I greatly admire the work of John Carney. But I was bemused on Friday by his Clusterstock headline:
New Worry: Mortgage Cram Downs Could Set Off Huge Bank Losses
This seems to me to be a very old worry--I've been voicing it for months. Why do people want to do mortgage cramdowns? They want to bail out distressed homeowners. And because this will be very expensive, they want to do so in a way that does not involve a direct government outlay of cash. The normal way that the government gets what it wants without paying for it is to make a rule forcing someone else to shell out.
We're fundamentally having an argument about who should bear the systemic risk of a fall in housing prices. There is a sizeable constituency that wants to put the entire price on the banking system, both because they think bankers are ultimately responsible for the problem, and because they think of banks as rich corporations that won't be hurt the way ordinary homeowners or taxpayers would.
This belief seems odd to me, given that we're in this mess in large part because the banking system is in such parlous shape. Moreover, the history of financial crises amply illustrates that ordinary taxpayers and homeowners can end up in pretty dire straits when the banks are too screwed up to lend.
How bad might it be? Carney links to a post from HousingWire
that lays out some of the potential pitfalls:
- Mortgage insurance is concealing risks that won't get covered in a cramdown. In order to make highly leveraged mortgage backed securities more attractive to investors, bankers bolstered them with mortgage insurance. These mortgage backed securities based on bundles of mortgages that had a thin collateral margin were able to get higher ratings from credit agencies because of mortgage insurance. Banks could carry these on their books as less risky assets, and weren't required to reserve much capital against them. They haven't been written down as much as other MBS because the insurance was supposed to protect them from losses. But most mortgage insurance won't cover bankruptcy modifications. That means that assets on the books of banks could have far higher loss risks than they appear to on balance sheets. When the losses kick in, banks will once again have to raise capital.
- Even the best, most highly rated paper will share in cramdown losses. Ordinarily, lower tier mortgage backed securities get whacked first by losses on the portfolio. High tier MBS with higher ratings can keep delivering revenues because they get the first claims on payments. But a good number of private-party MBS deals, pretty much every prime and Alt-A deal, have provisions that allow bankruptcy losses to be allocated among all investors. This means that highly rated paper, which has not been written down very much and is held on bank balance sheets at high levels could get hit. The capital reserved against these losses will be inadequate. You see where this is going, right?
- Federal Housing Administration insurance does not cover bankruptcy. If you were a conservative banker who only bought MBS backed by FHU insurance, you probably think you're in good shape. Think again. The cramdown legislation will make those assets unsafe. "The reason is simple: FHA insurance does not reimburse servicers for losses on bankruptcy cramdowns on loans that are not in foreclosure. If servicers have been stretched thin already by this mess, the cram-down proposal currently being considered might serve as the final nail in the coffin for some," Jackson writes.
Think of these kinds of government cramdowns as doing it on the faux-cheap. It looks inexpensive, because the government isn't shelling out directly. But making things artificially cheap by hiding the pricetag from yourself encourages you to do things you oughtn't--just ask the current holders of "investment" properties purchased with "innovative" mortgages. In the end, the bill always comes due--and the accrued interest is usually a killer.
This article available online at: