Estimating Floating-Rate Bond P&L from Discount Margin Changes
Summary
The discussion asks whether a floating-rate note’s discount margin (DM) change can approximate its credit-related return, in the way spread changes and duration are used for fixed-rate bonds. One proposed method estimates the bond’s price sensitivity to a one-basis-point DM move, then multiplies that sensitivity by the observed DM change. The calculation can use a rough price estimate based on DM, time to maturity, and coupon spread, though using more exact prices may reduce noise. The residual P&L captures effects not explained by the DM move.
The responses stress that total P&L also reflects carry, rolldown, interest-rate changes, and potentially credit convexity and cross-effects. They suggest a credit spread term structure may be preferable to DM when available. Another response identifies option-adjusted spread duration as a measure of spread sensitivity distinct from the short interest-rate duration of a floater. The example illustrates that spread exposure can remain material even when coupon resets limit rate duration. These are attribution and approximation methods; the discussion does not provide a worked numerical P&L estimate or claim DM alone explains returns.
Key ideas
- Estimate DM sensitivity as the price change for a one-basis-point move, holding other inputs constant.
- Multiply that sensitivity by the observed DM change to approximate the P&L attributed to spread movement.
- Account separately for carry, rolldown, interest-rate changes, and other effects.
- Option-adjusted spread duration can measure a floater’s credit spread sensitivity independently of its rate duration.
Tags
Full text
# Is there a way to estimate PnL of floating-rate corporate bonds using Discount margin? # Is there a way to estimate PnL of floating-rate corporate bonds using Discount margin? Is there a way to estimate PnL of floating rate corporate bonds using Discount margin? To illustrate the problem: For Fixed rate bonds I would use the weighted duration of a bond and multiply that with the spread change compared to similar maturity government bond (+ a carry term) to get a first order approximation of a duration-hedged credit long position. Can something similar be used for Floating rate notes using the change in discount margin? I am trying to avoid using dirty prices, accrued interest and coupon payments in the estimation process. For example, if in a months time the discount margin of a floating rate bond has tightened from 30bps from 200bps to 170bps over EURIBOR, can the positive return be approximated, without using prices? If I plot discount margin vs clean price of a floater it shows some inverse proportionality with a magnitude approximately matching the maturity, but not precisely and constantly. As the floater has almost no duration risk, but mostly credit risk, can the DM be used as some kind of spread to estimate pnL? Also appreciated if you know any good reads on this topic thanks, Alex ## Answer by Dimitri Vulis (score 0) https://quant.stackexchange.com/a/81791 Your P&L explain should include nor only credit, but also carry & rolldown, interest rates like any interest rates sensitive instrument. Yes, you can calculate a "discount margin delta" - the P&L if the DM moves 1 bp ceteris paribus, then see how many bps the DM actually changed, and multiply this change by the sensitivity to estimate the P&L attributable to this. I.e. at time $T_0$, you observe the price and the corresponding $DM_0$. (And by "corresponding", I mean that you could make very rough estimate of the price from the DM, time to maturity, and coupon spread, rather than calculate more exactly, but that would increase the noise.) You recalculate what the price from DM $\pm$ 1 bp, and call the price difference delta. At time $T_1$, you again observe the price and the corresponding $DM_1$; you multiply the DM delta by $DM_1 - DM_0$ and attribute this to the DM change, and the rest of the P&L is unexplained. However it would be better to use some credit spread that has a term structure, rather than DM. If you need to minimize unexplained P&L, then include credit gamma and various cross gammas. ## Answer by Alex (score 0) https://quant.stackexchange.com/a/81801 There is a field for this called option-adjusted spread duration. For fixed rate bonds this is the same as OAD, but for FRN this is different and can be used as the price sensitivity to spread or DM changes. For example for a quarterly EURIBOR+100 3 year FRN the OAD is between 0 and 0.25 years (floating index reset period) while the OASD is still somewhere around 2.5years
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.