With the price of the bailout rising and momentum for reform in Washington slowing, Fannie Mae and Freddie Mac remain vital to the functioning of the mortgage and housing markets. A framework for a new system is beginning to take shape, at the same time that analysts are looking to fully and adequately address flaws in the current system.
Mortgage Banking
May 2010
By Robert Stowe England
A giant question mark hangs over the mortgage finance industry. What is the future of Fannie Mae and Freddie Mac? That is part of a larger question. What is the future of mortgage finance?
The two pillars of the mortgage industry that own or guarantee payment on $5 trillion in U.S. mortgages and/or mortgage-backed securities (MBS) failed dramatically at the height of the Panic of 2008 and were placed into conservatorship by the Federal Housing Finance Agency (FHFA).
The significance of the failure of the two government-sponsored enterprises (GSEs) has been difficult to fully capture; but however you describe it, it is epic in scope. The Congressional Budget Office (CBO) has projected that the bailout of Fannie and Freddie added $291 billion to federal outlays in fiscal year 2009 (October 2008-September 2009). CBO predicts another $99 billion in outlays from 2010 to 2019, for a total loss of $380 billion--by far the largest federal rescue ever.
It’s also one for the history books. “It was the biggest failure of housing policy on the planet and throughout all of history,” says Alex Pollock, resident fellow at the American Enterprise Institute (AEI), Washington, D.C. The GSE failures dwarf the next largest spectacular failure, the savings-and-loan (S&L) crisis of the early 1990s, where only $1 trillion in assets were on the line, he says.
As a ward of the federal government, Fannie and Freddie were placed on financial life support in September 2008 by the U.S. Treasury, with a pledge by Treasury to provide up to $400 billion in funding ($200 billion for each) via the purchase of preferred equity shares in the two GSEs. The Treasury pledged to provide equity funding to boost investor confidence in GSE mortgage-backed securities and corporate bonds.
Without this ongoing commitment from Treasury, Fannie and Freddie would not have been able to continue their role in supporting mortgage originations at a time when mortgage lending outside the arena of government guarantees had virtually collapsed. Without an adequate supply of mortgage finance from Fannie and Freddie, the housing market’s decline would likely have been far deeper, with even more dire consequences for the overall economy.
On last Christmas Eve, Treasury announced that it was removing the $400 billion cap and, if necessary, would provide the GSEs an unlimited amount of funding support over the next three years to continue to maintain the confidence of investors in agency securities and debt and to ensure a functioning U.S. mortgage market.
The decision to place Fannie and Freddie into conservatorship--and not into receivership, which was another of the options available--keeps open the option that the two GSEs could continue in some form or another in the future.
While Fannie Mae and Freddie Mac represented 40 percent of all mortgage originations in 2006, by 2009 they represented 69.9 percent of the market, according to Inside Mortgage Finance. With the Federal Housing Administration (FHA) at 20.7 percent and the Department of Veterans Affairs (VA) at 4 percent, the federal government backed about 94.6 percent of all mortgage origination activity in 2009, says (ital) Inside Mortgage Finance. The remaining non-government part of the origination market was “portfolio lending--mostly jumbo mortgages,” says Guy Cecala, Inside Mortgage Finance's president and publisher.
The consensus for reform and reducing the role of government in mortgage finance is strong. Even though Fannie and Freddie were defended heatedly in the past, today it is widely agreed in Washington that the two GSEs got too deeply involved in risky lending, and made the housing bubble bigger and last longer. Despite this common understanding, the road to reform may be long and winding, as the fate of the housing sector and the economy depend on a strong mortgage market, which, for now, means Freddie, Fannie and FHA.
New York-based Standard & Poor’s (S&P) primary credit analyst Vandana Sharma and secondary credit analyst Daniel Teclaw noted in a report in January that the decision by the U.S. government to remove the $400 billion cap “was not unexpected, given the increasing role that they play in helping to stabilize the housing markets.” The analysts noted that the two GSEs have a combined $4.3 trillion mortgage portfolio with $289 billion in nonperforming assets.
“It’s hard for us to imagine how the $11 trillion U.S. residential mortgage market could attract enough capital to replace the GSEs’ historical [43 percent market] share” and, even less so, their current share, Sharma and Teclaw wrote in a Standard & Poor’s report.
Part of the reason the GSEs remain in demand, according to Sharma and Teclaw, is that the private-label mortgage-backed securities market remains moribund, and banks and thrifts are unwilling to hold new mortgages on their balance sheet.
Sharma and Teclaw, in fact, warn that the financial markets will continue to rely on the GSEs “to help cushion the pain of more delinquencies and a potential ‘double-dip’ in home prices.” S&P finds the potential for a double-dip in the fact that only 85.5 percent or 111.4 million of the nation’s 130.3 million housing units are occupied. That leaves 19 million unoccupied. About 6.6 million of the unoccupied units are being held off the market--“a likely sign that owners and institutions may not want to accept the current market price,” according to S&P.
