Tim Olson
47471b7700
- Flink update_bars debouncing - update_bars subscription idempotency bugfix - Price decimal correction bugfix of previous commit - Add GLM-5.1 model tag alongside renamed GLM-5 - Use short Anthropic model IDs (sonnet/haiku/opus) instead of full version strings - Allow @tags anywhere in message content, not just at start - Return hasOtherContent flag instead of trimmed rest string - Only trigger greeting stream when tag has no other content - Update workspace knowledge base references to platform/workspace and platform/shapes - Hierarchical knowledge base catalog - 151 Trading Strategies knowledge base articles - Shapes knowledge base article - MutateShapes tool instead of workspace patch
61 lines
4.5 KiB
Markdown
61 lines
4.5 KiB
Markdown
---
|
|
description: "Mortgage-backed security (MBS) passthrough trading strategy that buys MBS and hedges interest rate duration with interest rate swaps, leaving prepayment risk as the primary exposure and profit driver."
|
|
tags: [structured-assets, mbs, mortgage, prepayment, duration-hedge, interest-rate-swap]
|
|
---
|
|
|
|
# Mortgage-Backed Security (MBS) Trading
|
|
|
|
**Section**: 11.7 | **Asset Class**: Structured Assets | **Type**: Carry / Relative Value / Duration-Hedged
|
|
|
|
## Overview
|
|
MBS passthrough securities pool mortgage cash flows and pass them through to investors. The primary risk specific to MBS (beyond general interest rate risk) is prepayment risk: homeowners can prepay their mortgages when interest rates fall, refinancing at lower rates. This creates negative convexity in MBS prices. The core trading strategy buys MBS passthroughs and hedges the interest rate duration exposure with interest rate swaps, isolating prepayment risk as the residual P&L driver.
|
|
|
|
## Construction / Mechanics
|
|
**Position:**
|
|
- Long MBS passthrough (receive mortgage cash flows)
|
|
- Short interest rate swap notional calibrated to hedge the MBS duration (pay fixed, receive floating; or receive fixed, pay floating depending on direction)
|
|
|
|
**Prepayment mechanics:**
|
|
- When interest rates fall, homeowners refinance (prepay mortgages early)
|
|
- The MBS investor receives principal early (at par) but loses the above-market coupon stream
|
|
- This creates **negative convexity**: the MBS price rises less than a comparable bond when rates fall, and falls similarly when rates rise
|
|
- The 5-year swap rate is the key reference benchmark for prepayment behaviour
|
|
|
|
**Hedge ratio determination:**
|
|
The hedge ratios are model-dependent and require prepayment models. Two approaches:
|
|
|
|
1. **Model-based**: use a prepayment model to compute duration, then size the swap position to match the MBS dollar duration.
|
|
|
|
2. **Non-parametric / empirical**: using historical data, estimate the first derivative of MBS price P with respect to the 5-year swap rate R:
|
|
- Constrain P to be a non-increasing function of R (reflecting negative convexity)
|
|
- Estimate via constrained regression (e.g., Ait-Sahalia and Duarte, 2003; Duarte, Longstaff and Yu, 2006)
|
|
|
|
## Return Profile
|
|
Profits from:
|
|
1. **Carry**: MBS passthroughs typically offer a spread over Treasuries and swaps as compensation for prepayment risk.
|
|
2. **Prepayment mispricing**: if the market overprices prepayment risk (i.e., the MBS is cheap relative to the modelled value), the hedged position profits as spreads tighten.
|
|
|
|
The primary risk is prepayment model error and the negative convexity of MBS when rates decline sharply.
|
|
|
|
## Key Parameters / Signals
|
|
| Parameter | Description |
|
|
|-----------|-------------|
|
|
| Prepayment rate (CPR) | Conditional prepayment rate; key driver of MBS cash flows |
|
|
| Duration | Interest rate sensitivity of MBS; hedge-determining quantity |
|
|
| Hedge instrument | Interest rate swaps (typically 5-year or 10-year) |
|
|
| OAS (option-adjusted spread) | Spread over swap curve after adjusting for the embedded prepayment option; buy high OAS |
|
|
| Negative convexity | MBS price appreciation is capped when rates fall (prepayments accelerate) |
|
|
| Hedge ratio ∂P/∂R | First derivative of MBS price w.r.t. 5-year swap rate; model or regression estimated |
|
|
|
|
## Variations
|
|
- **TBA (to-be-announced) trading**: most liquid MBS market; trade generic pools before specific pools are specified.
|
|
- **Specified pool trading**: target pools with lower prepayment characteristics (e.g., low-loan-balance, geographic concentrations) that command a premium.
|
|
- **IO/PO strips**: trade interest-only or principal-only strips to isolate specific prepayment exposures.
|
|
- **Agency vs. non-agency MBS**: agency MBS (Fannie/Freddie/Ginnie) carry implicit government guarantee; non-agency MBS add credit risk.
|
|
|
|
## Notes
|
|
- Prepayment modelling is complex; the standard PSA (Public Securities Association) benchmark model and its descendants require significant calibration.
|
|
- Negative convexity is the key distinguishing risk vs. standard fixed income: as rates fall, the effective duration of MBS shortens (prepayments accelerate), creating re-hedging costs.
|
|
- The non-parametric hedge approach avoids model specification risk but requires substantial historical data and may be slow to adapt to regime changes in prepayment behaviour.
|
|
- The 2008 financial crisis demonstrated that non-agency MBS carries substantial credit risk in addition to prepayment risk; agency MBS largely avoided credit losses but still suffered spread widening.
|