In one of the fateful moments in the financial crisis, Fannie Mae and Freddie Mac were placed in conservatorship on September 6, 2008, one week before the earth-shattering collapse of Lehman. The panic that followed the Lehman bankruptcy overshadowed the crisis facing the two mortgage giants, but the idea that these two companies could both face insolvency would have been ridiculed in the years prior to the crisis. After all, housing was not a speculative investment, and two giant companies focused almost exclusively on buying home mortgages and issuing mortgage-backed securities would seem to have little risk—or, at least, that was the general belief prior to the collapse of the housing bubble.
It has now been almost a decade since these quasi-governmental corporations, known as government-sponsored enterprises (GSEs), were placed in conservatorship. There are few people in Washington who like the current arrangement, but it is nonetheless likely to persist for at least several more years. And with each passing year, the power of inertia becomes ever greater, as the companies’ profits help to reduce the federal budget deficit.
This article examines the merits of the various alternative paths being considered for the GSEs, as well as the political obstacles they face. Before examining the current prospects, however, it is worth briefly recapping some history.
The Evolution of Fannie Mae and Freddie Mac
Fannie Mae was the first of the two mortgage giants, established as the Federal National Mortgage Association in 1938 when the economy was emerging from the Great Depression. It was originally established as a government-owned corporation with the purpose of creating a secondary market in mortgages. It exclusively purchased mortgages that were already insured by the Federal Housing Authority (FHA), which meant that it faced little risk itself, since the FHA would cover almost all of the loss if a mortgage defaulted.
Fannie Mae is given much of the credit for the homeownership boom following World War II. Prior to the creation of Fannie Mae and FHA, long-term mortgages were rare. Most homes were sold on short-term contracts, usually issued by the seller, which had to be paid in full after three to five years. There also was not an effective national market for housing credit, which meant there could be large differences in interest rates between regions. Since Fannie Mae offered banks and savings and loans the opportunity to sell mortgages, they didn’t have to worry about their capital being tied up for long periods of time. Issuing thirty-year mortgages and selling them to Fannie Mae came to be a very profitable business. And with a generally strong economy and low unemployment, Fannie Mae could count on a modest profit, which was revenue to the government.
This model was changed in 1968 when Fannie Mae was made into a private company that was allowed to buy mortgages not insured by the FHA. The main rationale for privatizing Fannie Mae was not any inherent problem with the management of the company by the government. Rather, the primary motive was the desire of the Johnson administration to get Fannie Mae’s debt—the bonds issued to buy mortgages—off the government’s books, thereby reducing the size of the national debt. It is worth noting the absurdity of this concern: since the debt was fully offset by assets, the mortgages purchased had equal or greater value than the debt that Fannie Mae issued to buy them. Nonetheless, if the concern is exclusively debt, there is no doubt that the debt of a government-owned company is government debt.
Two years later, in 1970, the Federal Home Loan Mortgage Corporation, or Freddie Mac, was created by Congress. The intention was to provide some competition in the secondary mortgage market for Fannie Mae. Freddie Mac was originally owned by the eleven Federal Home Loan Banks, which were in turn owned by the banks within each of their districts, but in 1988 it was set up as a public company listed on the New York Stock Exchange.
In the years after Fannie Mae was privatized and Freddie Mac was created, it was understood that they would be run for profit, but also that the GSEs maintained a special responsibility for supporting home ownership. Both companies were subject to government supervision, not only to ensure their soundness but also to ensure that they were complying with their public mission of promoting homeownership. It is worth noting that securitization of mortgages did not begin until Freddie Mac began doing it in 1971, shortly after the company was created. Prior to that point, Fannie Mae had held its mortgages on its books, and it continued to hold most of its mortgages until it began to follow Freddie Mac’s lead in the 1980s, when it also started issuing mortgage-backed securities (MBS).
The two GSEs largely had the secondary mortgage market to themselves (though the Veterans Administration and Government National Mortgage Association, Ginnie Mae, also purchased mortgages) until the late 1990s when the investment banks started to get into the business. They were both very profitable companies that operated largely like other companies in the financial industry. They paid out dividends to shareholders and their top executives earned the same sort of salaries as top executives at large Wall Street banks. Any concerns about their public mission of extending homeownership did not stand in the way of profitability.
