The GSE Public-Private Hybrid Model Flunks Again: This Time It’s the Federal Home Loan Bank System (Part 1)

Introduction

In November of last year, the Federal Housing Finance Agency (FHFA) published a report¹ (the Report) recommending reforms for the Federal Home Loan Bank (FHLB) System.² The report came about at a time when the FHLBs have been increasingly criticized for neglecting their housing and community development mission and instead using their subsidies and privileges to benefit their financial institution owners and executives. Written in a neutral regulatory tone, the document largely validates these criticisms via facts and analysis, and then proposes extensive recommendations for change.³  

The release of the FHFA Report has also sparked a flurry of articles, op-eds, and think tank reports either supporting or countering some or all the recommendations.  The leadership of the FHLBs,  along with various beneficiaries of their subsidies, has also begun to aggressively lobby and advocate against virtually any proposed change that it views as detrimental to its bottom line.⁴ For those who have closely followed the housing finance industry for a long time, this should evoke memories of the efforts to reform Fannie Mae and Freddie Mac (F&F) in the years before the Great Financial Crisis (GFC), and how their management teams also responded so aggressively to counter them.⁵  

The similar key behaviors exhibited by both F&F and the FHLBanks should not come as a surprise.  That’s because both are government-sponsored enterprises (GSEs). This two-part article argues that the root cause of their need for reform is a fundamental flaw in Congress’ design of GSEs.  Part 1 covers some general background information on GSEs as well as the extremely relevant precedent set by the reforms imposed by the FHFA upon F&F in reaction to the most well-known case where the public-private hybrid GSE design had quite visibly flunked. In Part 2, we will shift our focus to the FHLBs and how the GSE design flaw has similarly manifested itself.  This is important context to more fully appreciate the importance of the FHFA Report as well as many of its specific findings and recommendations.  Finally, in the concluding section of Part 2, I will propose a solution for how the fundamental flaw of GSE design can be permanently fixed.  

The GSE Background 

Over a century ago, Congress began creating public-private hybrid corporations known as government-sponsored enterprises. The GSEs were authorized by congressional legislation to perform a very specific set of activities, primarily focused on providing loans to consumers or businesses for a designated purpose, and mostly related to real estate and housing. The private persona of a GSE is that it is owned and capitalized by the private sector and operated as a for-profit company.  The public persona of a GSE, meanwhile, is that it carries out a congressionally defined public purpose mission⁶ and is given subsidies and privileges to do so.  

One of the main reasons for this approach is that GSE borrowings to fund loans don’t directly show up as federal government indebtedness.⁷ Additionally, GSE subsidies and privileges are crafted to largely avoid showing up as a federal budget expenditure, thus avoiding competing with other priorities for scarce tax dollars.⁸ The most significant privilege and subsidy that the GSEs enjoy is an implied guarantee⁹ by the U.S. government, allowing them to borrow at near-Treasury rates, lower than any regular private sector company or bank could.  They also enjoy certain income tax exemptions, although specifically which ones varies by GSE.  

By far the largest and most well-known GSEs are F&F, with combined assets of roughly $7.5 trillion today; they are commonly, if inaccurately, called “the GSEs” as if no others existed.  The oldest GSE is the Federal Land Bank System (now part of the Farm Credit System), established in 1916, to support agricultural lending, currently with assets of about $0.5 trillion.  The Federal Home Loan Bank System, established in 1932, is another GSE, sometimes known as the “third GSE,” currently with about $1.3 trillion of assets.¹⁰

With the benefit of hindsight, it is obvious that designing such a public-private hybrid would be extremely challenging. First, it is difficult to ensure that the GSEs can only use their privileges and subsidies to support their congressionally-intended mission activities, instead of them also being exploited to produce stand-alone profits that primarily benefit the GSEs’ owners and executives.  Second, it’s unreasonable to assume the design could work effectively through many decades of often unpredictable changes in markets, legislation, and regulation.¹¹  

This core flaw of GSE design was described by then-Secretary of the Treasury Henry Paulson when F&F were placed into conservatorship in 2008. He referred to¹² the tension between their profit-making and their public sector mission as “the inherent conflict and flawed business model embedded in the GSE structure,” resulting in a “conflict of interest attempting to serve both shareholders and a public mission.”  

The Pre-2008 Fannie Mae and Freddie Mac 

F&F were relatively small prior to the Savings & Loan (S&L) Crisis of 1989, having only a single-digit market share of funding U.S. mortgage originations.  But with the collapse of the S&Ls and other thrifts culminating in 1989, and the development of a securitization business model,¹³ F&F rapidly gained market share through the early-to-mid 1990s.  Their market share then reached about 45 percent, which meant they financed almost half of all dollars going into first-lien home mortgages nationwide. 

