Financial Services US Mortgage Finance what should the future look like? Authors James Wiener, Partner Michael Zeltkevic, Partner
Introduction The cataclysm of 2008-2011 exposed fundamental deficiencies in the US mortgage finance system. Economic growth and financial stability require a comprehensive redesign of all aspects of the financing of single-family home purchases. A new approach should be driven by a coherent set of policy objectives, including: Stability of the overall financial system Explicit government support for low to moderate income borrowers Privatization of risk taking outside of low to moderate income borrowers Enhanced protection for consumers The new architecture will also need to address the large size of mortgage outstandings relative to the aggregate banking balance-sheet, changes to the risk profile of mortgages housed in the regulated financial system, and challenges in the current model for mortgage servicing. A central element of any redesign will be a re-conception of the role of the governmentsponsored enterprises (GSEs 1 ) Fannie Mae and Freddie Mac. Currently, the overwhelming majority of US mortgage market originations are funded by Fannie Mae, Freddie Mac, and the FHA/VA. In February of last year, the Treasury laid out three options to significantly reduce the role of government in the mortgage market 2 by phasing out Fannie Mae and Freddie Mac. The options maintain that the government s role should be limited to oversight and consumer protection, targeted assistance for lower-income homeowners and renters, and support for market stability under severe stress. The mechanics of the redesigned role are where the three options differ. The Treasury s plan to phase out the GSEs is meant to encourage more private sector involvement by increasing guarantee pricing to reflect the same capital standards as private institutions, primarily banks. This would encourage GSEs to pursue additional credit-loss protection from private insurers, adjust conforming limits on loan size and minimum down payments, and wind down investment portfolios. In February, the FHFA (conservator of the GSEs) 3 had taken first steps in this direction through its three-pronged strategic plan: Build lay the framework for the secondary market; Contract reduce the dominance of the GSEs; Maintain continue the offering of foreclosure alternatives and fostering credit availability. In this paper, we examine the options for the future of the GSEs and make recommendations for the design of a new US mortgage finance system. These include: Design of a new risk intermediation structure to provide liquidity and efficiently transfer risk to the capital markets Explicit guidelines for risk taking within the regulated financial system 1 For simplicity, we will use the term GSEs to refer exclusively to Fannie Mae and Freddie Mac. 2 Department of the Treasury and Department of Housing and Urban Development, Reforming America s Housing Finance Market: A Report to Congress (Feb 2011). 3 Federal Housing and Finance Agency (FHFA), A Strategic Plan for Enterprise Conservatorships: The Next Chapter in a Story that Needs an Ending (Feb 2012). Copyright 2012 Oliver Wyman 1
Creation of a centralized registry for all mortgage liens with names of the servicer Changes to servicer compensation to an activity based model Harmonization of regulator and investor servicing guidance and modification programs Greater clarity on reps & warrants risk and servicing liabilities The future of the secondary mortgage market: What are the options? In the current model (Exhibit 1), the GSEs, by standardizing mortgage terms, have created a valuable source of capital markets based funding. In doing so, they transferred interest rate risk from borrowers to investors (one of the under-appreciated aspects of the current GSE model is the efficiency of pricing the 30-year prepayment option to the borrower). As Exhibit 2 illustrates, US mortgage assets are far too large to be funded on bank balance sheets via deposits. Accordingly, any proposal for the future of mortgage finance needs to maintain a vibrant role for capital markets funding. Exhibit 1: Overview of current conforming mortgage market model and role of GSEs PRIMARY MARKET SECONDARY MARKET BORROWER (Potentially via broker) A CORRESPONDENT ORIGINATOR Warehouse line of credit Floor of loans to be aggregated FANNIE OR FREDDIE Investment portfolio Market trading MORTGAGE BANK Originator B Agreement to buy loans at specific price Delivery of loan pools (servicing retained) Securitizer & guarantor Sales of MBS with credit guarantee (e.g. TBA market) C INVESTOR SERVICING Collections/ modifications Monthly payments Claims Primary servicer Contractual payments D Aggregation Serving standards Master servicer Scheduled P&I payments MORTGAGE INSURER (for high LTV) KEY VALUE-ADDS OF GSES TO MORTGAGE MARKET PARTICIPANTS A Borrowers: Mortgage products with desirable features (long term, protection from inflation, no prepay penalties) through efficient pricing of credit risk and management of prepayment risk B C Originators: Investors: Standardization of underwriting standards, pipeline hedging, massive driver of added funding Protection from credit loss, increased liquidity through TBA market and investment portfolio activities D Servicers/collections: Standardizing of protocols Copyright 2012 Oliver Wyman 2
