Narrowing the Differences

On April 20 the Mortgage Bankers Association (MBA) published a 60-page white paper titled “GSE Reform: Creating a Sustainable, More Vibrant Secondary Mortgage Market,” that was “the product of more than a year’s worth of work by the MBA’s Task Force for a Future Secondary Market.” This paper follows a preview piece released by the MBA in January, “GSE Reform Principles and Guardrails,” which concluded by saying the “full paper, anticipated later this year…will also include a more detailed end-state reform recommendation as well as a transition plan.”

The members of the MBA task force apparently were unable to agree on the details of their end-state recommendation and its associated transition plan, or else chose not to specify them, because the current paper leaves virtually all of the important elements of both to be determined either by Congress in legislation or the regulator upon implementation. The paper instead expands considerably on the “principles and guardrails” the task force members believe should guide others in their decisions about how best to balance the “three major objectives” they set forth for reform: “protecting taxpayers, attracting capital to Guarantors, and ensuring consumers and borrowers have access to affordable housing.”

Taking the white paper for what it is, and with the caveat that the unspecified details of the MBA’s recommendations could change my opinion of them, I found much more to like than to dislike about it. I agree with the MBA in three important areas: its choice of a secondary market business model, its recommendation for the determination of credit guarantor capital standards, and its proposed approach to the use of credit risk transfer (CRT) mechanisms. I also am encouraged by its endorsement of the utility model—although we interpret and apply this concept differently—and I agree with its “three major objectives” for reform. My one serious disagreement is over the MBA’s contention that achieving those objectives requires legislation to replace Fannie and Freddie as they now exist. I believe that it does not.

Arguably the most significant positive aspect of the MBA’s April paper is its explicit recognition of the merits of the Fannie and Freddie equity-based model for credit guarantors, and its endorsement of that model. The biggest impediment to mortgage reform over the past eight years has been an insistence on attempting to replace Fannie and Freddie with untested mechanisms seemingly designed by people with no practical knowledge of the credit guaranty business. The Corker-Warner and Johnson-Crapo proposals epitomized this failing, with their inclusion of features like the creation of a Federal Mortgage Insurance Corporation (FMIC), whose sweeping powers would have replaced complex risk assessments and business judgments best made in real time at the transaction level with simple and inflexible rules from a remote and cumbersome bureaucracy; allowing the same FMIC to put government guarantees on private-label securities (at potentially ruinous cost to taxpayers), and forcing the separation of the insurance, securitization and loss mitigation functions whose interaction allows a credit guaranty operation to work efficiently. The MBA correctly rejects these and other risky and untested features in favor of the endorsement of a model that has been proven to work.

Second, unlike Corker-Warner and Johnson-Crapo, the MBA does not propose an arbitrary, fixed capital ratio for credit guarantors. Instead, “MBA’s proposal would give the regulator authority to set specific capital levels, both risk-based and overall leverage limits/ratios…Congress should have the regulator develop a stress loss capital standard rigorous enough that Guarantors meeting that standard could have withstood the Great Recession.” That is what I’ve advocated in “Fixing What Works,” “A Welcome Reset” and elsewhere, and in fact it’s the current law. The 2008 Housing and Economic Recovery Act (HERA) states in section 1110, “The Director [of FHFA] shall, by regulation, establish risk-based capital requirements for the enterprises to ensure that the enterprises operate in a safe and sound manner, maintaining sufficient capital and reserves to support the risks that arise in the operations and management of the enterprises.” FHFA has yet to implement this provision of HERA, and I agree with the MBA that it should. And while I do have some concern that the MBA might later urge FHFA to set “overall leverage limits/ratios” at the “bank-like” level of 4 percent, which would override and make moot the risk-based component, I won’t prejudge that; I’ll wait to see if it happens.

The third area of the MBA’s April paper that I believe it got right is the reversal of its recommendation in January that “(e)ach guarantor would be required to leverage private-capital investors to share credit risk above and beyond the loan-level credit enhancement provided by the primary market.” The MBA now says, “The capital base for the requirement should primarily be composed of Tier 1 capital, i.e., common and preferred equity, but should also provide capital relief to the Guarantors for distributing rather than retaining credit risk, so long as this is done on an economically sensible, equity equivalent basis [emphasis added].” To reinforce this point, the paper goes on to add, “Capital relief from CRT, either frontend or back-end, should be evaluated on an equity equivalent basis, i.e., the economic benefit of the transfer should be measured relative to another dollar of equity capital. Credit should be given only when risk-share capital is truly committed and targeted to cover losses ahead of the Guarantors.”

