The GSE Conservatorships: Fifteen Years Old, With No End in Sight
This week, September 6 to be exact, marks the fifteenth anniversary of Freddie Mac and Fannie Mae – the two government-sponsored enterprises (GSEs) - being placed into conservatorship by the Federal Housing Finance Agency (FHFA).[1] This occurred at the very height of the Great Financial Crisis (GFC), when market confidence in the value of mortgage assets and more generally in the liquidity of large financial institutions was collapsing.
I think it can fairly be said that no one at that time ever expected conservatorship to last so long, especially as Secretary of the Treasury Paulson publicly referred to it as a “time out,” with its short-term connotation.
To commemorate this anniversary, I have compiled a ten-point Q&A to explain some of the history and key events of the two conservatorships, which reasonably lead to the conclusion that they won’t likely end anytime soon.
Q1: What problem was the government trying to solve when it placed the two GSEs into conservatorship?
A: Freddie Mac and Fannie Mae provided financing for almost half of the outstanding first mortgages in America at that time. However, the financial markets began to lose confidence in them given how much such assets were then losing value versus the amount of their capital to absorb such losses. As background, the companies had long been thinly capitalized and relied upon the “implied guarantee” of the U.S. government (i.e., the market expectation that the government would not let the companies fail)[2] rather than their own capital strength to sustain the near-government quality market access needed for their business model to work.
Under the pressure of the financial crisis, in mid-2008, the market began to lose confidence in the implied guarantee given its informal and unwritten nature. As a result, the ability of the two companies to renew maturing debt obligations began to falter, which could have quickly led to insolvency. Because the two GSEs failing would have pushed mortgage values down even more as well as accelerated the collapsing overall market confidence in all larger financial institutions, the government absolutely did not want the two companies to fail.
The Bush administration, therefore, did two things. One, it placed the companies into conservatorship, an obscure legal status in which they would continue to operate, but with their regulator, the FHFA,[3] in operational control of the companies in lieu of their stockholders and boards of directors. And two, it replaced the unwritten implied guarantee with a specific written support agreement, called the Preferred Stock Purchase Agreement (PSPA), where the Treasury was legally obligated to inject equity funds into the two companies so each would never have a negative net worth. The result was a quick regaining of market confidence, and the PSPA[4] has proven strong enough to maintain their needed near-government quality market access ever since.
Q2: Why did policymakers want significant change in the GSE business model before the companies could exit conservatorship and be returned to private sector ownership?
A: Policymakers at the time believed the GSE business model had significant defects that needed to be fixed before any exit from conservatorship. While opinions on this topic vary, in my view, five core concerns needed to be addressed:
- The implied guarantee was a problematic method for government support, including that the taxpayers did not receive any compensation for providing it.
- The GSEs were historically undercapitalized.[5]
- The GSEs were able to use the implied guarantee to borrow unsecured funds – without limit and at near-Treasury level interest rates[6] – to build up extremely large, discretionary investment portfolios that became their largest source of profits. Policymakers considered this an abuse, as it was never intended for the implied guarantee to be used in this manner.
- The two companies, which were designed to be monolines, were naturally very excessively exposed to concentrated residential mortgage credit risk. This was an undesirable source of financial instability.
- The GSEs were also extremely political, using their large profits to fund very extensive lobbying and influencing activities that had stymied various efforts to reform one or more of the above-listed defects.[7]
Because Congress had set up the GSEs through legislation, the Obama administration believed it was up to Congress to develop revisions to that legislation to eliminate these defects.
Q3: Why was Congress unable to expeditiously pass legislation to correct those business model defects as a key step to ending conservatorship?
A: Many ideas for changes in the GSE business model, generally referred to as “GSE reform” at that time, were presented to Congress by the housing policy community. These came from academics, members of DC-based think tanks, members of Congress, industry associations, and more. The ideas almost always went beyond a list of specific changes to the existing two GSEs and instead proposed that they be replaced by “something else.” The suggestions of what that might be ranged from winding down the GSEs to almost nothing (the assertion being that private capital markets would replace their role), having many smaller GSEs (rather than just two) compete, having the primary mortgage lenders run the GSEs as one or two industry cooperatives, merging the two companies into a single government-owned corporation, and more. They were more or less high-level proposals rather than fully fleshed-out ones with all the needed details.
