Report: Ginnie Stress Testing Inappropriate for Smaller Issuers, Cautions CHLA

  • by

The Community Home Lenders Association (CHLA) is pleased to submit these comments in response to Ginnie Mae’s Request for Input on its Stress Testing Framework.

Ginnie Mae’s stress testing framework is designed for non-bank issuers (also known as Independent Mortgage Bankers or IMBs). CHLA submits these comments as the only national association that exclusively represents IMBs, and many CHLA members are community-based Ginnie Mae issuers. References to “smaller issuers” are intended to refer to small and small-to-mid-size issuers. Following are CHLA’s main recommendations:

1. Stress testing should not be used for smaller Ginnie Mae issuers as it would not accurately assess smaller issuer risk, would undermine Ginnie Mae’s access to credit duties, and would increase issuer concentration.
2. If applied in any capacity to any smaller issuers, stress testing should: (1) be fully transparent, (2) not impose an undue regulatory compliance burden on smaller issuers, (3) be fully beta tested for smaller issuer impact and negative bias before being applied to any smaller issuers, and (4) not supercede qualitative reviews or liquidity or net worth ratios that have historically been used to supervise smaller issuers.
3. Ginnie Mae should proactively communicate with warehouse lenders about whether and how it intends to apply stress testing to smaller issuers – in order to avoid warehouse lenders reducing credit to smaller issuers merely out of fear or lack of knowledge of any changes that Ginnie Mae might be considering.

Ginnie Mae Statutory Responsibilities: Access to Credit and Financial Soundness The Congressional statute that chartered Ginnie Mae (Section 301 of the National Housing Act) expressly states that Ginnie Mae’s purpose is to establish a secondary mortgage market, in order to: (1) provide stability in the secondary market for residential mortgages; (2) respond appropriately to the private capital market; (3) provide ongoing assistance to the secondary market for residential mortgages (including activities relating to mortgages on housing for low- and moderate-income families. . .); (4) promote access to mortgage credit throughout the Nation (including central cities, rural areas, and underserved areas) by increasing the liquidity of mortgage investments and improving the distribution of investment capital available for residential mortgage financing; (5) manage and liquidate federally owned mortgage portfolios in an orderly manner, with a minimum of adverse effect upon the residential mortgage market and minimum loss to the Federal Government.

Consistent with these dual statutory responsibilities, CHLA’s recommendations above reflect the following: • Stress testing of smaller issuers would be inconsistent with how stress testing has historically been used. • Applying stress testing to Ginnie Mae’s largest issuers statistically addresses most of its risk. • Ginnie Mae already has sufficient tools to regulate smaller lenders. • Ginnie Mae stress testing model is based on large issuers and ignores smaller issuer qualitative factors. • Problem resolution risk levels are higher for larger issuers than for smaller issuers. • Stress testing of smaller issuers would increase Ginnie Mae concentration and risk.

2 CHLA January Report on Ginnie Mae

In January 2019, CHLA released a comprehensive report on Ginnie Mae: Click for Report

CHLA’s report explicitly expressed support for a number of steps Ginnie Mae has taken to date to ensure it is adequately supervising non-bank issuers, including
• Stress testing of its 15 largest non-bank issuers.
• All Participants Memorandum 18-07, to strengthen counterparty risk management.
• Curbing churning of VA Loans (e.g. through APM 177-06).
• GNMA 2020 – particularly the emphasis on “Very Large Institutions,” which highlights the point that failure resolution costs and challenges are substantially higher for larger issuers than small ones.

A major focus of the CHLA report was to put Ginnie Mae’s financial condition and risk in perspective. Ginnie Mae noted in its 2016 Summit release that Ginnie has a ”last dollar loss” position on loans it insures, only incurring losses after the following first absorb losses: (1) Homeowner Equity, (2) Government Agency Insurance, and (3) Corporate Resources of Issuer. CHLA’s Report also noted that, “Ginnie Mae can even make money when an issuer goes out of business, as an issuer contractually forfeits its right to recoup previously made advances and Ginnie Mae takes over that asset.”

The CHLA Report highlighted Ginnie Mae’s strong historical financial results – noting that: (1) Ginnie Mae currently has $25.9 billion in equity to absorb losses. (2) The Administration budget projects Ginnie Mae will make $1.914 billion in net profits in the current year. (3) Ginnie Mae made $10 billion in net profits the last decade & was profitable during the 2008 housing crisis. (4) Ginnie Mae’s historical loss claims appear to be minimal or non-existent.