In S&P’s pessimistic scenario for the economy, national home prices could this year fall to a level as low as 45 percent below the peak in the summer of 2006. The baseline scenario is for a 35 percent decline from peak. According to the S&P/Case-Shiller Index, from the peak in June/July 2006 through the trough in April 2009, the 10-City Composite was down 33.5 percent and the 20-City Composite was down 32.6 percent. With improvements in pricing since the trough last year, peak-to-date figures through January 2010 were down 30.2 percent (10-City Composite) and 29.6 percent (20-City Composite), respectively.
The need to hold on to Fannie and Freddie in their current form may continue for years, given the slow rebound in housing and the lack of a private mortgage market as well as the slow pace at which policymakers have moved toward recommendations for the future of the mortgage finance system.
Last year and this, Congress virtually ignored the future of Fannie and Freddie as it focused instead on financial regulatory reform. Further, while the Obama administration had indicated it planned to release in February its own proposal for the future of Fannie and Freddie and the mortgage finance system in the new budget, it was not there.
Indeed, Howard Glaser, principal of the Glaser Group, Washington, D.C., said on CNBC in late March that Washington is “kicking the can down the road,” and he expects Fannie and Freddie to continue for another five years to a decade. “We’re stuck there right now because there is no political interest in discussing this in an election year,” he said.
Debate on the future
Since the failure of the GSEs nearly two years ago, there have been many proposals put forward to address both their future and the future of mortgage finance. The key question in all these discussions is what role the government should play, with some advocating no role whatsoever. Virtually no one is advocating keeping the status quo indefinitely and, thus, any discussions about the future also involve the question of what to do with Fannie and Freddie in a transition to a new system.
A common theme in the debate about the future of the GSEs and/or their successors is that the public/private combination simply does not work. “You can either be private or part of the government, but you shouldn’t be allowed to try to be both,” says AEI’s Pollock.
The government and the private sector impose such different disciplines, Pollock says, that you can be either one or the other but not both. “I still believe that Fannie and Freddie’s business, to the extent there is a business there, should become a truly private activity,” he says. The government-related activities, on the other hand, should be part of the government and merged into the Department of Housing and Urban Development (HUD)/Ginnie Mae/FHA complex, he adds.
One of the sharpest rebukes of the GSE public/private hybrid structure has come from Lawrence Summers, director of the White House’s National Economic Council and President Obama’s top economic adviser. Two years ago he was scathing in his denunciation of the GSEs in a post on the blog Creative Capitalism, founded by Michael Kinsley and Conor Clarke after a January 2008 speech by Bill Gates at the World Economic Forum. Gates, founder and former chairman of Microsoft® Corporation, said many of the world’s biggest problems cannot be fixed by philanthropy, but instead require free-market capitalism--so-called creative capitalism. The speech and blog led to a book of the same name by Kinsley and Clarke, while the blog was discontinued. “What went wrong? The illusion that the companies were doing virtuous work made it nearly impossible to build a political case for serious regulation,” he wrote, in reference to efforts to explore reform when he was Treasury secretary for President Clinton.
Summers faulted the hybrid nature of the GSEs for their failure. “When there were social failures, the companies always blamed their need to perform for the shareholders,” he wrote. “Government budget discipline was not appropriate because it was always emphasized that they were ‘private companies.’ But market discipline was nearly nonexistent given the general perception--now validated--that their debt was government-backed,” he wrote. “Little wonder, with gains privatized and losses socialized, that the enterprises have gambled their way into financial catastrophe.”
A conflict between their public and private functions was a cause of the GSEs’ biggest blunder: a decision more than five years ago to take on more than $1 trillion in subprime and other risky loans and securities.
Nearly all of those bad assets were added between 2005 and 2007, according to an American Enterprise Institute brief titled The Last Trillion-Dollar Commitment: The Destruction of Fannie Mae and Freddie Mac, by Peter J. Wallison, fellow in financial policy studies at AEI, and Charles W. Calomiris, professor of financial institutions at Columbia Business School, New York. The authors estimate, for example, that an astonishing 40 percent of the mortgages Fannie Mae added to its single-family book of business from 2005 to 2007 were “junk” loans.
“The most plausible explanation for the sudden adoption of this disastrous course--disastrous for them and the U.S. financial markets--is their desire to continue to retain the support of Congress after their accounting scandals in 2003 and 2004 and the challenges to the business model that ensued,” wrote Wallison and Calomiris.
“Although the strategy worked--Congress did not adopt strong GSE-reform legislation until the Republicans demanded it as the price for Senate passage of a housing bill in July 2008--it led inevitably to the government takeover and enormous junk loan losses still to come,” the authors stated.
Even if policymakers could agree that Fannie and Freddie’s business and government sides be separated, the rising losses at the two entities would represent an issue that would need to be resolved before the business part of Fannie and Freddie could be spun off into a purely private-sector affair. The bad loans and bad securities could be put into a bad bank and gradually liquidated over time, as taxpayers absorbed those losses, which at minimum are going to be hundreds of billions of dollars, according to Wallison and Calomiris.
As of the end of 2009, the reported losses at Fannie Mae since 2007 had reached $137 billion, with Freddie Mac adding another $80 billion for the same period, for combined losses of $217 billion. Edward J. DeMarco, acting director of the FHFA, declined to make public FHFA’s own internal projections of what it expects to see in total accumulated losses or even when the GSEs can return to profitability. (See “Q&A with Edward DeMarco” in this issue of Mortgage Banking.
