PAC 1 & Support

Please help out on a simple Q. I am curious if PAC 1 tranches may take away not just the principal but also the interest from support tranches if prepayments are much slower than expected? Thanks in advance and good luck all.

From my understanding it is only principal, but couldn’t find any clear guidance…

for a regular CMO yes but once you get into the ABS’s its a diff game, there are provisions where the principal, interest and prepay are redirected to the PAC and nonaccelerating senior trances

depends on the effective pac collar. ie 90 - 150 psa

epoh, why would it depend on the pac collar? i vaguely remember reading in schweser that interest can be redirected to PAC1 as well, not just the principal.

the only thing the collar does is determine the speed of the prepays, for a standard CMO, sequential order, the first tranche gets principal/prepays and interest and so on

it just seems funny for pacs to be taking interest away from supprot tranches. support tranches are supposed to see arisk in the prepayment they recieve. once they start losing their interest payments as well that seems like it goes beyond the scope of their purpose.

pacs just get principal and prepayments from support tranches. depending on defaults/prepayments, effective collars may be shifted up or down relative to the initial

Its true the pac collar only speaks of the avg expected life of diff tranches… as far as support tranches giving up their int to cover senior tranches is sounds unreasonable

tenten Wrote: ------------------------------------------------------- > Please help out on a simple Q. > > I am curious if PAC 1 tranches may take away not > just the principal but also the interest from > support tranches if prepayments are much slower > than expected? > > Thanks in advance and good luck all. If prepayments are much slower support tranches absorb the “excess” principal and face extension risk. The objective is to maintain the average life of the PAC tranches, which in turn is determined by PSA “allocated” (effective collar) to that particular PAC tranche. So, in a way, epoh is correct. In a sequential pay CMO, when senior tranches are backed by Accrual tranche, then the interest is re-directed from the accrual tranches to the senior tranches in order to sequentially pay-off the senior tranches. the interest equivalent is taken from the principle amount of the senior and moved to the accrual. So, we are talking 2 different things… hope this helps.

My understanding is that support tranches are “bouncers” and take all the punishment one way or another. So, based on my understanding they would take prepayments and give up interest in order to create the collar. I’m not sure how this would be a collar otherwise… Theorithecally, it could be set up in any way that the issuers or spv wishes, but I think I’m in line with the CFAI here… I would be pleased if some well versed FI poster could confirm…

From Schweser: PAC tranches can siphon interest slated for the support tranches, but do not redistribute it. In fact, PAC tranches are designed to provide a minimum guaranteed principal payment by collecting interest targeted for a support tranche when prepayments are low.

that is CRAP. Please find a statement in the CFAI text to support that, then I will buy it. the only place that the so-called siphoning interest happens is in the case of the z-tranche - which by design pays no interest - and the interest is moved up to the PAC tranches higher up in the schedule.

AFAIK Interest is never taken away except in the case of an accrual tranche whose principal is increased anyways. In this case I beleive if the prepayment speed falls below the collar, the live of the bond just increases…but not as much as the support tranches.

the show NY Wrote: ------------------------------------------------------- > From Schweser: > > > PAC tranches can siphon interest slated for the > support tranches, but do not redistribute it. In > fact, PAC tranches are designed to provide a > minimum guaranteed principal payment by collecting > interest targeted for a support tranche when > prepayments are low. This is not true! I agree with cpk here. z tranche or the accrual tranche is the only instance when this is done.