Option-Adjusted Spread OAS: Measure Spread Beyond TTM

Core Description

• Option-Adjusted Spread (OAS) is a model-based spread that aims to compare bonds with embedded options (calls, puts, prepayment) to option-free bonds on a more “apples-to-apples” basis.
• It backs out the estimated value of the embedded option by modeling how cash flows change across many interest-rate scenarios, then finds the constant spread that makes the modeled price match the market price.
• OAS is useful in MBS, ABS, and callable corporate markets, but it can be misleading if the rate curve, volatility, or behavioral assumptions (such as prepayment or call exercise) are inconsistent or unrealistic.

Definition and Background

What Option-Adjusted Spread Means in Plain English

Option-Adjusted Spread is the constant spread added to every point on a chosen benchmark yield curve so that the model value of a bond, after explicitly modeling embedded option behavior, equals its observed market price. The key phrase is “after modeling”. OAS aims to isolate the compensation investors receive for risks other than the option itself, such as credit risk, liquidity risk, and structure-related uncertainty.

If two bonds have the same maturity but one is callable, their quoted yields are not directly comparable because the callable bond’s cash flows can change when rates move. OAS is designed to adjust for that difference by treating the option as a priced feature rather than ignoring it.

Why OAS Became Standard in Modern Fixed Income

OAS gained importance as securitized products and option-embedded structures expanded:

• Mortgage-Backed Securities (MBS) and many Asset-Backed Securities (ABS) have cash flows that change when borrowers refinance or prepay.
• Callable corporate and municipal bonds can be redeemed early by the issuer, often when rates fall.
• Certain structured notes have explicit call or put schedules, or path-dependent features.

Market participants needed a spread metric that better reflected “option-free compensation”. Over time, OAS became common in relative value discussions (for example, “which bond pays more spread after adjusting for optionality?”), risk reporting, and performance attribution (separating rate moves, option value changes, and spread moves).

What OAS Is Not

OAS is often mistaken for a pure credit spread. It is not. Even after option adjustment, OAS may still include:

• Liquidity premium (hard-to-trade bonds can show wider OAS)
• Model risk premium (investors may demand compensation for uncertainty in the assumptions)
• Sector and structure effects (for example, refinancing friction, servicing, collateral quality)

OAS is best viewed as a model-implied spread measure, not a direct readout of issuer default risk.

Calculation Methods and Applications

How OAS Is Calculated (Process Logic)

In practice, OAS is solved numerically. A common workflow is:

1. Choose a benchmark curve (commonly an overnight-indexed swap curve, or a government curve depending on desk conventions).
2. Generate many interest-rate scenarios (rate paths) using a term-structure model.
3. For each rate path, project cash flows based on the embedded option:
   • For MBS: prepayment or refinancing behavior changes principal timing.
   • For callable bonds: the issuer may call when it is economically favorable.
4. Discount each path’s cash flows using the benchmark curve plus a trial spread.
5. Average discounted values across paths to get a model price.
6. Adjust the trial spread until the model price matches the market price. That constant spread is the Option-Adjusted Spread.

A commonly cited pricing identity for a bond price is the present value of expected cash flows. In OAS frameworks, the “expected” part is created by scenario simulation and option behavior rules:

\[P=\mathbb{E}\left[\sum_{t} \frac{CF_t}{(1+r_t+\text{OAS})^t}\right]\]

Here, \(P\) is the observed price, \(CF_t\) are cash flows that may vary by scenario due to option exercise, and \(r_t\) are discount rates from the chosen curve. Implementations typically use short-rate trees (lattice models) or Monte Carlo simulation. A practical point is that OAS depends on both the curve and the option model.

What Inputs Matter Most

Even when two platforms both report “OAS”, their results can differ materially if they use different assumptions. The most influential inputs are:

• Benchmark curve choice: OIS vs Treasury vs agency curve conventions can change the base discounting level.
• Interest-rate volatility: Higher volatility usually increases option value, which can change OAS meaningfully.
• Option exercise rule:
  • Callable corporates: issuer call policy (for example, optimal vs conservative vs notice or operational friction).
  • MBS: prepayment model (refinancing incentive, burnout, seasonality, turnover, servicing constraints).
• Calibration and scenario design: mean reversion, distribution shape, and correlation structure.

Because OAS is a solution to “make the model match the price”, changes in assumptions can move OAS even when the security itself did not change.

Where OAS Is Used in the Real World

OAS is widely used where embedded options dominate cash-flow timing.

MBS (Mortgage-Backed Securities)

Investors compare pools or coupons by OAS to evaluate compensation after accounting for prepayment risk and negative convexity. In many market discussions, OAS helps explain why two MBS with similar headline yields may behave very differently when rates rally.

