Spread to worst is a bond’s yield-to-worst minus the yield on a Treasury security of matching maturity, and it tells you the minimum compensation you would earn above the risk-free rate if the worst plausible scenario for the bond plays out. For a callable bond trading above par, that worst case is almost always an early call by the issuer. The subtraction itself is simple. The work is figuring out which scenario produces the lowest yield in the first place.
What Spread to Worst Actually Measures
When a bond carries embedded options, a plain yield-to-maturity figure can flatter the return. A callable bond might advertise a 5.5% yield to maturity, but if the issuer calls it in two years, your realized yield will be materially lower. Spread to worst removes that false comfort. It answers a narrow question: under the scenario that hurts the bondholder most, how much are you earning above a comparable Treasury?
For callable bonds, the worst case is usually an early call. Issuers call when rates fall, because they can refinance cheaper, which returns your principal precisely when reinvestment options are least attractive. For puttable bonds the logic reverses, since a put protects the holder and the “worst” scenario is the one where the put is not exercised. Either way, the worst yield is the lowest yield the bond can deliver to you.
The spread component captures everything above the time value of money: the issuer’s credit risk, the bond’s illiquidity relative to Treasuries, and the cost of the embedded option. A wider STW means more cushion for those risks. A narrower one means less.
Step 1: Calculate Every Possible Yield
Before you can find spread to worst, you need yield-to-worst, and before you can find that, you need every plausible yield the bond can produce. Depending on the bond’s features, that means computing up to three types.
- Yield to maturity is the annualized return if you hold the bond to its stated maturity and collect every coupon along the way. This is the baseline where no option is exercised.
- Yield to call is the annualized return if the issuer redeems the bond on a specific call date at a specific call price. You calculate a separate yield to call for every call date in the indenture, using the call date in place of maturity and the call price in place of par.
- Yield to put is the annualized return if you sell the bond back to the issuer on a specific put date at the put price. Calculate one for each available put date.
The underlying math is the same in every case. Find the discount rate that sets the present value of remaining cash flows equal to the current market price. What changes is the terminal date, the terminal payment, and the number of coupons collected before that date.
A bond with three call dates and no put feature produces four yields: one yield to maturity and three yields to call. A bond with two call dates and one put date produces four as well. Every figure matters, because the worst-case yield can sit in any of them.
Step 2: Pick the Yield to Worst
Line up every yield you calculated and take the lowest. That is the yield to worst. No probability weighting, no averaging, just the floor.
Where the yield to worst comes from depends heavily on where the bond trades relative to par:
- For premium bonds priced above par, the yield to worst is almost always a yield to call. The issuer has a strong incentive to call and refinance, and getting called at par or a small call premium after paying more than par drags the return down. The nearest call date usually binds.
- For discount bonds priced below par, the yield to worst is typically the yield to maturity. The issuer has no reason to call, so the call scenarios sit above the maturity yield and the maturity date is the binding constraint.
- For bonds trading right at par, the yield to worst is generally the yield to maturity, since call yields at par tend to sit just above it.
This pattern tells you something practical. Spread to worst earns its keep on premium callable bonds where the call is a live possibility. On deep-discount bonds without realistic call risk, the STW and a plain spread over Treasuries will look nearly identical.
The date that produces the yield to worst is sometimes called the worst-case date. Note it, because the next step depends on it.
Step 3: Subtract the Maturity-Matched Treasury Yield
The final step is subtraction. Take the yield to worst, subtract the yield on a Treasury whose maturity matches the worst-case date, and express the result in basis points.
Matching maturities is what makes the number meaningful. If the yield to worst came from a call date three years out, use the three-year Treasury yield, not the ten-year. Matching maturities isolates the spread to credit, liquidity, and option risk instead of mixing in differences along the yield curve.
The worst-case date rarely lands exactly on a standard Treasury maturity. Analysts interpolate between the two nearest points. If the worst-case date is 3.5 years out, you would interpolate between the three-year and five-year Treasury yields, weighted by proximity to each.
For U.S. dollar corporate bonds, the Treasury curve is the standard benchmark. Some desks use SOFR-based swap rates instead, which produces a tighter spread because swap rates typically sit above Treasuries. If you are comparing STW figures from different sources, confirm they use the same benchmark curve.
A Worked Example
Take a corporate bond with a 5.8% coupon, ten years to maturity, and two call dates: one in three years at $1,015 and another in five years at par. The bond trades at $1,040.
