In February, the Federal Housing Finance Agency (FHFA) issued a strategic plan for its ongoing conservatorship of Freddie Mac and Fannie Mae (the GSEs).  The plan had three primary goals:

1. Build a new infrastructure for the secondary mortgage market.
2. Gradually contract the Enterprises' dominant presence in the marketplace, simplifying and shrinking their operations.
3. Maintain foreclosure prevention activities and credit availability for new and refinanced mortgages.

Yesterday FHFA published a white paper proposing an infrastructure that attempts to connect these goals.  To maintain the efficient flow of mortgage credit, the existing antiquated and inflexible GSE infrastructures must be upgraded.  A transition to a future securitization framework would require a more flexible infrastructure than is currently available to accommodate future policy decisions. Given that the expenditures for the upgrades are necessary, it makes sense to direct them toward the development of a common flexible infrastructure to accommodate various structures and policy goals.

FHFA said it believes that work can begin on these elements even while the structure of housing finance and government's role in it are debated through an incremental approach, starting with the very basic elements and building with feedback from the industry and regulators and decisions from policymakers.

At present approximately 75 percent of mortgage securitizations is done through the GSEs and 25 percent through Ginnie Mae.  Two key goals of the proposal are to preserve liquidity while developing a framework that will include the reentry of private capital into risk taking.

Some core functionalities of a securitization infrastructure can be standardized and serve a utility function; issuance, master servicing, bond administration, collateral management, and data integration and these are where FHFA suggests focusing its efforts.  The securitization platform must be flexible enough to adapt to the evolving standards and requirements of stakeholders including originators/lenders, securitizers/issuers, trustees and bond administrators, investors, and regulators. 

The GSEs currently perform many securitization functions including issuance, trustee, guarantee, credit underwriting and pricing, master servicing, and credit loss management.  Their functions are essentially similar and the GSEs are working to align them so markets don't have to manage two distinct processes. Additionally, each Enterprise continues to make separate investments in maintaining its set of guides and contractual framework, increasing the cost to taxpayers.

Ginnie Mae administers the mortgage-backed securities program for FHA, VA, and the Department of Agriculture (USDA) where the pooled loans are issued as securities by the issuers.  The standards that govern the origination and default servicing of loans in the Ginnie Mae pools are established by the sponsoring agencies which also bear the credit risk.  The issuers are responsible for servicing and Ginnie Mae mainly has exposure to servicer counterparty risk while the GSEs also have exposure to mortgage insurers and borrowers.

The securitization platform in the Ginnie Mae model has the same elements as those of the GSEs, but Ginnie Mae is responsible for few of them, outsourcing all but policy-making and contract and non-contract resource and relationship management functions to third parties.

The Private Label Securities (PLS) market relies on private investment capital for support and private parties perform all of the functions shown above.  Credit risk is either retained by the aggregator or investor, guaranteed by a third party or re-packaged into new securitization structures.  This market relies on customizing its own rules rather than following the GSE model.  Since 2008 investors have largely avoided mortgage credit risk but there signs some capital is returning although not at significant scale.

Each of the three securitization platforms uses its own contracts and guidelines and while those used by the GSEs and Ginnie Mae result in fairly uniform and standard outcomes, many versions of Private Label PSAs have produced divergent business practices that led to ambiguity in interpretation and placed different market participants (i.e., sellers, borrowers, servicers, trustees, senior investors and subordinate investors) at direct odds with each other. 

The framework proposed by FHFA contains certain key principles it views as critical to the success of a functional secondary mortgage market; promoting liquidity, attracting private capital, benefiting borrowers and operating flexibly and efficiently, while minimizing market disruption during transition.  This framework focuses on functions repeated across the industry where greater standardization would benefit the overall market while not limiting market choices or valuable independent innovations.

The functions highlighted in Figure 3 could be accommodated via the securitization platform through standardized processing, and possibly offered to the market as a form of utility.

Functions are defined as:

• Collateral management, specifically centralized note tracking.
• Master servicing within the overall servicing related functions, including asset and cash management, standardized interfaces to servicers, guarantors, and aggregators, servicing metrics, data validation and reporting.
• Issuance, including eligibility rules, data quality standards, pool delivery, settlement and disclosure.
• Data validation (servicing and issuance) with the securitization platform storing loan level, pool level, and bond level data.
• Bond administration, including standardized investor and third party disclosures, bond processing, principal and interest distributions, securities monitoring, portfolio reporting and role of Trustee.

In addition to these key functions, greater consistency is necessary to a more stable, liquid, and efficient market so another proposed function is the creation of standardized documents including a new PSA incorporating a robust selling and servicing guide.

FHFA proposes that a new platform would include credit risk distribution designed to be flexible to accommodate multiple models of housing finance so that policy makers can ultimately chose those that could be less dependent on government assumption of risk. It could also support the distribution of credit risk to the private sector through various credit risk sharing arrangements and support credit structured securities where various classes of securities holders agree to receive cash flows based on contractual distribution rules.

The envisioned platform would bundle mortgages into an array of securities structures and provide all the operational support to process and track the payments from borrowers through to security creation to payoff. 

The platform must maintain existing liquidity, while enabling the entry and participation of private capital.  Thus is should be able to support the current TBA market and securities across a range of interest rates and should include the high level of automation. necessary to process the current volume of industry issuance - more than $100 billion per month.

