At the end of October 2018, Mark Calabria, who would just weeks later become the nominee to head the Federal Housing Finance Agency (FHFA), said the Trump administration was committed to ending the conservatorship of Fannie Mae and Freddie Mac. FHFA took control of the government-sponsored enterprises (GSEs) during the height of housing crisis more than a decade ago.
In order to evaluate what should be done with the GSEs, it makes sense to ask why they exist in the first place. By buying mortgages and using them to underpin marketable securities (a process called securitization), the GSEs support 30-year fixed-rate mortgages in the U.S. — making homeownership available to more people. This assumes that homeownership has value both to homeowners and to society in general. It also supports the homebuilding industry and the broader overall economy.
What are the possible solutions for untangling the GSEs from the federal government? One would be to get rid of the GSEs entirely and let banks find a way to securitize the mortgages they originate. Another would be to dissolve Fannie and Freddie and replace them with a group of private entities that would securitize mortgages much in the manner as the GSEs do currently.
The federal government could leave them as is, under control of the U.S. Treasury Department, with taxpayers either benefiting from the profits or covering the losses. Another option would be to privatize the GSEs with explicit backing from Treasury. Or, the GSEs could be privatized with no explicit backing from Treasury. Another option would be to keep them under Treasury, but wind them down slowly by steadily decreasing the size of their portfolio of purchased loans.
The GSEs bankroll around 45 percent of all residential mortgages. It serves mortgage originators well to understand the pros and cons of each potential path forward when it comes to GSE reform.
End, replace or status quo
Let’s take a look at ending the GSEs entirely. This appeals to those who believe that the government and taxpayers should not be involved in the homebuying process. It works for jumbo mortgages. Having more than two entities securitize conforming mortgages might lead to more competition and lower rates.
The downside is that banks may not step up to securitize loans. These institutions might decide that the reward (profit) is not worth the risk, both financially and reputationally. Some banks have suffered substantial reputational damage from making mortgages. Additionally, having each bank do its own thing creates a number of places where things could go wrong.
What about replacing the GSEs with a larger set of private entities? This enables the securitization of mortgages. Having more entities do this could create price competition and benefit borrowers.
This solution has its own problems. In order to attract investors and ensure an attractive price for these securities, they would have to be guaranteed by a government-sponsored entity such as Ginnie Mae. Each of these private entities also would have to be capitalized separately and regulated. Replacing the existing GSEs with several private entities, then, creates multiple possible points of failure, and the securities they issue would still be guaranteed by taxpayers.
Then there is the argument for leaving the situation as it is. This is working and uses the reliable infrastructure and guidelines created over many years. Having two entities provides a measure of choice and price competition. Investors who buy their paper do so with confidence because it is guaranteed by Treasury.
Being controlled by Treasury, however, means that the GSEs are subject to political whims. Part of the reason Freddie and Fannie were put in conservatorship to begin with was because they made too many bad loans. It was the Department of Housing and Urban Development (HUD) that mandated the GSEs loosen their lending guidelines, however.
When politics trumps fiscal responsibility, disaster can ensue. This also is counter to what FHFA nominee Calabria says the administration wants to do.
To back or not
Another option would be to privatize the GSEs with explicit backing from Treasury. Before 2008, Freddie and Fannie were owned and controlled by stockholders. They were private entities, but had public aspects. They were regulated by HUD. They had guaranteed credit lines from Treasury, and their paper, while not having an explicit guarantee from the federal government, was assumed to be backed by Treasury. (That was an assumption that proved to be correct in the wake of the financial crisis some 10 years ago.)
By 2007, Fannie and Freddie were required by HUD to show that 55 percent of their mortgage purchases involved loans to borrowers with low to moderate incomes. Moreover, 38 percent of all purchases had to be from underserved areas, usually inner cities, and 25 percent had to involve purchases of loans that had been made to low-income and very low-income borrowers.
After the crash, Freddie and Fannie lost $187 billion, which was covered by Treasury. After they were put into conservatorship, the GSEs worked to get their house in order and also raised their guarantee fees. Treasury has since recouped the entire $187 billion in losses — and nearly $100 billion more. In essence, everyone who has gotten a loan since 2011 is paying in the form of an increased interest rate or upfront points for the GSE losses and is now making money for Treasury.
Privatizing the GSEs would necessitate a legal settlement with the prior shareholders of Fannie and Freddie who feel that they are due some of the profits being paid to Treasury. This is a legal issue not a mortgage issue.
This approach — privatizing the GSEs with an explicit government backing — would essentially put things back to the way they were before September 2008. This gets rid of the awkward conservatorship relationship between the GSEs and Treasury. At present, the GSEs are cash cows for Treasury. In a sense, they are socialist entities in an otherwise capitalist economy.
Another option would be to privatize the GSEs with no explicit backing from Treasury. Under this approach, taxpayers are not legally on the hook for any losses. If losses were substantial, however, Treasury would likely have to step in anyway.
If there really were no backing by the government, mortgage rates would be higher because investors buying paper with an explicit non-guarantee from Treasury would want higher yields to offset the risk. A serious backlash would likely ensue if the government then had to step in once again to bail out entities that did not have an explicit government guarantee — an Occupy Wall Street 2.0.
The final approach to the GSE dilemma is setting the stage for their slow death. That would be phasing the GSEs out by slowly decreasing the conforming loan limit. That would be the same as ending them entirely, but in slow motion. The upside is this could give the banking system time to find ways to securitize mortgages on their own. Again, the downside is the banks would not have to do this and might simply generate fewer mortgages.
When all is said and done, only two of these GSE-reform options would likely work. These are the leave-as-is option or the option of restoring them to the way they were prior to the 2008 financial crisis. The latter is the best choice.
The way ahead
The Treasury should end the conservatorship, return ownership of the GSEs to the prior shareholders and add additional shareholders as providers of capital. Treasury would then need to acknowledge an explicit guarantee of the securities issued by the GSEs.
Treasury also would have to give back to the GSEs a portion of the profits it has collected that have exceeded the GSEs’ losses and also create a sizable contingency fund to cover potential future losses. This could be achieved by directing the current high guarantee fees into a loss-reserve fund. The guarantee fees could be reduced when the loss reserve is sufficiently capitalized.
The most important practical effect of mortgage securitization is that it allows for the 30-year fixed-rate mortgage. If there were no securitization system in place, these loans would have to be adjustable-rate mortgages (ARM) or have balloon payments.
Before 1930, most mortgages had balloon payments. In the early part of the 20th century, getting a mortgage usually required a 50 percent downpayment. The Federal Housing Administration (FHA) was created after the Great Depression. FHA then started backing 15- and 30-year fixed-rate mortgages. When the rate for a 30-year fixed-rate mortgage rose to more than 20 percent in the early 1980s, ARMs were created.
The mortgage industry exists to help cash-strapped folks finance the purchase of a home and become homeowners. A healthy system for securitizing 30-year fixed-rate mortgages is necessary to provide homeowners with predictable, manageable monthly payments until the mortgage is paid off. That’s important for the mortgage-origination business.
Restoration of the GSEs to their pre-2008 state appears to be the best way to achieve this. In addition, it is necessary that nothing causes the GSEs to relax their lending standards with respect to underwriting criteria. This includes credit scores, debt ratios and downpayments.