This point is important in understanding the role of the GSEs in the housing bubble, the collapse of which led to the financial crisis and Great Recession. Many on the right have argued that the housing bubble was driven by a push to extend homeownership to low-income families and minorities who were not really prepared to own a home. In this telling, lenders issued mortgages not out of a desire to make money, but because government regulators forced them to make loans to meet goals on homeownership. For private banks, the Community Reinvestment Act (CRA) was the main villain, with regulators requiring banks to make bad loans to ensure compliance.
In the case of Fannie Mae and Freddie Mac, the claim is that Congress and the regulatory agency overseeing the GSEs, the Office of Housing Enterprise Oversight, pressured the GSEs to make loans to meet housing targets that were not justified on economic grounds. In some accounts, Representative Barney Frank is the primary villain. And while Frank did play a major role in writing the Dodd-Frank financial reform bill that bears his name, he was a minority member of the House of Representatives until January 2007, at which point most all of the bad loans had already been issued.
Just as the CRA is not a plausible villain in explaining the behavior of the mortgage issuers in the bubble years (the worst loans were issued by mortgage companies like New Century and Ameriquest, which were not even covered by the CRA), it is also not plausible to claim that Fannie and Freddie were sunk because of their desire to extend homeownership, as opposed to increasing their profit. The reality is that the GSEs were being run first and foremost as profit-making corporations. Their top executives were drawing huge paychecks in the bubble years, and they were booking billions in profits, just like the major banks and issuers of subprime mortgages.
But the GSEs were followers rather than leaders in the securitization of questionable and often fraudulent loans. That honor went to the private investment banks that were aggressive buyers of subprime loans, which did not meet the standards required by the GSEs. This was widely reported at the time, as the investment banks were winning away market share from Fannie and Freddie. For example, a 2004 article in USA Today touted the explosion in subprime lending in the bubble years. The same article included a list naming the players and their market shares.1 Fannie Mae was a small actor in issuance, and Freddie Mac was nowhere to be found. Fannie and Freddie were losing market share rapidly in the bubble years, precisely because they were not securitizing the worst loans.
In fact, the loss of market share was a topic for concern among investors. A 2006 Moody’s report on Freddie Mac’s economic prospects gives the basic story from the standpoint of investors:
Freddie Mac has long played a central role (shared with Fannie Mae) in the secondary mortgage finance market. In recent years, both housing GSEs have been losing share within the overall market due to the shifting nature of consumer preferences towards adjustable-rate loans and other hybrid products. For the first half of 2006, Fannie Mae and Freddie Mac captured about 44 percent of total origination volume—up from a 41 percent share in 2005, but down from 59 percent in 2003. Moody’s would be concerned if Freddie Mac’s market share (i.e., mortgage portfolio plus securities as a percentage of conforming and non-conforming origination), which ranged between 18 percent and 23 percent from 1999 through the first half of 2006, declined below 15 percent. To buttress its market share, Freddie Mac has increased its purchases of private-label securities. Moody’s notes that these purchases contribute to profitability, affordable housing goals and market share in the near term, but offer minimal benefit from a franchise-building perspective.2
As Moody’s analysis makes clear, Freddie and Fannie were rapidly losing market share precisely because they were not involved in the subprime and exotic mortgages being securitized by the private issuers. As the analysis notes, Freddie bought stakes in private-label MBS at the end of 2006 to gain back market share. This ended up being an incredibly foolish business decision, but it is clear that it was done to keep up with private issuers. Freddie Mac was not driving this market; it was following it.
The superior quality of Fannie and Freddie loans showed up clearly in the delinquency data.3 Even at the worst points in the crisis, more than 90 percent of Fannie and Freddie’s mortgages were current and fully performing. By contrast, less than 70 percent of option adjustable-rate mortgages were fully performing.
There is not much room for debate on the question of who the leading bad actors were in securitizing subprime mortgages. Fannie and Freddie should have been more responsible in recognizing the bubble and trying to rein it in, as I argued at the time.4 But they were following the market, not driving it.
