Average Life and Weighted Average Maturity

Average Life and Weighted Average Maturity (Why Stated Maturity Tells You Almost Nothing)
A 30-year mortgage-backed security rarely behaves like a 30-year bond. Borrowers prepay. Defaults accelerate principal returns. Amortization schedules front-load payments. The stated maturity date -- the legal final -- is the outer boundary, not the expected reality. If you price an amortizing security off its stated maturity, you will systematically misprice duration, yield, and risk.
The metrics that actually matter are weighted average life (WAL) and weighted average maturity (WAM). They measure when you actually get your principal back, not when the bond's legal documents say you might.
The point is: for any amortizing or prepayable security, WAL and WAM replace stated maturity as the relevant time horizon for pricing, duration estimation, and portfolio construction.
Weighted Average Life (WAL): The Core Metric
WAL is the weighted average of the time to receipt of each dollar of principal. It tells you, on average, how long each dollar of your principal investment is outstanding.
The Formula
WAL = (1 / Total Principal) x SUM [t(i) x P(i)]
Where:
- t(i) = time (in years) to the i-th principal payment
- P(i) = the amount of the i-th principal payment
- Total Principal = sum of all principal payments (equals the bond's face value)
Worked Example: Simple Amortizing Bond
Consider a $1,000 bond that returns principal in three annual installments:
| Year | Principal Payment | Calculation Component |
|---|---|---|
| 1 | $200 | 1 x $200 = $200 |
| 2 | $300 | 2 x $300 = $600 |
| 3 | $500 | 3 x $500 = $1,500 |
| Total | $1,000 | $2,300 |
WAL = $2,300 / $1,000 = 2.30 years
Even though the bond's legal maturity is 3 years, its WAL is only 2.30 years because more than half the principal is returned before maturity. (Corporate Finance Institute, "Average Life," 2023.)
The durable lesson: WAL is always less than or equal to stated maturity for any amortizing security. It equals stated maturity only for a bullet bond (which returns 100% of principal at maturity).
Worked Example: 30-Year Pass-Through MBS
Now consider a more realistic case: a $10 million GNMA pass-through backed by 30-year fixed-rate mortgages with a 6.0% coupon, assuming 165% PSA prepayment speed.
Under the PSA prepayment model (developed by the Public Securities Association, now SIFMA), the standard 100% PSA benchmark assumes:
- Month 1: annualized CPR of 0.2%
- CPR increases by 0.2% each month
- Month 30: CPR reaches 6.0% and remains flat thereafter
At 165% PSA, prepayments are 65% faster than the benchmark. The CPR ramps to 9.9% by month 30 and stays there.
The impact on WAL is dramatic:
| PSA Speed | Long-Run CPR | Approximate WAL | Approximate Macaulay Duration |
|---|---|---|---|
| 0% PSA (no prepayments) | 0% | 20.5 years | ~12.5 years |
| 100% PSA | 6.0% | 11.8 years | ~7.2 years |
| 150% PSA | 9.0% | 8.4 years | ~5.5 years |
| 200% PSA | 12.0% | 6.5 years | ~4.4 years |
| 300% PSA | 18.0% | 4.3 years | ~3.1 years |
(SIFMA, "Standard Formulas for the Analysis of Mortgage-Backed Securities," Chapter SF; Wikipedia, "PSA Prepayment Model.")
Faster prepayments → shorter WAL → shorter duration → lower interest rate sensitivity.
The point is: a 30-year MBS at 200% PSA behaves more like a 6.5-year bond than a 30-year bond. If you hedge it as a 30-year instrument, your hedge ratio is off by a factor of three or more.
Weighted Average Maturity (WAM): The Collateral Metric
WAM is distinct from WAL. It measures the weighted average time to the final scheduled maturity of each loan in the underlying pool, weighted by each loan's outstanding principal balance.
The Formula
WAM = SUM [w(i) x M(i)]
Where:
- w(i) = loan i's principal balance / total pool balance
- M(i) = remaining months to maturity for loan i
Why WAM Matters
WAM tells you the contractual maturity profile of the underlying collateral -- how long the loans are scheduled to be outstanding, ignoring prepayments. It is a property of the collateral pool, not the securities.
For a pool of 30-year mortgages originated over several months, WAM might be 348-360 months at issuance (reflecting that some loans were originated a few months before the securitization closed).
WAM vs. WAL: The Critical Distinction
| Metric | What It Measures | Prepayment Assumption | Applies To |
|---|---|---|---|
| WAM | Weighted average of contractual maturities of underlying loans | None (scheduled maturity only) | Collateral pool |
| WAL | Weighted average time to principal receipt | Yes (requires prepayment model) | Securities/tranches |
WAM is a pool characteristic. WAL is a security characteristic.
