Going Cold Turkey
Three Ways to End Fannie and Freddie without Slicing Up the Taxpayers

Click here to view this Outlook as an Adobe Acrobat PDF.

Click here to view the press release.

September 2010

Most of the plans put forward for replacing Fannie Mae and Freddie Mac involve some continuing role for the government in housing finance. The reasons behind these plans are less than persuasive, and recent experience has shown that the moral hazard created by the government's presence in the housing-finance market can produce catastrophic results for taxpayers. Instead of trying to find ways that government can remain involved in housing finance, the new Congress should consider how to withdraw the government from any role in financing prime mortgages and implement various promising private-financing mechanisms--covered bonds, the Danish system, and a more focused and regulated securitization system.

Key points in this Outlook:

  • Government involvement in housing finance has caused huge losses to taxpayers and is not necessary to achieve the benefits claimed by its proponents.
  • Yet most proposals to reform Fannie Mae and Freddie Mac still call for government involvement in housing finance.
  • Private-sector solutions such as covered bonds, the Danish mortgage system, and controlling the quality of securitized mortgages can deliver the same benefits to the housing market without the risks to taxpayers.
  • As long as Fannie and Freddie remain under government control, a transition to a fully private market for residential mortgages should not be difficult.

The key question about the future of Fannie and Freddie--or, more properly, of government involvement in the mortgage market--is not whether the government guarantee should be explicit or implicit, or whether it should apply to the firm that issues mortgage-backed securities (MBS) or to the securities themselves. The key question is whether any government support for the secondary mortgage market makes sense as a matter of policy. If we look at the recent history of government involvement in the housing field, the picture is not pretty. Just within the last twenty years, the savings and loan (S&L) industry collapsed, with a loss to taxpayers of approximately $150 billion. Now Fannie Mae and Freddie Mac are operating under government conservatorships, with estimates of losses running from $400 billion to $1 trillion. Is it possible that Congress simply cannot learn from its mistakes?

Both the S&L industry and Fannie Mae were the products of Depression-era legislation to assist the housing industry. Fannie began life as a government agency with a simple mission: to buy mortgages from banks and other mortgage originators, providing these institutions with the funds to make more mortgages. Initially, Fannie was funded by appro-priations, and its activities added to the federal deficit during the Vietnam period. To address this problem, Congress "privatized" Fannie in 1968, allowing it to sell its voting shares to the public. But Fannie retained its government charter and mission, and these--together with various tax and other benefits--gave it a credible claim to implicit government support and to designation as a government-sponsored enterprise (GSE). Freddie Mac, Fannie's smaller competitor, was chartered in 1970 with the same benefits, and the two GSEs were seen in the capital markets as backed by the government.

Federal S&Ls had government charters and a government mission to originate mortgages, and they were eligible for federal deposit insurance. Thriving in the postwar boom, by 1965 S&Ls controlled 26 percent of consumer savings and made 46 percent of all single-family mortgages.[1] In 1966, when the industry had assets of approximately $130 billion, the government placed ceilings on the interest rates that could be paid on insured accounts at banks and S&Ls, but it allowed S&Ls to pay one-quarter percent more, assuring the industry of a steady flow of insured deposits to support the long-term mortgage assets they were expected to hold. With this government assistance, by 1979 the S&L industry had more than quadrupled in size to almost $580 billion.[2]

But the S&Ls were taking massive interest-rate risks. Their government mission required them to fund their long-term mortgage assets with short-term deposit liabilities. This was possible as long as the government controlled interest rates on insured deposits, but if the cost of their deposits ever increased substantially, they could not survive. When money-market mutual funds were developed during the inflationary (and high-interest-rate) period of the late 1970s and early 1980s, the S&Ls and banks suffered huge losses of low-cost deposits, and the government had to abandon its deposit-interest controls. As a result, by the early 1980s, most S&Ls were insolvent; it was impossible to fund thirty-year fixed-rate mortgages, paying 5 or 6 percent, when short-term rates in the money markets were as high as 20 percent.

The ability of the GSEs to continue operating when firms without government backing would have been shut down was in itself a major contributor to the bubble's size and destructiveness when it began to deflate in mid-2007.

