[author: Rich Jones]
As I read through my last commentary on the future of the GSEs and fielded some comments from readers, I realized I glossed over hard questions about how the governmental backstop would work, sort of like envisioning a plane and assuming there’s an engine.
Before I go on, let me reiterate what I said in connection with the original commentary that this is me talking and nobody else. See the last commentary for a more detailed apologia for my comments.
In my original commentary I talked about using a government warehouse for, among other purposes, guaranteeing P&I on “the senior bonds,” while allowing all securitizations to have non-rated tails. My thought had been that this would bring GSE pricing a bit closer to conventional execution. That was rather nonchalant, wasn’t it? I dashed past the hard questions with a casual disregard of how this would actually work.
So, what is executable?
What is executable is having the GSEs, in this new world, guaranty only the ultimate receipt of principal and have that guaranty attach only to the AAA class. This in large measure would replicate the practical economics of the AAA class in conventional securitizations where generally, AAA securities are almost always money good. The resulting bond structure would still be better than a conventional structure because every once in a while, when the blue moon puts in an appearance, the AAA class in a CRE deal does suffer an actual loss, but let’s agree it’s rather rare. The enhanced super AAA securities would go a long way to ensuring investor appetite even in a disruptive market.[1]
The way this guaranty would work would be through the GSEs’ obligation to repurchase non-performing loans using a combination of its warehouse facility and its private capital. That repurchase obligation would be sized (and periodically re-sized) to protect the AAAs with the rest of the classes being protected only through the customary conventional mechanics of subordination. Our customary advancing structure would clearly take care of the interest obligation with respect to the AAA classes.
If this is to work in an entirely private enterprise, we have to think about GAAP. We can get true sale treatment on these guaranties because it is reasonable to expect no losses would be experienced in covering only the AAA securities. Moreover, if necessary to satisfy the twitchiest of lawyers and accountants, this guaranty could probably be capped somewhere between 10-20% of OPB of the entire pool and achieve much of the same goal.
How to size the AAA class? Well, perhaps the easiest way is to do it through the ratings agency’s mechanics but as the market can largely reverse engineer those mechanics, the AAA sizing of the GSEs could be determined without an actual rating. Would the rest of the capital stack be rated? Probably would, and that would make sense.
These mechanics would avoid a material disruption of private market and avoid a material distortion of private market pricing. Yes, these AAAs would be better than other AAAs, but only by a skosh.
That amount of disruption is tolerable. Remember, the principal purpose of re-committing to a GSE structure should be liquidity and ensuring that between private capital and governmental warehouse capacity, there will be enough money in the system through thick and thin to provide mortgage finance to the multifamily marketplace. This modest distortion of pricing is a reasonable price to pay to get there.
So there. That’s how the sausage should get made. We need to ween our markets off the governmental teat of distorted pricing. We need to reject the notion that has been received wisdom that the government guaranty of the GSEs’ almost endless appetite to originate more multifamily debt is the only thing that makes finance markets function. That is a fallacy; that it is a solution in search of a problem. Let the conventional market do its job and satisfy multifamily borrowing demand. It will.
[1] This footnote will recognize something of an apology to Fannie, whose DUS program is a long way toward appropriately risking a pool of commercial mortgage loans, but hey, this is broad brush stuff and even in that program there is unnecessary pricing and credit distortion.