The private/public hybrid that has proved so dangerous to financial markets and a mounting burden to taxpayers was not the original business model set forth by Congress for either Fannie Mae or Freddie Mac. Fannie Mae was created by Congress in 1938 to buy FHA loans. (Pollock thinks that if and when the two GSEs are divided into public and private spheres, the public side of the GSEs could go back to that original 1938 model.)
In 1968, during the Johnson administration, Fannie Mae became a GSE partly to get its debt off the consolidated federal budget at a time of rising deficits in order to pay for expanded social programs and the war in Vietnam. Freddie Mac was created in 1970 in order to fund the mortgages of the Federal Home Loan Banking System. Freddie Mac was sold to private shareholders in 1989.
GSEs ‘contributed to the bubble’
Indeed, in his testimony before Congress on March 23, Treasury Secretary Timothy Geithner announced that Treasury and HUD are only in the initial stages of cobbling together a “comprehensive” mortgage finance reform proposal. Geithner said Treasury and HUD would seek input from all stakeholders in the mortgage finance system by submitting a list of questions for public comment. (See sidebar, “Geithner’s Questions” at the end of this article.)
The list of questions was being drawn up to solicit replies from mortgage market participants, academic experts, and consumer and community organizations, Geithner said.
The Treasury secretary also gave some hints about how the administration is thinking about potential reform and the transition period from the current bailed-out system to a new one.
There is, for example, a clear preference by the administration for continuing a role for the government in housing finance, although Geithner also just as clearly finds fault with the current hybrid public/private GSE model.
“To some degree the role the GSEs came to play was an extension of the original function of the FHA,” Geithner said in his March 23 testimony. “However, their ability to properly serve this function was undermined over time by the unhealthy combination of their pursuit of profits and their misuse of the perception of government support, which was condoned by a wide range of regulators and oversight bodies.”
Geither also finds much that is positive about the GSE experience. “For a long time, the GSEs supported a well-functioning, efficient mortgage market, and the existence of their underwriting standards acted as a guideline for responsible underwriting by lenders,” he said. “They played a central role in the development of securitizations of conventional mortgages. They established appropriate benchmarks for conforming loans and brought transparency and standardization to the housing finance system. Borrowers, lenders and investors benefited from the deep, liquid markets which were formed.”
Where did the GSEs go wrong? Geithner thinks that they were not content with earnings from the mortgage guarantee business and saw more potential for growth in earnings from their retained portfolio business. The GSEs had a lower cost of capital because of an implicit backing by the government and they were able to boost their profits even more by their very high leverage, which was allowed under the statutes governing the GSEs.
“While the activities of the GSEs in theory should have resulted in lower borrowing costs for homeowners, a significant amount of the subsidy was not passed on to homeowners, but instead benefited GSE shareholders, managers, mortgage originators and other stakeholders,” he testified.
The GSEs also went astray, Geithner testified, when they responded to the expansion of the private-label MBS market of increasingly risky and exotic products by purchasing private-label MBS for their retained portfolios to continue to grow their companies’ earnings. While the GSEs held very few private-label securities in 2000, by 2007 those securities represented 23 percent of the combined portfolios of the GSEs, he said.
Geithner told Congress that by pursuing shareholder interests so aggressively through the expansion of their retained portfolios, the GSEs were a “pro-cyclical source of capital for the housing market and contributed to the housing bubble.”
Geithner also testified that the GSEs lobbied for lax oversight and low capital requirements, and thwarted “several attempts to limit their scope and scale and risk profile.”
Powerful and passionate allies of the GSEs within Congress for many years were chief among those who opposed efforts to rein in Fannie Mae and Freddie Mac. House Financial Services Chairman Barney Frank (D- Massachusetts) was one of their strongest allies and most vocal in defending them, especially when it came to affordable housing, during his long career in Congress. “I do not want Fannie and Freddie to be just another bank,” the committee’s then-ranking minority member said in 2003. “I do not want the same kind of focus on safety and soundness that we have in the [Office of the Comptroller of the Currency] and the [Office of Thrift Supervision]. I want to roll the dice a little bit more in this situation toward subsidized housing.”
Nowadays, Frank seemingly has changed his tune. In January he said he was in favor of “abolishing Fannie and Freddie in their current form and coming up with a whole new system of housing finance.”
Affordable goals defended
Department of Housing and Urban Development Secretary Shaun Donovan vigorously defended affordable housing goals against the charge that they contributed to the failure of Fannie and Freddie in testimony before the House Financial Services Committee April 14.
“For Fannie Mae and Freddie Mac, the affordable housing goals expanded the reach of lower-priced mortgage credit to many families. Some have argued that these goals were a principal cause of Fannie and Freddie’s collapse and subsequent losses. But this argument is simply not supported by the facts,” Shaun testified.
“One of the primary factors driving GSE losses was the desire to recapture market share and increase profits. The housing boom saw a rapid rise in non-prime and alt-A originations and securitization outside of the GSEs,” Shaun testified. “To regain market share, and increase revenue, Fannie and Freddie made poor strategic decisions to take on greatly increased risk, notably in alt-A mortgages. Management made clear that increasing revenue was the motivation for purchasing non-prime and other alternative mortgage products that subsequently produced significant losses,” he said.