The CHLA Report noted that Ginnie Mae itself seems to regard its largest issuers as their greatest risk, and it has applied stress testing to their 15 largest non-bank issuers. An October 2018 HUD IG report “cited the risks of larger issuers, referring to the challenge of Ginnie Mae servicing mortgages absorbed in a default, saying this ‘might require additional funds from the U.S. Treasury to pay investors if a large issuer default occurs.”

Because of all these factors, and in light of Ginnie Mae’s dual statutory responsibility to facilitate access to credit, the CHLA Report concluded: “The purpose of this report is to warn that Ginnie Mae supervisory tightening of smaller IMB issuers that is disproportionate to the risk they pose could undermine GNMA’s primary statutory responsibility to facilitate access to mortgage credit.”

The Report went on to make a number of recommendations relevant to this RFI – including • Ginnie Mae stress testing and increased net worth and liquidity requirements should be primarily focused on issuers with the highest risk of Ginnie Mae loss, i.e.: (a) issuers with the largest portfolios and (b) issuers with complicated financial structures.

• Ginnie Mae should not have an objective of reducing the number of Ginnie Mae issuers or of eliminating smaller issuers – particularly on a basis: (a) of a general objective of shrinking the government footprint, (b) that smaller issuers are allegedly not as profitable as larger issuers, or (c) that Ginnie Mae does not have adequate staff to supervise all issuers [Congress should fully fund staff out of Ginnie Mae’s billions in annual net profits].

• Ginnie Mae should balance supervision of issuers between the dual objectives of: (a) Meeting access to credit needs through a broad borrower base, and (b) Protecting taxpayers through prudent financial management.

 Smaller Issuer Stress Testing is Neither Appropriate or Needed

Ginnie Mae stress testing of smaller IMBs would be inconsistent with how stress testing is generally applied – which is to a financial sector’s largest firms and/or those that are “too-big-to-fail.” The 2010 Dodd-Frank Act required stress tests for only the largest non-bank financial firms and for banks in excess of $10 billion in assets (and Congress has been considering raising even that threshold). Though mandated in Dodd-Frank, the NCUA started stress testing credit unions – but only those over $10 billion in assets – in 2015.

A 2015 FDIC paper [“Stress Testing Banks: Whence and Whither?”] explains the purpose of the Dodd-Frank stress testing, stating that this provision was “aimed at improving the stability of the U.S. financial system” and “calls for larger banks and bank holding companies to undergo so-called “stress tests” that require both bank managers and regulators to understand what would happen to banking institutions if they were subjected to ‘exceptional but plausible macroeconomic shocks.’”

The additional step of engaging in detailed analysis of how a financial firm would hold up in exceptional and adverse economic circumstances has been in large part driven by concerns about the shocks to the system – socalled “systemic risk” – of a large institution whose interrelationships can have an adverse economic impact on other substantial financial entities, on consumers, on the economy, or on all three.

In sharp contrast, smaller IMBs – like smaller community banks – meet none of these criteria. Failure of one or a number of smaller IMBs would have no such systemic impact – nor would it pose a grave risk of a taxpayer bailout for Ginnie Mae, with its tens of billions of dollars of reserves and is consistently profitable.

Further, as the FDIC paper cited above points out, none of these major stress-testing regimes of major banks include stress tests of bank liquidity (the main risk specifically identified by Ginnie Mae) – but instead are focused on analyses of a bank’s net worth.

Applying stress testing to only its largest issuers addresses most of Ginnie Mae’s risk. As noted in CHLA’s January report, “Large issuers constitute the bulk of Ginnie Mae’s risk exposure. Ginnie Mae’s largest 15 issuers constitute 75% of GNMA securities – while its smallest 144 issuers (almost half of the number of total issuers) constitute only .49% [less than ½ of one percent].”

Therefore, applying formal stress testing to only Ginnie Mae’s 15 largest issuers covers the great majority of their risk exposure – without the risks of inappropriately driving out smaller issuers from the program. Ginnie Mae already has sufficient financial tools to regulate smaller issuers. CHLA believes that the current bright line and transparent liquidity and net worth ratios that Ginnie Mae uses to establish program eligibility, combined with qualitative supervision and monitoring of an issuer’s management, are fully sufficient for smaller issuers with a traditional lending business and financial structure.

