During negotiations over the bill to reform the Government Sponsored Enterprises (GSEs) that was recently passed out of the Senate Banking Committee, some stakeholders and policymakers argued that the better course was to pursue long term GSE reform administratively. In this commentary, Jim Parrott explains why that cant be done. Leaving the enterprises in conservatorship permanently is unhealthy for the market and unduly risky for both the enterprises and the taxpayer. Yet bringing them out of conservatorship would be cataclysmic for the enterprises and market alike. The only path available on long term reform thus leads through the halls of Congress.
An important chapter in the debate over housing finance reform closed last week with the Senate Banking Committee's passage of the Housing Finance Reform and Taxpayer Protection Act of 2014. While negotiations over that bill will continue as its supporters work to build momentum to bring it to a vote on the Senate floor, this is a useful moment to step back and clarify the options for long-term reform.
One path that has been much discussed in recent weeks is pursuing long-term reform of Fannie Mae and Freddie Mac (the GSEs or the enterprises) through administrative action. Why bother going through the compromise and risk associated with a messy legislative process, the thinking goes, when the Federal Housing Finance Agency (FHFA) and administration can work together to provide the structural reforms needed to put the system we have on more stable and healthy footing for the long term?
The appeal of that view is understandable, but it is unfortunately based on a misunderstanding of the options. For better or worse, the only path available in long-term reform is through the halls of Congress.
Bringing Them out of Conservatorship
The most popular version of what I'll call the “administrative reform” view is that the administration and FHFA should address some of the flaws of the enterprises and bring them out of conservatorship. There are two versions of this idea: bringing them out with the backstop of the full faith and credit of the government, and bringing them out without that backstop. As we will see, neither is economically viable.
Exiting with a Backstop
Those who believe that the government guarantee is critical to maintaining a stable system with broad access to mortgage credit would presumably contend that the enterprises must exit with the backstop. However, if they do, section 3.2 of the Preferred Stock Purchase Agreements between the Treasury and each of the enterprises (PSPAs) requires that the enterprises pay the Treasury a fee equal to the backstop's value. To review the agreement, see here. A fee equal to the fair value of the Treasury's $265 billion line of credit would be astronomically high, well above what the enterprises could cover in revenues over the near term. For an explanation, see Box 1.
To cover the shortfall each quarter, the enterprises would have to borrow against their line of credit from the Treasury. As the draws continued, investors would quickly see that the enterprises were on track to exhaust their line of credit and would gradually begin to price in the risk that their investments would not be protected by the full faith and credit of the government. The enterprises would have to charge more for their guarantee, to cover the growing premium demanded by investors, driving up the cost of mortgages. Demand for their guarantee would plummet, setting off a downward spiral of more revenue shortfalls, more borrowing from the Treasury, and higher-priced mortgages, leading to still less demand for their guarantee, still lower revenues, and so on.
Exiting conservatorship with the backstop of the full faith and credit of the government is thus not an economically viable alternative, for either the enterprises or the broader housing finance system.1
Box 1: The Cost of the Backstop
Determining the market value of a fee sufficient to compensate the taxpayer for the Treasury's combined $265 billion line of credit to Fannie Mae ($125 billion) and Freddie Mac ($140 billion) would be difficult. No private institution would provide a line like that to such severely undercapitalized institutions, so there are no obvious corollaries in the marketplace to base the analysis on. The government is also not simply providing a line of credit, but a level of counterparty certainty that would give investors interested in taking interest rate risk, but not credit risk, confidence that the credit risk has been removed altogether from an investment in the enterprises' MBS. Difficult though it would be to come up with a defensible fee, it is easy to see how untenable any version of the fee would be for the enterprises, given their revenues. A fee of 5 percent, for instance, would cost them more than $13 billion a year.
Exiting without a Backstop
Nor, unfortunately, is exiting without the backstop. If the enterprises exit without a backstop, two things would occur immediately, each devastating for both the enterprises and the housing market.