As privatized and publicly-owned companies (Fannie Mae since 1968 and Freddie Mac since 1989), the two GSEs were incented to seek opportunities for profit growth to increase their stock prices. This would, of course, benefit their shareholders, which in turn would benefit their executives through stock options and bonuses.  These factors then led to several problematic behaviors, of which I chose four notable ones to review.

• Growing profits by exploiting the implied guarantee.  The F&F core business of securitizing and guaranteeing loans essentially takes the below-market cost of the funds raised by issuing mortgage-backed securities (MBS), which reflects the implied guarantee, and passes that subsidized cost through to primary market lenders.  Such lenders are expected, in turn, to pass almost all of that low cost on to actual homeowning borrowers.¹⁴ Thus, the history is that F&F shareholders and executives didn’t unduly benefit from the cheap MBS funding, earning just a conventional level of profit on this mission-centric activity, and profits only grew in line with the broader mortgage market.

Naturally, F&F’s management teams looked to grow profits further.  But because their congressional charters, which limit their activities quite narrowly, did not allow the companies to grow in a conventional private-sector manner,¹⁵ they turned to exploiting their key GSE advantage – the implied guarantee¹⁶ – to generate earnings.  They eventually discovered what critics rightly called a loophole when they began to generate additional profit growth by issuing unsecured debt and using these funds to invest in fixed-income securities related to the mortgage markets, especially their own MBS.¹⁷ Thus, the economic value of the implied guarantee subsidy on these investments did not flow directly to mortgage borrowers but instead went to F&F shareholders and executives.¹⁸ Over time, this investment portfolio grew to just over $1.6 trillion,¹⁹ which was about double the amount then held by the Federal Reserve,²⁰ to become the largest source of profit for the two companies prior to conservatorship.  It seems unlikely such a situation was ever contemplated when the two companies were privatized by Congress.  

• Denial of the subsidy.  In response to criticism regarding this exploitation of the implied guarantee, F&F took an unusual approach in responding:  they simply denied that the implied guarantee and the subsidy that flows from it existed.  This statement was made despite the fact that government support was widely known to be real and despite the credit rating agencies specifically stating they relied upon such government support to justify the AAA ratings then given to F&F.²¹ The implied guarantee also fully proved to be real when, in 2008, the companies were rescued with no losses to the creditors of both firms. 
• More political than commercial.²² While F&F often portrayed themselves as simply competitive, commercial companies, they were not.  As GSEs, they were highly limited in their permitted activities but given a preferred market position, along with the subsidies and privileges needed to deliver a public policy benefit to the public. This was all laid out by Congress in their charters (i.e. the legislation establishing them).  They didn’t have competition in the conventional sense, but mainly competed in a limited manner with other government units, such as the Federal Housing Administration and the Department of Veterans Affairs, and to a certain degree with banks and institutional investors.  To maintain their profitability and even grow it, F&F had to be heavily political. To that end, they turned to lobbying and advocacy to fend off any possible profit-reducing limitations that Congress might impose on them. At the same time, they hoped to gain new authorities to deploy their subsidies for profit growth.   

One aspect of this political nature was their false denial of the implied guarantee.  Another was their lobbying efforts, as the two GSEs became among the most powerful lobbying interests in Washington, D.C. This included employing ex-government officials from both parties to aid their cause.  As an example of their lobbying power, in 2004 and 2005 the George W. Bush administration and the Federal Reserve joined together to get legislation passed to place a limit on the size of the F&F discretionary investment portfolios.  However, F&F launched a major lobbying effort and successfully defeated the legislation in Congress.²³ To beat the combined power of a presidential administration and the central bank was indicative of how much F&F were political power players in Congress, not just conventional businesses.

Additionally, the GSEs did not commercially operate like efficient and competitive business organizations.  Despite their technical expertise about the mortgage markets, prior to conservatorship they had very rudimentary customer service capabilities, poor technology, and inadequate risk management.²⁴ 

• Excessive executive compensation.  In 2003, Franklin Raines, the highly visible CEO of Fannie Mae at the time, received compensation of $17.1 million,²⁵ which angered critics of the company for being so out of line for a public-private hybrid. This level of compensation made him one of the highest-paid financial institution CEOs in the country.  Fannie Mae’s board justified the compensation as being in line with comparable positions, which were other CEOs of large financial institutions with similar asset size, meaning mainly the largest global banks.  But that was clearly not a valid comparison,²⁶ since these comparable CEOs ran firms with at least 10 to 20 times the number of employees, dozens of major product lines rather than just two,²⁷ operations in hundreds if not thousands of locations rather than a handful, and dealt with dozens of regulators and multiple currencies associated with offshore business, which Fannie Mae did not have.  Additionally, Fannie Mae was given by Congress a privileged market position with limited competition and its entire business model was dependent upon the implied guarantee subsidy, which generated most of its large profits.  Freddie Mac’s board took a similar approach.  To their critics, this was concrete evidence of how much F&F and their boards had lost their way.

