I see the very basic idea of CMOs is to spread the principle prepayment risk among the various tranches. The idea is great, however, the ultimate collateral source is the pool of mortgage loans. And the loans permitted to be included in the pool must meet the underwriting standards that have been established by the entities such as Fannie Mae and Freddie Mac. Are Ginnie Mac, Fannie Mae and Freddie Mac the only entites to issue CMOs? It looks like there are a large amount of bad mortgage loans in the pool. I’m wondering what happened to the Underwriting Standards and who’s supervising these standards? With these numerous bad loans, the tranches seem lost their purpose of protecting investors from prepayment risks. Does the tranche with higher par value always have shorter maturity and lower risk than the one with lower par value? Does the holder of the support tranche be compensated with higher yield for absorbing more principle prepayment risk?
hyang Wrote: ------------------------------------------------------- > I see the very basic idea of CMOs is to spread the > principle prepayment risk among the various > tranches. The idea is great, however, the ultimate > collateral source is the pool of mortgage loans. > And the loans permitted to be included in the pool > must meet the underwriting standards that have > been established by the entities such as Fannie > Mae and Freddie Mac. > > Are Ginnie Mac, Fannie Mae and Freddie Mac the > only entites to issue CMOs? You have this a little mixed up. The importance of the agencies is that they securitize the mortgages and provide a credit back stop. The CMO can be structured by anyone. You could structure a CMO (well, maybe not yet, but you will be able to). The first CMO was issued by Salomon Bros and if you haven’t yet read Liar’s Poker for the inside story, you should do that. The CMO itself is some bankruptcy remote vehicle that is a legal entity completely separate from the structurer. If the CMO has underlying collateral which is GNMA’s there is 0 credit risk but GNMA won’t even know who is getting what income. > It looks like there > are a large amount of bad mortgage loans in the > pool. Yes > I’m wondering what happened to the > Underwriting Standards and who’s supervising these > standards? > It’s VirginCFAHooker and that’s why we have this problem. > With these numerous bad loans, the tranches seem > lost their purpose of protecting investors from > prepayment risks. Not so - if the agencies back the loans there isn’t a problem with credit (almost certainly). > Does the tranche with higher > par value always have shorter maturity and lower > risk than the one with lower par value? Par value has nothing to do with maturity > Does the > holder of the support tranche be compensated with > higher yield for absorbing more principle > prepayment risk? Absolutely which is why people buy the supports.
BTW - You can certainly structure crappy non-agency mortgage loans into a structure that is about credit risk instead of prepayment risk (or even about both) but such securities are usually called CDO’s not CMO’s.
It may eventually show that CMOs and CDOs are bad, unsafe bond instruments. They seem backed by the full faith and credit of Nobody. Since the principle prepayment amount and timing are unpredicatable, it’s uncertain for that part of cash flow. Therefore, the investor won’t know for sure how long the loan will be outstanding. So I assumed the maturity for CMO and CDO is uncertain. Any model to predict this could be flawed. The book example: $200M passthrough securities has 3 tranches - Tranche A ($80M); Tranche B ($70M) and Tranche C ($50M). Based on the CF distribution rule, tranche B will receive monthly principle payment after tranche A is paid off and tranche C will receive monthly principle payment after tranche B is paid off. The rule appeared based on each tranche’s par value. The higher the par value, the higher priority to get paid off. Does it always hold? The maturity of each tranche is unknown, right?
There is no reason to believe in that structure for tranches. In particular, you are being shown very simplified examples. A typical CMO now might have 40 different tranches. CMO’s have been around a long time (since the '80’s) and have shown themselves to be very good securities. The transference of prepayment risk has enabled much more diverse funding of mortgages. The are a great thing. CDO’s are also great at risk transference - the problem is that tranching doesn’t really take away risk and can’t replace good intermediation. That seems pretty obvious but we blew it this time around. CMO’s based on GNMA’s are backed by the full faith and credit and other agency debt nearly is (or at least until a couple of months ago we thought so). CDO’s of course might contain some treasuries for parking money or for credit guarantees but since they are about distributing credit risk, they aren’t backed by full faith and credit essentially by definition.
Thx. I think you are right to the point. CMO/CDO might be great to transfer risks among the tranches, yet the total risk is not eliminated. Many investors thought by buying the most respected firms (usually rated AAA) with the best underwriting, they could avoid the train wreck. It seems that’s not the case now…