Hedging Mortgage-Backed Securities with Rate and Volatility Sensitivities
Summary
The document outlines why mortgage-backed security derivatives require a broader hedge than a conventional bond. A basic bond hedge can use key-rate durations to represent exposure across points on the Treasury curve, but mortgage cash flows also depend on embedded prepayment options. The answer emphasizes that this makes MBS behavior more complex than that of a callable bond whose exercise depends mainly on rates and credit spreads.
A valuation model can estimate sensitivities to interest rates, implied volatility, and other parameters. The answer says rate risk is commonly hedged with over-the-counter swaps or exchange-traded interest-rate futures, while some large institutions hedge callable instruments’ rate vega with swaptions. These are dynamic hedges because sensitivities change over time, and duration alone is not presented as a convenient way to size hedge notionals. The response recommends several books and industry primers, but supplies no model specification, hedge example, or performance evidence; the quality and applicability of each resource must be assessed separately.
Key ideas
- MBS derivative risk includes interest-rate, volatility, and other model sensitivities.
- Prepayment options make mortgage-backed securities more complex than ordinary callable bonds.
- Interest-rate swaps and futures are cited as instruments for hedging rate exposures.
- Swaptions can be used to hedge interest-rate vega in some callable positions.
- Hedges need adjustment as modeled sensitivities change over time.
Tags
Full text
# Resources for MBS Derivative hedging? # Resources for MBS Derivative hedging? I was wondering if anyone could suggest resources that delve into how MBS deriv books are managed. I've read fabozzi but he only touches very briefly on OAS modeling which seems key to MBS derivative valuation and it seems there's multiple elements to hedging. From my understanding whereas for a traditional "regular bond" you would hedge the duration risk by calculating "key rate" durations at points across the treasury curve and using those numbers to hedge. But for hedging derivatives it seems there's a volatility component, there's also some element of mortgage spreads you would need to hedge. It seems to be a very niche topic and I can't find any good resources that really dive into it. Are there texts that give the topic a thorough treatment? ## Answer by Dimitri Vulis (score 1, accepted) https://quant.stackexchange.com/a/83606 Books: I recommend Andrew Davidson, Alexander Levin. Mortgage Valuation Models: Embedded Options, Risk, and Uncertainty. Oxford University Press (2014). Somewhat dated, but good: Andrew S. Davidson, Michael D. Herskovitz. Mortgage-Backed Securities: Investment Analysis & Advanced Valuation Techniques (1993) and its companion is The Mortgage-Backed Securities Workbook. Also: Niels Rom. Callable Mortgage Bonds: Numerical Methods and Valuation Models for Pricing and Risk Analysis Edit: also look (on Torrent etc) for: JPMorgan. MBS Primer (2006) Lehman Bros. Srinivas Modukuri. Mortgage Convexity Risk (2003) Lehman Bros. Mortgage Options: A Primer (2004) RBS Greenwich Capital. A Guide to US MBS (2007) Hedging: Yes, a model outputs the MBS's sensitivities to interest rates, to interest rate implied volatilities (vega), and to other parameters. MBS with prepayment options are much more complicated than just callable bonds, where the decision whether to call can be assumed to be driven just by the interest rates and credit spreads. Durations are not a convenient way to calculate interest rate hedge notionals. Many people hedge the interest rate sensitivities with with OTC interest rate swaps and/or exchange-traded interest rate futures. Some large institutions also hedge callables' interest rate vega with OTC interest rate swaptions. These are dynamic hedges that need to be adjusted as the MBS sensitivities change.
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.