Susan Gates, a long-time executive at the Federal Home Loan Mortgage Company (Freddie Mac), has written a riveting account (The Days of Slaughter) of her years at the company, leading up to and through the financial crisis of 2007-2008.  As you may recall, the two major Government Sponsored Entities (GSES) Freddie Mac and older and large sibling Fannie Mae, guaranteed or owned more than one half of the outstanding residential mortgages at the time of the crisis, and were “bailed out” to the tune of a $181 billion capital injection by the US Treasury.

Susan describes the history of Fannie and Freddie, and housing policy in the US more generally, beginning with the Great Depression of the 1930s during which 50% of mortgage borrowers defaulted.

Part of the legislative response to that debacle was to set up separate government entities to insure bank and thrift deposits, to establish standards and provide insurance for approved mortgage loans, and to create a secondary market by purchasing mortgage loans from primary market lenders.  The Federal National Mortgage Association (FNMA) was set up to purchase mortgage loans insured by the Federal Housing Administration (so-called “government” loans).  Subsequently, in 1968 FNMA was broken up into two entities – the Government National Mortgage Association (GNMA or Ginnie Mae), which took over the insurance of government loans, and Fannie Mae, which took on the function of purchasing and insuring “conventional” loans (conventional means non-government). In 1970, Freddie Mac was set up to compete with Fannie Mae.  Ginnie Mae is a government owned enterprise, while Freddie Mac and Fannie Mae are government “sponsored” entities (GSEs).  The difference is great in theory but modest in practice.  Loans or securities insured by Ginnie Mae are backed by the full faith and credit of the United States.  Loans or securities insured by Freddie or Fannie explicitly do not have this backing.  However, in practice market participants have acted as if there the GSEs had an implied government guarantee, and this assumption was borne out during the crisis.

Freddie and Fannie each have two primary business lines.  First, in return for a monthly insurance premium (called the guarantor fee or “g-fee”) they promise to provide timely payment of interest and principal to owners of Freddie Mac or Fannie Mae mortgage backed securities (MBS).  This is the “guarantor” business.  Secondly, Freddie and Fannie act as investors in mortgage loans and securities by building and managing large portfolios of MBS.  This is the “portfolio” business.  The implied government guarantee of Freddie and Fannie liabilities is particularly useful for the portfolio business as it enables very low cost funding.

Susan describes Freddie’s unusual corporate structure.  During most of her tenure, Freddie was a privately owned company with a publicly traded stock (symbol FRE) and yet a public mission to promote homeownership, particularly in underserved markets.  This created potential tension among top management between the goals of serving the company’s housing mission versus striving to maximize value for shareholders.  This tension takes its strongest form in 2005 as management debates the merits of purchasing and insuring new and untested mortgage products.

This sets up what I see as the key theme for the book – the ethical standards of the senior management team in place in 2005 (“New Freddie”) as compared to the ethical standards of the team that developed Freddie’s business in the 1990s (“Old Freddie”), but was largely replaced following accounting scandals in the early 2000s.  Susan strongly supports the Old Freddie team.  They served their public mission well and simultaneously built an extremely profitable business.  According to Susan, underwriting standards were based on deep and meticulous analysis of defaulted loans; great care was placed on maintaining high standards, even amidst tremendous pressure to lower standards in order to enhance home ownership.

By 2005, the combination of increasing competition from private sector securitization and increasing demands for affordable housing from HUD forced Freddie into a jam.  In order to meet fair housing goals and preserve market share, Freddie would have to purchase and insure so-called “non-traditional mortgage loans” (NTMs).  These included loans with poor credit (subprime), loans with non-standard amortization (Interest Only loans, or Pay Option loans), and loans with limited documentation.  This dilemma set up a conflict between the Chief Risk Officer, a holdover from Old Freddie, and the new CEO.  The CRO recommended not entering the NTM market because there was inadequate data to support a risk assessment.  There was no way to estimate how well these loans would weather a period of distress.  The CRO was over-ruled and fired.  Freddie, like sibling Fannie, moved aggressively into the NTM markets, just in time to run into a very stressful period through which the NTMs did not fare well.

In retrospect, Susan concludes that entering the NTM markets in 2005 was unethical.

Root Cause of the Crisis

Peter Wallison provides a different perspective in his book Hidden in Plain Sight.  He asserts that the root cause of the financial crisis was the accumulation of NTMs by the GSEs.  He agrees with Susan that key drivers of this accumulation were the HUD Affordable Housing goals.  However, in contrast to the Gates timeline, Wallison claims that the GSEs were purchasing and guaranteeing NTMs well before 2005.  To be fair, Wallison’s definition of NTM differs from Gates’ definition.  To him, an NTM is any mortgage that does not meet traditional mortgage underwriting standards of at least 10% down payment, FICO score 660 or greater and payment obligation ratio (also known as “Debt to Income” ratio or DTI) of no more than 33%.  Thus, Peter’s NTM definition includes subprime (FICO less than 660) and so-called “Alt-A” (low down payment and/or high DTI).  According to research conducted by Wallison’s American Enterprise Institute colleague, Ed Pinto, by 2007 NTMs constituted approximately 60% of all mortgages outstanding, and more than 70% of these were owned or guaranteed by the GSEs!  Wallison notes that the HUD affordable housing goals were created in 1992 and ramped up steadily thereafter.  He disagrees with Susan’s claim that 2005 was a major turning point.  Indeed, Pinto’s research shows that the GSEs had substantial subprime portfolios as early as 1999.

The problem with NTMs, as verified by Pinto’s research, is that they do not perform well in a stressed environment.  This was known by the brilliant credit researchers at Freddie and Fannie but was ignored by senior management.  Furthermore, the extent of GSE participation in NTMs was not revealed by Freddie or Fannie until after the onset of the crisis.  In Wallison’s story, if HUD had not pushed increasingly stringent affordable housing goals upon the GSEs, the overall decline in mortgage underwriting standards would have been much less prevalent. Furthermore, if the GSEs had been forthright about the extent of the deterioration in standards in the loans they owned or insured, other financial market participants would have been much less likely to invest in non-government insured (or “Private Label”) mortgage securities or to lend using these securities as collateral.

Wallison’s critics, notably his colleagues on the Financial Crisis Investigation Committee (FCIC), claim that his argument is bogus because the losses on loans behind Private Label Securities were greater than the losses on GSE insured loans.  This means, the critics claim, that the bad actors in the drama were private sector lenders and distributors of Private Label Securities.  However, in an “originate to distribute” world (where newly created loans are sold rather than retained by the lender), underwriting standards are set by the investor or the guarantor, not the lender.  Who were the primary investors?  Ed Pinto’s data shows that as of 2007, the largest owners of Private Label Securities were the GSEs!

So what happens next?  The most unfortunate part of Wallison’s story is that we have not learned the key lesson, namely that a financial system based on poorly underwritten loans is not stable.  Wallison argues that the GSEs should be wound down over time because it is inevitable that government will push for lax underwriting standards in order to boost the homeownership rate.  The GSEs continue to operate under conservatorship and remit all profits to the U.S. Treasury.  Although the Dodd Frank Act defined the concept of a “qualifying mortgage” (QM) that meets the minimal standard of ability to pay, the post-passage definition of a QM is a loan with DTI<43% and no unusual amortization feature or teaser rate.  There is no minimum down payment nor minimum FICO score for a QM.  Thus, in spite of the inglorious history there is today little constraint against creation of new NTMs.