In its 2020 document, Ginnie Mae made it clear that the principal risk to Ginnie Mae is not credit risk but advance risk – the need for an issuer to have liquidity to cover advance obligations. Basic financial ratios measuring liquidity and net worth already have as an underlying basis an implied purpose of being available to cover adverse economic circumstances – which is what stress testing is meant to address.

Ginnie Mae stress testing model is based on large issuers and ignores smaller issuer qualitative factors. Financial and managerial challenges of smaller issuers differ from those of larger issuers and are unlikely to be accurately captured by stress testing. Smaller community-based issuers have a deeper understanding of the communities and borrowers they work with, and because of their smaller size, can more quickly adapt their business to changing market realities and have more options in selling loans or portfolios than larger issuers.

A January 2017 Ginnie Mae report concluded that “. . . nonbanks have led the way in improving access to credit for low- and moderate-income borrowers. . .” Therefore, a broad diversified base of non-bank issuers is critical to carrying out its critical mortgage access to credit role.

After the 2008 crisis, as banks retreated from mortgage lending and Ginnie Mae issuance, non-banks stepped in to fill the gap, providing more personalized service. If stress testing overshoots, driving out small/mid-sized IMB issuers, who will be left to fill the gap?

Layering on complex calculations using adverse speculative economic scenarios on top of existing net worth and liquidity ratios for plain vanilla small issuers risks setting the bar so high and inflexibly that smaller lenders will be unnecessarily driven out of the Ginnie Mae program, either voluntarily or involuntarily.

CHLA believes such an approach would not properly balance Ginnie Mae’s dual statutory responsibilities of facilitating mortgage access to credit and limiting financial risk. If anything, Ginnie Mae should explore more ways to promote access to credit, such as not counting Ginnie Mae advances for loans in natural disaster areas as a default and restoring its Targeted Lending Initiative, which was targeted to low/moderate income families.

Problem resolution risk levels are higher for larger issuers than smaller issuers Another reason it is not appropriate to establish similar stress testing standards for larger and smaller issuers is that the degree of loss exposure is disproportionately higher for larger issuers than for smaller issuers, for a number of reasons specifically cited in CHLA’s January Report:

(c) “In the same way that it is much easier for the FDIC to resolve a smaller failed bank than a larger one, it is generally easier for Ginnie Mae to arrange the transfer of a failed portfolio of a smaller issuer than a large issuer. This is because the universe of servicers capable of absorbing a small servicing portfolio is much larger than for a large portfolio.

(d) The possibility that a single, fast growing issuer is going to engage in fraud or otherwise place FHA loans without proper insurance in Ginnie Mae pools is much higher for one large servicer than for a number of smaller issuers with the same total issuance exposure.

(e) Ginnie Mae proactively encourages smaller issuers starting to encounter financial problems to sell their servicing portfolio before financial problems worsen – thus eliminating or reducing Ginnie Mae financial exposure before it occurs. This is more difficult to do with larger issuers.”

Formal stress testing of smaller issuers would increase Ginnie Mae concentration and risk Use of formal stress testing to terminate smaller issuers could have the exact opposite result of its intended effect. An exodus of smaller lenders out of the program would inevitably increase Ginnie Mae issuer concentration and thereby increase its financial and systemic risk. As noted in CHLA’s January report: “Tightened supervision of small lenders disproportionate to their risk could even increase Ginnie Mae risk. To some extent, driving smaller issuers out of the program will result in fewer Ginnie Mae securities issued and fewer mortgage loans originated. However, to the extent such loans are made and securitized instead by larger issuers or correspondent lenders, the result is an increase in the concentration of Ginnie Mae securities among the largest Ginnie Mae issuers – precisely the types of issuers with higher level of risk.”