From roughly 2010 to 2016, GSE reform took the form of a search among these proposals to find a specific “something else” that could both (1) gather the broad political support needed for legislation to be passed in Congress, and (2) be workable in the very complex mortgage markets. In the end, none were successful. The proposal that got the furthest was the Corker-Warner bill (later renamed Johnson-Crapo), based upon the concept of many competing smaller GSEs. However, it never made it past Senate Banking Committee approval in 2014. In addition, detailed analysis of the proposal as it was fleshed out showed that it too was unworkable.[8]
Q4: Is there any likelihood of Congress passing legislation to reform the GSEs in the next several years?
A: It is broadly agreed, as best as I can determine, that there is no appetite in Congress, either at this time or in the foreseeable future, for devoting the considerable time and resources required to attempt full-scale GSE reform.
Q5: During the fifteen years of conservatorship, how has the business model or operations of the GSEs changed, if at all, and what has been the resulting impact on ideas for GSE reform?
A: Starting in 2012, the FHFA began to systematically correct the defects in the GSE business model while they were still in conservatorship, rather than maintaining the status quo year after year while waiting for Congress to someday pass legislation. This included:
- Developing an updated capital requirement broadly consistent with post-2008 bank regulatory requirements.[9]
- Putting strict and comparatively low limits on the size of the investment portfolios. Currently, the portfolios combined are well less than 20 percent of their pre-conservatorship $1.5 trillion amount.
- Reducing the extreme concentration of residential mortgage credit risk through developing and executing credit risk transfer transactions to spread the credit risk among institutional investors globally.
- Prohibiting lobbying by the GSEs as well as generally controlling their contact with Congress or other policymakers.
Broadly, the GSEs began to act more like normal, competitive firms: developing better customer service, becoming more sophisticated in risk management, and heavily investing in technology of all types. It is fair to say that the primary mortgage lending industry believes the GSEs operate better now than they ever did before conservatorship.
As a result of these improved GSE operations and also the fruitless search for “something else” to replace Freddie Mac and Fannie Mae, the primary mortgage lending industry’s fundamental view about GSE reform changed during the 2017 to 2019 period. The industry began to articulate that the “something else” approach, where the undefined replacement(s) for the two GSEs were in many ways starting anew and likely to generate problems of their own, was risky and would probably have a very disruptive transition and be unacceptably damaging to mortgage markets generally. Instead, the industry’s leaders began to publicly support that GSE reform should consist of the two current GSEs continuing to exist, albeit only if their significant defects were fixed. Industry also noted that nearly all of the needed reforms had already occurred during conservatorship,[10] and thus adopting this approach would be very natural and evolutionary, with little potential for market disruption. This support for reforming the already-existing GSEs instead of the more radical and uncertain “something else” approach quickly became widespread consensus as the only practical way forward.[11]
Q6: Is Congressional legislation required for the GSEs to be able to exit conservatorship with the reforms desired by policymakers?
A: Inside conservatorship, circa 2016, the concept of “GSE reform via administrative means,” i.e., one that would not require legislation, was developed. This was publicly disclosed by Treasury Secretary Mnuchin in 2019. It means that changes in the GSE business model could be made via requirements and terms specified in the PSPA agreement, which are binding on the GSEs, as well as new regulations by the FHFA.[12] Secretary Mnuchin indicated his preference that such an administrative approach to reform be undertaken only with the expectation and hope it would later be enshrined in legislation by Congress.
However, such administrative reform could not, by its nature, go as far as fundamentally changing the long-time business model of the GSEs, thus throwing aside all the more radical alternative reform proposals made over the years. This meant there would still be two GSEs named Freddie Mac and Fannie Mae and that they would still be monoline mortgage purchasers and guarantors working under their existing federal charters. They would have just been operationally reformed (as listed above, e.g., having strict and low investment portfolio limits). Naturally, this dovetailed very nicely with the new industry consensus that reforming the two existing GSEs rather than replacing them was the right approach.