In 2008, when F&F were put into conservatorship, the government – through the FHFA (then a newly-created independent regulatory agency) and Treasury – had the opportunity to restore F&F’s balance between mission and profit.  To that end, they consciously took measures to address the four problematic behaviors quite directly, with the first three being tackled immediately. Such measures included: 

• Shrinking the investment portfolio.  The agreements put into place to rescue F&F in September 2008 required that there be an orderly reduction in their large investment portfolios to eliminate that abuse of the implied guarantee.  Today, the investment balances are down about 90 percent, leaving only the small amount required to actually support the underlying securitization business of the companies. 
• Subsidy denial ends.  F&F went into conservatorship because their funding was drying up as market confidence in the implied guarantee that supported their borrowings was shaken at the height of the GFC.  As part of the rescue, the implied guarantee was replaced by a formal legal support agreement to regain the needed level of market confidence – which it has successfully done ever since.  As a result, the political requirement to deny the reality of the implied guarantee –and thereby admit that F&F needed government support to operate its securitization-centric business model – ceased.  Today, management at F&F will admit that government support was and is still needed for their business model to work.  
• Political is out, commercial is in. One of the terms imposed on the two companies during conservatorship is that they are prohibited from lobbying.  Any contact between F&F and elected officials is strictly controlled by the FHFA.  The two companies have no lobbyists or lobbying budget and make no political donations of any kind.  All significant political and policy decisions are taken by the FHFA on their behalf.  As a result, F&F’s management and boards are primarily focused on operating the companies to the highest standards.  In fact, an industry consensus emerged around 2016-7 that F&F have been operating far better in conservatorship in areas such as technology, customer service, and risk management than they ever did prior to 2008.
• CEO compensation at public-private hybrid levels.  In 2008, when F&F entered into conservatorship, the three reforms described above occurred immediately. However, it took three years for the FHFA, as conservator, to fully revise the approach to executive compensation.  During that time, a new public/private hybrid-consistent executive compensation system was established, which is still in place today.  The new system includes no stock options or large bonuses for the CEO, and the compensation level of their executives is somewhere between a pure private sector company and a government agency.  It well reflects how much F&F are a hybrid of the two.

In summary, the reforms that were called for by many critics of F&F prior to 2008, but which never happened, have been substantially implemented within conservatorship.  The result has been a great success, as the two companies now operate within both the letter and spirit of the charters that Congress put into place long ago, reasonably balancing mission and profit.  

In Part 2, the FHLBs will be discussed, showing how they also have lost the proper balance between mission and profit, demonstrating an almost instant replay of the problematic behaviors described above at F&F.  The FHFA Report is thus understood as an attempt by the agency to institute the reforms – mainly through regulatory actions and rulemaking – that were implemented for F&F only upon their entering conservatorship.  

Footnotes

[1] Federal Housing Finance Agency, “FHLBank System at 100: Focusing on the Future,” https://www.fhfa.gov/AboutUs/Reports/ReportDocuments/FHLBank-System-at-100-Report.pdf?utm_medium=email&utm_source=govdelivery.

[2] There are today eleven individual Federal Home Loan Banks.  They are, however, operated as a system, cross-guaranteeing each other’s debts and issuing their debt through a legally mandated single office (the Office of Finance).  In this article, I will focus on the system as a whole, referring to it variously as “the System,” “the FHLB System,” the “FHLBs,” or the “FHLBanks.” 

[3] Most of the reform recommendations can be implemented by the FHFA itself as the System’s regulator, although many will require the agency do so via an official rulemaking process.  Such a process requires a public comment period, and it can sometimes take up to a year or longer.  A few of the recommendations require legislation by Congress.  

[4] See Bloomberg’s article from Dec 20, 2023, “A $1.3 Trillion Home-Loan System Gone Astray Is Fighting an Overhaul.”  https://www.bloomberg.com/news/articles/2023-12-20/federal-home-loan-banks-why-lobbyists-are-fighting-housing-lending-reform.  

[5] Interestingly, the calls for reform of F&F proved ineffective, with virtually none happening prior to their being placed into conservatorship in 2008.  

[6] While all the GSEs will refer to their mission, actual congressional legislation is anywhere from being silent as to what exactly that is (letting the specific functional powers and limitations in the legislation define that) to being rather high-level and general (e.g., for Freddie Mac, the mission wording in its founding legislation includes phrases such as “respond appropriately to the private capital market” and “to provide ongoing assistance to the secondary market for residential mortgages”).  In no case is the mission statement specific enough that it can help prevent a GSE from using its subsidies and privileges to inappropriately pursue stand-alone profit.

[7] This means it is easier for the government to stay under congressionally-mandated debt ceilings.  

