The House Subcommittee on Capital Markets and Government Sponsored Enterprises took its microphones to New York City on Wednesday for a field hearing on "Facilitating Continued Investor Demand in the U.S. Mortgage Market without a Government Guarantee."  The representatives heard testimony from four secondary market participants: Marty Hughes, CEO of Redwood Trust, Inc., a publicly traded company that invests in mortgage credit risk; Jonathan Lie Berman of Angelo, Gordon & Co speaking on behalf of the Association of Mortgage Investors (AMI); Joshua Rosner, Managing Director, Graham Fisher & Co. and Ajay Rajadhyaksha, Managing Director, Barclays Capital.

Hughes told the committee members that the origination half of the private mortgage market is functioning well. The top ten jumbo mortgage lenders originated $25 billion in the first quarter of the year and $30 billion in the in the 4th quarter of 2010.  The segment of the private market that is not functioning well, he said, is private securitization. 

Fixing the mortgage market is simplified if looked at in its component parts, i.e., distinguishing between prime and the non-prime segments and the reforms needed for each.  We need to restore basic functioning to the prime segment, 90 percent of the market, and design reforms to prevent abuses in the subprime segment.  A regulation designed to do one can do great harm if applied to both, something now happening with Dodd-Frank rulemaking.  Regulators should first focus on the larger prime market and leave the complexities related to risk retention, premium capture, qualified mortgages and conflicts of interest to the reform of the subprime market.

Hughes Said there are three hurdles to restarting the private residential mortgage-backed securities (RMBS) market.  First, there is no financial incentive for bank originators to securitize mortgage loans. There are currently excess reserves in the banking system of $1.6 trillion and those reserves are earning at a rate of 0.25 percent.  Excessive reserves and low rates have reduced the cost of funds for the 50 largest banks to an average of 0.81 percent and given banks a strong incentive to hold loans that cannot be sold to the GSEs to earn the spread between mortgage yields and the cost of funds.  This lack of incentive will self-correct over time and with an improving economy, but it is critical that regulators and industry practitioners address structural issues so that there will be a functioning private RMBS market.

Second, the government is crowding out the private market through programs that make 90 percent of borrowers eligible for a below-market-rate guaranteed mortgage loan.  Private capital cannot compete with government subsidized mortgage programs.

Allowing the temporary increase in conforming loan limits to expire at the end of September would be a good first step toward reducing the government's participation in the mortgage market and scaling back government subsidies that do not pass on the full cost of the risk assumed would permit a private market to flourish.  Hughes said his company advocates testing the private market by first lowering the upper limit from $729,750 to $625,500 representing about 2 percent of industry originations.  There is ample liquidity and origination capacity to allow banks to step into this small breach.  As the market begins to recover there should be further measured reduction in the conforming loan limit. He noted that, with housing prices now down over 30 percent from their peak, it would seem logical to reduce the loan limit by the same amount over time.

Third, there is a need to resolve several key regulatory and market issues including reform of underwriting and servicing standards, greater investor protections, and addressing the second mortgage problem.  Hughes said that regulators took a well intentioned approach to crafting a new set of risk retention rules by applying it to cover the entire securitization market.  However the differences between prime and subprime markets make it very difficult to apply a one-size-fits-all set of rules and the new ones are "effectively subprime-centric" and will be overly and unnecessarily harsh when applied to prime securitization structures.   

Hughes said his company supports the horizontal slice form of risk retention which requires a sponsor to retain all of the first-loss securities and places the sponsor's entire investment at risk.  The other three forms result in substantially less of the sponsor's investment in the first risk position, reducing the incentive to sponsor quality securitizations.

Redwood Trust supports the intention of the proposed definition of a qualified residential mortgage (QRM) but believes it is too restrictive, supporting the concept of "common sense" underwriting, similar to those used by the GSEs prior to the period leading to the credit bubble. 

Berman told the hearing that there are a number of market problems which presenting obstacles for private capital to return to the securitization space: 

  • Market opacity, an asymmetry of information, and a thorough a lack of transparency;
  • Poor underwriting standards;
  • A lack of standardization and uniformity concerning the transaction documents;
  • Numerous conflicts-of-interest among servicers and their affiliates;
  • Antiquated, defective, and improper mortgage servicing practices; and,
  • A lack of effective legal remedies available to investors for violations of RMBS contractual obligations and other rights arising under state and federal law.