When the S&L industry collapsed in the late 1980s and early 1990s, Fannie and Freddie--with their implicit government backing--stepped in as two gigantic S&Ls and began a period of exceptional growth, doubling in size every five years. By 2008, they were responsible for the credit risk of mortgages with an unpaid principal of over $5.5 trillion. But while the S&Ls were destroyed by interest-rate risk, the GSEs were destroyed by credit risk. Beginning in 1992, they were required to purchase increasing numbers of subprime and Alt-A mortgages in order to meet the affordable-housing goals imposed by Congress and administered by the Department of Housing and Urban Development (HUD). When these high-risk loans defaulted as the housing bubble deflated in 2007, Fannie and Freddie suffered massive losses. By late 2008, they were insolvent and the government took them over; at that point, they held or had guaranteed 12 million high-risk loans, with an unpaid principal of $1.8 trillion.

Thus we have two business models, both operating in the housing field with government support, that eventually collapsed into insolvency with huge costs to taxpayers. What they had in common was their government backing, which substantially reduced market discipline, allowing them to grow at extraordinary rates and take risks that other financial intermediaries would not have been permitted to take--for far longer than other firms that are funded privately.

Neither the interest risks taken by the S&Ls nor the credit risks taken by the GSEs could have occurred if investors and lenders had been looking to the credit of these institutions--rather than the government's credit--when they provided them with funds. Moreover, it was government backing that allowed these institutions to continue operating and incurring losses long after ordinary private companies would have been shut down by their creditors. The ability of the GSEs to continue operating when firms without government backing would have been shut down was in itself a major contributor to the bubble's size and destructiveness when it began to deflate in mid-2007. As the bubble grew, rising home prices both disguised the risks that homeowners and lenders were taking and made greater risks necessary if homebuyers were to afford the higher housing prices.

Arguments in Support of a Government Role

Given this history, it is surprising that arguments for continued government involvement in the housing market are so resilient. Government support for housing resulted in two recent catastrophes for taxpayers, and the arguments in support of government involvement are remarkably thin. In effect, they boil down to three:

The Government Enables the Thirty-Year Fixed-Rate Mortgage. Frequently, supporters of government involvement in the mortgage market argue that only with the government's support can a thirty-year fixed-interest-rate mortgage be possible. This is a myth. Those who make this statement have apparently not bothered to check the market for mortgages larger than those the GSEs are permitted to buy (known as "jumbo" loans). Even today, when the private securitization market is still weak from the effects of the financial crisis, many lenders and brokers are offering thirty-year fixed-rate jumbo loans, which of course have no government backing.

Government Involvement Lowers Mortgage Rates. This issue was carefully explored by Federal Reserve economists before the financial crisis occurred; their summary was that the low funding rates that Fannie and Freddie received because of their implicit government backing did not result in reduced interest rates for homebuyers: "We find that GSE portfolio purchases have no significant effects on either primary or secondary mortgage rate spreads."[3] A more detailed description appears later in the paper:

We estimate that if the GSEs unexpectedly increase their portfolio purchases by $10 billion (about 3.7 percent of average monthly originations during 2004), the primary and secondary mortgage rate spreads would increase 1.4 and 1.3 basis points after one month, respectively. But if the GSEs instead unexpectedly increased their securitization activity (that is, their gross issuance of MBS) by $10 billion, we estimate that primary and secondary mortgage rate spreads would decline 0.6 and 0.5 basis points, respectively. Note that none of these effects is statistically different from zero.[4]

FSO Table FSO Table

The Housing Market Would Not Be Stable without Government Support. Finally, the argument is made that the housing market must have government support in order to remain stable in adverse financial conditions. This is a peculiar argument to make when the government's main efforts to promote housing--the S&Ls and the GSEs--have both collapsed with devastating consequences for taxpayers, and the government's own housing policies produced a vast number of high-risk mortgages that caused a 30 percent decline in housing prices when they defaulted. As shown in the table, the government was the principal customer for these subprime and Alt-A loans.

