A Mirror Up to the FHFA: Holding Washington Accountable for Fannie and Freddie’s Stress Tests
Dan Duray Oct. 28, 2010, 2:20 p.m.
More than two years have passed since Fannie Mae and Freddie Mac were brought under the conservatorship of the Federal Housing Finance Agency (FHFA). In September 2008, the weight of the government-sponsored enterprises’ housing-related losses threatened the stability of the two institutions, as well as any modicum of functioning in U.S. mortgage markets and the financial system.
At that time, policy makers may have anticipated that the government’s intervention in housing markets would run its course by late 2010. But the unprecedented and protracted weakness of the national housing market has, instead, entrenched the government’s role in buttressing residential mortgage finance.
Over the course of conservatorship, the government-sponsored enterprises’ preferred stock purchase agreements (PSPAs) with the Treasury Department have proven the most critical mechanism for stabilizing the enterprises. Through this channel, the FHFA has facilitated $148 billion in public investment in the GSEs, offsetting losses that otherwise have threatened the institutions with receivership.
Given the near-certainty of additional public investment, the FHFA last week released an analysis quantifying the GSEs’ cumulative draws under different scenarios for the housing market. Through 2013, the FHFA estimates that draws might range from $221 billion to $363 billion over the period from the onset of conservatorship through the end of the analysis period. Required dividend payments on the preferred investments account for $80 billion or more of this total.
To put these amounts in perspective, the federal government spent $188 billion in net interest on the debt in fiscal year 2010. Another $134 billion was spent on defense procurement; $21 billion, on mandated student financial assistance. We spend about $43 billion each year on our highways.
One might argue that the money used to stabilize the GSEs is structured as an investment and that dollars will be recovered. But in an era of ever more scarce federal dollars and an ever-growing set of priorities, education and roads and bridges are investments as well.
Ultimately, the problem with the FHFA analysis is not that funds are being deployed in support of the GSEs. Circumstances necessitate a government role in stabilizing these institutions and the housing market. How and why we ended up in this predicament is an argument for another day.
Rather, the challenge is in the potentially misplaced confidence that the analysis offers to anyone who might take it for granted that the estimation has been undertaken with the requisite degree of rigor.
IN EXPLAINING THE basis for the report, the FHFA offers that the “… approach taken in developing these projections is based roughly on the approach taken by the federal banking agencies last year in the Supervisory Capital Assessment Program,” referred to in the common tongue as the bank stress tests.
Across the three scenarios, which are carefully positioned as potential and not probable paths, house prices are the only dimension of variation. No explicit information is offered as to how the relationship between house prices and GSE losses is modeled, nor is there any scrutiny of the underlying house price projections.
In considering these issues, I return to my assessment last year of the basic problem with the bank stress tests, which holds even if the outcome for the banking sector has been to the upside:
“… Whether the stress tests accurately capture the outcome of the ‘what if’ scenario is of critical importance. Rather than being able to accept or dismiss the report, the absence of transparency around the inner workings of the tests themselves renders the exercise of limited relevance. Neither investors nor the public should accept policymakers’ and regulators’ assertions that the tests of real estate-related losses have been undertaken in a manner that is analytically sound.
The onus rests with policymakers to validate the tests’ conclusions and establish that they are credible, even as potential outcomes. In the most extreme case, the deterioration in the performance of mortgage pools may threaten the stability of specific institutions and, by extension, the stability of the entire financial system. … [T]he stakes are too high for us to acquiesce to an opaque determination of the tests’ robustness. …”
Policy makers and elected and appointed officials must be held accountable for some degree of transparency around their projections and forecasts. The government cannot be held to a lower standard than it would hold the private sector in this regard. The stakes are simply too high for us to take any one person or any one agency’s assessment at face value.
Barney Frank’s now-infamous September 2003 comments before the House Committee on Financial Services offer a stark parallel in understanding the need for greater public scrutiny. During the Congressional debate over regulation of the GSEs, Mr. Frank offered the following:
“… So let me make it clear, I am a strong supporter of the role that Fannie Mae and Freddie Mac play in housing, but nobody who invests in them should come looking to me for a nickel–nor anybody else in the Federal Government. … There is no guarantee, there is no explicit guarantee, there is no implicit guarantee, there is no wink-and-nod guarantee. Invest, and you are on your own. … The more people, in my judgment, exaggerate a threat of safety and soundness, the more people conjure up the possibility of serious financial losses to the Treasury, which I do not see. …”
Without greater transparency around the FHFA analysis, there is no reason to believe that its findings are any more valid or sound than Mr. Frank’s conjecture proved to be.
Sam Chandan, Ph.D., is global chief economist and executive vice president of Real Capital Analytics and an adjunct professor of real estate at Wharton.