Hughes pointed to the risk inherent in second mortgages and said that, when discussing retained risk, it is also necessary to make sure that the borrower has "skin in the game".  He recommended a Federal law that would prohibit any second mortgage on a residential property with the consent of the first mortgage holder or a requirement that a second mortgage would not push the combined loan-to-value over 80 percent. 

Berman suggested treating MBS separately from other asset classes in an effort to restore the housing sector.  "The problems impacting investors by the malfeasance of servicers and their affiliates are numerous," he said.  Many servicers are conflicted, such as when the servicer and the master servicer are the same.  Servicers may not be servicing mortgages properly, thus harming the interests of both investors and homeowners and originators and issuers may not be honoring their contractual representations about what they sold; sometimes the documents and their terminology are valueless or meaningless.

Rosner said securitization markets too often operate in a "Wild West" environment where the rules are often opaque, standards vary, and useful and timely disclosures  of the performance of the collateral at loan level is hard to come by.  "Asymmetry of information, between buyer and seller, is the standard."

Both Berman and Rosner offered similar recommendations to enhance transparency and securitization practices within capital markets:

  • Provide loan-level information for investors, ratings agencies and regulators to evaluate the collateral and its expected performance and provide monthly disclosure of collateral information post-market in an electronically manageable and standardized format.
  • Require a "cooling off period" so investors have sufficient time for loan level due diligence;
  • Make deal documents publicly available sufficiently in advance of investor decisions;
  • Develop standard pooling and servicing agreements with model representations and warranties as an industry minimum standard for all asset classes;
  • Develop clear standard definitions for securitization markets;
  • Directly address conflicts of interests of servicers by imposing direct fiduciary duties to investors and/or separating those economic interests.
  • Require substantive provisions to protect asset-backed security holders in securitization agreements.
  • Direct ratings agencies to use loan-level data on their initial ratings and update their assumptions and ratings over time.

Rosner said that even now we have not started a real discussion about housing policy or the recreation of the mortgage finance industry but, as they are two different subjects, legislators must not be permitted to use private markets as tools of social policy. He advocated for grandfathering the existing mortgage tax deduction but replacing it with a mechanism such as a tax-free housing personal savings account similar to Health Savings Accounts or 529 accounts which would be used for the future housing expenses of first-time homebuyers or first-time renters.  Another possibility is an equity principal tax credit for future mortgage originations which could target the most underserved households with a subsidy or tax credit based on the yearly reduction in mortgage principal balance.

Rajadhyaksha suggested that, on the issuance front we must rationalize various regulatory regimes that mandate capital requirements and take into account the investors' cost-basis in the security as well as expected losses when mandating those requirements. There must also be a reduction of legal uncertain especially with regard to repo and warranty enforcement mechanisms and the enforceability/transferability of the related mortgage notes. Risk retention and disclosure rules must be clarified and risk retention should focus on the origination point where credit decisions are truly made.

On the disclosure side the Mortgage Electronic Registration System (MERS) must be legalized and the process to correctly transfer loans streamlined and made uniform across states.  There must also be a timely and transparent way to enforce repo and warranties.

The uncertainty around servicing must be eliminated by creating standards similar to those needed by the FHA and FHFA.  There should be a mandated periodic release of information about the loss-mitigation efforts of servicers and standardized information on repurchases and requests for each securitizer/originator.  He would also mandate that the deal waterfall model be made available to all investors in an accessible manner such as an Index CDI and penalties be imposed for incorrectly modeled securities.

Rajadhyaksha said for decades the GSEs have hedged their interest rate risk actively in the capital markets but their bigger hazard has always been the credit risk in their guarantee pool which was not hedged even as the size of that pool grew.  He recommended that the GSEs sell a portion of the credit risk in their existing guarantee business to the private sector.   This would transfer some of the risk from the taxpayer to the private sector but more importantly would establish a benchmark against which the private sector can price mortgage credit.