This history is important to remember when we think about the future of the GSEs, since it is necessary to dispel the myth circulated by the Right that government entities designed to promote homeownership, like the GSEs, will inevitably be pushed by politicians into irresponsible lending of the sort we saw in the bubble years. In this respect, it is worth noting that the FHA, which has the explicit responsibility for promoting homeownership among moderate-income households, saw its market share plummet to just 2 percent at the peak of the bubble, down from an 8- to 10-percent share during more normal times. In fact, many conservatives argued at the time that the FHA no longer had a purpose since the private sector was doing such a great job fulfilling its role.
Unlike the GSEs, the FHA is an actual agency of the government rather than a for-profit corporation with an explicit public purpose. In the midst of a huge housing boom, the FHA maintained its lending standards and therefore lost most of its market share. This should give us confidence that publicly run entities designed to promote homeownership can resist pressures from politicians to engage in irresponsible lending practices in pursuit of this end.
Three Futures for the GSEs
While there have been a wide range of reform proposals for the GSEs, they can be lumped into three basic categories.
(1) Pure privatization route: In its strong form, this means getting the government out of the mortgage finance business. The private financial sector would have full responsibility for issuing and holding mortgages.
(2) Government-private partnerships: These proposals have taken a variety of forms, but most follow the general outlines of a bill introduced in the 113th Congress by Senators Tim Johnson and Mark Crapo. These proposals would have the government offer guarantees for most of the value of privately issued MBS.
(3) Preserving the GSEs in their current form: This is the idea of leaving Fannie Mae and Freddie Mac as fully public companies, possibly in merged form. In effect, it would turn the conservatorship into a permanent arrangement.
These three options are discussed below, but before getting into specifics, it is worth making a few points about the key policy issues. First, there is the question of moral hazard associated with giving a government guarantee to private corporations. The problem here is straightforward: the corporation has incentive to take large risks to maximize profits, since the government ultimately bears the costs associated with this risk.
The second point is the concern about corruption in meeting publicly set goals. This is an issue that arises from efforts to promote homeownership among lower-income households. The government may create incentives to issue dubious mortgages to meet goals in this area. The result is a large number of mortgages going into default, imposing large costs on the government while leaving the new homeowners with no wealth and a bad credit record.
The third issue is the desire to have an efficient market for housing finance. This means devoting as few resources as possible to the issuance, packaging, and servicing of mortgages. We have productive tasks that need to be done; we don’t need to make people shuffle mortgage-related documents to keep them employed.
The Privatization Route
The path for GSEs favored by many conservatives and libertarians is to simply phase them out and leave mortgage finance to the private sector. Their argument is that the problems of moral hazard and government corruption are too large to rein in effectively. This path does have a clear benefit in reducing the risks in these areas although, given the current structure of our banking system, it does not eliminate the moral hazard problem.
The collapse of Lehman Brothers and the near collapse of the other investment banks had little to do with government guarantees for the GSEs. These investment banks had loaded themselves up with questionable mortgages that they had planned to offload in MBS, but the housing market collapsed before they were able to do so. The government did allow Lehman to go bankrupt, with disastrous consequences for the financial system. The other major investment banks were rescued, although some smaller banks that got themselves into trouble with bad mortgages were allowed to fail.
The question going forward is whether there is reason to believe that the government would allow one of the largest banks to go under in the future. While the reforms in Dodd-Frank might reduce the risk of a major bank putting itself into bankruptcy with bad lending practices, there are still good grounds for questioning whether one of the country’s largest banks would be allowed to fail in a crisis.5 If the government would not allow a major bank to fail, then the issue of moral hazard would still be present with mortgages, even with no direct government involvement in the market.
If the government got out of the mortgage business altogether, there are also serious questions about what would happen to the price and availability of mortgages. There is some concern as to whether the thirty-year, fixed-rate mortgage as we know it would survive. This is in fact an unusual debt instrument. The standard practice is that homeowners can refinance at any time, assuming they are in a position to qualify for a new mortgage. This creates one-way interest rate risk where the lender cannot benefit from a rise in interest rates, since the rate is fixed. If interest rates fall, however, the borrower can pay off the loan early and take advantage of the new, lower rates.
