By Alison Frankel
NEW YORK It is an axiom of the financial crisis that Goldman Sachs realized before any of the other big banks that the mortgage-backed securities market was going to implode in 2007. Goldman dumped MBS and shorted the market, turning a profit in its mortgage department when every other major financial institution suffered record losses.
So what tipped Goldman to start off-loading its MBS exposure at the end of 2006? In a new brief in its securities fraud case against the bank, the Federal Housing Finance Agency has an intriguing theory.
According to the conservator for Fannie Mae and Freddie Mac, Goldman received a report on December 10, 2006 from the CEO of Senderra, a subprime mortgage lender partially owned by Goldman. Goldman had taken a stake in Senderra -- a relatively small mortgage originator -- to stay informed about the state of the mortgage market, according to FHFA.
In the December 10 emailed report, Senderra CEO Brad Bradley supplied grim market intelligence. "Credit quality has risen to become the major crisis in the non-prime industry," he wrote to Kevin Gasvoda, the head of Goldman's whole loan trading desk. "We are seeing unprecedented defaults and fraud in the market, inflated appraisals, inflated income and occupancy fraud."
Gasvoda forwarded the proprietary report to other Goldman executives, along with what he called a "very telling" follow-up email from Senderra that warned of "irrational" mortgage originators chasing "any loan which smells of quality."
Four days later -- after the confidential Senderra report circulated within the bank -- the head of Goldman's mortgage department met with senior bank officials, including CFO David Viniar, to discuss how Goldman could take advantage of the imminent mortgage meltdown. That very day, according to the FHFA brief, Goldman launched the sell-off of its MBS holdings.
"Goldman had unique and confidential access to the dire warnings directly provided by its own affiliate and proprietary originators that the sub-prime mortgage market was experiencing 'unprecedented defaults and fraud,'" the brief said. "Yet Goldman failed to respond, except to the extent it sought to profit through shorting the market."
FHFA's lawyers at Quinn Emanuel Urquhart & Sullivan contend that the Senderra report is just one of the reasons that Goldman should not be permitted to argue to a jury that it conducted a reasonable investigation of the mortgage loans underlying the securities it sold to Fannie and Freddie.
The agency has asked the judge overseeing the Fannie Mae and Freddie Mac litigation, U.S. District Judge Denise Cote of Manhattan, to grant summary judgment precluding Goldman from asserting a so-called due diligence defense when the case goes to trial on Sept. 29.
Goldman's diligence, according to the Fannie and Freddie conservator, was at best "ramshackle" and at worst a tool of deception. Either way, FHFA said, the bank did an inadequate job, as a matter of law, of assuring that underlying mortgage pools lived up to its promises to investors.
In addition to the Senderra email revelation, the other marquee accusation in FHFA's new brief is that Goldman did conduct independent investigations on the value of some of the properties whose mortgages it bought from originators -- but the bank supposedly used that information for its own benefit rather than for the benefit of MBS investors.
The alleged duplicity involved a key MBS benchmark, the ratio between the size of each underlying mortgage and the actual value of the mortgaged property, otherwise known as the loan-to-value ratio, or LTV.
If the property is worth less than it's mortgaged for, homeowners are likelier to walk away from their loans, which diminishes the stream of revenue to MBS noteholders.
In about 170 instances, according to the FHFA brief, after Goldman ran its own checks on properties with apparently outsized mortgages, it kicked those loans back to the mortgage originators from which it had purchased them.
FHFA's brief said that Goldman then repurchased 11 of those mortgages from originators at a discounted price, based on its finding that the value of the property didn't justify the size of the loan.
But according to the FHFA brief, when Goldman included the repriced loans in MBS pools, it did not tell investors that the loan-to-value ratio was out of whack. "Goldman used these 'final values' to negotiate lower prices for itself," FHFA said, "but it reported LTV ratios to investors based on the originators' stated values -- which Goldman believed to be false."
I'm sure Goldman's lawyers at Sullivan & Cromwell will have lots to say about FHFA's allegations when they file their response to FHFA later this month. For starters, Goldman will probably emphasize that FHFA has claimed it renegotiated prices on a grand total of 11 underlying loans, of the tens of thousands at issue in the case.
You can also expect the bank to argue that the Senderra report in December 2006 was old news: The mortgage industry was already rife with talk of loose underwriting standards and borrower fraud.
In fact, according to a joint summary judgment brief filed in June by three of the remaining FHFA defendants, including Goldman, Fannie Mae and Freddie Mac knew those rumors as well as anyone in the MBS marketplace. (Unfortunately, most of the banks' specific evidence of what Fannie and Freddie were told in 2006 is redacted.)
Summary judgment is supposed to rest on undisputed facts, and Goldman will certainly claim that the facts here are in dispute. On the other hand, Judge Cote, as you know, has consistently sided with FHFA in this litigation, which is why banks have already coughed up $15 billion in settlements with the agency.
With a trial date only weeks away, is Goldman willing to risk adverse summary judgment rulings that will increase FHFA's leverage?
Goldman still has one potential trump card -- a motion that could end the litigation and make all of the banks that have settled with FHFA wish they'd stuck it out too. But you'll have to wait until my post tomorrow to find out what Goldman's Hail Mary is!