The effect of such a large number of subprime and other nontraditional mortgages on the U.S. financial system is shown by data on foreclosure starts following the deflation of recent housing bubbles.[5] After the bubble that ended in 1979, when almost all mortgages were the traditional type, with substantial down payments and borrowers with good credit histories, foreclosure starts reached a high point of only 0.87 percent in 1982. After the next bubble, which ended in 1992 and in which the vast majority of loans were still the traditional type, foreclosure starts reached a high of 1.32 percent in 1994. After the bubble that ended in 2007, however, when almost half of all U.S.mortgages were subprime or otherwise high risk, foreclosure starts for all loans reached the (thus far) unprecedented level of 5.37 percent in 2009, despite government efforts to prevent or delay foreclosures. The lesson is that where residential mortgages are of good quality, the deflation of occasional bubbles will not cause a serious financial disruption--certainly not enough disruption to warrant continuous government involvement in the housing field--but if government policy or moral hazard erodes underwriting standards, the private housing-finance system can be devastated.

FSO Figure FSO Figure

Certainly, the fact that Fannie and Freddie--even though insolvent--are today the mainstays of the housing-finance market will be used to support the view that the government needs to be involved in the housing market, if only to be there in the event of a colossal failure such as the 2008 financial crisis. If we leave aside the fact that the government's involvement in the housing-finance business actually caused the colossal failure, this argument is superficially attractive. However, the data in the figure show that the prime residential MBS market was no more seriously affected by the deflation of the housing bubble in 2007 than any of the other securitized-asset classes. All these classes suffered losses in value as a result of the recession that followed the financial crisis, but prime residential MBS--which were almost entirely prime jumbo mortgages--declined and recovered along with all other asset-backed securities. The real losses were in the collateralized debt obligations (CDOs) and the mezzanine (BBB) asset-backed CDOs. These were the lowest-quality asset-backed securities available in the market and included mainly subprime and Alt-A mortgages. The fact that prime mortgages behaved the way all other assets behaved in the financial crisis and afterward is a significant observation; it shows that government involvement in the housing market is no more necessary for housing assets than it is for any other asset class, and--perhaps more important--that if the assets themselves are high quality, like prime mortgages, a nongovernment market for these assets is as stable as any other asset market.

Still, some argue that only a government-backed organization will be able to keep interest rates stable when the private markets are disrupted. This, however, turns out to be another myth. Federal Reserve economists also reviewed this issue, by looking at what happened in an earlier financial crisis. They concluded that Fannie and Freddie did not add appreciably to the stability of interest rates:

We examined the empirical connection between mortgage interest rates and GSE secondary market activities, especially GSE purchases of mortgages for their own portfolios. If GSE portfolio purchases affected mortgage rates, they could stabilize mortgage markets. This benefit would flow to all mortgage market participants, not just GSE shareholders. . . . We found that portfolio purchases have economically and statistically negligible effects on both primary and secondary mortgage rate spreads.[6]

More important, enlisting the government to keep money flowing to the housing market when it is tight elsewhere would be bad policy. If participants in the housing market are insulated from the vicissitudes of the market, they will take more risks and be less prudent in their investment decisions. This is what helped create housing bubbles in the past. The possibility that financing for housing could be subject to disruption or financing restrictions is, of course, one of the risks that the housing industry fears, but that fear itself will reduce the overbuilding and excessive leverage that have caused volatility and repeated housing bubbles in the past.

The Problem with Current Solutions

Despite these myths and potential policy errors, most of the proposals for replacing Fannie and Freddie involve a government role of some kind.[7] How many times do these housing-based disasters have to happen before policy-makers realize that there is no way for the government to participate in the financing of mortgages without distorting the market and making it vulnerable to bubbles and collapses? To be sure, the taxpayer costs that resulted from the insolvency of Fannie and Freddie came from a very special kind of government support--an implicit guarantee of their obligations--that enabled them to function for many years with lower funding costs than any other market participants. For this reason, many of the plans that have been advanced to supplant Fannie and Freddie specifically avoid any guarantee of the firm or firms that would operate in their place. Instead, they propose only that the government guarantee the MBS that these privately funded firms would issue.

If we are serious about preventing another round of government-induced losses in the housing-finance system, we should begin to look seriously at wholly private-sector solutions.

