Even as the world contends with the most recent political-economic upheavals, from the aftermath of Brexit to mass migration, unfinished work remains from the 2008 global financial crisis. If left undone, the same stresses that led to the last U.S. housing crisis will continue to dog the financial system.

Eight years have passed since Congress put Fannie Mae and Freddie Mac under conservatorship controlled by the Federal Housing Financial Agency (FHFA), meant to be a temporary salve to calm the market in the immediate aftermath of the financial crisis. In 2014, legislative efforts for long-term housing finance reform gained momentum under the bi-partisan Johnson-Crapo bill, but the bill never came to a floor vote.

Now, new thinking on how to restructure the $4.5 trillion secondary mortgage market marks a growing consensus and renewed opportunity for a solution, according to a research symposium sponsored by the Penn Wharton Public Policy Initiative and the Penn Institute for Urban Research and held in June at the U.S. Capitol in Washington, D.C. The symposium examined two proposals — one from Andrew Davidson, founder and president of Andrew Davidson & Co. Inc., and one from housing finance and policy experts Jim Parrott, Lewis Ranieri, Gene Sperling, Mark Zandi and Barry Zigas.

By addressing issues that stalled legislative action in the last go-round, the two proposals move the debate forward, according to Susan Wachter, Wharton professor of Real Estate and organizer of the symposium, in a Penn Wharton Public Policy Issue Brief co-authored with Patricia McCoy, professor at the Boston College Law School. In particular, while the Johnson-Crapo bill proposed dividing Fannie and Freddie’s functions among many players to create a competitive industry, the new proposals both advocate centralizing the system. The reason: Both plans posit that large-scale efficiencies in mortgage securitization make it a natural monopoly, and any competitive market may just naturally lead again to a duopoly or oligopoly that’s too big to fail. (At present, Fannie Mae and Freddie Mac, along with Ginnie Mae, another agency owned by the U.S., either own or insure about 60% of all mortgages outstanding in the U.S.)

But there are ways to structure the new monopoly or duopoly in ways that maintain efficiency and competition in the housing finance market, while safeguarding taxpayer liabilities, according to the plans’ authors. In particular, the proposals would consolidate the purchasing, pooling, master servicing and risk control functions, still enabling “competition where it should occur” in mortgage origination, innovation and customer service, write Wachter and McCoy. The centralization of those securitization infrastructure functions would allow lenders of all sizes to compete and the secondary market to access a variety of private capital sources, including the capital markets, reinsurers, mortgage insurers, lenders and others. A primary difference between the two proposals is the structure chosen to house the new centralized functions.

A New Government Corporation

Representing his co-authors at the Wharton symposium, Mark Zandi, chief economist of Moody’s Analytics, recommended that government-sponsored enterprises (GSEs) Fannie Mae and Freddie Mac be merged into a single government corporation, a closely regulated monopoly tentatively named the National Mortgage Reinsurance Corporation (NMRC). By separating out mortgage securitization as a government function and mortgage origination as the province of private lenders, the system helps avoid perverse incentives by the mortgage securitization entity to make excessively risky loans to feed the creation of securities. In addition, lenders of all sizes will have equal access to securitize their loans at the NMRC, and competition between lenders will help keep mortgage rates down, increase access to mortgages and promote innovation in products and services, noted Zandi.

Both plans posit that large-scale efficiencies in mortgage securitization make it a natural monopoly, and any competitive market may just naturally lead again to a duopoly or oligopoly that’s too big to fail.

The NMRC would take on Fannie and Freddie’s core functions of purchasing and pooling home loans, issuing securities backed by the loans, providing master servicing of the loans, and furthering affordable housing goals and duty-to-serve requirements, funded by a 10 basis points fee included in the NMRC’s guarantee fee. The entity would be required to transfer non-catastrophic risk to the private sector. In addition, the NMRC’s retained portfolio of mortgage-backed securities would shrink significantly, and the new government corporation would be prohibited from using it for investment purposes.

To help inject market discipline into the government corporation and protect taxpayers, the NMRC would raise capital from the private capital markets. The NMRC would be backed by 8.5% capital altogether — with holders of common equity raised from the markets covering the first 3.5% in losses and holders of fixed-dividend securities raised from the markets covering the next 2.5% of losses. If losses exceed 6%, a mortgage insurance fund (MIF), akin to federal bank deposit insurance and funded by the 10 basis point fee referenced above, would cover the next 2.5% of losses. The 6% level is double the realized GSE losses during housing crisis but under the 10% required under the Johnson-Crapo bill, noted Wharton’s Wachter.

Zandi highlighted five benefits of the NMRC. First, it solves too-big-to-fail by helping to eliminate incentives towards making riskier loans to feed the securitization engine. Second, it provides access to all lenders, big and small, including those lending in underserved communities. Third, its capital structure helps protect taxpayers while maintaining mortgage rates at current levels, rather than raising them. Fourth, the new system promotes competition by giving every lender access to the system on the same terms, creating “pure competition in the primary market, key to creating innovation in the mortgage market and to addressing demographic changes,” said Zandi. Last, transition costs are minimal, leveraging off a transition already under way, as the FHFA has been requiring the GSEs to transfer more credit risk to the private market since 2013.

