It asks which of the four FNMA tranches in following is most appropriate for Minna’s portofolio Coupon Maturity Yield Spread Duration Effective PSA PAC Bond A 6.25% 7/15/28 155 4.6 95-105 PAC Bond B 6.25% 1/15/30 170 4.85 90-170 Ignore those support and accrual info since we all know that’s not appropriate than PAC A. PAC Bond A B. Support Bond A c. Accrual Bond The answer states PAC Bond A has a narrow effective prepayment collar and it well protected from extension and contraction risk. I thought larger effective prepayment collar is better, so is PAC B better than PAC A with almost same duration?
anyone mind to give some inputs?
I think it is because of the narrow band as stated… given that the prepayment speed at 90 PAC B will start prepaying before PAC A…i.e. more contraction risk than PAC A… in the upper hand you have the same thing happenning at higher PSA speeds PAC A be will start prepaying hence have less extension risk…
thank you for your answer. I get it like this. Since PAC bond needs to protect both extension and contraction risk, so it can’t go too high or too low on this extent In this case, PAC A has better protection on contraction and extension risks than PAC B. I am just confused by one of the text on CFAI textbook They gave an example PAC Tranche X PAC Tranche Y Initial PAC collar 180 PSA - 350 PSA 170 PSA - 410 PSA Effective PAC collar 160 PSA - 450 PSA 240 PSA - 300 PSA The text is like this, notice that at issurance PAC tranche Y offered greater prepayment protection than PAC tranche X as indicated by the wider initial PAC collar. However, that protection is irrelevant for an investor who is considering the purchase of one of these two tranches today. Despite PAC tranche Y’s greater prepayment protection at issurance than PAC tranche X, trache Y has a much narrower effective PAC collar than PAC tranche X and therefore LESS prepayment protection
What page is this on, kevincwang? I have no idea what to think now except that the shorter the collar, the better…
Page 364 second last paragraph I read it three times and still didn’t get it… I can’t let PAC has any chances to screw me again…
I think it has to do with initial collar and effective collar. The effective collar changes every month; just because the initial collar may be large, the effective collar for a seasoned PAC may change if the PSA speed changes. Say you have 115-300 INITIAL PSA collar. If the prepayments come in at 200, which is less than the 300 upper PSA, you can raise the 300 to, say, 400, because you came in at under the maximum allowed and WELL within the upper collar, 300, so you have more protection (less contraction risk) from the support tranches than you thought you would. So your collar went from initial of 115-300 to 115-400 and you can breathe easy cause you came in at fewer prepayments than you thought you would. Likewise, if you have prepayments that come in at less than the lower bound of 115, you’ve got a problem (extension risk) that arrives from fewer payments coming in than you hoped, so you have to raise the lower bound from 115 to, say, 200. Now the EFFECTIVE collar is 200-300, whereas the INITIAL collar was 115-300. This is a narrower collar because you are worried about the payments having to come in faster now to make up for their slowness before so that investors concerned about extension risk don’t get pissed cause they’re not getting their payments fast enough.
CFAI is saying that you don’t use the initial collar to judge seasoned CMOs. You use the effective collar, which is constantly adjusting. You do in fact want a large collar I think, not a narrow one.
So whether we want narrow or larger collar, it is all depends on which risk we are worrying about, extension and contraction I read the question in BSAS again, and it says they are looking for a straight corporate bond, that may be the reason why they like PAC A better because it like a straight bond most?
Kevin we’re worring about both extension and contraction. We want to limit both. If the collar is above the maximum PSA (the upper of 300 PSA) we panic and if it’s below the lower PSA of 115 then we panic. The PAC collar’s goal is to limit all prepayment risk, contraction and extension. We are worried about both.
So out of that question, just purely select the best PAC bond against prepayment risk, we would choose PAC B which has more effective collar, is that right?
Well I think so. I’m just going off the same reading you saw, which helped me immensely. I would say if there’s a seasoned CMO, use the effective collar with the highest collar, and the initial collar is meaningless for seasoned CMOs.
I know, that thing is a beauty, I probably would get back in details after the exam. Thanks for helping me through it. 
Now I could think several different ways CFA could throw a question for this.
rellison…does that mean that you wouldn’t choose the effective collar with the narrowest band???