I was completing question 26 of Reading 29 in the CFA books and the answer states…“OAS has limited use in the analysis of speculative grade bonds because default risk is excluded from the calculation.” No doubt we all know the formula OAS = Z-spread - option cost. I was of the belief that it inherently includes a consideration of default risk. A simple google search provides multiple sources that state the OAS can be viewed as the compensation an investor receives for assuming a variety of risks (e.g. liquidity premium, default risk, model risk), net of the cost of any embedded options. Any of you gurus got any thoughts?
not sure of what context they provided it, but i dont see how it would not provide an embedded default premium. Z spread is the spread over the whole treasury curve, so theoretically wouldnt the z-spread of a treasury be zero? To get a meaningful number you must incorporate default risk in my mind.
agreed, if you assume market pricing captures the various risk exposures such as liquidity, default etc then the z-spread would similarly incorporate these. I think this will be one to shelve and store in the WGAF (who gives a f**k ) section of my brain.
OAS covers all risk elements in one measure. But, in order to evaluate whether a junk bond will be a good investment, the primary factor to consider should be if it deserves its rather low rating (=high default spread) when compared to other same-rated junk bonds; and OAS will not help at that point.
nice one, that clarifies it. cheers
Since OAS has limited use in the analysis of speculative grade bonds , which other spread shall be used for junk bond analysis?
Is the default risk factored in the prices of junk bonds?
Junk bonds could have embedded options; OAS is recommended for spread analysis of bonds with embedded options. Kind of confusing to me. IMO, the default risk has been factored in the prices of junk bonds(high yield bonds), so OAS shall reflect the default risk. This is not contradictory with what’s in CFAI book…The default risk is not included in the OAS calculation, just like model risk not included in the OAS calculation.
OAS cannot cover default risk , because default is highly firm specific. Plus in L2 if I remember right, the OAS calculation methodology mainly was applied to examples from MBS and used prepayment related data .i.e. the option part had to do with convexity and not default possibility. Default was not even a big deal up until 2008 anyway
I think we’re talking about two different things here: 1. The actual adjustment made to the z spread in order to take into account the effect of options. 2. The spread that remains after the adjustment has been made. In regard to #2, I would think it would still reflect default risk otherwise it would be equivalent to the risk free rate.
If OAS is inappropriate for speculative grade bonds, which other spread shall be used?
I read in the CFAI books that it doesn’t cover default risk, but I still think that’s bogus. I think even in LII if i remember it said it covered default risk. I guess for the test we will have to put our dunce caps on and just answer the way they want and say it doesn’t.