The three monkeys of mortgage bonds

By Felix Salmon
November 18, 2010
mortgage-bond scandal yet?

" data-share-img="" data-share="twitter,facebook,linkedin,reddit,google" data-share-count="true">

Have you forgotten about the mortgage-bond scandal yet? I’m sure a lot of bankers are hoping that you have. But Adam Levitin hasn’t, and his written testimony today to the House Financial Services Committee is well worth reading in full. He concludes:

The foreclosure process is beset with problems ranging from procedural defects that can be readily cured to outright fraud to the potential failure of the entire private label mortgage securitization system…

In the worst case scenario, there is systemic risk, as there could be a complete failure of loan transfers in private-label securitization deals in recent years, resulting in trillions of dollars of rescission claims against major financial institutions. This would trigger a wholesale financial crisis…

A critical point in any global settlement must be removing mortgage servicers from the loan modification process. Servicers were historically never in the loan modification business on any scale, and four years of hoping that something would change have demonstrated that servicers never will manage to successfully modify many loans on their own. They lack the capacity, they lack the incentives, and the lack the will…

For many, the preferred course of action is not to deal with a problem until it materializes and certainly to avoid any loss allocation that might threaten US financial institutions. But if we pursue that route, we may well be confronted with an unmanageable crisis. We cannot rebuild the US housing finance system until we deal with the legacy problems from our old system, and these are problems that are best addressed sooner, before an acute crisis, then when it is too late.

David Dayen reports on the consequent follow-up, in oral testimony, between Levitin and Brad Miller, who’s arguably the most sophisticated member of the House when it comes to mortgage finance.

Levitin said that we don’t have the full data sets from the servicers, or any comprehensive data to see whether there is a full-on crisis of unclear title and improper mortgage assignment. In other words, we don’t quite know the full extent of the problem. Levitin said, essentially, “The federal regulators don’t want to get info from servicers, because then they’d have to do something about it.” They don’t want to recognize the scope of the problem because it would require them to act.

And Levitin in particular singled out the Treasury Department. “The prime directive coming out of Treasury is ‘protect the banks’ and don’t force them to recognize their losses.”

Essentially, there’s a three monkeys act going on here: the servicers will hear no evil, the trustees will see no evil, and Treasury will speak no evil. And so long as they all remain deaf, dumb, and mute, they can ignore the problem as it slowly approaches systemic levels.

One very suggestive datapoint here comes from Tomasz Piskorski, Amit Seru, and Vikrant Vig, who have written a very dense but important paper on what happens to houses and mortgages when borrowers are delinquent on their loans. One finding is that banks are much less likely to foreclose on delinquent loans if they own the mortgage themselves than if they’re simply servicing the mortgage on behalf of RMBS investors. Foreclosure destroys value for the owner of the loan—but, crucially, it does not destroy value for the servicer, who therefore has very skewed incentives.

And then there’s this chart, at the end of the paper:


What you’re looking at here is the percentage of delinquent mortgages which are put back to the lender after they turn delinquent. All of these mortgages had early pay default clauses, which allowed investors to put back any loan which went delinquent within three months.

The investors, of course, can’t put back the bonds themselves: they have to rely on their trustee to do so for them. But the trustees are so otiose that they don’t do that.

Look at the y-axis: the highest number there, about 14%, is the fraction of loans which went delinquent in the very first month, and which were returned within three months. For loans which went delinquent in the third month, less than 5% were ever put back.

By rights, 100% of those loans should have been put back: that’s the whole point of having an early pay default clause in the contract. But the otiose trustees instead simply do nothing.

Levitin makes an extremely strong case that it’s much better to bite the bullet now, even if that involves socking the banks with losses, than to wait for the situation to continue to deteriorate to the point at which a devastating crisis is unavoidable. But Treasury, it’s clear, is not going to act, any more than the servicers or trustees are. Maybe because the technocrats at Treasury don’t really mind seeing pain being borne by homeowners and investors. Just so long as the banks are OK.

More From Felix Salmon
Post Felix
The Piketty pessimist
The most expensive lottery ticket in the world
The problems of HFT, Joe Stiglitz edition
Private equity math, Nuveen edition
Five explanations for Greece’s bond yield
2 comments so far

If the house goes into foreclosure then the servicer doesn’t get paid until the house is sold ( hint who is paying the servicer those fees if the owner isn’t even making his mortgage payments? ). It also has to justify the expenses to the investor. Of course back in the day when houses were being sold for more than the loan the investor didn’t really care. Now, you can bet they are all over those accounts given those expenses are being paid out of their pocket. So a servicer has to make upfront payments to foreclose, has to wait until the property is sold and has to haggle over every penny when claiming. Maybe you are confusing them with the foreclosure lawyers who ARE coining it out of this.

Servicers are very profitable when someone is simply paying his mortgage – money for old rope – and when people are slightly late paying – all those yummy late fees – where they most certainly are not profitable is when there are massive numbers of foreclosures with a significant number being disputed and where the equity in the house is around or below the loan value.

Foreclosure doesn’t “destroy value” for the owners of the loan. If the loan is in foreclosure then the owner of the loan is getting zero. When the house is foreclosed and sold he gets the amount it is sold for minus expenses up to the value of the loan which is typically non-zero which is more than zero.

As for the title issues, surely with all the excitement, one of the intrepid reporters would be doing some investigating. Ask the courts. When this came out Alphaville posted a wholely unsatisfying article from WaPo where they interviewed a NY judge who threw out around 40% of cases when they first arrived. He knew the exact number but weirdly couldn’t even vaguely guess how many were resubmitted and passed second muster. It seemed to me that a halfway decent journalist would have asked the judge – how representive he was, how many passed second muster, how many had fundamental issues and how many had just procedural issues.

Personally, I will happily bet that in the vast majority of cases there are zero issues with title because with all the focus and all the lawyers swarming around if cases were **regularly** being thrown out over **title** then it would be getting covered regularly – and not just a here and there story.

Posted by Danny_Black | Report as abusive

Danny_Black, this is just to let you know you are read and appreciated.

Posted by walt9316 | Report as abusive
Post Your Comment

We welcome comments that advance the story through relevant opinion, anecdotes, links and data. If you see a comment that you believe is irrelevant or inappropriate, you can flag it to our editors by using the report abuse links. Views expressed in the comments do not represent those of Reuters. For more information on our comment policy, see