Donovan’s rosy assessment of the impact of the affordable goals on the GSEs is not shared by most if the analysts of the failure of Fannie and Freddie.
Risky business
One of the people who have done the most to pin down the extent of risky lending and investing that was done at the GSEs is Edward Pinto, the former chief risk officer for Fannie Mae and currently a real estate financial services consultant. In testimony before Congress in December 2008, he charged that the affordable-housing goals imposed on Fannie Mae (and Freddie) are the primary reason the two GSEs loaded up on $2.25 trillion of subprime and alternative-A loans--and classified most of them as prime loans.
“Fannie and Freddie went from being the watchdogs of credit standards and thoughtful innovators to the leaders in default-prone loans and poorly designed products,” Pinto said in prepared testimony in December 2008. Some of the new products included 97 percent and 100 percent loan-to-value (LTV) products. The GSEs had to know these products would turn out to be disastrous, Pinto contended, as Freddie Mac had published information in the late 1990s that its 95 percent LTV loans had about six times the default rates of 80 percent LTV loans--which the GSEs knew and had stated in the past.
Pinto excoriates the GSEs for their reckless lending. “They introduced mortgages which encouraged and extended the housing bubble, trapped millions of people in loans that were unsustainable, and destroyed the equity savings of tens of millions of Americas,” he testified.
The GSEs also ignored the fact that the default rates for borrowers at a given FICO® score were rising over the 1990s, according to Pinto. In 1992, “a mortgage borrower with a FICO of 620-659 was seven times more likely to experience a serious delinquency over the next two years than a borrower with a 720-759 FICO. By about 2004, the 620-659 borrower was now 12 times more likely and the default propensity of the 720-759 borrower was unchanged,” he testified.
The GSEs’ risky behavior in the 1990s was in stark contrast to their prudent behavior in the 1980s, according to Pinto. In the 1980s, he testified, Fannie Mae had worked with local banks engaged in Community Reinvestment Act (CRA) lending “to keep the banks on the hook for a substantial portion of the risk.” That program of risk-sharing with local S&Ls was, however, abandoned in early 1989, “because the requirement was slowing down the desired ramp-up of Fannie’s affordable-housing initiatives,” Pinto testified.
Pinto reports that James Johnson, at this point a politically connected consultant from Lehman Brothers, advised Fannie Mae that in order to protect its charter from political attack, it needed to ramp up affordable lending. Johnson later became Fannie’s CEO in early 1990 and then its chairman in 1991, according to Pinto.
“The new team at Fannie either forgot and/or ignored its recent brush with disaster in the early 1980s when foreclosures ballooned out of control,” Pinto testified. “It embarked on a massive affordable-housing effort (mandated and encouraged by its mission regulators--HUD) that eventually promoted subprime, ultra-high LTV and alt-A loans (many were NINJA loans--no income/no job or assets).”
HUD adopted its first set of affordable-housing goals in 1993, and Johnson reciprocated in 1994 when he announced a new goal of $1 trillion for Fannie’s “Opening the Doors to Affordable Housing” initiative, Pinto testified.
Over time Fannie and Freddie would announce more than $5 trillion in affordable housing initiatives, according to Pinto. “This unprecedented abandonment of underwriting principles, coupled with the fact that the GSEs were permitted to take on $5.6 trillion in credit risk and maintain portfolios of $1.5 trillion, has put America’s homeowners at risk,” Pinto stated. “Their high-risk activities were allowed to operate at a 75:1 leverage ratio--much higher than that of the recently bankrupted Lehman Brothers.”
Last year Pinto updated his estimate of the level of risky loans and mortgage-backed securities held by the GSEs based on new information, some of it from Securities and Exchange Commission (SEC) filings of Fannie and Freddie. When FHFA’s DeMarco was before the Senate Finance Committee last Oct. 8, he was asked by Sen. Robert Menendez (D-New Jersey) how much subprime was loaded onto the balance sheets of the GSEs.
DeMarco replied that while he knew it was a substantial amount, he would have to get back to the senator with a more specific number. However, Pinto released publicly his own estimate that the GSEs had added $1.73 trillion in subprime loans between 2002 and 2007. In response to a query from the blog Mind Over Market (http://mindovermarket.blogspot.com) about how much alt-A was on the balance sheets, Pinto said that an additional $1.44 trillion was added by the GSEs over the same period of time, for a total of $3.17 trillion for both subprime and alt-A.
“I think the problem of this half-public, half-private organization was the crux of the problem. It was made worse by the affordable-housing goals they’re ever ratcheting up until they reached 55 percent in 2007,” Pinto told Mortgage Banking.
Pinto says that the affordable-housing requirements need to be separated from Fannie and Freddie and put into a government program where they can be monitored and the subsidies can be transparent. If a private entity is given affordable-housing goals, he says, “you never know what their unintended consequences might be.”
Did the GSEs aggravate the market’s problems?
Pinto thinks the private sector would have performed better if there had been no GSEs in the mix. “The private sector would not be able to go like they did without Fannie and Freddie as a large part of the market,” he says.
