Schweser says: The process for valuing callable (or putable) bonds is similar to the process for valuing a straight bond. However, instead of spot rates, one period forward rates are used in a binomial tree framework.
Either this is wrong or I’m missing something huge here. Don’t binomial trees for straight bonds use one period forward rates as well?
Good question, and you are only missing something small here.
Recall first of all, that for straight bonds we do not need the binomial tree framework, we only need trees for bonds with embedded options. Put differently, for option free bonds performing valuation discounting with spot rates produces an arbitrage-free valuation ( Method 1, simply discount all payments with the corresponding spot rates). The trees become necessary once we introduce embedded options.
However, we can still use trees for option free bonds as well ( Method 2 ), i.e. as long as the tree is calibrated correctly (i.e. we derive the forward rates from the same spot rates as in Method 1 ) we should arrive at the exact same price as we do with the discounting with the spot rates. See Blue Boxes 2-4 in Reading 36.
To summarize: dem Schweser folks are referring to Method 1, while you are thinking of Method 2.
I do think their wording is a bit unfortunate though, they should have added that they are referring to a different method, because if you use method 2 (i.e. trees) both use 1 period forward rates.
Thank you for this response, makes perfect sense. The wording of “instead of spot rates, one period forward rates are used in a binomial tree framework” made it seem as if they were saying spot rates are used in a binomial tree but its clear now that they were referring to Method 1.
I was a little worried that I was completely wrong haha
You are welcome, glad I could help.