Again I return......, Recapping on the section 6 of chapter 15 the notes cover how to measure credit risk. one particular approach considers using asset swaps for this purpose. Based on what I have read in the notes and in the Hull text book I am led to believe that under a swap of this nature the known coupons from a reference corporate bond (or index) are paid away (irrespective of whether the coupons are actually paid by the issuer) and in return a floating rate of LIBOR + a margin is received. I believe that at outset an adjustment is made so that both sides of the swap are effectively based on a par value of 100. I.e. if the underlying corporate bond is trading at below 100 then the party paying the floating rate will increase their spread (such that the pv of these additional payments equals 100 - MV of bond). Conversely the margin is decreased if the bond is trading above par. My question relates to whether this is actually an accurate way to measure credit risk? Subject to my understanding being correct an asset swap takes no allowance for the principle amount defaulting. Should the happen the party paying the coupons would be severely disadvantaged. Assuming the steps set out above are correct it this measure therefore sufficiently accurate to measure the credit risk associated with a specific corporate bond issue?
I don't know a lot about these, but my understanding of how these asset swaps work (after reading p502 in the 7th edition of Hull) is that the coupon payments are guaranteed and the asset swap is structured to compensate for any loss of capital. So the 150 basis points spread only reflects any potential loss of coupons on the actual bond. I think the extra initial "adjustment" payment allows for both (a) the difference between the actual coupon on the bond and the current risk-free swap rate and (b) any potential capital loss at redemption due to credit risk. For example, if the coupons on a junk bond were the same as the current risk-free swap rate but the market thought there was only a 75% chance of it paying 100 at redemption (and you'd get zero otherwise), the market price of the bond would be 75 discounted. But the adjustment payment would be 25 discounted, which would make up for the expected capital loss.