“The private sector would have run into problems before Fannie and Freddie got to where they did” in terms of bad lending. “They would have been more subject to market forces” that would put a brake on risky lending. “There’s a point in which all booms get exhausted,” he says.
“To the extent you have mostly the private sector, as opposed to a government entity like Fannie and Freddie, you rely more on market rates and market conditions. There’s a point where various lenders can see that things are getting dicey and start to pull back.”
He notes that the private sector pulled back in the periods from 1983 to 1985 and again from 1989 to 1991. “We didn’t see it in the 1996 to 2006 and early 2007 period because Fannie and Freddie were able to plow through that entire period,” he says. Thus, the public/private hybrid was not just bad for the GSEs, but bad for the private sector, which would have functioned better had there been no GSEs, according to Pinto.
Pinto acknowledges that returning to a pure private-sector mortgage finance system would be difficult to do at this time. “Ultimately, however, we have to go back to a much more private-sector approach,” he says.
This could be through a private-sector securitization arrangement or covered bonds. The private sector could be nudged along by phasing back the loan limits for Fannie and Freddie, and thereby “create more space for the private sector to get a toehold and grow.”
Pinto estimates that Fannie and Freddie bring only about a quarter percentage point advantage to the conventional mortgage market over where mortgage rates might be without an implied or explicit government guarantee. “We’ve got to get away from the idea that the quarter percent [advantage] is the be-all and end-all,” Pinto says. “Fannie and Freddie have demonstrated that in order to get the quarter percent, you put yourself on the hook for hundreds of billions [of dollars] of losses.”
Subsidized credit
Arnold Kling, former senior economist in Freddie Mac’s financial research department from 1986 to 1994, and a member of the Financial Working Group at the Mercatus Center at George Mason University, Arlington, Virginia, questions the very basis of the GSEs, which is subsidized credit.
Kling also questions the extent to which a new mortgage finance system in the United States should rely on securitization, and sees securitization as partly to blame for the housing bubble and the mortgage market troubles.
“Mortgage securitization is not inherently efficient,” Kling testified before the House Committee on Oversight and Government Reform in December 2008. Instead, “it owes its growth to accounting and regulatory treatment,” he testified.
The anomalies that led to more securitization are even stronger with high-risk loans, which require even higher capital requirements for banks than safer mortgages, according to Kling. “However, it turns out that when the loan has been laundered by Wall Street, it can come back into the banking system in the form of an AA-rated security tranche,” Kling testified. Thus, banks could hold the mortgage-backed securities of the same risky loans they sold into the secondary market for far less capital. “Most of the true risk is still there, but that risk is now hidden from capital requirements,” Kling testified.
Kling said, “I think securitization always required some kind of government support. In the most recent period, private-label [mortgage-backed securities, while not guaranteed by the government], benefited from peculiar capital regulations,” he explains. The regulations enable banks to hold risky loans inside [private-label] securities with less capital than the low-risk loans they originated themselves,” he says. “I’m not sure securitization can survive in the wild without government subsidies.”
Kling does not advocate outlawing securitization--only making sure that there is not a public policy to keep securitization alive. “Then, if it survives in the marketplace, it really is more efficient,” he says.
Capacity and competition
In spite of Kling’s argument against securitization, few of the proposals gaining traction in Washington omit a role for the government or omit a role for securitization.
One interesting proposal that has emerged is to expand the use of covered bonds. Covered bonds--a popular form of mortgage finance in some nations in Europe--are debt instruments that provide funding to regulated banks or mortgage issuers, relieving them of the longer-term interest-rate risk. The bank, in return, retains the residential mortgage assets and related credit risk on its balance sheet.
So far, the idea of creating a covered bond market has not really caught on in Washington. But, given that it addresses many of the weaknesses associated with securitization, it may yet find support and become part of any long-term reform.
For now, the idea of abandoning securitization is simply off the table, according to S&P’s Sharma.
“It’s hard for us to imagine how the $11 trillion U.S. residential mortgage market could attract enough private capital to replace the GSEs’ historical share,” she says, which was 43 percent at the end of 2007. To do without securitization, you need a combined balance sheet of enormous proportions from all the participating lenders to hold all the loans that would not be securitized, she explains. “Even if private mortgages did come back, we don’t have a balance sheet large enough to take the whole thing,” Sharma says.
Then there is a question of whether or not banks will want to be in the business of holding mortgages on the balance sheet--a business that is “not very attractive,” according to Sharma. If you originate and retain 30-year mortgages, “they just sit on the balance sheet with 30 years of interest-rate risk,” she says.
If Congress were to approve covered bonds as a funding option for mortgages originated by depository institutions, it would be easier to create more capacity in lending and help reduce the amount of lending that would be securitized.
Sharma also doubts that the 30-year mortgage would exist without the GSEs. Homeowners can also enjoy the efficiencies the GSEs bring to mortgage funding. “How else could we have a mortgage rate slightly above Treasury rates?,” Sharma asks.
The pure mortgage banking model would also be in serious jeopardy without securitization. Those loans could still be sold to aggregators who could fund them with covered bonds, for example, but the absence of an ability to sell directly to the GSEs is likely to constrain overall supply, according to Sharma.
