Wednesday, August 11, 2010

Five Points On The Fed Announcement

There's a number of comments that need to be made regarding the Fed's announcement yesterday that it will start reinvesting principal payments from its mortgage portfolio in longer-dated Treasuries.

1) The amount of such proceeds is probably in the $250 billion range in the course of the next 12 months, but its exact size can vary appreciably from that number as it will depend on changes in the current prepayment rate of the mortgage portfolio. If the rally that was set off yesterday afternoon in the long end of the Treasury market has legs and MBS spreads over Treasuries remain tight, then the amount that the Fed will have available for reinvestment can quickly rise to $300 billion, or possibly higher.

2) There is no explicit promise by the Fed that this program will remain in effect over a full 12-month period ahead, but a reasonable assumption is that it probably will, as it allows for a gradual return of the Fed's portfolio to the normal "Treasuries only" content. However, if at any point along the way, the mortgage market shows evidence of serious strain as a result of the Fed's actions, then the program could be either slowed or suspended, as the Fed can hardly afford to destabilize a market that had only recently started healing from a highly traumatic experience in the previous couple of years. Any such mid-course correction in the Fed's plans would obviously result in a smaller total amount of Treasury purchases than the "penciled-in" $250 billion number above.

3) The reinvestment of the proceeds of maturing mortgages in coupon Treasury securities will be roughly in like with the overall lengthening of the duration of the Fed's Treasury portfolio, which currently stands at approximately 7 years from 3 1/2-years that prevailed in the pre-crisis period. However, no further extension of that duration is likely, as for that to happen the Fed would need to be buying exclusively 10-year paper- which is a highly unlikely prospect. In fact, the average duration of the Treasury portfolio may shorten very slightly, as a result, over the coming months.

4) Inasmuch as the average duration of its Treasury holdings will not be altered dramatically from its current level, the implication of those additional purchases ahead is that, when the moment down the road comes for the Fed to actually start shrinking its overall portfolio, there will be a more massive amount of selling of longer-dated coupon securities. It is reasonable to argue that this, viewed in a vacuum, would have a steepening effect on the Treasury curve. However, such sales intended to shrink the Fed's overall portfolio would by definition be taking place in an environment where the Fed is finally implementing its exit strategy, that is to say, tightening. As a result, the curve at that time would be coming under severe flattening pressure and the massive sales of the Fed's longer-dated Treasury portfolio would probably act more as a factor mitigating the degree of such curve flattening rather then cause an outright steepening per se.

5) All in all, the Fed's decision to gradually shift the composition of its overall portfolio away from MBS is arguably one of the least dramatic, or disruptive, measures that could have been adopted in response to the recent loss of momentum in the economic recovery. Its true effectiveness in helping the economy is very much an open question, once the Treasury market's initial reaction and front-running of the announcement plays out. But it allows the Fed to deflect the steadily growing pressure in recent weeks to show some responsiveness to the softer economic data on the ground, while quietly working toward restoring at least a more normal composition in its portfolio holdings.

Anthony Karydakis