September 11th, 2012, 3:40 pm
Hi there,I searched for answers to my question elsewhere on the forum but wasn't able to find one: if I overlooked something then apologies, please just let me know or point me in the right direction. (I also wasn't able to locate anything useful in Hull.)What I am interested in is understanding when it is optimal to post different types of collateral under different CSAs, in particular for IR hedging. Suppose for example I have a CSA with counterparty 1 which allows the posting of only cash as (2-way) collateral, and a CSA with counterparty 2 which allows the posting of either corporate bonds or cash. I now want to put on a hedge with a negative DV01: for example, a swap where I receive fixed and pay floating. A colleague of mine has mentioned that there is a big difference between when I want to hedge with counterparty 1 (cash) or counterparty 2 (cash/corps), and I want to understand this a little better.Suppose the hedge moves in my favour: that is, rates fall, bond prices rise. If I have hedged with counterparty 1 then I get cash as collateral, which I can now only reinvest at a lower rate than the return I would be getting for the bonds. On the other hand with counterparty 2 I would get bonds as collateral, which have risen in value as rates fall, so that seems like the better option. If rates rise, bond prices fall, and with counterparty 1 I would post cash at a higher rate, whereas with counterparty 2 I post bonds which are now paying at (relatively speaking) a lower rate. Again, counterparty 2 seems like the better choice, and similar with a positive DV01 hedge it seems like counterparty 1 is the better choice whatever happens to rates.However, is this effect not negated by the fact that, for example if receiving bonds with rates falling, the bond prices will have risen so we will get fewer bonds to collateralise this gain? Likewise if delivering bonds with rates rising, the bonds are worth less but we should have to post correspondingly more of them to compensate? I guess what I'm saying is that it seems like the benefits of getting the collateral which is "better" in each rates movement seem like they would be counterbalanced to some extend by the fact that the better collateral should go up in price, so you get correspondingly less of the better collateral. It seems like these effects should balance out to make the collateral choice not that significant, so either my logic is wrong (probably) or I have misunderstood my colleague. I hope that made sense - guidance would be greatly welcomed, thanks!