But would things have been different if the government had only guaranteed the GSEs' MBS? Not significantly. The GSEs needed government support to borrow the huge sums that were necessary to carry their portfolios, which amounted to about $1.5 trillion by 2008. If their activities had been limited to issuing and guaranteeing MBS, the mere fact that their government backing was limited to MBS would not have prevented the huge crash in home prices resulting from the default of the poor-quality mortgages they had guaranteed. It was not the failure of the GSEs to pay their investors that caused the financial crisis; it was the fact that they were able to make low-quality mortgages that drove down housing values when they failed. This was only possible because investors relied on the government's backing of the GSEs' guarantees; without that, no one would have taken these guarantees seriously. For good reason, none of the reform plans under consideration allow any of the replacements for Fannie and Freddie to accumulate large portfolios, but that does not mean that these plans eliminate the moral hazard the government guarantee creates.

Some plans recognize this and seek to address it with a fee or premium for the government's support. This has a nice private-sector ring, but it is not a solution. The government has no way to set a risk-based premium, and any premium it does set will be viewed as a tax on homebuyers and thus be subject to political control. If the premium is supposed to be risk based, the government agency that imposes it faces a difficult political problem because it will have to discriminate among specific borrowers, with low-income and other risky borrowers subject to the highest rates. This is politically infeasible, but it is also largely impractical; if there is any question about whether the government can set risk-based premiums, we need only look at the record of the Federal Deposit Insurance Corporation (FDIC), which has been required to do that with bank insurance premiums for years. It is questionable whether the agency has been able to develop a truly effective risk-based system, but the FDIC's current negative-capital position demonstrates that whatever premiums it thought it might have been setting were inadequate for the risk involved. Now is the time to be realistic about what government can and cannot do and what we can expect from a government role in the housing market.

If we are prepared to face facts, recent experience should tell us that there is no way to avoid the moral hazard arising from the government's involvement in housing finance. Housing risks will always gravitate to the place where they are covered by the government guarantee. If they do not start there, over time Congress will move them there, if only to benefit constituents. A good example is the increase in the conforming-loan limit (the maximum size of a mortgage that the GSEs can buy) that Congress adopted in 2008. There was then what seemed to be a quarter-point difference between the interest rates on mortgages within the conforming-loan limit and the interest rate on the jumbo mortgages outside it. This was unacceptable to Congress, even though the higher rates were applicable in communities where homes could cost upward of a million dollars. So as the GSE reform legislation moved through Congress, a higher conforming-loan limit was legislated for those areas where houses were expensive. That moved these higher-cost mortgages under the GSE guarantee that was ultimately backed by taxpayers. This process will never stop unless politics is taken out of housing finance, and that can only happen if the government's role is eliminated.

Options for the Private Financing of Mortgages

If we are serious about preventing another round of government-induced losses in the housing-finance system, we should begin to look seriously at wholly private-sector solutions. These fall into three categories: (1) those that rely on depository institutions, like covered bonds; (2) those that rely on a separation of the credit and interest-rate risks, like the Danish model; and (3) those that rely on a senior-subordinated or collateralized structure, like the securitization model.

Covered Bonds. In 2008, the Treasury Department developed a model covered-bond program. It was modest in size but was intended to demonstrate how such a system would work.[8] As proposed by the Treasury, banks would be authorized to sell bonds backed by a specified pool of mortgages. The mortgages would remain on the banks' balance sheets but would be segregated from their other assets. The banks would be able to use the bond proceeds for any banking purpose, but the principal and interest on the mortgages would flow to the investors. The key to the success of the program would be an obligation on the part of the banks to top up the mortgage pool (which would be overcollateralized in any event) whenever defaults or delinquencies jeopardized the quality or sufficiency of the collateral backing the bonds.

Structures like this have been used for many years in Europe and have largely succeeded in creating a high-quality debt security without any government guarantee. One of their advantages is getting the incentives right: the banks that issue covered bonds have a strong incentive to make good-quality loans because they are ultimately required to keep replenishing high-quality collateral for the bonds. If they do not, the bondholders have a claim beyond the mortgages against the bank itself.