Privately guaranteed securities could be similar to the guarantee provided by the GSEs today but different in the future and the platform must to be flexible enough to support all potential options while bringing transparency to investors and facilitating the return of private capital by supporting various options for credit investors to participate in the market.

The housing finance lifecycle, shown in Figure 4 below, frames the scope and functions that a platform would serve, and how it would benefit the industry and the housing finance system:

It is at the final phase that the automated securitization platform could serve both the GSEs and the wider securitization market of the future by providing a discrete set of services at the time of issuance and on an ongoing (typically monthly) basis in support of principal and interest payments to investors. 

The initial core services proposed by the paper are intentionally selected as foundational due to their place within the housing finance lifecycle. The initial core services of the proposed platform have the following characteristics:

• They lend themselves to straight-through, highly automated processing of large volumes with limited manual intervention;
• They accommodate setting and adjusting market standards and assist in market and data transparency.
• They are rules driven and can readily adapt to changes in standards and policy.
• They are flexible to service many industry models.
• They can enable the private sector to continue to drive security selection, loan pooling and underwriting, and other non-GSE functions.

The platform scope would include the issuance of securities as requested by a user, the monthly master servicing of loans while in the securities, the payment of monies to securities investors, and the tracking and disclosure of securities balances, payments and underlying loan performance data.  The data validation service would verify that the request conforms to standards and the format for loans and securities is correct and complete. 

The master servicing functions performed by the platform include asset and cash management activities including collecting and processing primary servicer loan activity and verifying that principal and interest payments are correct for each reporting cycle.

The platform could monitor and direct document custody, and monitor primary servicer performance for adherence to standards as well as all other compliance directives requested by the trustee or other governing documents.

Bond administration would be responsible for calculating investor payment factors and making data available to the disclosure function for ongoing investor reporting for each payment cycle.  Bond administration would support both first level securitizations and second level re-securitizations. Bond administration would also ensure payments to investors and other parties are managed in a timely manner.

Disclosures describe a security and underlying loans or pools published to the marketplace before and when the security settles describing the securities' fundamental characteristics and detailed information on the pool and security structure.  On-going disclosure occurs monthly, or when relevant changes occur.  Disclosure documents proposed by FHFA would include the prospectus supplements and supplemental oversight documents. Disclosure information would be published by the platform on the Internet and possibly in other forms.

There a handful of other, more specific design principles that will guide how the platform is designed and built. 

• Open architecture: standard external interfaces would be established and the platform would leverage existing data standards
• Functional modularity - internal components would communicate via standard interfaces to ensure that changes can be accomplished with minimal impact.
• Scalability - Integration architecture would be based on standard technologies with proven scale in financial services and multiple industries.
• Data transparency - Architecture would provide data traceability and accessibility via common infrastructure and standards.

Under previous FHFA guidance, the Enterprises are transitioning their single-family loan delivery data formats to one that leverages the industry-recognized standard.  The platform would expand upon this model, allowing the primary market to deliver bundles of mortgages into any of an array of securities structures, including those that support PLS and other innovative credit transfer facilities.  It would also provide operational support to process and track the payments from borrowers through to the investors, by leveraging the Uniform Mortgage Servicing Dataset Initiative to define a standard servicing dictionary and data exchange.

This, in conjunction with the proposed PSA framework, could improve service to borrowers, reduce financial risk to servicers, and provide flexibility and data for guarantors to better manage non-performing loans.  The Uniform Mortgage Servicing Dataset Initiative will make standard servicing interactions (such as servicing transfers, loan removals from securities, etc.) consistent and transparent across the industry.

The proposed integrated infrastructure cannot function without the appropriate legal agreements, rules and allocations of responsibilities, including a framework for an effective contractual PSA.

The key elements for private capital to return to the housing finance market hinge on clear rules and standards about: 1) the integrity of mortgage originations in terms of data and seller responsibilities; 2) the adequacy of servicing; 3) servicer compensation; and 4) disclosures to investors. In addition, there are examples of ongoing work at the GSEs that can be incorporated into the model PSA: 1) alignment and enhancement of the existing policies, practices, and legal documents of the GSEs; and 2) standardized servicing, disclosure and other practices of the GSEs.

The proposed PSA framework would leverage the existing structure.  The framework would include a short PSA containing lender/seller specific requirements and variances, general trust provisions and incorporate robust program guides which would set out the requirements for underwriting, disclosure, servicing and loan delivery and setting forth the minimum standards for eligibility to use the platform, specifying duties and responsibilities, reflect a compilation of "best practices" and applicable regulatory requirements for servicing and the other transaction participants, and address certain shortcomings of the Private Label PSA.

 The design of the common securitization platform would support a government guarantee while not assuming that the future housing market will incorporate one and will have the flexibility to accept different levels of credit standards and mortgage products.  The structure could help enforce Qualified Mortgage (QM)/Qualified Residential Mortgage (QRM) guidelines, screening for QM/QRM eligibility, and identifying any corresponding risk retention requirements.

Here are a few of the questions about the proposal for which the FHFA is soliciting comments.

1. Will the four core functions (issuance, disclosure, bond administration and master servicing) provide an efficient and effective foundation for the housing finance system going forward?  Are there additional functions that should be considered as core?
2. Will the framework for a model PSA described herein provide the foundation for a standardized contractual framework for the housing finance system going forward?
3. Are there additional elements/attributes that should be included in a model PSA?
4. What enhancements to the role of trustee should be considered in order to better attract private capital to the housing finance system?

Directions for submitting comments which are due December 2, 2012 can be found here.