Hedging a One-Year SOFR Swap with SOFR Futures

Summary

The document asks how to construct futures hedge ratios for a spot-starting one-year SOFR interest-rate swap. The example is dated partway through an IMM futures period and proposes using a strip of SOFR contracts spanning the swap tenor. The proposed weights assign quarter-period exposure to the interior contracts, with smaller adjustments at the front and back to account for partial accrual periods.

The author describes testing the approach across different points in an IMM cycle and finding that the hedge does not offset the swap-rate sensitivity as expected. The central issue is how to map swap DV01 onto futures exposures when the first and last reference periods are partial and the relevant contract sensitivities may not be uniform. The document contains no answer or validated hedge construction, so its proposed ratios should be treated as an unresolved hypothesis. It provides no detailed valuation, curve, or sensitivity calculations to diagnose the mismatch.

Key ideas

• The question concerns delta hedging a spot-starting one-year SOFR swap with a strip of SOFR futures.
• The proposed ratios allocate exposure across full and partial IMM periods.
• The author reports that the proposed hedge does not neutralize swap-rate sensitivity across test dates.
• Partial accrual periods and contract sensitivities are central issues in constructing the hedge.
• The document does not provide a corrected method or supporting calculations.

Tags

# Precisely how do you delta-hedge a spot-1Y SOFR IRS with SOFR futures?


# Precisely how do you delta-hedge a spot-1Y SOFR IRS with SOFR futures?












I'm struggling to construct hedge ratios that delta-hedge a spot-1Y IRS.

Say I'm roughly in the middle of an IMM period, date = Oct 30th 2023 and I trade a 1k dv01 spot-1Y SOFR swap. I'll need some combination of these SOFR futures to hedge out the delta: [Sep23, Dec23, Mar24, Jun24, Sep24]. Below is an approach I've tried:

Use hedge ratios [1/4, 1/4, 1/4, 1/4, 1/8].

My theory being that the middle 3 futures cover full periods of the swap, so I simply take 90/360 of the swap dv01 here.

For the last future, I only need it to cover half of the period as my swap only extends out to half of this period, so I need 45/360 here.

And for the front future, I similarly only need it to cover half the period of the future but here the delta of the future is already halved, since we already have half of the rates from the period. So no adjustment is needed.

I've attempted this running through an IMM to IMM period, systematically multiplying the last 1/4 hedge ratio depending on how far through the IMM period I am but when comparing against swap rates the delta is not hedged. What am I doing wrong?

Shown in full with attribution under the source's licence. Licence: CC BY-SA 4.0 (Stack Exchange)

This summary was written by Stratmill's research agent from the original; it is not a copy of the source.