House under construction

Housing Perspectives

Research, trends, and perspective from the Harvard Joint Center for Housing Studies

Temporarily Ending the GSE Net Worth Sweep: A Limited but Important Step Towards GSE Reform

On Monday, September 30, the US Treasury and the Federal Housing Finance Agency (FHFA), the independent agency that is the regulator and conservator of Freddie Mac and Fannie Mae, the two government-sponsored enterprises (GSEs), announced a start to their much-awaited ending of the “net worth sweep.” While ending the sweep is only temporary and challenging decisions needed to make it permanent were put off until the future – very possibly until after the upcoming presidential election – this is nevertheless a momentous event as finally, after more than a decade, the first concrete step has been taken to actually implement GSE reform and end the conservatorships.

As a reminder, in September 2008 the two GSEs were placed in conservatorship – a legal status where they essentially operate as wards of the government – after losing the confidence of the financial markets that they would meet their obligations.  To restore that confidence, Treasury pledged to maintain their net worth at a positive level by injecting funds into the companies whenever needed, in exchange for senior preferred equity shares. (By structuring the support in this manner, Treasury keeps the confidence of the debt markets but also does not allow the owners of pre-conservatorship common and preferred shares to get any value from the rescue until, at least in theory, Treasury’s senior preferred is fully paid off. Treasury took an option on 79.9 percent of the common equity as well.) This was done in a legal agreement called the Preferred Stock Purchase Agreement (PSPA).

Initially, the GSEs had to pay a 10 percent dividend on the amount of the preferred shares outstanding; today that would be almost $20 billion of annual dividends on the almost $200 billion of senior preferred currently outstanding. But, at the end of 2012, this was revised – it eliminated the fixed 10 percent dividend but substituted a payment equal to all the earnings of the companies (after allowing a small capital buffer of $3 billion each). This clause, which became known as the net worth sweep, was designed to end the problem wherein the two GSEs had to take funding from the Treasury under the PSPA only to turn around and give it back to meet the 10 percent dividend requirement (which became known as “circular” payments).

This revised structure relieved the GSEs of ever having to draw under the PSPA to make dividend payments, but it also prevented the GSEs from building up capital in their own name. As a result, they continue to be creditworthy to the financial markets only because of their ability to call capital from Treasury via the PSPA at any time needed and in very large amounts.

With the release by Treasury of its Housing Reform Plan on September 5, there was significant disappointment that the plan had many generalities but few specifics, and no deadlines. But immediately thereafter, statements by Secretary of the Treasury Mnuchin and FHFA Director Calabria – both of whom must agree to any changes to the PSPA – indicated that they were prioritizing the ending of the net worth sweep so that capital building by the GSEs could finally begin (which the industry had expected to be part of the announced plan), and that they looked to complete it by the end of the month.

So, in terms of timing, they met their committed deadline. But, in terms of what was announced on September 30, the net worth sweep is only being ended temporarily, so it is just a partial step forward in ending the conservatorships of the two GSEs.

Specifically, the PSPA agreements between Treasury and Freddie Mac and Fannie Mae are being revised. This revision has two key features:

First, the capital buffer of each GSE, as described above, is being increased from its current $3 billion level to $20 billion and $25 billion, respectively, which allows them to retain an additional $17 billion and $22 billion of earnings. These amounts, probably not coincidentally, are enough so that at recent earnings levels the new caps will only be hit after the November 2020 election.

And second, Treasury will be compensated for its support to the GSEs by an increase in the value of the senior preferred shares (technically known as the “liquidation preference”) dollar for dollar for any sweep payments foregone. Conventionally, such support would be paid for via some sort of fee tied to the amount of GSE obligations being supported (e.g. the $5 trillion-plus of their balance sheets), so this unusual treatment stands out more than a bit. This mechanism is also now the third incarnation of the payment to Treasury – the first having been the 10 percent dividend requirement and the second the now temporarily-discontinued net worth sweep.