Why this matters: two MBS tranches backed by the same pool (same WAM) can have vastly different WALs depending on their position in the waterfall. A sequential-pay Class A tranche receives principal first and has a shorter WAL. A Class Z (accrual) tranche receives principal last and has a longer WAL. Same pool, same WAM, different WALs.
The test: if someone quotes you a "maturity" for an MBS without specifying whether they mean WAM, WAL, or legal final, ask which one. The answer changes everything about how you evaluate the security.
WAL in Non-MBS Asset Classes
WAL is not exclusive to mortgage products. It applies to any security with scheduled or expected principal amortization:
CLO Tranches
A CLO's AAA tranche might have a stated maturity of 12-13 years but a WAL of 4-6 years (because the reinvestment period ends after 4-5 years, and sequential amortization then pays down the senior tranche rapidly).
Auto ABS
Auto loan ABS (backed by 60-72 month auto loans) have short stated maturities. A senior tranche's WAL is typically 1.0-2.0 years because amortization and prepayments return principal quickly. (Corporate Finance Institute, 2023.)
Credit Card ABS
Credit card ABS have revolving periods (during which principal collections are used to buy new receivables) followed by amortization periods. WAL depends heavily on the length of the revolving period.
CMBS
CMBS are backed by commercial mortgages that are typically interest-only for most of their term with a balloon payment at maturity. WAL for a conduit CMBS senior tranche is heavily concentrated around the balloon dates (typically years 7-10), giving WAL values of 8-10 years despite 30-year legal final maturities.
Prepayment Sensitivity: The WAL Table
Practitioners produce WAL tables (also called yield tables or prepayment sensitivity tables) that show a security's WAL, yield, and duration across a range of prepayment speeds. This is the standard analytical output for any MBS.
Example WAL Table: FNMA 30-Year 5.5% Pass-Through (Hypothetical)
| PSA Speed | WAL (years) | Yield at 99.50 Price | Effective Duration |
|---|---|---|---|
| 75% | 14.2 | 5.58% | 8.9 |
| 100% | 11.8 | 5.55% | 7.4 |
| 125% | 9.9 | 5.53% | 6.3 |
| 150% | 8.4 | 5.51% | 5.5 |
| 175% | 7.2 | 5.49% | 4.8 |
| 200% | 6.5 | 5.48% | 4.4 |
| 250% | 5.1 | 5.45% | 3.5 |
| 300% | 4.3 | 5.43% | 3.0 |
(Values are illustrative, based on standard MBS analytics; SIFMA, Chapter SF.)
Lower rates → faster prepayments → shorter WAL → lower duration → negative convexity.
The durable lesson: this table is the primary analytical tool for MBS. If you cannot produce or interpret a WAL table, you cannot trade mortgage-backed securities. Period.
Why WAL Matters for Portfolio Management
Duration Estimation
For amortizing securities, Macaulay duration is bounded by WAL. You cannot have a duration longer than your WAL (because you cannot have more interest rate sensitivity than the time horizon of your cash flows). Portfolio managers use WAL as a quick proxy for duration when full analytics are unavailable.
Reinvestment Risk
A shorter WAL means you receive principal earlier and must reinvest it at prevailing rates. In a falling rate environment, this is reinvestment risk -- you get your money back faster but have to put it to work at lower yields. Prepayment risk and reinvestment risk are two sides of the same coin, and WAL quantifies the exposure.
Extension Risk
Conversely, if prepayments slow (because rates rise), WAL extends. A pass-through you expected to have a 7-year WAL might extend to 12 years. This is extension risk -- the security stays outstanding longer than expected, and you are locked into a below-market coupon. WAL sensitivity tables (as shown above) quantify this risk across scenarios.
The point is: WAL is not a static number. It is a function of the interest rate environment. Every WAL you see is conditional on a prepayment assumption. Change the assumption, and the WAL changes with it.
CPR vs. PSA: The Two Prepayment Languages
Practitioners express prepayment speeds in two equivalent formats:
Constant Prepayment Rate (CPR)
CPR is an annualized rate of unscheduled principal prepayment, expressed as a percentage of the remaining principal balance. A 10% CPR means that 10% of the remaining pool balance is expected to prepay each year (assuming the rate is constant).
The monthly equivalent is the Single Monthly Mortality (SMM): SMM = 1 - (1 - CPR)^(1/12).
For a 10% CPR: SMM = 1 - (1 - 0.10)^(1/12) = 0.874% per month.
Public Securities Association (PSA) Benchmark
The PSA model (now maintained by SIFMA) provides a standardized prepayment ramp:
- 100% PSA: CPR starts at 0.2% in month 1, increases by 0.2% monthly, reaches 6% in month 30, and remains at 6% thereafter.
- 200% PSA: Double the ramp -- CPR reaches 12% in month 30.
- 50% PSA: Half the ramp -- CPR reaches 3% in month 30.