A proposed new framework
The Mortgage Bankers Association formed the Council on Ensuring Mortgage Liquidity in October 2008 with the mission to look beyond the current crisis to what a functioning secondary market should look like for the long term.
In August 2009, the council published a set of recommendations for a framework for government involvement in the single-family and multifamily secondary mortgage markets (see www.mortgagebankers.org/ceml).
The new system would have a two-tier guarantee arrangement for mortgage-backed securities--one for securities and one for loans that back the securities. The council proposes 1) a security-level, federal government-guaranteed (GG) “wrap” like that on a Ginnie Mae security; and 2) a private loan-level guarantee from privately owned, government-chartered mortgage credit guarantor entities (MCGEs).
The council recommendations rest on the understanding that the federal government’s role should be to promote liquidity for investor purchase of mortgage-backed securities, not to attempt to provide the capital for or absorb the risks.
The government would guarantee timely interest and principal payments to bondholders and the instruments would explicitly carry the full faith and credit of the U.S. government. The MCGEs would, in turn, rely on their own capital base, as well as credit-risk retention from originators, issuers and other secondary market entities such as mortgage insurers.
Under the proposed framework, “there would be separate credit-risk and interest-rate risk,” according Michael Berman, chairman-elect of the Mortgage Bankers Association, and president and chief executive officer of CWCapital, Needham, Massachusetts, who has also served as co-chair of the council. The credit risk of the underlying mortgages would be removed from the securities issuer, while the interest-rate risk would remain with the security investor, he explains.
The guarantees in the proposed system would be supported by federal insurance that will be funded by risk-based fees charged for the securities at issuance and on an ongoing basis. “Some have likened it to an FDIC [Federal Deposit Insurance Corporation] insurance fund,” explains Berman. “That fund would be there to protect taxpayers in the event of losses at an MCGE, if the capitalization of the MCGE is insufficient and it fails or goes under.”
There would be a new government regulator for the MCGEs similar to the role now played by the Federal Housing Finance Agency.
Further, the council recommends that the MCGEs be limited to core mortgage products and not be involved in risky loans products. Subprime and no-documentation loans will not be permitted, while 30-year fixed-rate loans at 80 percent LTV “clearly will be permitted,” Berman explains.
Three MSGEs
The council’s recommended new system would start by taking Fannie and Freddie and dividing them into a good bank for the new system and a bad bank to unwind old problem loans and securities. The good-bank part of Fannie and Freddie would form the basis of two new MCGEs, each of which would be capitalized by the private sector. A third MCGE would be set up as a co-op of banks. Thus, there would initially be three MCGEs, but there could be more, according to Berman.
Berman says that the council is now modeling what the capitalization of the banks would look like, as well as what the mortgage insurance premium would look like. The costs of the mortgage insurance would be passed through to borrowers.
Default expectations in the modeling for the new system are based on historical data, according to Berman. There are two separate models--one for the single-family market and one for the multifamily market.
Aside from the modeling project, the Council on Ensuring Mortgage Liquidity is also working out “a draft of a road map for how this system could be implemented,” says Berman. This includes indentifying what legislation is needed; how and when that could be accomplished; what it would take to do a good-bank/bad-bank transaction; and what it would take to set up a new regulator, the insurance fund and new government-guaranteed securities. The road map also looks at whether to recommend Ginnie Mae as the guarantor and FHFA as the regulator, Berman says.
Berman says the council is not suggesting the MCGEs would have the kind of affordable-housing goals that were imposed on Fannie and Freddie.
“We want to be careful that affordable-housing goals do not distort the secondary market,” he says. “What I mean by that is that any securities need to be properly priced. There needs to be proper transparency. So, we’re not distorting subsidies in the way these new securities would be brought to market. There would be proper risk-adjusted pricing for riskier loans.”
According to Berman, the council is well aware of the problems associated with the fact that Fannie and Freddie had to be bailed out because they are too big fail. That’s one reason the recommendations start with at least three MCGEs. It would be up to the regulator, Berman says, to determine how many MCGEs there should ultimately be.
In defining the core products for the MCGEs, the idea is to have basic conservative products that could withstand “the hundred-year flood,” Berman says. At the same time, the new system should not be set up in any way that would crowd out private-label mortgage-backed securities, he explains.
Berman says the council has talked with the Center for American Progress (CAP), Washington, D.C., about the design of a new system. CAP has its own draft proposal and “there is a lot of commonality with ours,” Berman says. The council has also met with up to 15 different trade groups about the structure of a new system.
“Not everyone agrees with everything we said,” Berman says, “But there was a pretty good consensus around a number of elements,” The council has also talked with mortgage market experts at the American Enterprise Institute and The Brookings Institution, Washington, D.C.
Will the new system provide lower interest rates? Berman expects it will lower them by a quarter percentage point (25 basis points), although that will be offset by the cost of the premiums, reducing the quarter point by as much 15 to 17 basis points for higher-risk loans and offset the quarter percentage point by as low as 6 to 7 basis points for lower-risk loans.
New York-based Credit Suisse’s fixed-income research division has also done research on how to model a reformed GSE system that has many similarities to the recommendations from the council.
Credit Suisse described its proposal in a report last October by mortgage strategists Qumber Hassan and Mahesh Swaminathan. Hassan says that he and Swaminathan propose “rebuilding the GSEs rather than reinventing the wheel.”
