# CMO/PAC question

\$400 million of mortgage pass-throughs will be used as collateral for five tranches. The first two tranches are planned amortization tranches: \$260 million of bonds of tranche U and \$50 million of bonds of tranche V. Tranche U is a PAC I tranche and tranche V is a scheduled tranche. The third tranche is a scheduled support tranche—the holders of the \$40 million of bonds in tranche W receive principal repayments according to a schedule as long as prepayment speed is between 190 and 240 PSA. The last two tranches are unscheduled support tranches—\$25 million of X bonds in a floating-rate support tranche and \$25 million of Y bonds in an inverse floating rate support tranche. Which of the following statements regarding the prepayment risk of the bonds is most accurate? The: A) W bonds have less prepayment risk than the U bonds, which have less prepayment risk than the Y bonds. B) U bonds have less prepayment risk than the Y bonds, which have less prepayment risk than the X bonds. C) U bonds have less prepayment risk than the V bonds, which have less prepayment risk than the Y bonds. D) W bonds have less prepayment risk than the X bonds, which have less prepayment risk than the Y bonds. Your answer: D was incorrect. The correct answer was C) U bonds have less prepayment risk than the V bonds, which have less prepayment risk than the Y bonds. Even though U and V are sequential pay tranches, they are also planned amortization class tranches (PACs). Once the PSA rate is exceeded by the tranche collars the support tranches absorb all the prepayment risk first and then the prepayment risk is absorbed by the tranches in reverse order with the lower unscheduled tranches absorbing more prepayment risk before the higher scheduled tranches. Therefore tranche V would have more prepayment risk than tranche U. Support tranches X and Y have the greatest prepayment risk because they are unscheduled and lowest in the order of tranches. Tranche W will have less prepayment risk than X and Y followed by V and lastly U since once again prepayment risk for PACs is highest at the lower tranches and lowest at the upper tranches. *********************** I get why C is correct, but why is D wrong?

As I understand, 1st part: W will have less prepayment risk, no doubt in this. 2nd part: coming to second part, we are not sure which tranche, x & y, is more risky because, both are structured in two separate themes. X on floating rate support and Y on inverse floating rate support. We know both X&Y will act inversely depending on some scenario. That’s why we are not sure which is more risk X or Y. So we cannot accurately assume that X is less risky than Y tranche. Hope this is right and helps.

That makes sense. I was focusing on the fact that both were unscheduled support tranches. I am not familiar with how the how floating rate vs inverser floating rate tranches work. This is not detailed in schweser. Thanks.

Can someone explain the floating rate vs inverse floater to me? Is it just like it sounds? Also, is the spread on the collar of the PAC related to the par amount of the support tranches? Will the spread be wider on the collar (thus providing less prepayment risk) if the support tranche amounts (par value) is larger?

Floating rate tranches are directly connected to an interest rate index, most commonly the London Interbank Offered Rate (LIBOR). The interest rate on these tranches resets periodically to account for changes in the underlying index rate. Because of the rate reset feature, these tranches bear less interest rate risk than most fixed rate CMO tranches.

mwvt9 – your 2nd question – you are correct, the bigger the par on the support tranches of a CMO, the more protection you have for the PACs. think about it like a rainy day fund – the more coins in the piggy bank the better off you are if you lose your job for a little further info on the supports x and y – its hard to say which one is riskier, unless the question specifically says so. especially when you have a floater/inverse floater relationship, they are usually stripped from fixed rate collateral, so they are two pieces of the same pie anyway. D isnt a terrible answer, but it did say which is MORE accurate, and C definitely is MORE accurate

This question came up last year and there was a big discussion about it. The ambiguity here is the word “prepayment risk” which the answer assumes means “contraction risk”. If prepayment risk includes both contraction and extension risk, this question is ambiguous.

That is how schweser defines prepayment risk - the combination of contraction and extension risk.

Me too which means the question is ambiguous because the answer above only deals with contraction risk.

That said, I have to come up with some pretty bizarre stories for the answer not to be C, particularly because I would say that X and Y have the same prepayment risk which rules out B and D.

JoeyDVivre Wrote: ------------------------------------------------------- > Me too which means the question is ambiguous > because the answer above only deals with > contraction risk. Why? The PAC structure enables the PAC tranches to deal with both contraction and extension risk. Of course this shifting of risk is taken on by the unscheduled tranches…a zero sum game. That is why I thought my answer could be right based on what interest rates did, but C would be right under all circumstances. This stuff is new to me so I could be way off.

I think C is the right answer and, yes, PAC’s can deal with both contraction risk and extension risk. This question actually doesn’t look that ambiguous to me on second read. I’ll bet the question we discussed ad nauseum last spring was a different but similar looking question.