The GSEs and the Secondary Mortgage Market—Three Questions I Wish I’d Been Asked During the Hearing
On February 11, 2026, I testified before the House Financial Services Committee in a hearing titled “Homeownership and the Role of the Secondary Mortgage Market.” I was the only witness who argued that Congress should move America away from the existing housing finance system.
The existing system is dominated by federal guarantees of all kinds, including those in the secondary mortgage market through Fannie Mae and Freddie Mac. Prior to the 2008 financial crisis, only an implicit federal guarantee backed the securities of these government-sponsored enterprises (GSEs). But since 2008, they have both been in government conservatorship.
During these past 18 years, several members of Congress have offered proposals to “fix” the system by making those implicit federal guarantees explicit. The logic behind these proposals is highly flawed.
Fannie Mae and Freddie Mac were government-sponsored corporations, not purely private companies. Their federal backing was widely known and acknowledged prior to the 2008 crisis. The bond market has always referred to their bonds as “agency debt,” a reflection of the (correct) belief that the federal government would back the companies (or at least their securities) if they were in financial trouble. (See footnote 22 of this testimony for statements made in Congress during 1968.) The 2008 crisis demonstrates that the GSE system provides too much leverage, too much risk-taking, distorted prices, and system-wide fragility. It clearly does not provide the safety and stability that GSE proponents repeatedly claim.
Unsurprisingly, very few members of the committee asked questions about shrinking the role of the GSEs, much less about shrinking the broader role of the federal government. And while I very much appreciated the opportunity to testify, there are quite a few questions I wish members had asked. In the spirit of furthering the debate over the appropriate federal role in the housing finance system, here are three of the questions I had in mind, along with the answers I would have offered.
1. Is it necessary to have GSEs and federal guarantees in the mortgage market?
Anyone taking on financial risk would love for the federal government to step in and say, “Don’t worry, if anything goes wrong, we’ll cover your losses.” The housing finance industry is not unique in this regard; many of its participants enjoy federal backing and want even more. But this desire for federal backing simply does not justify providing federal guarantees. Nor does this desire outweigh all the negative consequences associated with providing those guarantees. Objectively, this federal backing is not necessary, only desired by those with something to gain.
More broadly, if it were necessary to have government-sponsored businesses and to provide federal guarantees, then it would imply that private markets do not work. The problem here, of course, is that we know private markets do work.
Thousands upon thousands of financial instruments exist without any federal guarantee at all, including both mortgage insurance and mortgage securities. Similarly, countless businesses function very well without any sort of government sponsorship. In most cases, competing with the federal government makes it more difficult for private companies to satisfy their customers, and the mortgage market is no exception. Private companies have less incentive to insure mortgages (or mortgage securities), for instance, when the federal government does it.
Thus, for GSEs and federal backing to be necessary in the mortgage market, there must first be some unique problem with the mortgage market compared to all other areas of the economy. The metric normally used for this sort of problem is whether there is a market failure, and the housing market doesn’t pass that test. In other words, there is no market failure in housing that justifies widespread guarantees or government backing. Some will surely disagree with this position, but the fact that some people think housing costs are “too high” is not a market failure. (Question #3 elaborates on this issue.)
For roughly 50 years prior to the 2008 crisis, the US homeownership rate hovered around 64 percent—approximately where it stands today. The explicit shift to higher GSE use in the mid-1990s, with the goal of raising the ownership rate to 70 percent, failed spectacularly. The two housing GSEs, Fannie Mae and Freddie Mac, imploded in 2008 and have been in government conservatorship ever since. In the meantime, taxpayers have been forced to provide hundreds of billions to prop up the GSEs, even with the Federal Reserve purchasing more than $3 trillion of the companies’ mortgage-backed securities.
Factoring in the economic losses associated with the 2008 recession, it’s much easier to argue that the GSEs should no longer be a part of the US housing finance system. Regardless, the burden of proof should fall on those who believe GSEs and federal backing are necessary. Without a market failure, meeting that burden is exceedingly difficult, if not impossible.
As a result, many supporters claim that GSEs are needed to ensure “liquidity” or “stability” in the secondary mortgage market. These arguments also fail, though, because the GSEs were a primary cause of instability and illiquidity leading to the 2008 financial crisis. The truth is that there is no outcome that can only be achieved with GSEs or an explicit guarantee—private enterprise can provide the same outcome.
2. Why are US policymakers so fixated on the 30-year mortgage?
Many advocates of maintaining the existing GSE system argue that eliminating Fannie Mae and Freddie Mac would imperil mortgage financing in the US. They’re especially worried that it would mean the end of the 30-year fixed-rate mortgage (FRM). This argument regarding the 30-year FRM is weak for several reasons.
