No, it’s not another luxury liner. There are enough of those floating all-you-can-eat buffets already.
I’m talking about the successor to Quantitative Enhancement.
This is big for amREITs, and warrants an alert.
The higher the premium at which they own their portfolios, the more risk to principal the amREITs are about to feel.
See the dividend histories of 2003/2004 to get an idea what happens during a strong bond bull market to those “riskless” amREITs. Even though the value of the MBS they hold kept going up, the prepayments by people refinancing cause several of the amREIT favorites to take big losses and cut their dividends.
I’m sure a few people who read this blog post through the break were laughing at their computer screens and saying to me “What makes you think these suckers are going to be able to refinance?” You’d be right, as far as that goes.
On the other hand, the real issue is early return of principal on premium bonds, no matter what caused that early payoff.
My buddy Jody Shenn over at Bloomberg just sent me his latest article. It describes how they are going to deal with the logjam of future foreclosures in the Fannie/Freddie MBS programs.
Of course, some will start thinking this is horrible when they see the big numbers. My response as a taxpayer is “about time.” The last thing I wanted was to subsidize either investors or former borrowers living rent free.
Let’s make no mistake — paying out 5% or 6% on MBS when a goodly chunk of the borrowers were delinquent is a pure transfer of wealth from the taxpayer to the investor.
When Fannie or Freddie “buy out” a loan from an MBS pool, all they are doing is fulfilling their contractual obligation to pay the principal on the note, and shutting off the obligation to pay month interest. At that point they are free to work out something with the borrower, foreclose and sell, or even rent (I hope they don’t do the latter, as tempting as it might sound to get our money back with maybe even a positive return over time.)
Agency MBS have had an amazing run since last spring. I guess that happens when the Fed is in there buying more than the originators can produce. Needless to say, we were all worried about where things might go when the Fed stops buying, just the way the gold bugs and Dollar bears point to the fact that China doesn’t even have to sell to make things tough on the debt front.
If Fannie and Freddie stop the bleeding on just 10% of their outstanding book of MBS, that would remove hundreds of billions of Agency MBS from the market, and, on average, turn into a 0.3% loss on any portfolio that has a current cost basis of 103%.
Sounds like no big deal, right? Wrong. That could easily amount to a third of the net carry for a quarter at an amREIT that was collecting a net interest margin around 300 basis points. It might also leave them overhedged, which can spread additional margin pressure over the subsequent quarters or even years.
By the way, the latter is less likely, simply because all the people managing these amREITs have lived through the prepayment storms already. I don’t know of any seasoned professionals that would be foolish enough to be “100%” hedged. The most we carried when I was responsible for the risk management on a trading portfolio was around 70%.
There are other technical reasons for underhedging MBS related to the effect that has the mathematical misnomer “negative convexity,” but I did promise to hold the math to a minimum in case you’ve been imbibing spirits of the holidays.
I just wanted to get this alert out there, because there may be consequences when the institutional investors consider the risk of large scale loan buy-in at the Agencies.
PS: On the plus side, removing that much product from the MBS market (forever) will definitely mitigate the negative effect of the Fed finishing up its Quantitative Enhancement program. Won’t that be a surprise for all those bond bears positioned for rates to rise in the long end of the curve?