Shortly after the Treasury's proposal was advanced, Representative Scott Garrett (R-N.J.) reintroduced legislation he had previously sponsored to create a larger covered-bond program. The legislation was marked up in the House Financial Services Committee and sent to the House floor in July 2010 as HR 5823.[9] It emerged different from the Treasury's approach in several respects, the most significant of which allowed covered bonds to be issued by any financial firm, not just banks and other regulated depository institutions. This is potentially important because deposits in the U.S. banking system today are not large enough to provide funding for the U.S. housing market.[10] Whether other mortgage lenders will be able to compete with banks as covered-bond issuers is a question for the future, but permitting nondepositories to issue covered bonds may increase their usefulness as a funding mechanism for mortgages.

While covered bonds do not require any direct or indirect government support, the issuers under both the Treasury and Garrett proposals would be regulated by a covered-bond regulator, and each issuance of covered bonds would have to be approved by the regulator. The Treasury proposal requires, among other things, that the mortgages in the cover pool have a loan-to-value ratio of 80 percent at the time they are included in the pool. The Garrett plan only requires that the mortgages be in compliance with supervisory guidance at the time the loan is originated.

The Danish Mortgage System. This system is unique in the world, although Mexico appears to be adopting the same structure, and whether it would be compatible with the way mortgages originate in the United States remains to be seen. One of the distinguishing features of the Danish system is that it separates the credit and interest-rate risk on a mortgage. This permits some interesting options for mortgage borrowers. In Denmark, mortgages are arranged with a small group of specialized mortgage banks. When the terms are settled, the mortgage is added to a pool of mortgages all issued at the same interest rate, so all the mortgages in the pool are identical (except for size). For example, on any given date, a hypothetical issue of thirty-year fixed-rate callable bonds will go to market at whatever the market rate for mortgage bonds with these terms is on that day. Borrowers who want thirty-year fixed-rate loans will authorize the mortgage bank to add their loan to the pool backing the bonds. Loans can also be added to the pool after it is established. Interests in the pool are purchased by capital-markets participants, which of course could be investors from anywhere in the world. The specialized mortgage bank that arranges the mortgage is responsible only for the credit risk, and in effect functions like a mortgage insurer. If the mortgage defaults, the mortgage bank must buy it back from the pool. Again, as with covered bonds, the incentives are well aligned. The mortgage bank has a strong interest in assuring that the mortgages it adds to any given pool are of good quality, since it is bearing the risk of default. It also earns a fee for its services in arranging for the mortgage and placing the mortgage in the pool. Under this system, there have been no mortgage-bank defaults in Denmark in the past two hundred years.

One of the most interesting elements of the Danish system is that the issuance of identical mortgages at the same interest rate makes it possible for homeowners to buy the bonds that are backed by their mortgages, or any portion of them, and thus create equity in the home that did not exist when the mortgage was issued. This becomes an attractive option, for example, if market interest rates rise. In that case, the market value of the mortgage bonds falls. The homeowner can then enter the market and--by arranging for a new mortgage at the higher rate--purchase at a discount any principal amount of the bonds backed by the pool of which his mortgage is a part. This portion of the mortgage obligation is then canceled. In effect, he is refinancing as someone might do in the United States, but with a major difference. Although he will be paying a higher interest rate for the new mortgage, he has in effect reduced the principal amount of the debt on his home by acquiring--at a discount--all or a portion of the debt represented by his original mortgage.

It is also possible for the homeowner to have his mortgage included in an issue of callable bonds. In this case, he also has the option of calling the debt on his home at par when interest rates fall. This transaction is then identical to what occurs in the case of a U.S. mortgage refinancing. So the Danish system offers the homebuyer the same opportunity to refinance as occurs in the United States when interest rates fall, but it also offers the unique opportunity to buy down the loan when interest rates rise. The use of a market rate also makes the Danish system completely transparent; homeowners know the interest rate they have received is the market rate at the time their mortgage was financed.

Again, like covered-bond systems, the Danish mortgage system is highly regulated but does not require any government financial support. All loans are made with recourse to the borrower, loan-to-value ratios may not exceed 80 percent, and mortgage banks are also regulated.[11] However, the Danish system could be adapted to provide a wider variety of options for U.S. homebuyers through, among other things, the use of supplemental mortgage insurance, improved disclosure to investors, and higher capital requirements for mortgage banks.