Working from the current price and the semiannual coupons, the three yields come out as follows:
- Yield to maturity, ten years: 5.30%
- Yield to call, first call in three years at $1,015: 4.25%
- Yield to call, second call in five years at $1,000: 4.80%
The lowest is 4.25%, so that is the yield to worst. It comes from the first call date, which fits the pattern for a premium bond: the issuer would save money by calling at the earliest chance.
The worst-case date is three years out, so you need the three-year Treasury yield. Assume it is 3.90%. The spread to worst is 4.25% minus 3.90%, or 35 basis points.
Those 35 basis points are the minimum you earn above Treasuries if the worst case plays out. If the issuer never calls, your realized spread will be wider. STW assumes the worst, and that is the point.
Why the Worst Case Can Change
Yield to worst is not fixed at issuance. It moves as rates and market prices move, and the worst-case scenario itself can shift.
Picture a bond bought at a premium with a yield to worst driven by the first call date. Rates rise sharply, the price falls below par, and the call scenarios stop making sense for the issuer. The call yields climb above the yield to maturity, and the yield to worst flips to the maturity yield. The worst-case date moves from three years out to ten years out, and the maturity-matched Treasury changes with it.
An analyst tracking STW over time has to watch which yield is driving it, not only the number. A jump in STW might reflect deteriorating credit, or it might mean rates moved enough to change which date is binding. For callable bonds priced near par, small rate moves can bounce the yield to worst between call and maturity scenarios, which makes STW more volatile there than for bonds deeply above or below par.
How STW Compares to Z-Spread and OAS
Spread to worst is one of three spread measures analysts use on corporate bonds. Each answers a different question.
Z-Spread
The Z-spread is the constant number of basis points you would add to every point on the Treasury spot curve to make the bond’s discounted cash flows equal its market price. It uses the whole curve rather than a single yield and a single benchmark, which makes it more precise for non-callable bonds where cash flows are certain. The Z-spread ignores embedded options, so it treats a callable bond like a bullet bond with the same coupons and maturity. That overstates the true spread on a callable security. Use it for bonds without options.
Option-Adjusted Spread
Option-adjusted spread strips out the value of the embedded option to isolate the spread attributable to credit and liquidity alone. It relies on simulating many interest rate paths, valuing the bond under each, and backing out the spread that equates the average modeled value to the market price. OAS is the cleanest measure of credit compensation on a bond with options, but it depends on an interest rate model and volatility assumptions, so different models produce different numbers for the same bond.
For a callable bond, the three measures typically rank in this order: the Z-spread is widest because it ignores option cost, STW sits in the middle because it embeds option risk in the spread, and OAS is tightest because it removes the option cost.
When Each Metric Fits
STW works well as a quick, conservative screen. If you are ranking a universe of callable bonds by worst-case compensation, STW gets you there without a model. It also fits when you actually care about a floor return, such as managing to a minimum yield target.
OAS is better for relative value between bonds with different option structures. Comparing a bond callable in two years to one callable in seven years using STW is misleading, because the embedded option costs differ. OAS levels that comparison.
Z-spreads belong in the toolkit for non-callable investment-grade debt, where they measure the credit and liquidity premium against the full Treasury curve.
Reading an STW Number in Market Context
A spread figure means nothing on its own. Whether 35 basis points is generous or stingy depends on the issuer’s credit, the bond’s structure, and where the broader market sits in the credit cycle.
As of late March 2026, the investment-grade corporate bond spread sat at roughly 88 basis points above Treasuries, while the high-yield index showed an option-adjusted spread of about 321 basis points.1FRED. ICE BofA US High Yield Index Option-Adjusted Spread Those are tight levels historically. Investment-grade spreads have typically ranged from around 80 basis points to over 200 depending on conditions, and high-yield spreads have moved from roughly 265 basis points in calm markets to well over 450 during periods of stress.2European Central Bank. Challenges to the Resilience of US Corporate Bond Spreads
When market spreads are compressed, the margin for error shrinks. A callable BBB bond offering 90 basis points of STW might look reasonable in isolation, but if the whole investment-grade market sits at 88, you are barely being paid above the index for a bond with real call risk. In a dislocation, the same issuer might trade at 200 basis points of STW, and the wider level may more than cover the embedded option.
Comparing a bond’s STW to rating-matched peers is the standard relative-value screen. A bond trading meaningfully wider than comparable credits deserves a closer look. Sometimes the market is flagging risk you have not identified. Sometimes the bond is simply less liquid or recently issued. Telling the two apart is where the analysis begins.