First, the return demanded by investors in the enterprises' post-conservatorship MBS would increase immediately and significantly. Investors interested only in bearing interest rate risk would no longer be willing to purchase the GSEs' securities, concerned that because the government no longer stood behind the credit risk, the interest-rate investors would be bearing it instead. As sovereign funds, mutual funds, and other large investors with little to no interest in bearing this level of credit risk depart the market, demand for the GSEs' securities would plummet. This drop in demand—coupled with the much higher returns that the remaining investors would demand to take the credit risk inherent in the securities—would force the enterprises to charge much less for their securities and thus much more for their guarantees. As the cost of their guarantees shot up, demand would fall dramatically along with their revenues, triggering a death spiral similar to that which would follow their exit with a backstop.
Second, the enterprises' creditors would behave much as their MBS investors would. With trillions of dollars of liability, no backstop from the government, and little to no capital cushion, the enterprises' ratings would drop, most potential creditors would balk at providing them credit, and those that did provide it would do so only at a prohibitive premium. Whether the enterprises collapse under these conditions is not the question, only when. If they were unable to use the Fed's discount window for short-term liquidity or roll over their short-term loans, it could be a matter of days not months.
Exiting conservatorship without the government backstop is thus also not economically viable, for the enterprises or for the market.
Box 2: Why the GSEs Can't Build Capital with the 10 Percent Dividend
If Fannie Mae and Freddie Mac retain the combined $5 trillion in MBS obligations they have today, in order to hold 5 percent equity against their outstanding liabilities they would need to build a cushion of $250 billion. Assuming that they continue the mandated wind-down of their portfolios to a combined $500 billion of their most liquid assets and maintain their current guarantee fee schedule, getting to $250 billion is formidable.
If, going forward, they earn $15 billion a year on their single-family guarantee business (an overly optimistic after-tax, after pay-for estimate of 30 bps on $5 trillion), $2.5 billion on their multifamily business, and $7.5 billion on their investment portfolio (an overly generous estimate of 150 bps on a portfolio of $500 billion), and pay $8 billion in overhead ($3 billion in operating expenses and $5 billion in credit losses), then they would earn $17 billion a year. This is short of what they need to cover the 10 percent dividend owed to the Treasury, which today would be about $20 billion.
The only two ways they can increase their revenues significantly would be to raise their guarantee fees dramatically or increase the size of their portfolios dramatically. Both of these steps are deeply problematic—substantively and politically—but even if they took them, the steps would not be sufficient to generate enough profits to create the needed buffer anywhere near quickly enough.
Take guarantee fees as an example: if the enterprises were to increase them so dramatically that the average guarantee fee on their entire book approached 40 bps—an enormous increase given that it would need to average out over the existing $5 trillion book already priced in at 25 bps—this would give them $22 billion in profits, or $2 billion after they paid the dividend. If they put all of that profit toward building their capital cushion, every year, it will take the enterprises the better part of a century to build a capital cushion of 5 percent.
An exit without a backstop is also virtually impossible under the terms of the PSPAs. Under section 2.5, to bring the backstop to an end the enterprises would need to pay the Treasury back the entirety of the $188 billion they have borrowed, a likely insurmountable hurdle, given their modest profits going forward (see Box 2). And even if they could pay off the obligation, under this same section of the agreement one of three additional things must also occur to end the backstop for either enterprise:
- the enterprise uses up their $265 billion line of credit with the Treasury;
- it is put into receivership and liquidated; or
- it pays off or completes all of their outstanding liabilities, which means paying off all of their debt-holders and either prepaying its outstanding MBS or allowing their 15–30-year terms to expire.
Each of these raises its own mix of prohibitive economic and political issues, and, when coupled with the requirement that the enterprises pay back the Treasury for the draws to date, make it virtually impossible for the enterprises to break free from the support of the government, absent an act of Congress. The PSPAs were in part intended to give future MBS investors confidence that the government would not remove its backstop of the enterprises or their securities, and the agreements did so by making it practically impossible for the government to remove the backstop.
Leaving Them in Conservatorship
Because the enterprises cannot be brought out of conservatorship, we turn to leaving them in, permanently. Few advocate for this as a viable alternative, but it is worth recalling why. It would leave us with a quasi-nationalized duopoly in which the two institutions that dominate the mortgage market lack adequate incentive to serve it efficiently or effectively. They would face no competitive pressure because of insurmountable market advantages: the government backstop, access to cheap funding, and ownership over much of the market's core infrastructure, to name but a few. And although they have a duty to serve the market, it is much less robust and comprehensive than that required of other government-backed monopolies, like public utilities, which are given dominance over key sectors of the economy.2 Facing neither competitive pressure nor a broad fiduciary obligation, they are uniquely poorly positioned to determine who in the nation should have access to a mortgage and on what terms.