With this change, the MBA breaks decisively with the proposals made by the Urban Institute and the Milken Institute, which erroneously assert that CRT mechanisms or securities can be used instead of equity capital, rather than serve as a supplement to it when they are economic on an equity-equivalent basis. In my last post (“Risk Transfers in the Real World”) I noted that the Urban Institute’s Jim Parrott included the MBA’s proposal with his own and that of the Milken Institute as variations on the same theme, with each relying either extensively or exclusively on risk transfer mechanisms rather than upfront equity to absorb credit losses. The MBA has now removed itself from this group, leaving the Urban Institute and the Milken Institute to defend their theoretical and dangerous CRT-based proposals by themselves.

Where I believe the MBA errs badly, however, is in its contention that legislation is required to achieve its “three major objectives” of reform (which I support). The MBA states that legislation is needed to accomplish three of the initiatives it cites in its paper: to “Empower FHFA or a successor regulator to grant charters to…new Guarantors;” to “Create the Mortgage Insurance Fund (MIF) to guarantee eligible mortgage-backed securities,” and to “Establish a new, explicit government guarantee that stands behind the MIF.” That undoubtedly is true. But what also is true is that none of these initiatives are necessary, or even desirable, in order to accomplish the MBA’s reform objectives, obviating the need to legislate them.

I’ll start with the first element: chartering new credit guarantors. It is perplexing that the MBA would combine an embrace of “utility-like regulation” with the idea that the government should be able to grant charters to new entrants. The utility concept is based on a carefully designed balance of restrictions and benefits that gives private capital a financial incentive to provide an important public service. In my version of the utility model for mortgage credit guarantors, the main restrictions are a limit to a single line of business and a cap on the returns that can be earned, while the compensating benefits are Fannie and Freddie’s existing federal charter, a backstop arrangement from Treasury should the new government-imposed capital requirements turn out to be insufficient, and protection from having new entrants erode their economies of scale.

The MBA offers little justification for granting new credit guarantee charters other than “(a) credible threat of additional entrants would encourage dynamism and spur the [existing] Guarantors to provide better service to their seller/servicers and ultimately to consumers.” That would be a decent argument if there were only a single credit guarantor, but with two (Fannie and Freddie), there are sufficient market share incentives to lead to the innovation and service benefits to lenders and consumers the MBA seeks. Currently there is virtually no capital in the conventional secondary mortgage market (Treasury has taken it out), so our first priority must be to ensure we can attract sufficient capital for the $5 trillion in credit guarantees that already exist. Upsetting the balance of restrictions and benefits in the utility model by opening the door to new entrants in the future would make it much more difficult to attract the capital needed to back the credit guarantees we have today.

The other two initiatives in the MBA’s plan that would require legislation are linked. They are to “Replace the implied government guarantee of Fannie Mae and Freddie Mac with an explicit guarantee at the mortgage-backed security (MBS) level only,” and to create “a federal insurance fund with appropriately priced premiums [the Mortgage Insurance Fund, or MIF].” Yet here too, the key elements of these initiatives—creating an explicit government guarantee, guaranteeing MBS rather than companies, and creating a Mortgage Insurance Fund—are not good ideas.

Taking them in reverse order, the Mortgage Insurance Fund (MIF) is not insurance. True insurance works by charging a large number of people (or institutions) small dollar amounts of premium in order to be able to pay a small number of people (or institutions) large dollar amounts when they have losses. With two or at most a handful of credit guarantors—subject to the same risk-based capital requirement, utility-like return limits and regulation—their credit performance will differ only slightly. In a stress environment, therefore, all guarantors either will have the right amount of capital or they won’t. If we believe the dollars the MBA proposes to go into the MIF are necessary to meet our updated standard of taxpayer protection, that amount should be built into the credit guarantors’ capital requirements by FHFA. Separating it, funding it differently and calling it an MIF is wholly cosmetic.

