stop using modified duration and switch to effective duration?
How should I think about the relationship between Macaulay duration and modified duration instead of memorizing two separate definitions?
Why do hedge calculations often use dollar duration or DV01 instead of just modified duration?
When should I prefer historical simulation VaR over delta-normal VaR?
Join the Discussion
Ask questions and get expert answers.