[8] The privatization of Fannie Mae in 1968 by the Johnson administration was explicitly done to take government spending off the books under the pressure of Vietnam War era “guns and butter” budget strains.

[9] The implied guarantee is defined as where the marketplace believes the government will have no choice, should a GSE get into financial trouble, but to rescue that GSE with no losses to bondholders.  Mechanisms to show that there is an implied guarantee include, among others, direct lines of credit from Treasury.

[10] The other GSEs and their established dates are (1) the Federal Agricultural Mortgage Corporation (known as Farmer Mac, 1987) and (2) the SLM Corporation (known as Sallie Mae, 1972; its GSE charter was eliminated in 1995). 

[11] The history makes clear that the various GSEs often engage in lobbying and advocacy for favorable changes to legislation and regulation, potentially biasing the outcome.

[12] See “Statement by Secretary Henry M Paulson, Jr. on Treasury and Federal Housing Finance Agency Action to Protect Financial Markets and Taxpayers,” September 7, 2008.  https://home.treasury.gov/news/press-releases/hp1128.

[13] The securitization business model – (1) buying in mortgages, (2) packaging them up into pass-through mortgage-backed securities, (3) guaranteeing the securities owners against the economic impact of credit losses on the underlying mortgages (with that guarantee backstopped by the federal government) – also solved the interest rate risk problem that felled the thrifts during the 1980s.  With respect to the government backstopping the guarantee against credit-related losses:  prior to 2008 this was done via the implied guarantee; since then, in conservatorship, it has been via a specific legal support agreement known as the Preferred Stock Purchase Agreement. 

[14] They are not expected to do so out of civic spirit.  Instead, it is the belief that competition within their industry will force them to pass through virtually all of it, thus benefitting the underlying homeowner.  

[15] For example, they could not go into a related business line, such as credit cards; they could not go international; they could not integrate forward into being a primary mortgage market lender, as they were only authorized to be secondary market firms, etc.

[16] See my article “The Role of the Implied Guarantee Subsidy in FHLB Membership,” July 20, 2020, Harvard’s Joint Center for Housing Studies, for more on the implied guarantee and how it works.  Another advantage they held was exemption from state and local income taxes.  https://www.jchs.harvard.edu/sites/default/files/media/imp/harvard_jchs_COVID_nhlb_politics_and_policy_layton_2020.pdf

[17]  This resulted in a “snake eating its own tail” economic picture.  F&F purchased mortgages, issued and guaranteed pass-through MBS on those purchases, and then bought back a material share of those same MBS, funding it with unsecured debt that carried the implied guarantee. 
              

[18] Later, upon becoming CEO of Freddie Mac, I had an analysis done of these investment portfolio profits.  It showed that substantially all the profits were due to the below-market cost of the funding (because of the implied guarantee), and virtually none to the company’s investing prowess.  
    

[19] It peaked in 2004 at just over $1.6 trillion.  See the 10-K annual reports for Fannie Mae and Freddie Mac.  The calculation is shown in the Freddie Mac 10-K balance sheet as “retained portfolio,” and for Fannie Mae two categories of assets are added up to get the equivalent calculation.  It is now commonly known as the retained portfolio, as the Freddie Mac usage was adopted by the government in setting certain limits in 2008 and after.  

[20] The Federal Reserve Bank’s combined balance sheet in 2006 was just $.81 trillion.  See https://www.federalreserve.gov/boarddocs/rptcongress/annual04/audits04.pdf.

[21] I also experienced the denial directly and personally.  In a meeting with the then-CEO of Fannie Mae (Franklin Raines), when I was working for a large bank, he very vehemently denied that any such subsidy existed despite the evidence to the contrary.  

[22] Reflecting their government and political nature as GSEs, the companies were set up to have on their boards a number of presidential appointees.  This practice was ended in the early 2000’s but contributed to their political nature prior to that.

[23]  For a summary description of this effort, see the report by the U.S. House of Representatives Committee on Oversight and Government Reform, “The Role of Government Affordable Housing Policy in Creating the Global Financial Crisis of 2008,” May 12, 2010, starting on page 20.  https://oversight.house.gov/wp-content/uploads/2012/02/20100512affordablehousingpolicyandthefinancialcrisis.pdf. 

[24] I can attest to all this personally, as this was true upon my arrival at Freddie Mac in 2012 as CEO, in comparison to what I was familiar with in the non-GSE financial institution private sector.  As but one example, Freddie Mac measured customer service quality mostly by a once-per-year survey.  Private sector banks would use everything from such surveys down to daily measures of error rates, how many customer tasks were completed in an established time allowance, and so on.

[25] He resigned in 2004 due to accounting irregularities.

[26] Having worked both at Freddie Mac and, previously, at one of the largest American banks for almost 30 years, I can also say this from personal experience.  

[27] The two product lines would be single-family mortgage guarantee and securitization, and multifamily mortgage guarantee and securitization.