First, there is no legal requirement that ties the GSEs to the 30-year FRM. Changing economic conditions and various government policies instituted over many decades account for the prevalence of 30-year FRMs in the US. Without the GSEs, people would still be free to issue 30-year FRMs. In fact, the 30-year FRM currently exists in the US without any form of government backing, even outside of the GSE system. These (nonconforming) loans generally charge higher interest rates than those purchased by the GSEs, but GSE-purchased, 30-year FRMs generally charge higher rates compared to 15-year FRMs and adjustable-rate mortgages.
This “higher interest rate” critique is even stranger because the predominance of the 30-year FRM means that US borrowers typically pay higher interest rates than borrowers in many developed countries. Yet, in many cases, the US homeownership rate is lower than it is in other countries. (See Table 1 for a few examples.)
Perhaps even stranger is that members of Congress have attacked the idea of using 50-year mortgages. According to some (even at the February 11 hearing), these longer-term mortgages too heavily burden borrowers with debt. It is not at all clear, though, that a 30-year FRM does not also “too heavily burden” borrowers with debt. Indeed, the interest cost on a 30-year FRM is more than twice that of a 15-year FRM. Moreover, without rapid home price appreciation, borrowers who use 30-year FRMs run a higher risk of having little to no equity in their home for many years.
All these problems are exacerbated by low down payment 30-year FRMs, a trait shared by most government-backed long-term FRMs. Put differently, 30-year FRMs place a high debt burden on borrowers, and that’s risky, especially for someone with lower income or uneven earning potential. From a wealth-building standpoint, it is incredibly risky for lower-income individuals to place all their future financial hopes on a single, highly leveraged asset. Indeed, many people would be better off renting and investing rather than taking on a 30-year FRM. If anything, the housing system should be geared toward a more diverse set of loan products, not any single FRM product.
3. What should we do about the current housing crisis?
Many people, including members of the committee and witnesses at the hearing, accept and repeat that the US is in the middle of a drastic housing crisis. It is characterized, supposedly, by a housing shortage and high housing prices. The best thing that members of Congress can do to address this situation is to stop referring to it as a crisis.
It is true that in the aftermath of the COVID-19 pandemic, the US experienced a much higher rate of inflation than normal. The prices of many goods and services, including housing, rose rapidly. The median US home price, for instance, jumped from less than $350,000 to a peak of almost $450,000 in 2022.
However, by October 2025, prices were about $70,000 lower than the peak in 2022. Currently, the number of weeks of wages needed to cover the full price of a home is at its lowest point since 2020, and it’s now even lower than during much of the 2010s. In October 2007, it took 324 weeks of wages to afford the full price of a home. In October 2025, that figure was 312. In fact, across several categories, median housing cost as a share of income has been flat or decreasing since the pandemic.
Of course, these figures represent national averages, and some local housing markets, in various districts throughout the country, still have elevated housing prices. Some of those local districts need to reform their permitting and regulatory processes. In those markets where regulatory restrictions and regulations make housing too costly and keep the supply suppressed, local officials and residents must work to solve the problem. Still, the best policy for the federal government is to stay out of those local markets. A blanket federal solution is likely to worsen both those local markets and the broader US market.
Members should also carefully question the estimates of the so-called housing shortage. These estimates vary widely, from about one million to 20 million. That range itself is enough to question the usefulness of those estimates. A closer look shows that these figures typically rely on highly subjective inputs, such as the desired housing stock and vacancy rates. A simple analysis of the sensitivity of these models to various assumptions shows that small changes in these inputs can easily reverse a model’s estimate from a shortage to a surplus.
More broadly, there is not a national shortage of housing in America. A shortage is when buyers and sellers can’t buy or sell at the prevailing price, resulting in people being left without a place to live. It is true that many people currently find prices “too high,” but that fact alone does not equate to a shortage. Most people view the price of Porsche sports cars as “too high” as well, but it would make little sense to characterize the market for Porsches as in some kind of crisis. It would be stranger still if Congress were to start a national effort to force Porsche to build twice as many cars to push prices down, thus enabling more people to buy Porsches.
Conclusion
Since at least the 1960s, members of Congress have used various versions of a “housing crisis” narrative to justify expanding the federal government’s reach into housing. Yet, after decades of expansion, members still claim that more federal backing is needed to solve a housing crisis. It is long past the time for Congress to take a different approach to housing policy, one that pares back federal involvement.
America does not need government-sponsored corporations, especially not in the housing market. Nor does it need the federal government to provide mortgage insurance or to guarantee mortgage securities. If the federal government remains so embedded in the housing finance system, Americans will have plenty to remain unhappy about.