“Bond buyers are disadvantaged by the call provision and have more reinvestment risk because their bonds will only be called (redeemed prior to maturity) when the proceeds can be reinvested only at a lower yield. For this reason, a callable bond must offer a higher yield (sell at a lower price) than an otherwise identical noncallable bond. The difference in price between a callable bond and an otherwise identical noncallable bond is equal to the value of the call option to the issuer.”
Would you clarify for me how the value of the call option to the issuer is measured when it is first issued? Shouldn’t we know how interest rate will fluctuate to measure that?
So in practice, bonds have standard call features.
For instance in high yield, 5y bonds are not callable for axyear (NC1) typically. 7y bonds are typically NC2 or NC3.
Because bonds are priced based on comps, and the call feature is standard, no one bothers pricing the option incrementally, at least on an analytical basis. Its already included.
Bloomberg attempts an OAS estimate - option adjusted spread, using a binomial model to value the option separately. But it doesnt work well, and market people dont focus much on it.
There are quite a few theoretical models that may or may not make use of the volatility of the interest rate. Mind you, I never said zero volatility but a static or a dynamic one. Models like, Binomial, BSM, Comparable like Put Call parity all are deployable with reasonable success.
It is futile to argue what works or what does not. Bloomberg does a perfect job of letting you make an educated guess. If you fully rely on any model to give you solution consistently, the fault and naivety lies with you. It is not the limitation of the model itself.
Thanks a lot to you both! Much appreciated!