Q7: What is the process for the GSEs to become properly recapitalized, as is necessary for eventual conservatorship exit?
A: The path of the GSEs being recapitalized has been fairly convoluted, although it has gotten to a relatively good place in the last few years. A very short synopsis is:
- From 2008 to 2012, the net worth of the two GSEs was wiped out by GFC-era losses. Under the terms of the PSPA, capital could then be rebuilt from earnings that were in excess of a required preferred stock dividend payment to Treasury under the terms of the PSPA. That dividend was 10 percent of the cumulative equity Treasury had placed into the companies (a combined $191 billion).[13] [DL1] [CZ2] Earnings were never good enough during this period for any large amount of net worth to be built.
- From 2012 to 2019, via an amendment to the PSPA, the net worth of the two GSEs was then reduced on a quarterly basis to eventually become near zero by having almost all profits paid to Treasury in lieu of the previous 10 percent preferred stock dividend.[14]
- From 2019 to today, via another amendment to the PSPA, the GSEs are retaining all their earnings by not paying any preferred stock dividends. This has resulted in them building up net worth and doing so quickly as pandemic-era profits have been very high. Today, their net worth, compared to the near-zero amount from before 2019, is a very considerable $111 billion.[15]
This retained capital can be supplemented, at least in theory, by issuing new common equity in public markets. This is probably not practical until conservatorship exit is nearing actual completion.
Q8: What actions has the Biden administration taken to move forward on the GSEs exiting conservatorship?
A: None. It is worth noting that it has also not taken any steps to end or modify the continuing retention of earnings, which was instituted under the Trump administration. Thus, net worth continues to grow.
Q9: If the decision was made to exit conservatorship today, how long would it take to complete?
A: It is very hard to generalize how such an exit would occur, given that nothing like this has ever really happened before. But in the most likely scenario, in my view, it will take at least three to five years. This is a roughly estimated time needed for net worth to grow via earnings retention to reach a level of capital that might be deemed sufficient for conservatorship exit and for all the other mechanics to be put into place for post-conservatorship regulation that retains the GSE conservatorship reforms made in the last fifteen years.
Also, importantly, the end of conservatorship does not mean the end of government control, as Treasury would, upon such exit, own the vast majority of the company's equity. [16] It would then take at least another several years for Treasury to sell its shares and resolve various stockholder-related issues. As a rough estimate, these two steps add up to six to ten years before the GSEs would once again truly be public-market shareholder-owned companies.
Q10: Is there any reason the conservatorships could not go on for many more years?
A: In the early days of conservatorship, government officials would refer to its “unprecedented scale and scope,” which was true enough. They would also say that conservatorship was meant to be short-term (also true) and therefore could not last a long time. However, the latter has not proven to be true, as it has lasted fifteen years to date. With the two companies operating so well, it does indeed seem that they can, at least under the right circumstances, be in conservatorship for a very extended period of time.
Looking back, Freddie Mac and Fannie Mae in conservatorship have arguably been a source of strength to the financial and housing markets more than they were pre-2008, with the companies also operating far better. Given how much time it will take for them to exit conservatorship and also then graduate from Treasury being their dominant shareholder, they will very likely be under government control for many years to come. And, as mentioned above, the Biden administration has not even begun to execute on any such exit plan. Possibly, this is because voices inside the administration support the status quo, as it enables the execution of many housing policies through the GSEs with no requirement for Congressional authorization.
It would obviously be consistent with the intended nature of conservatorship (i.e., that it not be perpetual) if a well-designed exit [17] from conservatorship and subsequent Treasury control could be developed and its implementation begun, even if it would take many years to complete. Unfortunately, there is just no sign of that being anywhere on the horizon at this time. Thus, this seemingly never-ending GSE conservatorship is likely to remain the status quo for the foreseeable future.
Footnotes
[1] There are two separate conservatorships, legally speaking, one for each of the two GSEs. Despite this, it is very common to reference conservatorship as if there was just one for the two companies.