A second problem with leaving the enterprises in conservatorship permanently is that the institutions have almost no capital cushion to protect them from downturns in the market. Without a cushion, they will have to draw from the Treasury when the market inevitably softens again. This exposes the taxpayer to significant risk and means that the enterprises' regulator will eventually have to push them to raise their guarantee fees and tighten credit to protect against future draws, which will constrain demand when it is most needed, further softening the market and exacerbating the enterprises' own losses.
To leave them in conservatorship, then, policymakers would have to let the enterprises build up capital. But there are only two ways to do this, and both would take too long to provide the needed check against the risk of a downturn.
Both alternatives would require the parties to amend the third amendment of the PSPAs, which converted the quarterly dividend that each enterprise owes the Treasury from a payment equal to 10 percent of their draws upon the Treasury to date (including the Treasury's initial $2 billion commitment), to one equal to all of the quarterly profits of each enterprise.
In the first alternative, they would simply reverse the change, reverting to the prior 10 percent dividend. If the parties did revert to the prior dividend, however, it would take an extraordinarily long time for the enterprises to build up the needed capital. If one backs out from their recent earnings the sources of revenue that are drawing to a close in the coming years—legal settlements, tax reversals, and the profits on the declining portfolios—it becomes clear that the GSEs' profits going forward will be modest. If one then accounts for the 10 percent dividend that they would pay the Treasury, they would have little if any profits to put toward building their capital, leaving them with a perilously thin cushion for decades. For an explanation, see Box 2.
Under the second alternative to build capital for the enterprises in conservatorship, the Treasury and FHFA would amend the PSPAs to suspend the dividend altogether, temporarily, allowing them to build capital up to a certain level after which the dividend would be reinstated. This would no doubt be a quicker path to building a capital cushion, but it too would take entirely too long. Using the assumptions in Box 2, it would take 15 years for them to build the capital needed to get to a cushion of 5 percent.
For all of these reasons, leaving the enterprises in conservatorship is not a viable alternative for long-term reform either.
To recap, leaving the enterprises in conservatorship permanently is unhealthy for the market and unduly risky for both the enterprises and the taxpayer. Yet bringing them out of conservatorship would be cataclysmic for the enterprises and market alike. We are, as it were, in a box.
This, of course, does not mean that that the FHFA can't provide important reforms of Fannie and Freddie. They can. They can take steps to ensure that the enterprises function more effectively, at less risk to the taxpayer, and to greater benefit to the mortgage market. FHFA Director Mel Watt mentioned many such reforms in his May 13 speech (For the speech, see here, and for an analysis, see here.) What the FHFA and the administration cannot do is overhaul the enterprises in a way that makes the system sustainable over the long term.
For that, we have no choice but to turn to Congress.
1. Just how bad this would be for the broader housing market depends on one's view of the resiliency of the FHFA, the private label securities market and portfolio lending. Even under the most optimistic views of these segments, however, transitioning away from the enterprises under these circumstances would be deeply destabilizing.
2. There is an argument that such a robust and comprehensive duty to serve the market is actually embedded in the various statutes that define the scope and conditions of the enterprises' charters, even if it is never actually fulfilled.
Copyright © May 2014. The Urban Institute. All rights reserved. Permission is granted for reproduction of this file, with attribution to the Urban Institute.
The Urban Institute is a nonprofit, nonpartisan policy research and educational organization that examines the social, economic, and governance problems facing the nation. The views expressed are those of the authors and should not be attributed to the Urban Institute, its trustees, or its funders.
The Housing Finance Policy Center's (HFPC) mission is to produce analyses and ideas that promote sound public policy, efficient markets, and access to economic opportunity in the area of housing finance.
We would like to thank The Citi Foundation and The John D. and Catherine T. MacArthur Foundation for providing generous support at the leadership level to launch the Housing Finance Policy Center. Additional support was provided by the Ford Foundation and the Open Society Foundations.