The primary reason for taking this approach is the preserve TBA (to be announced) liquidity that the Fannie and Freddie brands have for decades brought to the mortgage securities market, the authors state in the report. They argue that the technology and know-how of the GSEs are invaluable and need to be preserved, which would be more difficult if a new system were created and brought with it “potential problems associated with the transfer of knowledge and technology.”
“Although superficially similar to a recent Mortgage Bankers Association proposal in terms of a two-tier credit wrap, our idea differs in key respects that stem from our preference to building on the existing GSEs rather than dismantling and redistributing the entities,” the Credit Suisse report states.
Under the Credit Suisse proposal, the GSEs would be divided into good and bad banks. The bad bank would work through the existing credit and portfolio book of problem loans and securities. The good bank would operate “a well-capitalized, privately-held mortgage guarantee business with a full faith and credit government reinsurance wrap on the MBS,” the report states.
The Credit Suisse proposal would also improve the risk management of the GSEs and minimize operational involvement by the government. The authors of the proposal envision setting it up in a way that it would continue to operate in the event of a catastrophic credit loss.
The new GSE business would have equity capital of 0.85 percent, higher than the current 0.45 percent, with a reserve gradually raised to 1.70 percent over 10 to 20 years. The 0.85 percent risk-based capital cushion would absorb the first loss, with the government wrap being tapped in the event of catastrophic losses. The guarantee business would also take on the guarantee obligations of the good banks carved out of the existing GSE books. The 85 basis points capital standard “is twice the pre-crisis standard and multiples of realized losses in the 1990s,” according to Hassan.
Hassan and Swaminathan estimate the fair value of guarantee fees under this framework at 45 to 55 basis points, assuming catastrophic credit events occur between 20 and 50 years. That compares with a base fee of around 15 basis points, which existed before the current housing crisis, the report states.
“We considered alternative approaches as well,” says Hassan--including such things as privatization, covered bonds and a fully government-backed and -owned system, “and concluded that a two-tier guarantee and reform of existing GSEs may be the most practical approach.” The idea of adopting covered bonds, for example, “is an idea that could take a long time to gain critical mass,” he says. “We see it as a supplementary solution that could provide funding for only a small fraction of the mortgage market,” Hassan says.
A “back-to-basics model” that Credit Suisse analysts describe is “a three-part system with a clean split of mandates,” Hassan says. It involves the GSEs serving the traditional prime credit borrower for a conforming, conventional mortgage, as well as multifamily. FHA would continue to serve weak-credit and low-down-payment borrowers, and the private-label market would serve jumbo borrowers, Hassan explains.
A dozen GSEs
The idea of having many more GSEs, as long as they are private entities, has been advanced by Mark Calabria, director of financial regulation studies at the Cato Institute, Washington, D.C. “Whether purely public or purely private, having only two Fannie/Freddie-like institutions guarantees that these entities will be bailed out if they become insolvent,” Calabria testified to the House Financial Services Committee in March.
The “ultimate goal of any GSE reform should be to create a system where, in time of mortgage stress, a GSE can fail without cost to the taxpayer or significant disruption to the financial and mortgage markets,” he said.
“The only way to make failure a credible option is to have several [GSEs]. I would suggest breaking up Fannie/Freddie into about a dozen equal-sized entities,” Calabria testified. He would also limit the GSEs to the issuance of mortgage-backed securities without any guarantees and prohibit them from holding a portfolio or issuing unsecured debt, thereby reducing the impact on debt holders of any default.
Addressing the issue of regulatory arbitrage by banks, Calabria would require that bank regulators treat holdings of GSE debt the same as they treat non-governmental corporate debt, which would require banks that acquire them to hold more capital against the debt. He would also require that all mortgages purchased by a GSE have a minimum cash down payment of 10 percent and no piggyback loans would be allowed. GSEs should also be subject to the bankruptcy code, he advised. Further, he adds, GSEs should be chartered by the regulator and not by Congress.
Calabria also argues against the idea of subsidized credit pricing for mortgages. “It is worth remembering that most homeowners are taxpayers, so simply moving interest-rate risk from homeowners to taxpayers does not make homeowners as a group better off,” he says.
The taxpayer bears some of the interest-rate risk of the 30-year fixed-rate mortgage (FRM), allowing it to be artificially low through federal guarantees. “Were the taxpayer no longer bearing this risk, I believe financial institutions would still offer 30-year fixed-rate mortgages,” but at a higher interest rate, Calabria testified. He estimates that without any GSE or federal guarantee, 30-year fixed-rate mortgages (FRMs) would be 130 basis points higher than adjustable-rate mortgages (ARMs), or just modestly higher than the 100-basis-point historical spread.
Even as people remain aware of how difficult it is going to be to replace or even reform the GSE model, awareness of the dangers they pose remains part of the policy debate in Washington.
Thomas Stanton, a fellow at the Center for the Study of American Government at Johns Hopkins University, Washington, D.C., testified before Congress in December 2008 that the historical experience of Fannie and Freddie demonstrate the “shortcomings” of the GSE model as an organizational model. Stanton is the author of the 1991 book, A State of Risk: Will Government-Sponsored Enterprises Be the Next Financial Crisis?
