'RMBS 3.0' Takes Another Stab at Setting Post-Crisis Standards
After the downturn in 2007 and 2008 decimated their returns, investors in private-label mortgage securities demanded greater disclosure from issuers. Well, they got more, but not always better.
Among the many things that continue to dampen investor interest in the asset class are 75-page term sheets and "soul-sucking" offering circulars, says John Vibert, a managing director and portfolio manager at BlackRock Advisors.
Transparency should be less about the volume of words in disclosure and more about clearly defining key terms such as "material" and "adverse," Vibert said at a recent conference put on by the analytics vendor CoreLogic. "It's got to be something that's not left open to interpretation"—clarity, not clutter.
Such continuing investor discontent is one of the many reasons that the market for post-downturn private-label residential mortgage-backed securities, known as RMBS 2.0, has been slowing. Now the Structured Finance Industry Group thinks it's time to start preparing for RMBS 3.0.
The consortium is seeking, among other things, a consensus among buy and sell side members on the "perfect architecture for the trust" in a securitization, said Eric Kaplan, managing director of mortgage finance at Shellpoint Partners, a lender and securitizer. The group also wants more standardization in disclosures, to make them an easier read for investors. And it aims to provide "transparencies where there are differences" between deals, Kaplan said at the CoreLogic meeting.
Meager as post-crisis issuance has been, compared to pre-crisis levels, it grew a bit between 2010 and 2013. Issuance from pioneering seller Redwood Trust totaled about $238 million in 2010, according to Fitch, a New York-based ratings agency. By 2013, there were eight issuers in the market and about $11.8 billion in issuance. However, issuance this year is running at about half the rate of last year so far.
Given the PL-RMBS industry's limited size, and the constraints imposed on it by market conditions and public policy, the extent to which standardization can revive private investment appears limited for now. Recent changes to representations and warranties at Fannie Mae and Freddie Mac, aimed at encouraging more sales of mortgages to the government-sponsored enterprises, present a further potential challenge to standardization. In recent years, when the GSEs loosened rep and warranty requirements, private issuers have followed suit and departed from the original RMBS 2.0 model.
Still, an improved set of standards for the market could better set the stage for a larger scale return of private capital to the mortgage market—something policymakers and recent legislative proposals continue to target as a long-term goal.
"The thinking likely is, 'Let's be ready so that should the economics shift the other way, we'll be ready to execute quite cleanly and efficiently,'" said Dave Hurt, a vice president for industry data in CoreLogic's Washington office who specializes in global capital markets.
In RMBS 2.0, trade groups tried to set best practices for the PL-RMBS market based on consensus between buyers and sellers, with some success at the outset. But some issuers later failed to adhere to certain of these standards.
The first and most notable example of this was in 2012, when Fannie and Freddie loosened their representations and warranties to include a sunset provision and Credit Suisse followed their example. (Some PL-RMBS issuers that loosened reps and warrants have said that if investors voted with their feet they would reconsider the move, and some securitizers later returned to tighter buyback provisions.)
While private-label RMBS investors have been able to sell deals with differing reps and warrants and the product has enjoyed some demand when its yield is attractive enough, the market has been fairly slow more recently, not only because of GSE competition but also competition from bank whole loan buyers.
For collateral, the PL-RMBS market counts on jumbo-sized home loans that fail to meet government guarantors' size restrictions. These loans have been in shorter supply since rates rose last year in response to the Federal Reserve's plan to taper its asset purchases. Top jumbo lenders on average in the first quarter of this year were originating about 14% less than during the first quarter of last year, according to National Mortgage News' Quarterly Data Report. Banks that favor the assets have as a result grown more interested in buying jumbo loans, because they are harder to originate. That diminishes the product available for securitization.
Even if more PL-RMBS paper were available, Vibert says investors like him don’t have any appetite for the product and only have “dabbled” in it to test it out. They cite various frustrations with it, including disclosure overload.
Previous attempts at buyer-seller consensus on RMBS 2.0 standards have tended to break down in three main areas: representation and warranties, and enforcement thereof; data, due diligence and disclosures; and bondholder and trustee communications.
Issuers that departed from previous consensus standards did so most prominently in terms of reps and warrants. Life-of-loan reps and warrants, for example, had been the industry standard until Credit Suisse added a three-year "sunset" provision in a 2012 PL-RMBS transaction. The move followed a proposal by the Federal Housing Finance Agency that year to add similar sunset provisions to the GSEs' rep-and-warranty policies. This suggests that the most recent tweaks to the GSEs' rep and warrant policies could play a role the next round of revisions to PL-RMBS reps and warrants as well.
Investors in the PL-RMBS market likely want stronger reps and warrants than agency MBS investors, though. Government guarantees largely insulate agency MBS investors from credit risk (unless they choose to participate in a new type of agency security that allows them to share this risk).
PL-RMBS investors are not only more directly exposed to credit risk, they have relatively little control over how it is handled. Often, they lack the power to compel securitization trustees to take particular actions (unless they join with enough fellow buyers to represent a certain percentage of the transaction, such as more than 25%). Vibert says he's been unhappy with how servicers handle collateral in many instances as well.