A Co-op

In contrast, Davidson proposes putting the functions of mortgage securitization, “central to the government-sponsored enterprises’ mission of creating liquidity in the secondary market,” into new entities that are mutually owned by mortgage originators. Similar to Zandi’s proposal, these functions would include the buying and pooling of loans, the issuance of mortgage-based securities, and master servicing of the loans. (Unnecessary functions include a retained portfolio of mortgage-backed securities, which would be significantly cut or eliminated.) In Davidson’s plan, profits paid out to the mutual owners in the form of dividends ideally would be recycled into their mortgage businesses. Each mutual would also have the freedom to fulfill its duty to the secondary market through means of its own choosing.

Under the plan, Fannie Mae and Freddie Mac each would have the opportunity to convert into a mutual, regulated by the FHFA. Down the road, if the two new mutual companies find they cannot coexist, they could merge. But that more radical transformation isn’t necessary in the near term to move Fannie and Freddie out of conservatorship.

To protect taxpayers and enable the mutual companies to continue operating through housing crises, Davidson recommends that up to 75% of risk on new GSE loans be “reinsured” through credit risk sharing programs with mortgage insurers, reinsurers, REIT’s, banks, loan originators and others. The mutual companies’ exposure to losses would be subject to a cap. In addition, the entities’ mortgage-backed securities then receive an explicit government guarantee, akin to that of Ginnie Mae securities, paid for by a fee to the government that helps fund affordable housing and other housing policy goals. Where private capital is not enough, Treasury can assume part of the credit risk.

Davidson maintains that unwinding the GSEs altogether is unnecessary, but the challenge is to find the governance and capital structure to maintain stability in the housing finance market. “Mutual ownership of GSEs by mortgage originators offers the opportunity to align incentives and protect taxpayers from future losses, while limiting the incentives to siphon off the benefits of government guarantees to private shareholders,” he said. In his proposal, “the GSE is a mutual owned by the originators and doesn’t create another group of investors whose goal is to extract money from the housing system…,” he explained. “You don’t want monopoly profits pulled out of the market.”

The proposal has two benefits: “It’s innovative but not too innovative,” he said. “You don’t want to disrupt the competitive equilibrium of the members.” In addition, he noted, in a co-op, “there’s always someone who wants to tell on people,” and that can be an additional protection against risk.

Different Views

Reflecting on the two proposals from Zandi and Davidson at the Wharton symposium, panel participant Peter Wallison, chair of Financial Policy Studies at the American Enterprise Institute in Washington, D.C., took issue with the premise of the Wharton symposium: “There is not a consensus in the world at large. There’s a group that doesn’t agree; I think they’re called Republicans, and they have to be brought along.” Instead, Wallison suggested that the private market take over all the risk in the housing finance market from the government, starting with a requirement for the private market to provide first loss coverage at 10% to 15% and doubling the amount at, say, two or four year intervals. “It would take 10 years to phase out full faith and credit of the government,” said Wallison. “But you would end up with a private secondary market with no government guarantee. The financial crisis was caused by government involvement in the first place.”

“Mutual ownership of GSEs by mortgage originators offers the opportunity to align incentives and protect taxpayers from future losses, while limiting the incentives to siphon off the benefits of government guarantees to private shareholders.” –Andrew Davidson

Fellow panelist Michael Berman, principal at Michael Berman Consulting and former senior advisor at the Department of Housing and Urban Development in the Obama Administration, observed that neither plans addressed the multi-family housing market. As a governance structure, he posited a utility model, an entity owned by the public and where returns and operations are closely regulated.

Finally, panelist Joshua Rosner, managing partner at Graham Fisher & Co. in New York and author of the 2011 book Reckless Engagement on the 2008 housing finance crisis, questioned scrapping the entire model when the “full faith and credit [of the government] still works in many ways.” He, too, favors the utility model in an industry that tends towards natural monopoly and one that performs a social function with high infrastructure costs. “A utility takes it out of politics, and you have a public utility commission” to oversee the entity, he said.

In addition, he noted, “affordable housing is overly political because of the way we deliver it — ‘x’ basis points for every origination goes to affordable housing. That creates an incentive to move up risks to do more volume and more affordable housing, or vice versa. We can narrow the debate to how big affordable housing should be, so if you put that in a utility, you can do that. The fundamental duty to serve is built into a public utility, and the rate of return [which goes to affordable housing] is capped.”

Moving Ahead

While the two plans proposed by Zandi and Davidson represent many points of consensus, concluded Wachter, they overlook some remaining unresolved issues. Among them is the pro-cyclicality of the proposals, where private investors in the proposed new entities would demand higher returns for their risks in times of economic stress, driving up guarantee fees and mortgage rates. Still, these remaining issues can be resolved, she noted.

More important, said Wachter in the Wharton Issue Brief: “With each passing year, the ‘temporary’ conservatorship of the GSEs becomes more and more untenable,” and the current consensus on the centralization of core GSE functions is a promising starting point for a more long-term solution that paves the way for a stable housing finance system in the U.S.