The close correlation in the credit performance of a small number of guarantors with the same capital requirements, return limits and regulation also is the reason it is not a good idea for the government to guarantee only mortgage-backed securities and not the guarantors themselves. The government will set the stress standard for guarantors, so they should have only a remote chance of failure. But if one does fail, it is quite likely that both (or all) will fail, in which case a policy of guaranteeing only existing MBS but not guarantors would leave no secondary market mechanism that could finance the new mortgages required to keep the system from melting down.

Which leads me to the government guarantee itself. I believe every effort should be made to reform the secondary market in a manner that does not require an explicit guarantee from the government. Conservatives in Congress, and particularly in the House, will fiercely oppose adding a government guarantee to privately insured mortgages. And one is not essential. In “Fixing What Works” I proposed that “in an ‘exchange for consideration’ [for the imposition of utility-like return limits] the government would commit to provide temporary support to the companies should their capital ever prove insufficient (which by design would be highly unlikely).” Such a backstop would benefit homeowners through lower resulting MBS yields, while benefitting taxpayers by ensuring the continued survival of the companies even in a crisis. And the moral hazard of this arrangement would be minimized by stress-based capital standards, capped returns, and the fact that the companies would not benefit from repayable government loans unless they suffered losses extreme enough that their stock prices plummeted, top management lost their jobs, and the companies had to recapitalize themselves (and repay their loans) by raising large amounts of new equity at very low stock prices, badly diluting their existing shareholders.

Should a short-term backstop of this type prove to be politically unacceptable I still would not favor an explicit government guarantee on MBS (or credit guarantors). Binding risk-based capital standards, set to withstand a 25 percent nationwide decline in home prices and endorsed by the government, should be conservative enough for the MBS of a guarantor subject to this standard to find broad and deep acceptance among international capital market investors, at competitive yields.

Taken as a whole, the MBA’s April white paper significantly narrows the differences between its previous position on mortgage reform and the administrative reforms I advocate. The MBA now endorses the Fannie and Freddie equity-based model for secondary market guarantors (and rejects untested and unworkable alternatives like Corker-Warner, Johnson-Crapo and those of the Urban Institute and Milken Institute), agrees that credit guarantors need to be given risk-based capital requirements updated to current standards of taxpayer protection, and sees merit in the concept of utility-like regulation and returns for guarantors. And in the one critical area in which we differ—the need for legislation—strong arguments can be made against each of the MBA’s reasons for it: enabling new credit guarantors (inconsistency with utility model principles), authorizing a Mortgage Insurance Fund (it’s not insurance and should be included in required capital), guaranteeing MBS and not companies (it would expose the system to collapse) and putting an explicit government guarantee on mortgages (it’s politically controversial and there are better alternatives).

In my view, all of the MBA’s stated objectives for mortgage reform can be achieved administratively, without incurring any of the complex risks involved in a transition from the existing Fannie and Freddie to new (and essentially identical) companies created through legislation. The MBA acknowledges these transition challenges, but does so only in concept, offering principles it says should guide Congress, FHFA and Treasury in addressing them. That is not enough. Before any legislation is proposed, let alone passed, advocates for replacing Fannie and Freddie must show that they have identified and understand all of the related transition issues, have concrete and detailed solutions for them, and can demonstrate that the benefits of what they propose outweigh the risks.

The most daunting transition obstacles are ones the MBA does not mention: the complications created by Treasury and FHFA’s management of Fannie and Freddie in conservatorship, and the plethora of lawsuits filed as a consequence of that management. Shareholders of the existing companies met all of their regulatory requirements in September of 2008 yet still had conservatorship forced on them, and once they overcame the effects of inflated expenses put on their books by FHFA, Treasury and FHFA conspired to take all of their profits in perpetuity—an action the D.C. District Court of Appeals recently ruled was lawful. New investors will not supply the capital required to move $5 trillion in mortgage guarantees from Fannie and Freddie to their envisioned replacements (or to recapitalize the companies if they are brought out of conservatorship) until the uncertainties surrounding the rights of shareholders in the existing companies are definitively resolved.

The lawsuits and the issues attendant to them that would be a serious impediment to a legislative reform process can be addressed and dealt with in an administrative one. Indeed, that is probably the only way to resolve them. The MBA has gotten almost to the right place in its April white paper; it could get all the way there by dropping its insistence on unnecessary and unlikely legislation, and embracing an administrative reform process that can succeed in delivering on the objectives it says it supports.