Securitization. The figure shows that a securitization market for prime mortgages can survive even in the most adverse financial conditions in seventy years. The key is to make sure that the system securitizes mortgages of high quality. In one sense, securitization is inferior to both covered bonds and the Danish system because it does not align incentives as well as they do. The credit support in the securitization system comes not from a pool of mortgages backed by a well-capitalized issuer as in the covered-bond system or a well-capitalized financial institution as in the Danish system. Instead, it comes from a senior-subordinate credit structure that in effect--by requiring lower tranches to take the first losses in any mortgage pool--reduces the likelihood that the senior tranches will suffer losses. In this way, the senior tranches can reach investment grade. The failure of the securitization system in 2007 may lead some observers to believe that it does not have a future, but that judgment is highly premature. It occurred almost entirely in the securitization of subprime and Alt-A mortgages; as shown in the figure, securitizations of prime-quality mortgages performed as well as any other asset-backed security in the recessionary period that followed the deflation of the housing bubble.

Securitization has some distinct advantages. In addition to using proven technology and structures that are already in place, securitization is not subject to size limits. As the mortgage market grows, a soundly based securitization market grows with it; the capitalization of the mortgage insurer (the mortgage bank in Denmark or the covered-bond issuer in Europe or the United States) is not an issue. Finally, the securitization market has functioned satisfactorily for thirty or forty years, produces interest rates that Fed economists found to be competitive with those offered by the GSEs,and is still working well for many types of assets.

If the GSEs remain under government control, there is time to make the transition to a housing-finance system that relies principally on the private financial markets.

The principal deficiency of private securitization as a replacement for Fannie and Freddie is the absence of proper incentives. This, however, does not mean that the whole securitization system has to be discarded. The recently adopted Dodd-Frank Act requires that participants in a securitization chain divide up (as the regulators decide) a 5 percent retention amount on any securitization; this is intended to make them hold "skin in the game." The system is probably unworkable, but it is unlikely to be used in any event; there is a statutory provision that permits the Securities and Exchange Commission and other regulators to eliminate any retention if the mortgages involved in the securitization meet certain quality tests that the regulators themselves are authorized to establish. This provides a lot of leeway, for good or ill. But assuming that the regulators can establish strong quality standards--including a substantial down payment, good credit from the borrower, adequate documentation, mortgage insurance, adequate disclosure to investors, strong reps and warranties, and required terms for put-backs to earlier participants in the securitization chain--requirements such as these could cure many of the inherent ills of the securitization structure.

Of course, at this point, investors are still avoiding the mortgage-securitization market because of their experience during the financial crisis, but other asset-backed markets are beginning to function normally. This means that Fannie and Freddie will continue to be the mainstay of residential-mortgage finance until investors have the confidence to return to the mortgage-securitization market. The recoupment of some of the value of prime mortgages initially lost in the financial crisis, as shown in the figure, suggests that the jumbo securitization market is gradually returning to normal, although it might take several more years to reach the level of operation common before 2007.

The Transition to a Private Market

If we assume that during this period the GSEs remain under government control, there is time to make the transition from a government-dominated housing-finance system to one that relies principally on the private financial markets. As the securitization market for mortgages gradually returns to normal, Congress could authorize the GSEs' regulator, the Federal Housing Finance Agency, to reduce the conforming-loan limit in incremental steps. As this limit declines, the private securitization market should enter the newly vacant space, while Congress and others watch to make sure that adequate funds are available for mortgages. This process could be continued until the conforming-loan limit is so low that virtually all mortgages are being financed privately. At this point, the GSEs could be either closed down or allowed to restructure into fully private firms. Ideally, the government would no longer be in the business of financing prime mortgages.

Of course, there is no reason for securitization to be the only nongovernmental system for financing mortgages. Congress can always adopt legislation authorizing covered bonds, the Danish system, or any other financing system it can agree on. However, in the absence of such congressional action, securitization is a satisfactory fallback option so Fannie and Freddie can be eliminated or privatized.

