Restructuring Structured Finance
Dow Jones Newsletters: Daily Bankruptcy Review July 11, 2007
In the aftermath of the Amaranth meltdown, we've seen a few more hedge funds destroyed when their mark-to-model valuations proved to be out of sync with something as crass as the actual market. Notwithstanding the tendency of financial players to dismiss every meltdown as just another hiccup, the recent struggles of a pair of turbo-leveraged Bear Stearns hedge funds are raising concerns that all the happy chatter may have been just that.
Certainly, Bear Stearns' miserable experience with subprime securitizations and derivatives is another reminder of the risks of investing in hedge funds specializing in illiquid, opaque investments. But it has even more ominous implications for fiduciaries across the country who may have more pain coming their way then they ever realized. Why? Because the big revelation in this latest crisis is the significant modeling imperfections of the rated structured-finance world. The enormous recent ramp-up of securitized assets based on new products, unvetted structures and models that collectively are behaving in ways unanticipated by the rating agencies is increasing scrutiny of the agencies' efficacy, and even our entire ratings system.
The issues with Bear Stearns' particular bête noir, subprime, are instructive across today's structured-finance world. There was a time when the securitization of residential mortgages was a plain-vanilla and wholly wonderful vehicle for funding home ownership in this country. You could put a whole bunch of housing loans in a pool and pass through their payments to investors, or use them as collateral for a special purpose vehicle whose tranches would be sold to buyers willing to hold risk to a greater or lesser degree. Losses on tranched loans would first be absorbed by those investors holding the riskier tiers, while those desiring safety would rest comfortably in their investment grade perches. But, as industry observers have pointed out, what made it all work was the fact that those initial securitizations were populated by mortgages originated under the auspices of Government-Sponsored Enterprises ("GSEs") Fannie Mae and Freddie Mac, whose underwriting standards encompass pages and pages of detailed requirements for both the loan and borrower. These were the underpinnings of the famed conventional loans and, thanks to the GSE conformation process, these assets were pretty homogenous and quite susceptible to predictable modeling. Of course, the rates on the resulting securities were nothing to get too excited about but at least you got what you paid for.
After an unhappy initial run at subprime in the mid-nineties, non-GSE packagers began to pursue its securitization again in earnest after 2000. Notwithstanding the fact that the rating agencies were now in unfamiliar territory, modeling behavior of loans and borrowers who weren't conforming to anyone's standards, let alone the GSEs', they happily played their part, charging issuers a pretty penny for all the work they were required to do to tranche subprime pools up.
For a while, thanks to a charitable interest- rate environment and home prices rising to the sky, it all seemed to work. Emboldened by their seeming success, mortgage originators grew ever more creative and generous and loans and borrowers began to deviate substantially from those of old. We began to see massive securitization of products with characteristics that if they existed at all before did so in only small numbers and certainly not in combination: affordability loans - either interest-only or 40-year amortization, no doc/low doc (so-called "liar") loans; and silent seconds which allowed many borrowers to buy a house with absolutely no skin in the game.
And, of course, most of these subprime loans were 2/28 or 3/27 adjustable rate mortgages with low teaser rates set to adjust in two or three years. Those ARMs, offered in conjunction with some of the other loan features, attracted many otherwise marginal homebuyers including a class of borrowers who qualified for financing virtually on their own say so, making a bet that by the time their rate was to reset, they would have sold their mortgaged property for a quick profit, or at a minimum be able to refinance advantageously. The party ended during the waning months of 2005 as interest rates began an upward march and housing prices stagnated and even dropped. Many, many borrowers found themselves faced with refinancing or reset options at considerably higher cost than what they had been paying. With newspapers trumpeting deterioration in the housing market, mortgage defaults began occurring fast and furiously, certainly well above the expectations of rating agencies who are still trying to figure out how and why their models went so far wrong. Not only are the assets causing fits; the structures themselves are proving to be problematic. Some subprime vehicles are bumping up against tax and accounting restrictions on the ability to modify loans, an issue structurers had never contemplated since, under historic default levels, they never hit those limits. Others vehicles have the reverse problem. To the chagrin of the rating agencies, the ability to modify large numbers of bad loans may cause the premature release of collateral cushions because the modified loans, which will likely have higher default rates than unmodified ones, will be considered "good" loans for purposes of determining vehicle performance. There are countless operational problems as well, since the rapid transfer of paper from originators to warehousers to SPVs and beyond is creating a paper jungle crowded with examples of servicers unable to produce the documentation they need to manage a loan.
One more thing. With something like $1 trillion of two- and three-year adjustable rate, mostly securitized subprime mortgages set to reset in the next few years, it is a bit imprudent to say that subprime problems are anywhere near resolution (and we haven't even begun to address the billions of dollars of nearly as troubling alt-A loans). One also has to wonder if the same sorts of issues, modeling and structure errors and overly optimistic valuations are prevalent in commercial real estate and corporate loan securitizations as well since those sectors were certainly subject to the same relaxation of lending standards and accelerated securitization that we've seen in subprime.
Already the unfortunate fate of investors is spurring lawsuits and governmental investigations into potential fraud and malfeasance in the structured finance markets. But, it is clear that these troubles also raise important questions about the integrity of the rating system and its interplay with the insurance, banking and pension industries whose regulatory structures are based on the premise that ratings mean something. Should rating agencies be held more accountable for their pronouncements and, if so, how?
What are the implications of having them compensated by the issuers? How should regulators deal with the ability of structured finance to move credit extension out of the regulatory sphere? Expect these questions and others to get strong attention from regulators and investors as the problems in structured finance play out over the coming months.