Exhibit 2: US mortgage outstandings versus commercial bank balance sheet components, year-end 2011 4 $ TN 14 12 2.5 10 8 6 4 2 0 11.1 Residential mortgage debt outstanding 10.2 $1 TN Commercial bank deposits 1.3 2.9 7.3 Commercial bank assets Other assets Cash Securities Loans and leases (approx. 1/3 residential In the following subsections, we consider three potential options for restructuring mortgage securitization: 1. Private securitizers, with government playing reinsurer role 2. Industry consortium 3. Covered bonds Option 1: Private securitizers, with explicit government reinsurance This is effectively the third of the three options that has been laid out by the Treasury 5. A small set of large banks or private insurers could step into the void left by the GSEs and provide guarantees of timely principal and interest payments on MBS. These private guarantors would price credit risk and demand risk mitigation mechanisms (e.g. PMI) as they see fit, and the existence of multiple guarantors would foster competition. However, this structure involves significant wrong way risk. When mortgages go bad, there is a high chance that financial institutions will also be in trouble, regardless of how high their credit rating. Therefore, ensuring today s levels of market liquidity would require a government backstop. The government s role would be that of a reinsurer of last resort. It would charge a fee for its role and it would be called upon only once considerable private guarantor capital had already been consumed. In the rare event of private insurer failure, the government should have some flexibility to adjust fees as appropriate to offset losses and maintain zero taxpayer loss over a reasonable time horizon (as the FDIC does). The government s main role during times of calm would be the oversight of private mortgage guarantors and underwriting standards and 4 Federal Reserve statistical release on Mortgage Debt Outstanding and FDIC Statistics on Depository Institutions (all insured commercial and savings banks). Residential mortgage includes both single-and multi-family loans. 5 Department of the Treasury and Department of Housing and Urban Development, Reforming America s Housing Finance Market: A Report to Congress (Feb 2011). Copyright 2012 Oliver Wyman 3
reasonable pricing of reinsurance. The last of these is important in order to prevent guarantors from using the system and taking excessive risks because reinsurance is underpriced. Implications of this option: Borrowing costs are likely to increase somewhat once the GSEs ability to leverage without paying a risk-premium is reduced. However, the approach should maintain a high level of investor demand for mortgages to help mitigate increases in borrowing costs and allow for the origination of mortgages which banks may not want to hold on balance sheets en masse (e.g. prepayable 30-year FRM). This approach would require strong regulation; without it, the following could result: Increased fragmentation and different rules by guarantor, which is detrimental to investor transparency. (There needs to be a de facto setter of underwriting and servicing standards) Bias towards larger institutions as smaller lenders and community banks could face unfavorable pricing from private guarantors due to lower volumes As noted by the Treasury, an added benefit is that the government reinsurer s broad presence in the market would allow it to more effectively provide additional support during times of severe market stress. We should also note that this option would not require massive changes in the way the industry operates. Much of today s existing operational and business mechanics could be preserved under this option. Option 2: Industry consortium The government could remove itself from the conforming space by mandating that a private consortium of banks and mortgage companies (and potentially even investors) pool assets to provide the credit guarantees on low-risk loans, similar to the Freddie Mac structure pre-1990s when it was owned by the FHLB system. Mortgage market participants sharing credit risk may drive prudent behavior, resulting in a guarantor entity that is safer than any individual institution from an investor s perspective. The detailed mechanics of this consortium must be structured to incentivize owners to maintain the guarantor function as a low-risk, efficient service provider rather than simply a provider of profit. To prevent riskier lenders from arbitraging the system, each lender s costs would likely be tied to the performance of their own loans. Over time, guarantee fees charged could be adjusted as needed to maintain an appropriate capital base. Advantages relative to Option 1: Better inclusion of smaller institutions Copyright 2012 Oliver Wyman 4
May help promote better risk-taking Greater consistency could be achieved across the industry as the consortium will be able to set industry-wide standards (with regulatory oversight) Implications of this option: As noted in the Treasury report, a downside of such an approach is that it may be difficult for the government to step in if needed as a final support mechanism in a severe crisis. Borrowing costs would go up in general, and possibly more than in Option 1 depending on whether the market buys into the solidity of the consortium entity. Again, the approach would help maintain a high level of investor demand to allow for mortgages, such as 30-year FRMs, which banks may not want to hold on their balance sheets. Option II would also not represent a dramatic departure from the operational infrastructure of the industry. Option 3: Covered bonds The covered bond structure for future securitizations could be combined in the two options presented above, but given that this concept has been raised as a possibility by a number of market observers since the crisis, we want to address it separately. The covered bond has been a popular financing tool in Europe for a long time but has not made much headway in the US. Covered bonds are corporate debt securities secured by a pool of assets (in this case mortgages) on the issuer s balance sheet. The investor has recourse to both the pool and the issuer. This provides the issuer lower cost funding than unsecured corporate debt (these bonds are usually rated triple-a) and provides investors with a way to make