This one-sided interest rate risk is the basis for the argument that ending the government guarantee would lead to the end of the thirtyyear, fixed-rate mortgage. There is good reason to question this assessment, most importantly because of the existence of a jumbo market for loans that are too large to be purchased by the GSEs under their guidelines. These mortgages typically have interest rates that are 25 to 50 basis points higher than the conforming mortgages that can be purchased by the GSEs. There are differences in borrower characteristics in that purchasers of jumbo mortgages have higher incomes and generally have other assets, but research from the Fed suggests that, after properly controlling for differences in characteristics, the gap is less than 10 basis points.6
This suggests that the government guarantee provides a modest but hardly essential subsidy to the mortgage market. After all, the current interest rate on thirty-year mortgages is roughly 4 percent. Two decades ago, the interest rate on these mortgages was over 7 percent. If the government’s withdrawal from the market caused rates to rise from 4.0 percent to 4.1 or even 4.25 percent, it would hardly shut down the market.
There is also a reasonable question as to whether a thirty-year, fixed-rate mortgage should be the gold standard for mortgages. The United States is the only country in the world where this is a common mortgage instrument. In most other countries, mortgages are for much shorter periods of time and often at adjustable rates. Many of these countries have higher rates of homeownership than the United States. While the security provided by a fixed rate is desirable, it is possible to argue that a shorter term with a lower interest rate would be better for most homeowners.
At current interest rates, the gap between the rate charged on a thirtyyear mortgage and a fifteen-year mortgage is around 75 basis points. For a middle-income homebuyer with a mortgage of $200,000, this would translate into an interest savings of $1,500 a year. Obviously, affordability would be a problem in many cases, but if a family earning $75,000 a year could save $1,500 a year on interest by taking out a fifteen-year mortgage rather than a thirty-year mortgage, this would help them tremendously in accumulating home equity. In many cases, it could make sense for people to buy less expensive homes in order to be able to take advantage of the lower interest rate available on a shorter-term mortgage.
There are certainly other factors that need to be considered in the trade-off between shorter- and longer-duration mortgages, but the point here is that we should not assume that a thirty-year, fixed-rate mortgage is a necessity for high rates of homeownership. The fact that it has historically been the preferred instrument for homebuyers does not mean that it is necessarily the best way for them to finance the purchase of a home. The argument that a fully private market could not support thirty-year, fixed-rate mortgages is almost certainly wrong, but even if it were accurate, that would not necessarily be a fatal argument against going the route of full privatization of the GSEs.
The Government-Private Partnership Route
The path that has enjoyed the most support in the years after the GSEs were put into conservatorship is a route that would gradually downsize the GSEs and replace them with a system in which private banks issue MBS, but with substantial backing by the federal government. There have been several versions of this proposal, but the one that has advanced most politically was incorporated in the Crapo-Johnson bill that was first put forward in 2014.
Under this plan, the GSEs would be wound down and replaced by an institution that would be collectively owned by the financial institutions that use its services. This institution would set up a framework for issuing MBS and would guarantee the MBS that met its standards. To limit the moral hazard problem, the holders of an MBS would be liable for the first 10 percent of losses due to defaults. The guarantor would be responsible for larger losses.
This plan was supported by a bipartisan 13–9 vote in the Senate Banking Committee but never came to the Senate floor for a vote. Nonetheless, this is the only proposal for reforming the GSEs that has made it out of committee. In addition to its relative success in Congress, this proposal has enjoyed the support of a number of policy groups in Washington, most notably the Center for American Progress, which has been closely associated with the Clintons.
At the same time, there are several reasons for questioning the wisdom of this sort of public-private partnership, compared with either pure privatization or maintaining the GSEs as effectively government-owned companies. The most obvious is that there is good reason to question whether the structure it proposes would address the moral hazard problem where financial institutions are able to profit by passing off risk to the government.
Its proponents assume that assigning first-dollar losses to the holders of MBS will provide them with sufficient incentive to scrutinize their risk. But this would seem to fly in the face of the practices we saw during the housing bubble years. Private issuers were able to sell hundreds of billions of dollars of poor-quality MBS with no government guarantee. They were also able to effectively buy investment-grade ratings from bond-rating agencies that did not want to risk losing business by making honest assessments of the issues they rated.
It is difficult to see why we would expect a better outcome in a situation where the government is prepared to cover 90 percent of potential losses. In this case, banks and other issuers could truthfully say that, even in a worst-case scenario, an MBS could only lose 10 percent of the value of the asset. If they failed to adequately scrutinize the quality of MBS when they were liable for 100 percent of any losses, why would we think investors would carefully evaluate the quality of MBS when the most they could possibly lose is 10 percent of their face value? This seems like a structure that is virtually certain to create very serious problems of moral hazard.
