Skip to content
All library documents

RFR Swap Floating Legs: Compounded Rates in Arrears

Article Quant Q&A · Author: JakcieJnr

Summary

The document asks how the floating leg of an interest rate swap referencing a risk-free rate is determined over a multi-month accrual period. It contrasts using a forward-looking rate known at the start of the period with compounding daily observed overnight rates through the period, so the final coupon is determined retrospectively. It also asks whether this latter convention makes such swaps equivalent to overnight indexed swaps and how uncertainty in a forward rate should be understood.

The text is a question and offers no answer, market convention, or contract-specific details. The key distinction it raises is between setting a coupon from a term rate at the period’s outset and calculating it from realized overnight fixings over the accrual period. The terminology and mechanics depend on the benchmark, product documentation, and any observation shift or payment lag. The question therefore identifies a useful rates-market distinction but does not establish that all interest rate swaps follow one convention or that reference to an overnight rate alone makes products identical.

Key ideas

  • The question contrasts forward-looking term-rate fixing with compounding overnight fixings through the accrual period.
  • A compounded in-arrears coupon is based on overnight observations made during the period.
  • The document asks whether that structure is equivalent to an overnight indexed swap but does not answer it.
  • The applicable convention depends on the benchmark and contract terms.

Tags

Full text
# Do RFR swaps fix in advance or arrears?


# Do RFR swaps fix in advance or arrears?












Consider the floating leg of a IRS on the RFR which is effective today at $t_1$ and has a payment at $t_1 + 3M$.

My question is, when the payment occurs at $t_1 + 3M$, is this the $3M$ forward rate observed at time $t_1$ (forward-looking), or is it the daily-compounded RFR rates observed throughout the $3M$period (backward-looking).

If the answer is the latter, then isn't that what we do for OIS swaps? Meaning all IRS are now OIS?

If the answer is the former, isn't there a lot of uncertainty about the $3M$ forward RFR rate?

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.