Yield to worst (YTW) is the lowest yield an investor can receive on a bond, assuming the issuer makes every payment, given all the dates on which the issuer may redeem it early. It is found by calculating the yield to maturity and the yield to call for each possible call date, and for any other early redemption at the issuer's option, and selecting the smallest. For a bond with no call features, yield to worst is simply the yield to maturity.
The measure exists because a callable bond gives the issuer a choice, and issuers tend to exercise it when it suits them rather than the holder. Yield to worst gives a conservative estimate of return that does not depend on guessing what the issuer will do. It is widely quoted in the municipal bond and corporate bond markets and in reports on bond funds and indexes.
How yield to worst is calculated
Each candidate yield is the discount rate that equates the bond's price with the present value of its coupons up to a redemption date plus the amount repaid on that date. Yield to worst is the minimum of these:
Here is the yield to the -th call date at the price set out in the bond's call schedule, and is the number of call dates considered. Bonds that are callable continuously after a certain date are usually evaluated on each coupon date. Mandatory or optional sinking fund redemptions can also be included where they apply to the bond being valued.
Only options held by the issuer enter the calculation. A puttable bond gives the investor the right to sell it back, and the investor would use that right only when it helps, so put dates do not lower the yield to worst.
Worked example
Consider a bond with a face value of 100 and a 5% annual coupon, maturing in eight years and trading at 104. It can be called after two years at 103, or after four years at 100. The three candidate yields are:
| Redemption assumed | Amount repaid | Yield |
|---|---|---|
| Call in year 2 | 103 | 4.34% |
| Call in year 4 | 100 | 3.90% |
| Maturity in year 8 | 100 | 4.40% |
The yield to worst is 3.90%, to the second call date. The example shows that the worst case is not always the first call: the early call returns a premium of 3 points, which offsets part of the premium paid, while the later call at par spreads the full 4-point premium over only four years. An investor relying on the 4.40% yield to maturity would overstate the likely return if rates stayed at their current level.
As a rule, yield to worst falls on a call date for a premium bond, because calling it is attractive to the issuer, and on the maturity date for a bond trading well below its call prices.
Where yield to worst is used
Dealers commonly quote callable bonds on a yield-to-worst basis, and municipal bonds in particular are often priced "to the call" when trading at a premium. In the United States, confirmation rules for customer trades in callable debt securities generally require the yield shown to be the lower of the yield to call and the yield to maturity. Rules for municipal securities are written by the Municipal Securities Rulemaking Board.
Bond index providers and fund managers commonly report a portfolio's yield to worst as a summary of the income it is likely to generate. For a fund, it is a weighted average across holdings and differs from the standardised 30-day SEC yield, which reflects recent income net of expenses. The measure is also used in setting the duration of a callable bond, by measuring it to the date that gives the worst yield.
Limitations
Yield to worst inherits the assumptions of the yields it is built from: that coupons are reinvested at the same rate and that the bond is held to the chosen date. It also treats the issuer's decision as fixed today, whereas the date that proves to be the worst depends on how interest rates move. If rates rise, a bond expected to be called may instead remain outstanding until maturity, a shift that reflects call risk and extension risk.
Because it picks a single scenario, the measure ignores the value of the call option itself. Analysts who need a more complete view use option-adjusted measures based on interest rate models, which average across many paths for rates. Yield to worst also says nothing about credit risk: it assumes the issuer pays in full, so a high yield to worst on a weak credit is not a guaranteed return.