What to Do about Affordable Housing

Affordable housing is one other alleged benefit of government involvement in the housing-finance market. It was not discussed earlier for two reasons. First, there now seems to be a recognition, even among some of the most ardent supporters of affordable housing in Congress, that imposing an affordable-housing mission on the GSEs was a mistake. In an interview on Larry Kudlow's television program in late August, Representative Barney Frank (D-Mass.)--the chair of the House Financial Services Committee and previously the strongest congressional advocate for affordable housing--conceded that he had erred: "I hope by next year we'll have abolished Fannie and Freddie . . . it was a great mistake to push lower-income people into housing they couldn't afford and couldn't really handle once they had it." He then added, "I had been too sanguine about Fannie and Freddie."[12] Accordingly, most of the groups submitting plans for the reform of the housing-finance system do not cite affordable housing as a reason for government involvement.

Second, there is no reason why policies on affordable housing should be introduced into the conventional market. The idea that low-income groups should have access to mortgage credit makes sense for many reasons, but this is social policy, not economic policy. It was the confusion of the two that led Congress in 1992 to adopt legislation that added an affordable-housing mission to the GSEs' existing obligation to foster and maintain a liquid secondary market in conventional (that is, nongovernment) mortgages. The government already had an affordable-housing agency, the Federal Housing Administration (FHA), which was authorized to insure mortgages for borrowers who could not meet the standards for prime loans. The problem was that FHA was an on-budget agency and thus was limited in how much credit it could provide to underserved groups. The affordable-housing mission for Fannie and Freddie put them into competition with FHA--as well as one another--for high-risk mortgages. The result was that millions of these mortgages were made and insured or guaranteed, with immense losses for taxpayers.

The right decision, if Congress wants to pursue a social policy in the housing field, would have been to provide the necessary funds to FHA, rather than distorting the conventional market by hiding the subsidies in the GSEs' off-budget activities. There are certainly reasons for encouraging homeownership, and Congress may want to subsidize the purchase of homes by individuals who cannot meet the standards for prime loans. But as shown by the experience with Fannie and Freddie, mixing social policy with what is essentially economic policy for improving housing finance was a costly mistake that should not be repeated.


The policy case for government involvement in housing finance is weak. All the major arguments in its favor have little or no merit. Moreover, recent history provides two examples of colossal losses for taxpayers coming directly from the moral hazard that government support of the housing market inevitably engenders. Instead of looking at more plans and proposals to replace Fannie and Freddie with another government-backed vehicle, Congress should begin to look seriously at ways to withdraw the government from the housing-finance market and rely solely on the private market.

Peter J. Wallison ([email protected]) is the Arthur F. Burns Fellow in Financial Policy Studies at AEI.

Click here to view this Outlook as an Adobe Acrobat PDF.


1. David Mason, "Savings and Loan Industry (U.S.)," available at http://eh.net/encyclopedia/article/mason.savings.loan.industry.us (accessed September 20, 2010).

2. Ibid.

3. See, for example, Andreas Lehnert, Wayne Passmore, and Shane M. Sherlund, "GSEs, Mortgage Rates, and Secondary Market Activities" (working paper, Federal Reserve Board, Divisions of Research & Statistics and Monetary Affairs, Finance and Economics Discussion Series, Washington, DC, September 2006), available at www.federalreserve.gov/pubs/feds/2006/200630/200630pap.pdf (accessed September 20, 2010).

4. Ibid., 23.

5. Mortgage Bankers Association, "National Delinquency Survey," available at www.mbaa.org/ResearchandForecasts/ProductsandSurveys/ NationalDelinquencySurvey.htm (accessed September 22, 2010).

6. Andreas Lehnert, Wayne Passmore, and Shane M. Sherlund, "GSEs, Mortgage Rates, and Secondary Market Activities," 34, note 5.