slightly higher yields relative to similarly-rated securities. The assets in the cover pool are subjected to monthly monitoring and, should one of the loans become nonperforming, the issuer is obliged to remove it and replace it with a performing loan. Usually there will be overcollateralization that is, more mortgage collateral than the bond s notional value as an added measure of protection for investors. Bond payments generally come from the issuer s cash flows and in the event of a default by the issuer, the cover pool of assets is segregated and used to pay principal and interest payments on the associated bond. This product has been noted as an interesting alternative structure to the current securitization model. It does not involve the sale or resale of loans, offers fixed terms desirable by investors, and incentivizes good underwriting. Of course, the success of the product for an issuer depends on the institution s ability to evaluate and price the credit risk of the pool. While the covered bond framework works in some markets, US issuances have been severely limited 6, mainly due to lack of clarity and agreement on how the bonds behave in the case 6 Two recent US issuances have occurred: one by Washington Mutual in 2006 and one by Bank of America in 2007. Copyright 2012 Oliver Wyman 5
of bankruptcy. In the event of a bank failure, assets pledged to a covered bond including overcollateralization cannot be used to make depositors whole, which presents a risk to the FDIC. Given the dearth of other alternative investments, several foreign mortgage players have taken advantage and issued dollar-denominated covered bonds, in particular to US institutional investors. The US Congress is working toward enacting legal rules to allow for covered bonds to be viable for US issuers 7. If legal details are resolved in the US to make covered bonds a viable alternative to securitization structures, the investor demand means it could work for certain segments, such as jumbos. However, there are considerations which will hinder its usefulness in the mainstream conventional space. Although lower in risk, covered bonds present a more complex risk for investors to understand and analyze. Residential Mortgage-Backed Security (RMBS) analysis was complicated enough, but for covered bonds hybrid corporate and mortgage risks need to be understood to assess potential losses. Given the focus on investor transparency, this could be an obstacle and it may hinder smaller investors from entering the market. With this option, the system-wide cost of credit risk will increase due to covered bonds being on banks balance sheets, unless leverage limits are changed, which will lead to higher borrowing costs. Given the sheer magnitude of the US mortgage market, moving any sizable portion to the balance sheet would effectively be contradicting the push against too big to fail institutions. Since the end of 2010, only 25% of the over $11 BN of US single and multifamily mortgage debt outstanding was on commercial and savings bank balance sheets 8. In addition, Option III would represent the greatest departure from today s operational state. Much of the existing capital markets and funding infrastructure would have to be fundamentally re-thought and re-built; this would be no small undertaking. Proposed Approach We believe that the Option I has the most promise for maintaining efficient access to capital markets-based funding and effective risk aggregation and intermediation. In this model, we expect the existing GSEs to continue to function but to rely on private sector capital and insurance and the government reinsurance on explicit rather than implicit terms. We believe that this model is the best way to leverage the advantages of the current system where standardization, risk transparency, and government ownership of the tail risk allows a highly liquid and efficient market for mortgage securities. 7 See House of Representatives bill H. R. 940 (March 8, 2011). 8 Federal Reserve Board statistical release: Mortgage Debt Outstanding (release date: March 2011). Copyright 2012 Oliver Wyman 6
Managing risk in the system: Guidelines for Regulated Financial Institutions Addressing the secondary market alone will be insufficient. The origins of the financial crisis lay in the collapse of the US single family home prices and the associated mortgage and real estate derivatives defaults. One empirical fact from the crisis is that different segments of mortgage credit experienced widely varying performance. At a high level, the most predictive factor was mark-to-market LTV (MLTV), the origination LTV adjusted for current changes in the home price since origination. All other factors being equal, mortgages with MLTV < 90 experienced a low level of credit losses through the crisis. Mortgages whose MLTV increased to between 95 and 100 experienced rapidly escalating losses and mortgages with MLTV in excess of 100 experienced inordinately high credit losses. A similar pattern can be observed by looking at credit performance by consumer behavior score. In addition, other aspects of risk layering such as negative amortization products, low or no documentation of income and assets, and payment options dramatically worsened credit performance. Our recommendation is that guidelines be established for the risk profile of single family first mortgage loans originated and/or held within the regulated financial system. Our recommended starting point would be: Full documentation of income and assets Origination LTV < 90 FICO > 680 DTI < 40 Loans outside of these risk parameters should be originated either via FHA programs to support low to moderate income borrowers or by non-bank financials to finance subprime borrowers. Overhaul of servicing: Changing the structure and aligning incentives A striking aspect of the crisis is that it is repeating itself the first time as tragedy and the second time as servicing. The immense volume of seriously delinquent 1st mortgages (through the crisis over 5 MM loans have been seriously delinquent) that require time and resource intensive actions has overwhelmed the system. Added to the sheer volume are a long list of new mortgage foreclosure alternatives that need be implemented and a high level of legal and regulatory scrutiny with respect to the integrity of the foreclosure process. The servicing crisis has highlighted a number of issues: Legal uncertainty if using the Mortgage Electronic Registration System (MERS) to stand in the place of debenture holders in foreclosure proceedings Quality control around all aspects of the foreclosure process Copyright 2012 Oliver Wyman 7