This system also fails the efficiency test. By all accounts, it would increase the costs associated with issuing MBS relative to the current system. The Congressional Budget Office (CBO) put the additional cost of the Crapo-Johnson system, relative to the current system, at 20 basis points, with other estimates considerably higher.7 This additional cost means real resources would be used up in the financial sector. Even the estimate of 20 basis points (which is probably realistic) would imply an additional $5 billion a year being spent just for the process of issuing mortgages. This might be a reasonable expenditure if we thought it were creating a more robust system less susceptible to bubbles and panics, but there is no reason to think that this is the case. We effectively would be spending more money to get a system that has a considerably greater problem of moral hazard than the current one.
Maintaining the GSEs as Government-Owned Companies
The third route, which probably enjoys the least political support in Washington, is to simply leave the GSEs as government-owned companies. The opposition seems to stem primarily from the general hostility to the idea of government-run businesses and, perhaps more importantly, to the private financial sector’s desire to secure greater profits. The GSEs also add very modestly to the budget deficit (roughly $30 billion over a decade) because of the risk accounting used by the CBO. Effectively, the CBO assumes some risk of another bailout of the GSEs in future years and assigns the implicit cost of this risk to the federal budget. If it turns out that we don’t need a bailout, then this risk will have proved costless and the deficit will be less than projected by the amount attributed to the risk. As noted earlier, the profits of the GSEs have actually reduced the budget deficit by around $15 billion a year.
Having the GSEs as purely government-owned companies eliminates the moral hazard problem altogether. There is no one in a situation to profit by taking on inappropriate risks. As noted earlier, the GSEs ran into trouble in the housing bubble years because they were trying to take back market share from private issuers of MBS. They were lowering their standards out of a concern for profits. This is in contrast to the situation of the FHA, which is purely public, and saw its market share dwindle to almost nothing in this period. As mixed public-private companies, the GSEs had serious problems with moral hazard. These problems go away if the companies are run simply as government agencies.8
While there is always the risk of corruption and waste in any government enterprise, the track record of the GSEs in this area has been reasonably good. The fear that they would be forced by politicians to buy up large numbers of bad mortgages to meet political ends has proven to be unfounded. Of course, this risk always exists, which argues for maintaining transparency and strong oversight, but this has not historically been a major problem at the GSEs. It is also important to recognize that keeping the GSEs as government-owned companies does not mean they will hold a monopoly on the secondary mortgage market. While the backing of the federal government clearly gives them an advantage over private competitors, if incompetence and/or corruption caused costs to rise, presumably private financial companies would move in and seize market share.
If we consider the original purpose of a government role in financing housing, the purely public model is likely best suited for the task. One of the main goals in establishing Fannie Mae was to provide a secondary market, which would allow lenders to issue mortgages without the fear of being too heavily tied to the real estate sector. While financial markets are much more fluid today than in the 1930s, they are still subject to cyclical ups and downs, which the GSEs can help to counteract, creating more stability in the mortgage market.
The other issue is that, by offering a government guarantee, the GSEs are effectively subsidizing home mortgages relative to other forms of credit. We can argue whether this is desirable. After all, if we begin from the standpoint that the economy is near full employment (a questionable assumption, but the starting point for most economic analysis), reducing the cost of credit for housing implies an increase in the cost of credit for other purposes. If we do decide to subsidize housing credit, however, it makes sense to do it in the most efficient possible way.
Government-owned GSEs clearly fit this bill. They have much less overhead than private sector banks, and the senior executives at a government agency do not get paychecks that run into the millions or even tens of millions, as is the case for Wall Street firms. Eliminating these expenses is a gain from the standpoint of making the mortgage market more efficient.
In fact, if the goal is to maximize the efficiency of the mortgage market, it is difficult to see the benefit of issuing mortgage-backed securities. The resources devoted to constructing and marketing MBS are a needless waste from the standpoint of the economy as a whole. The savings from not employing accountants, auditors, and others involved in this process can be passed on in savings to borrowers. As pointed out earlier, Freddie Mac first created MBS in the 1970s. Fannie Mae did not start issuing MBS on a large scale until well into the 1980s, so we do have extensive experience with the GSEs holding mortgages in their portfolios.
