callable bond has larger OAS than noncallable BOOK4 P212

Hi all,

On Book4 Reading 24 Page 212, it says callable debt often has larger OAS than otherwise non-callable debt.

This is also confirmed by the Question 12 comment 1 (said smaller, hence wrong)

My understanding is for callable, OAS < Z Spread and option cost is positive. so given z-spread=OAS for noncallable, OAS of callable should be smaller than a comparable noncallable.

Any thoughts?

thanks in advance

I agree with your approach, also see the discussion above. Maybe S2000 will chime in.

Hi, I think the confusion may come trying to use z-spread as a reference point. The z-spread of the callable and the non-callable bond, because of the difference in YTM, will be different. Do not consider the OAS in terms of the z-spread. Consider only the OAS itself. For a non-callable bond, the spread is the difference between yield of the bond and the yield of the risk free equivalent. With callable bonds, the call option benefits the bond issuer, not the bond holder (or buyer). Because the issuer can at any time buy back the bonds at par (or a strike price set in the contract), the bond holder or buyer will require a higher spread (lower price, or higher yield). Hence, the OAS is higher for a callable bond, because the yield is higher than that of an equivalent non-callable bond.

Hi will660,

Thanks for your reply.

My understanding to your reply is you are saying because Higher G spread(YTM-rf) is higher, so OAS is higher?

I am not sure about this… I think G spread assumes no change in CF, but when it is callable, the CF changes, so comparing G spread seems to me might not be valid and cannot be used to infer difference in OAS.


The OAS should be the same for a callable bond, a comparable putable bond, and the comparable option-free bond; if it isn’t, something in the analysis is wrong.

Hi Magician,

Thanks for the reply.

do you think the book’s statement might be incorrect? It is a simple statement and it is not a conclusion from any case or scenario.

I think that it’s probably correct; it wouldn’t surprise me in the least that the analysis of a bond with embedded options is incorrect. (By this I mean the analysis by the investing public. Put another way: is wouldn’t surprise me if a bond with embedded options is over_ under _priced.)

Hi Magician,

Just to clarify the very last ’ overpriced’ part: when the OAS of the callable is larger than the “true” OAS or the comparable noncallable, wouldn’t this mean the bond is underpriced? When choosing between bonds with different OAS, we would prefer the one with higher OAS and this is because it is cheaper/underpriced.

Please advise/confirm

Thanks again.

Good catch.

I shouldn’t reply on these threads after a long, tiring day. My mistake. I corrected it.

Many thanks for the reply and help, Magician. Really appreciated.

My pleasure.

Could you explain some more? I ran into the same theoretical misconception as fairplay…

OAS abbreviates option-adjusted spread, and it really means option-_ removed _ spread: the spread appropriate for the underlying straight bond (i.e., the bond that has all options removed).

Thus, you can compare the OAS of a straight bond, and otherwise identical callable bond, and an otherwise identical putable bond; they should be equal because they have the identical underlying straight bond.