Stories >> Political

Arpit Gupta: Fannie and Freddie’s Final Makeover

With the U.S. Treasury’s recently released plan to revamp Fannie Mae and Freddie Mac, an important piece of government-sponsored enterprise (GSE) reform is falling into place. Over a decade has passed since the 2008 subprime mortgage meltdown and the bailout of Fannie and Freddie, so you may be asking: What took so long? The answer is that the Treasury plan—developed with input from the Federal Housing Finance Agency and the Department of Housing and Urban Development—reflects the legacy of steady administrative reforms at the GSEs that have lowered the systemic risk of these institutions.

The Preferred Stock Purchase Agreement (PSPA) regulating the Treasury’s bailout of the GSEs called for a sharp reduction in the portfolio holdings of Fannie and Freddie. Though these GSEs are commonly known for their securitizing business, at their peak prior to the crisis they held an investment portfolio totaling 20% of the entire outstanding stock of mortgage-backed securities. These enormous holdings were largely the product of the GSEs leveraging their implicit bailout guarantee, and their resulting low cost of funding, in a form of regulatory arbitrage. The resulting subsidized and expansive portfolio helped to fuel the housing bubble, even if it was not the only factor, and added to substantial systemic risk in the financial system. Because of the PSPA, this portfolio has been wound down to a third of its formal level.

The introduction of credit-risk transfer (CRT) notes has additionally substantially de-risked Fannie and Freddie’s portfolio. These derivative structures are tied to Fannie and Freddie’s credit losses, and ensure that private companies bear much of the first loss in mortgage-backed securities, leaving Fannie and Freddie in the position of insuring only catastrophic losses, rather than other routine mortgage defaults. Through 2018, Fannie and Freddie have insured a total of $4 trillion in mortgage balances against credit risk through a variety of risk-transfer tools. The CRT program has been successful both in reducing taxpayer exposure to mortgage losses, as well as in establishing a market price of mortgage risk which can assist in other reform efforts going forward.

An important missing piece in the administrative reform plank of the agencies was the role of capital buffers themselves. Prior to the crisis, Fannie and Freddie only needed to hold 0.45% of capital against their mortgage guarantee business. This extremely low capital buffer allowed the agencies to inaccurately claim inflated returns on equity for Fannie and Freddie’s shareholders and pay out high executive compensation. The PSPA did not immediately address this issue, as Fannie and Freddie’s earnings were “swept” back to the Treasury to compensate the government for the expense of the bailout.

The federal government’s most recent announcement addresses this issue by allowing the agencies to retain $45 billion in earnings as a reserve and begin rebuilding their capital base.

Though a dizzying array of legislative proposals envisioning wholesale reform have come and gone, these steady administrative reforms have resulted in stealthy but effective reform of the GSEs which have lowered their systemic risk and costs to taxpayers without materially affecting the performance of mortgage markets.

This backdrop is essential to keep in mind in evaluating the Treasury’s recent plan for a larger-scale reform of the GSEs. While the two sets of proposals call for a variety of legislative changes which would require congressional approval; many are realizable through administrative changes alone and would further the long-term goal of ensuring a safe and sustainable model for housing finance.

The two main planks of the proposal are initiatives to expand the supply of private capital, and steadily withdraw the presence of the GSEs themselves. Private capital would be expanded by increased support for offering the credit support guarantee to private players. This would allow other mortgage originators to access the successful CRT program, and see the lower mortgage risk reflected in their capital requirements. The plans also envision greater competition by new private player entrants, who would have access to the federal default guarantee under standardized terms. Though it is unclear how many companies would ultimately be interested in this option, increasing the number of possible competitors is a valuable goal which would serve to lower the dependence of the mortgage financial system on these oligarchic giants.

Meanwhile, the GSEs would also see, under this proposal, a smaller mandate. Market participants have complained for years of unfair treatment through the qualified-mortgage (QM) patch, which allows loans purchased or guaranteed by Fannie and Freddie to comply with weaker regulations. In particular, these mortgages have frequently involved lower down payments (higher loan-to-value) and higher debt-to-income levels. This meant an unequal regulatory playing field, in which riskier origination simply flowed to the federal government.

The reform proposals call for eliminating this QM patch, which would limit the Agencies’ involvement in loans with less money down and higher debt-to-income levels. It would also reduce support for cash-out refis and second homes, which are harder to justify based on the agencies’ mandate for making primary homeownership more accessible. Additionally, support would be cut back for multifamily apartment financing; for which there is an active non-agency securitization market. The combination of these loan segments would narrow the federal government’s intervention into more narrowly defensible market segments, while allowing those mortgage needs to be met by private lenders.

Obviously, a key concern with the removal of the federal government footprint, and these efforts to revive private mortgage origination and securitization, is the legacy of the financial crisis. While people are too quick to minimize the role of Fannie and Freddie, particularly on the investment side, there were many problems with non-Agency securitization in the eve of the financial crisis.

However, it is important to remember that the failures here were often regarding incomplete securitization. For instance, the table below shows the residual holders of mortgage product pre-crisis. Strikingly, despite the supposed promise of securitization; the financial sector was left holding large amounts of mortgage products. It was ultimately losses on these real estate-linked investments that depleted the capital of financial institutions and contributed to subsequent drops in credit availability during the financial crisis.

Holders of mortgage debt, 2008. Source: Lehman Brothers Report (cited in “The Financial Meltdown: Data and Diagnoses” by Arvind Krishnamurthy, 2008.)

Especially important were specific financial structures such as asset-backed commercial paper and collateralized debt obligations (CDOs), which held mortgage bonds through short-term funding structures. These failures suggest that it is the risk-bearing capacity of the end holder of mortgage product, and the financing for these holdings, that are critical to financial solvency; not just securitization by itself. The success of the covered bond securitization system in Europe, as well as a range of other asset-backed securitization in the U.S. also suggests that private securitization is not infeasible as such, but does require prudential risk management and effective regulation.

Another important concern going forward is an entity not as much addressed by the reform—Ginnie Mae, which securitizes mortgages from the FHA, VA, and other sources. In recent years, Ginnie has grown to be even larger than Fannie, and is the home of lending from borrowers who put very little money down and have high mortgage payments relative to income. The bulk of the lending that gets securitized by Ginnie comes from dodgier originators (as highlighted in a recent Washington Post article article). The Treasury proposed reforms for Fannie and Freddie will likely have the effect of shifting even more risky mortgage origination to Ginnie. This agency, too, needs an effective overhaul.

Arpit Gupta is an assistant professor in finance at NYU Stern School of Business.

Click to Link

Posted: October 15, 2019 Tuesday 07:42 AM