“However sound the accountability structure may be when the organization begins, the incentive to satisfy private owners will lead a GSE to try to weaken safety-and-soundness oversight and lower capital standards,” he testified.
Stanton issued a warning in 2008 that still applies. “The enabling legislation for any surviving GSEs should contain a 10-year sunset provision so that policymakers can periodically revisit questions of their public benefits and public costs in the context of changing markets and public priorities,” he said.
Kenneth Posner, former head of New York-based Morgan Stanley’s financial services research group and author of the book Stalking the Black Swan, contends that the GSE model itself is unstable. “The underlying instability of business models is one of the causes” of a so-called black swan event--“the seemingly unpredictable but massively consequential surprise,” says Posner. That surprise for the GSEs was a decline in housing prices.
Posner recommends that the rights to guarantee mortgage-backed securities “be auctioned off to a small number of large and relatively healthy banks” to preserve the vital securitization market. At the same time, Fannie and Freddie should be wound down and their enormous liability, which he estimates at $2.3 trillion ($1.6 trillion at the GSEs plus $800 billion from the Federal Home Loan Bank system), should be removed from the federal government’s balance sheet.
Such reforms would be doubly beneficial because they would also reduce the leverage of the U.S. government. “Leverage is a source of extreme outcomes. Just ask the Greeks,” Posner says. “We need to avoid anything like that scenario for us. Irrespective of losses [from the bad loans], it would be a good thing” to get them off the government balance sheet, he says.
In summary, specific proposals are moving forward that retain both a government guarantee and a central role for securitization, and attempt to more clearly separate credit risk and interest-rate risk. The proposals also keep some of the credit risk with the originators of mortgages and continue to transfer the interest-rate risk to the future replacements for Fannie and Freddie, where it may continue to present problems, depending on how the future entities fund themselves and manage their duration risk.
Critics remain determined, based on the rising catalogue of failures they cite at the GSEs, that reforms need to address the full litany of concerns: inadequate capitalization and high leverage, inadequate regulatory oversight, potential taxpayer bailouts from too-big-to-fail institutions, susceptibility to intense political interference, and the potential that government subsidies will flow to market players and not mortgage holders.
Finally, any proposal that emerges from Congress needs not just the support of the “stakeholders” in mortgage finance, but the broad support of the public, too. Ultimately, both the fate of Fannie and Freddie, as well as reform of the mortgage finance market, will likely need to respond to the considerable public backlash against government over-reaching, rising deficits and debt, and wariness--if not weariness--about markets, companies and arrangements that involve government guarantees.
MB
* * * * * *
Geithner’s Questions
On April 14 the Department of Treasury and the Department of Housing and Urban Development released a promised list of questions they are posing to the public on how best to reform mortgage finance. “The questions have been designed to generate input from a wide variety of constituents, including market participants, industry groups, academic experts, and consumer and community organizations,” a Treasury press release stated.
In addition to seeking public comment to these questions, the Administration is holding a series of public forums across the country. Below are the questions, as released by Treasury, along with suggestions from Treasury on what respondents’ commentary might address.
1. How should federal housing finance objectives be prioritized in the context of the broader objectives of housing policy?
Commentary could address: policy for sustainable homeownership; rental policy; balancing rental and ownership; how to account for regional differences; and affordability goals.
2. What role should the federal government play in supporting a stable, well-functioning housing finance system and what risks, if any, should the federal government bear in meeting its housing finance objectives?
Commentary could address: level of government involvement and type of support provided; role of government agencies; role of private vs. public capital; role of any explicit government guarantees; role of direct subsidies and other fiscal support and mechanisms to convey such support; monitoring and management of risks including how to balance the retention and distribution of risk; incentives to encourage appropriate alignment of risk bearing in the private sector; mechanisms for dealing with episodes of market stress; and how to promote market discipline.
3. Should the government approach differ across different segments of the market, and if so, how?
Commentary could address: differentiation of approach based on mortgage size or other characteristics; rationale for integration or separation of functions related to the single-family and multi-family market; whether there should be an emphasis on supporting the production of subsidized multifamily housing; differentiation in mechanism to convey subsidies, if any.
4. How should the current organization of the housing finance system be improved?
Commentary could address: what aspects should be preserved, changed, eliminated or added; regulatory considerations; optimal general organizational design and market structure; capital market functions; sources of funding; mortgage origination, distribution and servicing; the role of the existing government-sponsored enterprises; and the challenges of transitioning from the current system to a desired future system.
5. How should the housing finance system support sound market practices?
Commentary could address underwriting standards; how best to balance risk and access; and extent to which housing finance systems that reference certain standards and mortgage products contribute to this objective.
6. What is the best way for the housing finance system to help ensure consumers are protected from unfair, abusive or deceptive practices?
Commentary could address: level of consumer protections and limitation; supervising agencies; specific restrictions; and role of consumer education
7. Do housing finance systems in other countries offer insights that can help inform US reform choices?
bio: Robert Stowe England is a freelance writer based in Arlington, Virginia. He can be reached at
Copyright © 2010 Mortgage Bankers Association of America. All Rights Reserved. Reprinted With Permission.