[2] The market based the implied guarantee on many factors, including their federal charters (i.e., Congressional legislation establishing the companies), special lines of credit directly to Treasury, and the importance of the role they played in mortgage markets. Over time, other government actions indicated the opposite (e.g., requirements for their debt to be clearly marked as not guaranteed by the U.S. government). In the end, the assumption of an implied guarantee proved accurate, as in 2008 the government absolutely rescued the GSEs, and did so in a manner that protected the full value of their debt.
[3] The GSEs have had several regulators over the decades, as Congress every so often lost confidence in the regulator of the day and replaced it. The FHFA was, in fact, only created in mid-2008 and given stronger powers than its predecessors enjoyed.
[4] As with the conservatorships, there were legally two PSPAs, one for each GSE. But often they are discussed as if there were just one.
[5] It is now accepted wisdom that large financial institutions at the time were broadly undercapitalized, so that today’s requirements are considerably higher, especially for large banks. The GSEs pre-conservatorship were actually capitalized to even lower standards, supposedly to “help housing” by enabling mortgage interest rates to be as low as possible.
[6] The implied guarantee, until the GFC began, was so powerful in the marketplace that the GSEs borrowed funds at interest rates just slightly above what the U.S. Treasury paid. This was lower than even what large private sector financial institutions (e.g., large banks) could do. An analysis of profitability of the pre-conservatorship investment portfolios that I directed be done at Freddie Mac in 2012 revealed that the profits were virtually all due to the resulting below-market cost of the funding rather than any particular investing prowess.
[7] The most well-known such effort was the 2004 defeat by the GSEs of a joint push by the administration and the Federal Reserve to cap the amount of their investment portfolios.
[8] I participated at the time in efforts to analyze all the leading proposals to see which ones were, indeed, workable as promised by their supporters. Actually, none were, reflecting how incredibly complex the mortgage markets are and how strongly the two existing GSEs are embedded in them.
[9] The first version of this was completed circa 2017 for use inside conservatorship. It was later revised, and meant also to be used post-conservatorship, in 2020.
[10] This approach is often called the “utility model” because it would have GSE pricing set by the FHFA rather than the two firms themselves, much as occurs for electric and other utilities. Such utility-style price setting already happens as part of the conservatorship operations of the two companies
[11] This became known, colloquially, as the “reform rather than replace” approach.
[12] The downside of such an approach is that it would not be as durable as reform achieved via legislation, as a future administration could change it as desired.
[13] This equity injection into the two companies took the form of Treasury purchasing preferred shares, rather than common.
[14] This rather unusual capital situation aligned with the Obama administration’s view that the GSEs would be wound down on their way to be replaced by “something else,” thus not requiring any capital in the future. There were lawsuits about the legality of how this was decided and implemented. The government won all of them until just recently, when investors initially won a relatively small amount that may or may not be appealed by the government.
[15] How much net worth would be needed for GSE conservatorship exit is a matter of great debate. To be considered “adequately capitalized” under today’s regulations (which are quite controversial), it would require roughly three times the current level.
[16] Treasury does not engage in such rescue financing without proper compensation on behalf of the taxpayer. With some simplification, in the case of the GSEs, this compensation includes warrants equal to 79.9 percent of the common shares, which were awarded upon the original signing of the PSPAs in 2008; plus senior preferred shares equal to (1) the cumulative investment Treasury has put into the companies to ensure the net worth of each never went below zero, (2) since 2019, an additional amount equal to the cumulative earnings that the GSEs retained rather than what they would have had to pay to Treasury under the terms of the 2012-2019 era PSPA, and (3) some additional one-time adjustments. Today, the senior preferred shares of both companies owned by Treasury have a face value of $297 billion. It is the general expectation that, upon exit from conservatorship by administrative means, the PSPAs would continue to provide financial support to the companies and there would then be an ongoing fee to compensate taxpayers for this risk. This fee remains unknown, as it has yet to be developed or specified by Treasury.
[17] In my view, a well-designed exit means one that is both (1) fully workable and practical, with little potential for disruption to mortgage markets, and (2) conventional in its terms, avoiding unique or unusual features that could be grounds for lawsuits by the owners of the common and equity shares of the companies that have remained outstanding from pre-conservatorship days.
