Shedding some light on scandal hitting the US mortgage market

We thought we’d clear up some of the misconceptions and present our view on this issue. [1].

The deluge of media attention was initially kicked off by the big mortgage banks placing a temporary ban on foreclosures, initially in 23 states with judicial foreclosure processes, and then later expanding to all 50 states.

At the heart of the issue was ‘robo-signing’; in other words, the bulk processing of foreclosures without thorough reviews of the relevant facts and documents.

Many commentators got very excited about this, seeing robo-signing as proof of a conspiracy and cover-up of massive mortgage origination fraud. To make matters more confusing, the financial press has managed to muddle three separate issues (robo-signing, chain of title, and mortgage put-backs) into one. We view these as distinct issues, each with unique ramifications for the market.

Robo-signing

The current process for foreclosing on a home after a mortgage default pre-dates computerization. It is, quite rightly, a painstaking process that has ensured a minimum of mistakes. However, the process was never designed to handle today’s volumes.

The chart above shows that from 1980 to 2006, the home foreclosure rate in the US ranged between 0.5% and 1.5%. With the advent of the credit crunch, that rate ballooned to reach a level of 4.5% today. This number would actually be much higher without the government- and bank-sponsored loan modification programs that have temporarily rehabilitated delinquent borrowers. (We believe that most of these rehabilitated borrowers will still eventually default, as interest rate is not their main consideration when they are in serious negative equity).

Furthermore, the problem will only get worse. Amherst Securities [2] estimates that 11 million mortgages, or about 20% of total mortgages outstanding, are at risk of default based on delinquency status and/or the amount of negative equity. Given that liquidations of foreclosed properties are running at about 90,000 per month (about 1 million per year) [3], this implies that it could take over 10 years to clear the full backlog of the foreclosures to come.

The reality is that the sheer numbers involved with today’s defaulted mortgages is what led to robosigning, with banks attempting to put in place procedures that streamlined the process. Clearly they cut corners, and clearly this expedited process resulted in mistakes in some cases.

We expect the solution will be some combination of better practices by the banks and legislative changes by a Congress that, no matter which party is dominant, can usually be counted on to help the banks clean up their mess. In the end, this is a process problem that can be fixed, not the mass fraud that some are asserting.

Whatever the solution, we expect there to be a slowdown in foreclosures over the next few months, if not longer. It was already going to take years to work through the existing pipeline of troubled loans, delaying a meaningful housing recovery. Further delays to the foreclosure process will inevitably push the housing recovery out further.

Chain of Title

Some of the more conspiracy-minded commentators have seen the robo-signing issue as a coverup for something much deeper – that the parties foreclosing don’t actually have the mortgage note that proves they have the right to foreclose. We believe this is a red herring – except in the case of outright fraud, there’s a mortgage note out there for every borrower.

We have no doubt that corners were cut in the mortgage securitization scramble of the mid-2000s, and many mortgage notes probably haven’t been properly assigned as they should have been along the way. Once again, however, this is a procedural problem, not large scale fraud. It will be remedied by banks going back and doing what they were supposed to have already done. There will be additional costs for fixing the chain of title and this too will add to foreclosure delays. However, in the end, if a borrower signed a mortgage note, he still owes the money to somebody.

Will there be lots of ‘show me the note’ lawsuits from individuals trying to take advantage of the system? Undoubtedly; at least at first. However, the courts are already so overwhelmed that they are not likely to be sympathetic to such claims. We doubt there will be many borrowers getting free houses out of this.

Loan put-backs

In our view, this is by far the most significant of the three issues and yet, is the one that has been downplayed in the furore over robo-signing. When banks securitize a pool of mortgages, they make certain representations and warranties (‘R&Ws’) about the pool. If certain mortgages in the pool are subsequently found to violate the R&Ws, those loans can be ‘put back’ to the bank at par.

Poor practices during the mortgage frenzy resulted in many R&W violations. The mortgage agencies (GNMA, FNMA, FHLMC) and the monoline insurers (AMBAC, MBIA, etc.) began aggressively putting back mortgages in 2009, and accelerated their efforts in 2010. Private label MBS have been slower to get involved in this process due to investor participation and materiality thresholds not present in the case of the agencies or monolines. However, there are currently a number of initiatives to bring investors together, which will lead to more put-back claims from private label MBS.

JPMorgan estimates that losses to banks from put-backs could range from US$55-120 billion [4]. Similarly, Barclays Capital makes a base case estimate of US$75 billion [5]. These are truly big numbers, but would be spread over several years.

Clearly the subprime mortgage mess is the financial industry’s asbestos. It will take many years to play out and continue to be a drag on bank earnings, enrich trial lawyers and distort the housing market. That said, this does not seem to be a systemic or existential problem for the banks.

We should point out at this point that we don’t have a horse in this race. We sold our private label (non-agency) mortgage-backed holdings in 2008/2009, so our investors won’t be directly affected by these issues (delays to foreclosures can have a material impact on many poorer quality private label MBS as much of their cash flow now comes from property liquidations, rather than loan repayments). Our main purpose is to shine some light on issues that we feel have been quite poorly analyzed by the media.

No quick fix for US housing

In conclusion, we can draw some key implications from all of this:

 First, any process that only delays foreclosures, rather than curing them (including interest rate modifications on underwater loans, politically induced moratoriums, and additional foreclosure procedure hurdles) simply delays the recovery in housing. Painful as it may be, as pine cones need a forest fire to release their seeds, the foreclosure process needs to be broadly completed before recovery can truly begin.
 Second, house price movements over the next few months will be heavily influenced by the lack of foreclosure activity (and therefore forced liquidations). Any indication of a housing recovery as a result of improving home prices should therefore be met with scepticism. Policymakers will hopefully factor this in when evaluating the strength of the economic recovery, and investors should certainly not assume that US housing is out of the woods if there are some strong home price prints.
 Finally, to the extent that the government wants and is able to continue down the bailout path, the only modification program that will actually work for borrowers in negative equity is principal forgiveness. Anything else is just kicking the can down the road. However, leaving aside the moral hazard and fairness implications of mass principal forgiveness, the magnitude of the cost of solving the foreclosure crisis in this fashion means it is unlikely to happen. A weak housing market (and its related drag on consumer activity) is likely to be a persistent feature of the US economy for several years.