ABS and Structured Credit

Some ABS tranches have structural features that create option-like behavior (for example, triggers, clean-up calls, revolving periods). OAS helps align valuation when cash-flow timing depends on path and structure, not just maturity.

Callable Corporate Bonds

A callable corporate bond may offer a higher nominal yield than a non-callable bond from the same issuer, but part of that extra yield can be compensation to the investor for effectively selling a call option to the issuer. OAS attempts to separate that option value from the remaining spread.

Risk and Performance Attribution

Risk teams often monitor OAS alongside:

• option-adjusted duration,
• convexity (including negative convexity in MBS),
• spread duration,

to differentiate “spread widening” from “option value changes driven by volatility”.

Comparison, Advantages, and Common Misconceptions

OAS vs Other Spread Measures

Different spread metrics answer different questions.

If a bond’s cash flows can materially change because of optionality, Z-spread can be misleading because it assumes fixed cash flows. OAS exists to address that mismatch.

Advantages of Option-Adjusted Spread

• Improved comparability: OAS aims to compare a callable bond to a non-callable bond on an option-neutralized basis.
• Scenario-aware pricing: It links valuation to multiple rate paths rather than a single yield assumption.
• Risk decomposition: When used with option-adjusted duration and convexity, OAS can support clearer attribution narratives, such as distinguishing spread moves from volatility-driven option value changes.

Limitations and Pitfalls

• Model dependence: Different models can produce different OAS values for the same bond on the same day.
• Volatility sensitivity: If the assumed rate volatility changes, the option value changes, and OAS can move even if the market price is unchanged.
• Behavioral assumption risk: Prepayment or call exercise is influenced by human and operational behavior, and it can shift in stressed markets.
• Liquidity and stress regimes: In illiquid markets, the observed price may include large liquidity discounts or premia. OAS can reflect that, but it may be misread as a change in fundamental risk.

Common Misunderstandings

“OAS equals credit spread”

Not necessarily. OAS can include compensation for multiple risks, and it is anchored to a specific curve and modeling framework. Two dealers can produce different OAS values for the same security without any change in credit.

“Higher OAS always means cheaper”

A higher OAS might indicate higher compensation, but it might also reflect:

• unrealistic prepayment assumptions,
• an inconsistent volatility surface,
• stale prices or liquidity distortions,
• a curve mismatch versus peers.

“I can compare OAS across any sectors”

Comparing OAS across very different liquidity regimes and structures can be misleading. For example, comparing an agency MBS OAS to a thinly traded callable corporate OAS without adjusting for liquidity and model differences can lead to unreliable conclusions.

Practical Guide

A Step-by-Step Checklist for Using Option-Adjusted Spread

Align the benchmark curve

Before comparing Option-Adjusted Spread values, verify:

• which curve is used (OIS, Treasury, or another internal curve),
• day-count and compounding conventions,
• whether curves are interpolated similarly.

If two OAS values come from different curves, the comparison can be unreliable even if the numbers appear precise.

Use a consistent volatility framework

For option-embedded bonds, implied volatility (or a modeled term volatility) is a key driver. A small change in volatility assumptions can create a large change in OAS for highly callable or prepay-sensitive instruments.

A practical habit is to ask: “What volatility surface and calibration is behind this OAS?” If the answer is unclear, treat the OAS as an indicative output rather than a decision variable.

Sanity-check the option behavior

• Callable bonds: confirm call schedule, call protection, make-whole features, notice periods, and whether the model assumes economically optimal calling.
• MBS: check whether the prepayment model reflects current refinancing frictions (credit availability, loan size, seasoning, servicing constraints).

Compare within similar liquidity and structure buckets

OAS is generally more informative in relative value comparisons within the same product family and similar liquidity tier. Examples include:

• an MBS pool vs a similar MBS pool,
• a callable corporate vs a similar callable corporate (same currency, rating band, and call structure).

Stress-test the OAS story

Rather than treating OAS as a single truth, treat it as an output under assumptions. A common discipline is to:

• re-run (or request) OAS under higher and lower volatility,
• evaluate the impact of a rate rally vs a selloff,
• check whether the “cheap or rich” conclusion remains consistent.

Case Study: A Hypothetical Callable Corporate vs Non-Callable Corporate (Illustrative Only)

The following is a hypothetical example for education only and is not investment advice.

Setup

• Bond A: 5-year non-callable corporate bond, price 100.00, yield-to-maturity 4.60%
• Bond B: 5-year corporate bond callable at par after year 2, price 100.00, yield-to-maturity 4.95%
• Same issuer and seniority, same liquidity tier (assumed), same day

At first glance, Bond B offers an extra 0.35% yield. However, because Bond B is callable, the investor is exposed to reinvestment risk. If rates fall, the issuer may call the bond and refinance, limiting the investor’s upside.