It is understandable that private banks would want to issue MBS or otherwise limit their exposure to the mortgage market, but since the GSEs guarantee the MBS they issue, they are not reducing their exposure to the mortgage market by issuing MBS.9 If we were to see another housing crash, like the one from 2007 to 2010, the GSEs would again face massive liabilities as the MBS they guaranteed lost value. Issuing MBS does protect the GSEs from interest rate risk, but from the standpoint of the government, this protection is virtually meaningless.
The issue here is that the value of mortgages falls when the interest rate rises. The market value of a thirty-year mortgage issued at a 4-percent interest rate is worth less when the prevailing mortgage interest rate is 6 percent. While the monthly payment is unchanged by a rise in interest rates, and the ultimate amount of principal collected is unaffected, if there were some need for liquidity in the short term, the mortgage would have to be sold for less money in the high-interest-rate scenario.
Interest rate risk is a real problem for a bank or other private financial institution, but it is not one for the government or a government-owned enterprise. The government is not going to be liquidity constrained and required to suddenly sell off large numbers of mortgages. The government can print money if there is some reason it would need cash to pay immediate expenses, so it is difficult to see the gain from having a government-owned enterprise sell mortgage-backed securities rather than just holding mortgages.10
Long Live the Status Quo
In the years immediately following the financial crisis, there were few people who would have bet that, in 2018, the GSEs would still be in a conservatorship in which they are effectively publicly owned companies. Nonetheless, inertia has proven to be a powerful force. In this case, I would argue that it has been a force for good in the housing market. The current system is one that minimizes the problem of moral hazard in housing finance, which was so important in the run-up in prices in the housing bubble years. It also is the most efficient mechanism for financing mortgages, as it requires fewer resources to be wasted on housing finance. We could make the system more efficient by getting rid of MBS altogether and having the GSEs just hold their mortgages, but keeping the GSEs as public companies is a good start for now.
This article originally appeared in American Affairs Volume II, Number 1 (Spring 2018): 42–55.
2 Analysis of Federal Home Loan Mortgage Corp., Moody’s Investors Service, December 8, 2006.
3 “OCC and OTS Mortgage Metrics Report,” Office of the Comptroller of the Currency, Fourth Quarter 2009.
4 Dean Baker, “The Run-Up in Home Prices: Is It Real or Is It Another Bubble?,” Center for Economic and Policy Research, August 2002.
5 A report from the Government Accountability Office in 2014 argued that evidence for the market’s belief in an implicit too-big-to-fail guarantee had largely disappeared based on a reduction in interest rate spreads. See “Large Bank Holding Companies: Expectations of Government Support,” Government Accountability Office, July 2014, no. GAO-14-612. This analysis, however, also finds little evidence of the belief in a too-big-to-fail guarantee in 2006, before the crisis.
6 Wayne Passmore, Shane Sherman, and Gillian Burgess, “The Effect of Government-Sponsored Enterprises on Mortgage Interest Rates,” Finance and Economics Discussion Series, Divisions of Research & Statistics and Monetary Affairs, Federal Reserve Board, Washington, D.C., 2015–16.
7 “Transitioning to Alternative Structures for Housing Finance,” Congressional Budget Office, December 2014.
8 It is worth noting in this respect that Fannie Mae had a major accounting scandal in the early part of the last decade. Franklin Raines, who was CEO of the company, along with other top officials, manipulated the reporting of profits in order to reach targets set for bonuses. Here also the issue is clearly one that resulted from Fannie Mae being run for profit rather than the public service goals of the company.
9 The GSEs can explicitly share the risk by selling “credit risk notes,” which have the investors bear a portion of the losses on a pool of mortgages. They GSEs have begun experimenting with this risk-sharing, but it still only applies to a relatively small portion of the MBS they have issued.
10 For those troubled by my dismissal of interest rate risk as a problem, consider the fact that the government has trillions of dollars of debt outstanding in the form of long-term bonds. If the market value of the mortgages held by the GSEs has fallen due to higher interest rates, so too has the market value of the long-term debt the government has issued. The net asset position of the government almost certainly would improve when interest rates rise.