Importantly, by temporarily ending the net worth sweep in the manner announced, the Treasury and FHFA are able to avoid having to design all the required specifics of a permanent cessation. This both enabled them to move fast to meet the September month-end deadline and also to push into the future – maybe even conveniently until after the November 2020 election – making the following difficult decisions that require detailed analysis and which would also likely generate political criticism:

  • Support Fee. As Secretary Mnuchin explained in Senate hearings on the reform plan, with the ending of the net worth sweep, the GSEs instead would have to pay a fee to compensate taxpayers for their ongoing support to the companies. In the announcement, because the cessation is temporary and ongoing support by Treasury is paid for by the unusual feature that the PSPA preferred shares increase in value in an amount equal to the foregone sweep amounts, Treasury and FHFA did not need to specify the amount or structure of the fee to be paid by the companies for Treasury’s support. No matter what is eventually permanently chosen for such a fee, it predictably will be criticized as too low by those mostly concerned about the risk to taxpayers of this backstop from the federal government and simultaneously as too high by those concerned with the cost of mortgages for homebuyers.
  • Reserve fund. The Treasury also wants a reserve fund to be built up, so that if post-conservatorship the GSEs generate losses beyond what each company’s net worth can absorb, the reserve fund will then in turn absorb such losses before the taxpayer might be called on to do so. The announced temporary end to the sweep is able to avoid saying anything about the reserve fund’s size, how quickly it will be built up, or how it will be funded. Again, it is predictable that no matter what level is chosen it will at least be criticized as too little and being built up too slowly.
  • Windfall to the “investors.” By having the amount of the PSPA’s senior preferred stock increase in the exact same amount as the foregone net worth sweep payments (a conventional fee would have almost undoubtedly been a smaller amount), Treasury cannot be accused of transferring any economic value to the “investors,” i.e. the owners of the pre-conservatorship common and preferred shares. Since those investors are now dominated by institutions, such as private equity or hedge funds, that purchased their shares at steep discounts after the two companies were placed in conservatorship, anything that looks like a windfall to them will be subject to substantial criticism by virtually every Democrat in Congress, and also probably many Republicans as well.

As I wrote in the paper Treasury’s GSE Reform Plan: My Top Ten Political Economy Insights last week, inside the Trump administration are three groups which together are driving GSE reform. One of them consists of White House-centered staff who are focused on the election in November 2020, and who do not want GSE reform to become an election issue through controversial actions. The overall direction of Treasury’s plan is therefore to defer controversial decisions, maybe even until after the election. And, as listed above, many of the necessary features of a permanent cessation of the sweep would definitely generate controversy.

There are two big political economy takeaways from this announcement beyond its specifics.

First, there is an old phrase “well begun, half done.” It says in just four words how getting started on a task is one of the biggest challenges to accomplishing it, even though starting is in reality nowhere near actually doing half the work. In this case, after more than a decade, the release of the GSEs from conservatorship has been finally “begun” (it’s in the eyes of the beholder if it has begun “well”), with something actually moving forward after many years of talk but precious little action. The temporary ending of the net worth sweep represents a very small portion of the total work to be done, but, even with the avoidance of so many decisions needed to make its ending permanent, it is a giant step forward, hopefully breaking the reform logjam that has been in place for so long.

Second, when policy experts make proposals for housing finance reform, they usually present a soup-to-nuts plan (albeit almost always at a high level with few specifics). Similarly, if housing finance reform were to be undertaken via legislation, the specific legislation would have to present a comprehensive program, although many details would be left for various government agencies to fill in. But for something large and complicated, like the reform of America’s $11 trillion system of housing finance, it is difficult for Congress to be comprehensive and not produce legislation with missing or ill-considered pieces, or which creates many unintended consequences. By contrast, with housing finance reform being done via administrative means, this is all being turned on its head. What we are seeing is Treasury and the FHFA looking to take discrete steps to “move the ball down the field” a bit each time. (The announcement also included a comment that further revisions to the PSPA to implement reform would follow, implying without inordinate delay.) This is all perfectly fine, even a good thing – there is likely to be better consideration of the right way to do each step and minimize unintended consequences. So, expect this pattern to continue, with specific steps announced from time to time, adding up over some years to become a full and comprehensive program – and one more likely to work well than would otherwise be the case.