Quick conversion: Long-run CPR = PSA% x 0.06. So 150% PSA = 9% long-run CPR; 250% PSA = 15% long-run CPR. (SIFMA, Chapter SF; CCB Financial, "Prepayment Rate Assumptions: The Difference Between CPR and PSA.")
Why this matters: when someone says "priced at 150 PSA," they are telling you the prepayment assumption embedded in the yield calculation. You need to evaluate whether that assumption is realistic given the current rate environment, the pool's characteristics (loan age, coupon, geography), and borrower behavior.
WAL for Structured Tranches (CMO/REMIC)
Collateralized mortgage obligations (CMOs) and REMICs carve a pass-through's cash flows into tranches with targeted WAL profiles. The same collateral pool produces tranches with very different WALs:
| Tranche Type | Typical WAL | Mechanism |
|---|---|---|
| PAC (Planned Amortization Class) | Stable within a band | Sinks prepayment variability into companion tranches |
| Companion/Support | Highly volatile | Absorbs excess or deficient prepayments |
| Sequential A (short) | 2-4 years | Receives all principal first |
| Sequential Z (long) | 15-20 years | Accrues interest; receives principal last |
| TAC (Targeted Amortization Class) | Stable at one speed | Protection against one-directional prepayment risk |
(FINRA, "MBA Dissemination Attachment A," 2021.)
The point is: CMO structuring is the art of redistributing WAL uncertainty. PAC tranches get stable WALs; companion tranches absorb the volatility. Somebody always bears the prepayment risk -- the question is who.
Historical Case: The 2020-2021 Refinancing Wave
When the Federal Reserve cut rates to near zero in March 2020 and began purchasing MBS at scale, prepayment speeds on 30-year conventionals spiked to 40-50% CPR (roughly 700-800% PSA) for in-the-money coupons. (SIFMA mortgage prepayment data, 2020-2021.)
The consequences for WAL:
- A FNMA 3.5% pass-through that had been priced at a 7-year WAL (assuming ~175% PSA) saw its realized WAL compress to roughly 2-3 years as borrowers refinanced aggressively.
- MBS portfolio managers who had built duration targets around expected WALs found their portfolios dramatically shorter than planned.
- Companion tranches in CMO structures -- designed to absorb prepayment variability -- saw their WALs collapse from 10+ years to under 2 years, producing significant negative convexity losses.
The durable lesson: the 2020-2021 episode demonstrated that WAL assumptions can be catastrophically wrong when rate environments shift violently. Your prepayment model is a forecast, not a fact. Scenario analysis across a wide range of PSA speeds is not optional.
Practitioner Checklist: WAL and WAM Analysis
Essential:
- Confirm whether a quoted "maturity" is WAM, WAL, or legal final
- Obtain WAL under at least three prepayment scenarios (slow, base, fast)
- Use WAL (not stated maturity) for duration and yield calculations on amortizing securities
High-impact:
- Build or obtain a full WAL sensitivity table across PSA speeds (or CPR scenarios)
- Compare current prepayment speeds (from factor reports) to the WAL assumption in your pricing model
- For CMO tranches, understand which tranche type you own (PAC, companion, sequential) and its WAL stability
Optional:
- Model WAL under path-dependent prepayment scenarios (not just constant CPR, but varying speeds over time)
- Analyze the "WAL drift" of your portfolio over time -- how realized WALs compare to projected WALs
- For CLOs, model WAL separately for the reinvestment period and amortization period
Your Next Step
For every amortizing security in your portfolio, find the WAL under three scenarios: current speeds, 50% faster, and 50% slower. Calculate the dollar duration change across those scenarios. That spread -- the range of possible duration outcomes -- is your prepayment risk exposure, and managing it is the central challenge of MBS portfolio management.
Sources:
- Corporate Finance Institute, "Average Life -- Overview, Calculation, Risks," 2023.
- Wikipedia, "Weighted-average life."
- SIFMA, "Standard Formulas for the Analysis of Mortgage-Backed Securities and Other Related Securities," Chapter SF.
- Wikipedia, "PSA Prepayment Model."
- CCB Financial, "Prepayment Rate Assumptions: The Difference Between CPR and PSA."
- FinanceTrain, "MBS Weighted Average Life."
- FINRA, "MBA Dissemination Attachment A," February 2021.
- Financial Analyst Guide, "Weighted Average Life (WAL) Calculations," CFA Level 1.
Related Articles
Extension Risk in Rising Rate Environments
Extension risk amplifies losses in rising rate environments; structured products require proactive duration management to mitigate cash flow instability.
Stress Testing Securitized Portfolios
In the Federal Reserve's 2025 severely adverse stress test scenario, house prices decline 33%, commercial real estate drops 30%, and the unemployment rate rises from 4.1% to a peak of 10.0% (Federa...
How Structured Products Trade in Secondary Markets
If you have ever tried to sell a mezzanine ABS tranche and received a bid 3-5 points below where you thought the bond was marked, you have experienced the central reality of structured product seco...