Comparing Index-Linked Bonds with different indexation lag

renikus · March 21st, 2007, 4:15 pm

Does anyone know of a robust method to compare inflation-linked bonds with different indexation lags?I.e. how to compare the relative value of a 30yr bond with a 3m inflation lag to a 30yr bond with a 6m inflation lag?I notice that in some research pieces they compare RPI ASW levels, what is the intuition behind doing this? Surely the lag is still an issue?Rgds,R.

Martinghoul · March 22nd, 2007, 7:53 pm

Hehe, let me guess, you're looking at the old gilts vs the new Canadian-style ones? I think the right way is to look at the actual mechanics of the calculation for the different lags. You can then calculate all the characteristic of the bonds, yields, carry, bei vs swap or appropriate nominal bond, etc. From then on, it's the wonderful world of good old bond RV for you... As to your question of comparing asw level, surely you see that asw is independent of indexing methodology, assuming you perform all the calculations correctly?

renikus · March 23rd, 2007, 10:43 am

dkosunsky, that is correct, I understand the mechanics of both the 3m and 8m very well, but because of the indexation, how can you possibly compare the breakeven or real yield of one a 3m against an 8m given that they have different cash flows that reference a different period of inflation? I have seen in some literature that for RV type analysis, they compare RPI asw levels, but im not sure that i do understand how the asw level is different from the indexation type? can you elaborate further?Regards,R.

Martinghoul · April 6th, 2007, 12:58 pm

renikus, sorry for the delay...You're generally right, i.e. asset swaps are the best way to look at these buggers. However, there's many types of asset swaps, par/par, proceeds etc. Of those the best method (it makes the least number of assumptions and is the most accurate) is to use z-spreads, which is the level of the zero-spread asset swap. Z-spreads are the way to go for you...

renikus · April 13th, 2007, 6:08 pm

thanks, i still dont really understand why looking at z-spreads solves the problem though...intuitively why is the z-spread ASW independent of the indexation lag length? Given a bond with a 3m lag and a bond with an 8m lag of same maturity, how can you draw conclusion that one is richer than the other in asw given that they reference different periods of inflation?Anything you have that i could read etc would be a great help as this really doesnt sit with me!Rgds,Ren.