Modeled results (hypothetical)

• Bond A (non-callable): Z-spread about 145 bps. OAS is close to Z-spread because there is no embedded option.
• Bond B (callable): Z-spread about 175 bps, but OAS about 135 bps after modeling the call option value under assumed volatility.

Interpretation

• Z-spread suggests Bond B offers more compensation than Bond A (175 bps vs 145 bps).
• OAS suggests Bond B offers less option-adjusted compensation than Bond A (135 bps vs 145 bps), because part of the extra yield is compensation for the embedded call option sold to the issuer.

How to use this in analysis

• If you rely only on yield-to-maturity or Z-spread, you may overstate Bond B’s relative value.
• If you rely on OAS, you still need to validate assumptions. If volatility is higher than assumed, the call option is typically more valuable, and Bond B’s OAS could be lower.

Case Study: A Hypothetical MBS Pool Comparison Using OAS (Illustrative Only)

This is a hypothetical example for education only.

Setup

• Pool X and Pool Y: similar coupon and vintage, both priced near 102.
• Pool X has more borrowers with smaller loan balances (often associated with different prepayment patterns).
• Pool Y has fewer borrowers with larger balances.

Modeled output (hypothetical)

• Pool X: OAS 20 bps
• Pool Y: OAS 45 bps

Possible interpretation

Pool Y appears to offer a higher Option-Adjusted Spread, which could indicate higher compensation after adjusting for prepayment optionality. However, the conclusion depends on whether the prepayment model appropriately captures:

• refinancing incentive and frictions,
• turnover vs rate-driven prepayment,
• borrower characteristics and servicing behavior.

This example illustrates why OAS is widely used in MBS relative value discussions, while also highlighting that the result depends on modeling assumptions.

Resources for Learning and Improvement

Foundational References

• CFA Institute fixed-income curriculum: term structure, spread measures, and structured product risk are covered in a systematic way.
• Investopedia: useful for quick intuition and terminology cross-checking (best used as a starting point, not a final authority).

Market and Research Reading

• Federal Reserve research on mortgage markets and MBS: helpful for understanding how refinancing waves and rate regimes influence prepayment behavior and risk transmission.
• SEC filings and offering documents for structured products: useful for identifying embedded options, call schedules, triggers, and cash-flow waterfalls.

Practical Skill-Building

• Practice reading a bond’s cash-flow rules directly from official documents, then compare what the OAS model assumes.
• Track how Option-Adjusted Spread changes when rates move sharply, especially for securities known for negative convexity or call risk.

FAQs

Can Option-Adjusted Spread be negative?

Yes. A negative Option-Adjusted Spread can occur when the market price is rich relative to the model’s valuation under the chosen curve and option assumptions. This can reflect strong demand or liquidity effects, or it may indicate that model inputs (especially volatility or behavioral assumptions) are misaligned with current conditions.

Is Option-Adjusted Spread the same as a credit spread?

No. OAS aims to remove the embedded option value, but what remains can still include liquidity premium, sector effects, and model uncertainty. It is not a pure measure of default risk.

Why does Option-Adjusted Spread change when volatility assumptions change?

Because volatility affects the value of the embedded option. Higher volatility typically increases option value (for calls and many prepayment options), which can change the Option-Adjusted Spread required to match the same market price.

Should I use OAS for a non-callable, option-free bond?

Often, OAS is not necessary for a truly option-free bond, and Z-spread may be sufficient. However, some systems still report an OAS number. In that case, it is typically close to Z-spread if cash flows are fixed and there is no material optionality.

Can I compare Option-Adjusted Spread across dealers or data vendors?

You can, but only with caution. Different vendors may use different benchmark curves, volatility surfaces, and option behavior assumptions. If the framework is not consistent, differences in OAS may reflect modeling choices rather than relative value.

Does a higher Option-Adjusted Spread automatically mean better value?

Not automatically. A higher Option-Adjusted Spread can indicate higher compensation, but it can also reflect higher liquidity risk, higher model uncertainty, or an unrealistic assumption set. It is typically used together with scenario checks and a review of key modeling inputs.

Conclusion

Option-Adjusted Spread is widely used for analyzing bonds whose cash flows can change because of embedded options, such as MBS, ABS structures, and callable corporate bonds. By modeling option exercise and discounting across interest-rate scenarios, OAS aims to produce an “option-free” spread that is more comparable across securities than yield-to-maturity or Z-spread alone. The trade-off is model dependence: curve choice, volatility, and behavioral assumptions can materially affect the result. When used consistently within comparable buckets and supported by sensitivity checks, Option-Adjusted Spread can help structure relative value discussions and clarify what is driving price and risk.