Capacity and flexibility to implement new modification programs and foreclosure alternatives Lack of clarity around allocation of responsibility in rep and warranty agreements Severe misalignment of servicer, investor, regulatory, and policy maker incentives in the default servicing process These issues are leading to negative consequences for the housing recovery. Uncertainty with respect to these large potential costs and liabilities has led originators to exceptionally conservative underwriting standards (most tellingly tighter standards for GSE products than GSE requirements where they bear no credit risk but retain the servicing) and extreme caution in implementing new public policy initiative where they might waive their rights or incur new liabilities with respect to these issues. At a more strategic level, most large mortgage players are planning to downsize their commitment to the business. A contributing factor to these issues has been the structure of servicer compensation. Servicers are compensated by receiving an interest only strip of cashflows equal to 25 bps of unpaid balance (UPB). Servicers have service level agreements with investors but the compensation structure still incents them to limit investment in infrastructure. In addition, the servicing strip is highly volatile and must be capitalized on their balance sheet. It creates significant risk management challenges for the servicer. We believe that a number of changes to the structure and compensation of servicing are necessary to restore the economic viability of mortgage businesses. Creation of a central repository of mortgage lien and title information: This will facilitate greater transparency and efficiency throughout the default servicing process and potentially facilitate the replacement of MERS in the foreclosure process with the actual debenture holder. It will also allow first lien holders to force restructuring of second liens (when they exist) before impairing first liens Streamlining foreclosure alternatives: Agreement on a small standard set of modifications and foreclosure alternatives along with government incentive payments to allow servicers to optimize the efficiency and quality of these programs Aligning incentives: Change the structure of servicing compensation to 12.5bps of UPB and a series of activity based payments. This will greatly reduce the financial risk in the servicing asset, keep investor and servicer incentives aligned, and create incentives to invest in complex and resource intensive default servicing activities Clarifying reps and warrants: Standardizing representations and warranty contracts to clarify legal treatment and ensure comparability of loans with the same/ similar characteristics. This would significantly reduce the occurrence of loan buyback disputes (as are being observed in this crisis) Restoring the health of the US mortgage market will require changes across the board from origination to securitization to servicing over the life of the loan. Both the public and private sector will need to be involved in moving the agenda forward. Copyright 2012 Oliver Wyman 8
Oliver Wyman is a global leader in management consulting that combines deep industry knowledge with specialized expertise in strategy, operations, risk management, organizational transformation, and leadership development. For more information please contact the marketing department by email at info-fs@oliverwyman.com or by phone at one of the following locations: AmericaS EMEA Asia Pacific +1 212 541 8100 +44 20 7333 8333 +65 6510 9700 About the Authors James Wiener is a Partner and Head of the Americas Public Policy Practice Michael Zeltkevic is a Partner and Head of the Americas Retail & Business Banking Practice www.oliverwyman.com Copyright 2012 Oliver Wyman All rights reserved. This report may not be reproduced or redistributed, in whole or in part, without the written permission of Oliver Wyman and Oliver Wyman accepts no liability whatsoever for the actions of third parties in this respect. The information and opinions in this report were prepared by Oliver Wyman. This report is not investment advice and should not be relied on for such advice or as a substitute for consultation with professional accountants, tax, legal or financial advisors. Oliver Wyman has made every effort to use reliable, up-to-date and comprehensive information and analysis, but all information is provided without warranty of any kind, express or implied. Oliver Wyman disclaims any responsibility to update the information or conclusions in this report. Oliver Wyman accepts no liability for any loss arising from any action taken or refrained from as a result of information contained in this report or any reports or sources of information referred to herein, or for any consequential, special or similar damages even if advised of the possibility of such damages. The report is not an offer to buy or sell securities or a solicitation of an offer to buy or sell securities. This report may not be sold without the written consent of Oliver Wyman.