7. The following seven proposals, all involving some form of government support, are representative of many more: Donald Marron and Phillip Swagel, "Whither Fannie and Freddie? A Proposal for Reforming the Housing GSEs," e21, May 24, 2010, available at www.economics21.org/commentary/whither-fannie-and-freddie-proposal-reforming-housing-gses (accessed September 20, 2010); James B. Lockhart, "Private Sector Should Take Over GSE Role," American Banker, September 14, 2010; Mortgage Finance Working Group, "A Responsible Market for Housing Finance" (white paper, Center for American Progress, Washington, DC, December 2009), available at www.americanprogress.org/issues/ 2009/12/pdf/housing_finance.pdf (accessed September 22, 2010); House Financial Services Committee, Testimony of Rick Judson on Behalf of the National Association of Home Builders, 111th Cong., 2d sess., April 14, 2010, available at www.house.gov/apps/list/hearing/financialsvcs_dem/judson_4.14.10.pdf (accessed September 20, 2010); House Financial Services Committee, Testimony of Jack E. Hopkins on Behalf of the Independent Community Bankers of America, 111th Cong., 2d sess., April 14, 2010, available at www.house.gov/apps/list/hearing/financialsvcs_dem/hopkins_testimony_4.14.10.pdf (accessed September 20, 2010); House Financial Services Committee, Statement of Michael D. Berman, CMB, Chairman-Elect, Mortgage Bankers Association, 111th Cong., 2d sess., March 23, 2010, available at www.house.gov/apps/list/ hearing/financialsvcs_dem/berman_testimony.pdf (accessed September 20, 2010); and House Financial Services Committee, Testimony of Anthony T. Reed on Behalf of the Housing Policy Council of the Financial Services Roundtable, 111th Cong., 2d sess., April 14, 2010, available at www.house.gov/apps/list/hearing/financialsvcs _dem/reed_4.14.10.pdf (accessed September 20, 2010).

8. The Treasury proposal was quite small, allowing covered bonds to form only 4 percent of the liabilities of a depository institution. This limitation might have been imposed to placate the FDIC, which does not like the idea that some of the assets of a bank might be placed beyond its reach in the event of the bank's insolvency.

9. United States Covered Bond Act of 2010, HR 5823, 111th Cong., 2d sess., July 22, 2010. The bill was considered in committee on July 28, 2010.

10. Deborah Solomon and Nick Timiraos, "New Fees Weighed for Mortgage Industry," Wall Street Journal, August 24, 2010.

11. A good summary of the Danish system can be found in Linda Lowell, "Can a Danish Import Fix U.S. Housing Finance?" HousingWire Magazine, June 2010, 32-38.

12. Larry Kudlow, "Barney Frank Comes Home to the Facts," GOPUSA, August 23, 2010, available at www.gopusa.com/commentary/2010/08/kudlow-barney-frank-comes-home-to-the-facts.php#ixzz0zdCrWpCY (accessed September 20, 2010).

Also Visit
AEIdeas Blog The American Magazine
About the Author


Peter J.

What's new on AEI

AEI Election Watch 2014: What will happen and why it matters
image A nation divided by marriage
image Teaching reform
image Socialist party pushing $20 minimum wage defends $13-an-hour job listing
AEI on Facebook
Events Calendar
  • 20
  • 21
  • 22
  • 23
  • 24
Monday, October 20, 2014 | 2:00 p.m. – 3:30 p.m.
Warfare beneath the waves: The undersea domain in Asia

We welcome you to join us for a panel discussion of the undersea military competition occurring in Asia and what it means for the United States and its allies.

Tuesday, October 21, 2014 | 8:30 a.m. – 10:00 a.m.
AEI Election Watch 2014: What will happen and why it matters

AEI’s Election Watch is back! Please join us for two sessions of the longest-running election program in Washington, DC. 

Event Registration is Closed
Wednesday, October 22, 2014 | 1:00 p.m. – 2:30 p.m.
What now for the Common Core?

We welcome you to join us at AEI for a discussion of what’s next for the Common Core.

Thursday, October 23, 2014 | 10:00 a.m. – 11:00 a.m.
Brazil’s presidential election: Real challenges, real choices

Please join AEI for a discussion examining each candidate’s platform and prospects for victory and the impact that a possible shift toward free-market policies in Brazil might have on South America as a whole.

No events scheduled this day.
No events scheduled this day.
No events scheduled this day.
No events scheduled this day.