Understanding Option-Adjusted Spread (OAS)
The option-adjusted spread (OAS) measures the yield spread a fixed-income security offers over a risk-free benchmark (typically Treasuries) after adjusting for any embedded options—such as callable or mortgage-backed securities (MBS) features—that can alter the bond’s future cash flows. OAS helps investors estimate the compensation they receive for credit, liquidity and other risks once option-related uncertainties are removed.
Key takeaways
OAS isolates the spread attributable to credit and other non-option risks by stripping out the value of embedded options.
It produces a more accurate relative-value measure for securities with embedded options than yield-to-maturity alone.
Estimating OAS requires modeling future interest-rate paths and option exercise (e.g., prepayment) behavior, often via Monte Carlo simulations.
OAS differs from the Z-spread: the Z-spread is a static constant added to each point of the yield curve and does not reflect option value.
How OAS is determined
Separate the security into two pieces: a vanilla bond and an embedded option.
Model many possible future interest-rate scenarios (and, for MBS, borrower prepayment behavior).
Value the bond’s cash flows under each scenario, net out the option value, and find the spread (over the risk-free curve) that equates the option-free bond’s present value to market price.
The resulting OAS represents the spread attributable to non-option risks.
How options and volatility affect OAS
Embedded options change expected cash flows:
Call options allow issuers to redeem early, reducing future coupon income for investors.
Put options allow holders to sell the bond back, limiting downside for investors.
Two primary volatility sources for OAS calculations:
Interest-rate volatility, which affects the likelihood of option exercise.
Prepayment volatility (for MBS), driven by borrower behavior and economic incentives.
Greater volatility typically increases the value of options and therefore alters the OAS necessary to match observed prices.
Limitations: OAS depends on model assumptions and historical inputs; inaccurate assumptions about future rate dynamics or borrower behavior can misstate the spread.
OAS versus Z-spread
Z-spread: the constant spread added to each point on the Treasury curve that makes a bond’s discounted cash flows equal to its market price. It is a static measure and does not account for embedded option value.
OAS: adjusts the Z-spread by removing the value of embedded options, yielding a dynamic, model-dependent measure that reflects option-related risks (e.g., prepayment or call risk).
* Use Z-spread for option-free comparisons and OAS when embedded options materially affect cash flows.
Example: mortgage-backed securities (MBS)
MBS include embedded prepayment options because homeowners can refinance or repay mortgages early.
When rates fall, prepayment risk rises—borrowers refinance, shortening cash flows to MBS holders and reducing expected yield.
* OAS quantifies the spread investors require after accounting for expected prepayments. A larger OAS indicates higher compensation for risks (credit, liquidity, model risk) beyond option-related effects.
Practical uses and cautions
Uses:
Relative-value comparisons among securities with differing option features.
Assessing compensation for non-option risks once option value is removed.
Scenario and sensitivity analysis (rate shocks, different prepayment speeds).
Cautions:
OAS is model-dependent—results vary with interest-rate models, prepayment assumptions and volatility inputs.
Illiquid markets and wide bid-ask spreads can distort implied OAS.
Complement OAS analysis with credit research, liquidity assessment and stress testing.
Bottom line
OAS is a vital tool for valuing and comparing fixed-income securities that include embedded options. By adjusting spreads for option value and modeling interest-rate and prepayment variability, OAS provides a clearer view of the compensation for non-option risks. However, because it relies on models and assumptions, OAS should be used alongside other analyses and sensitivity checks.