Why Former Investors Despair of Returning to Mortgages

Investors are likely to say "fool me once…" which may explain why the volume for newly issued private label residential mortgage securitizations is a fraction of what it was a few years ago. Through bitter experience, bond purchasers learned about the moral hazard embedded in private RMBS and their grossly inadequate legal protections.

"The private RMBS market was at the heart of the financial panic and the Great Recession that followed," writes Mark Zandi of Moody's Analytics. Indeed. At year-end 2007, private RMBS totaled $2.2 trillion. According to Zandi's tally, those bonds realized losses, during the 2006-2012 period, of $449 billion.  That amount exceeded the total losses on $9 trillion of mortgage debt financed by everyone else, all depository institutions and all government-backed lenders, including Fannie Mae and Freddie Mac. Also, Zandi excludes synthetic subprime collateralized debt obligations, which financed nothing tangible and which lost more money than Fannie and Freddie combined.

More pointedly, the heart of the financial panic and the Great Recession that followed was an epidemic of fraud and sloppy recordkeeping facilitated by the originate-to-distribute model for private RMBS.  Parties complicit in fraud – borrowers, mortgage brokers, originators, rating agencies, investment banks and servicers – calculated that the odds of being held fully accountable were close to nil. That assessment has stood the test of time.

How bad was the fraud? The Federal Reserve Bank of New York found that, in the bubble states at the peak of the market, "almost half of purchase mortgage originations were associated with investors. In part by apparently misreporting their intentions to occupy the property, investors took on more leverage, contributing to higher rates of default."  Fitch reviewed the documentation of a small sample of subprime loans in default in late 2007, and found some kind of misrepresentation in almost every file. In early 2007, before home prices began to tank, BasePoint Analytics found that 70% of early payment defaults were tied to mortgage fraud. 

Consider what U.S. Bank, N.A., a trustee of one Countrywide RMBS deal, HarborView Mortgage Loan Trust 2005-10, uncovered after it hired an underwriting consultant to review some nonperforming loans. The consultant looked at a small but meaningful sample, 786 loans out of total pool of 4,484 mortgages, and found that two-thirds of the loan sample, 520 out of 786, contained one or more breaches of representations and warranties. About 90% of those breaches involved fraud or violations of Countrywide's underwriting guidelines.

But U.S. Bank could not review every loan file. Remember, reps and warrants – which traditionally claim that no document in connection with the transaction contains untrue statements of material fact – have been used as the primary source of legal comfort that the seller stands behind the purported accuracy of loan files in a pool of 2,000 mortgages.

However, to get access to the loan files, investors generally need to attain a certain percentage of votes among investors, and that takes time and lobbying. And then it takes time to actually review the files. Both banks and trustees have been uncooperative in past cases.

The current owner of Countrywide, Bank of America, rejected the notion that aggrieved investors might be able to review loan samples for 530 other Countrywide deals, which were originally worth $424 billion and are now expected to lose about $100 billion. According to courtroom testimony, one B of A lawyer said that type of statistical sampling could involve hundreds of thousands of loan files, so that "our grandchildren would have grandchildren before a trust would receive a dollar of recovery."

B of A calculates the average Countrywide defect rate to be 36%, which is good enough for bond trustee Bank of New York Mellon, which has refused to review the actual loan files. If investors failed to accept B of A's $8.5 billion take-it-or-leave-it offer for settlement of all putback claims, the bank reportedly told investors it would put its Countrywide subsidiary into Chapter 11.

As for seeking recourse under federal securities laws, RMBS investors face daunting challenges. To bring a lawsuit, a plaintiff needs to show what statements in the public disclosures were false or inaccurate within three years of the initial offering. But, once again, it's frequently hard to get access to incriminating loan data. 

Poor record-keeping by all parties also provides barriers to pursuing and winning litigation. And how bad was recordkeeping? Nevada's experience suggests that it was really bad. Once the state passed a law making parties criminally liable for committing fraud as part of a foreclosure process, home foreclosures came to a virtual standstill. A coincidence?

Two RMBS investors, Fannie and Freddie, were fortunate, because their regulator could access data more easily, and because the Home Equity and Recovery Act of 2008 states that, for the government-sponsored enterprises, the statute of limitations begins to run when they are taken into conservatorship.

If Washington wants to reform housing finance, it should bolster investor protections by extending statutes of limitations for securities fraud, and by mandating greater investor access to loan files related to any mortgage default.

David Fiderer has previously worked in energy banking for more than 20 years. He is currently working on a book about the rating agencies.