CHLA is not opposed to the general concept of stress testing CHLA is not opposed to individual issuers voluntarily using the concepts of stress testing. In fact, CHLA encourages every IMB to make every effort to identify, analyze, and develop contingency plans to be prepared for potential adverse developments such as (1) increased delinquency rates, (2) decreases in values of MSR holdings, (3) changes in interest rates, (4) reduced firm loan origination volume, (5) changes in the availability and terms of warehouse and MSR-backed loans, and (6) others.
Moreover, as noted, CHLA has supported stress testing for Ginnie Mae’s largest issuers (consistent with the broader financial services use of stress testing). And, CHLA is not opposed to Ginnie Mae using some form of general stress testing analysis as another tool in their overall statistical analysis of the financial risks they face.
However, for the myriad of reasons outlined in this letter, CHLA opposes Ginnie Mae use of stress testing for smaller issuers – and would be particularly concerned if it is used to determine or eliminate program eligibility or limit commitment authority for individual smaller issuers.

Any Use of Stress Testing on Smaller Issuers Should be Done Carefully
Any Use of Stress Testing Should be Transparent and Understandable
Program confidence and participant buy-in to any use of stress testing would be severely undermined if the calculations and results are not fully transparent and objective. This is also essential for stress testing to have value as a planning tool for issuers. RFI details of how stress testing would work are extremely complex and somewhat opaque. While more clarity will likely emerge as this initiative becomes more fully developed, CHLA urges Ginnie Mae to strive for an open, collaborative process and a transparent, objective result.
Use of Stress Testing Should Not Impose New Compliance Costs and Burdens on Small Issuers
Smaller IMBs already experience disproportionate compliance cost burdens relative to larger IMBs and to banks (e.g. small IMBs are subject to supervision by every state they do business in, as well as the CFPB, while 99% of banks are exempt from CFPB supervision and enforcement). These factors are already driving consolidation and industry concentration within the industry. It is not clear to what extent resources would have to be dedicated to a new Ginnie Mae financial stress test. However, Ginnie Mae should be mindful of this impact in imposing any new stress testing calculation or compliance requirements for smaller IMBs.
Any Use of Stress Testing Must be Fully Beta Tested Before Any use with Small Issuers
Any use of stress testing with respect to smaller issuers should be fully beta tested for its impact, including any potential unintended bias or impact on smaller issuers relative to larger ones. Ginnie Mae’s stress testing model was based on large issuers, and as noted does not appear to adequately reflect important qualitative differences between larger and smaller issuers. Implementation before this is fully analyzed and beta tested could result in a reduced number of smaller issuers, which in turn would be difficult to reverse.
An example of this is consideration of geographic diversification and potential metrics which might assume a higher risk of defaults due to lack of geographic diversification of an issuer. For the traditional use of stress testing – on large financial entities – this is appropriate. A large entity whose business is concentrated in an area that could experience a sudden downturn (e.g. the oil patch in the 1980s) – AND which is large enough to have a significant financial impact and systemic risk to the system – should be scrutinized for this factor.
In contrast, Ginnie Mae is a national program. Thus, the geographic concentration of one smaller issuer does not pose the same risk to Ginnie Mae. In fact, diversification of hundreds of smaller issuers, as now exists, serves to reduce Ginnie Mae’s risk. A reduction in the number of smaller issuers would diminish geographic diversity, as smaller issuers’ portfolios and business inevitably end up in the hands of mega-servicer/issuers. Proactive Communication with Small Issuer Warehouse Lenders

IMBs are subject to a wide range of regulators – including FHA, VA, RHS, Fannie Mae, and Freddie Mac to the extent they participate in those programs. What is arguably less understood is that IMBs are also subject to significant financial scrutiny (a kind of private sector regulation) by their warehouse and/or MSR lenders (who are in turn scrutinized by their banking regulators regarding such loans).

Warehouse (and MSR) lenders are important to the financial stability of IMBs and the withdrawal or reduction of credit lines could have negative consequences. Therefore, it is critical that Ginnie Mae proactively communicate their plans with the warehouse lending community regarding any stress testing impact on smaller lenders. Failure to do so could actually increase risk to Ginnie Mae. As noted in CHLA’s January report:

“Moreover, excessive Ginnie Mae focus on the risk of non-bank issuers could increase its risk by becoming a self-fulfilling prophecy. Ginnie Mae should of course be prudent in its supervision and transparent about the risks it perceives. But, if warehouse lenders become overly concerned that Ginnie Mae will make it hard for smaller issuers to remain in the program or will reduce the value of Ginnie Mae issuance authority, warehouse lenders could clamp down on or stop lending to smaller IMBs. This could increase Ginnie Mae’s risk in this sector of the market.”