While the Wall Street reform bill passed this summer did not directly address the GSEs, it nonetheless affected future prospects for restructuring Fannie Mae and Freddie Mac. As I wrote last month, the bill created a series of impediments to any timely revival of the private-label MBS market, making the housing markets highly dependent on agency-backed funding. As a result, GSE reform will be limited by the need to avoid short-term disruptions that would freeze financing for the housing markets.
A timely report recently released by the Federal Housing Finance Agency on the Enterprises' Financial Condition is useful in understanding their components. The report stated that the single-family guaranty business (the business line most central to their activities) accounted for 73% of the $226 billion in combined capital reductions for Freddie and Fannie since the end of 2007. This means that the majority of losses taken by the GSEs stemmed from their core activity of providing loan guaranties and not from their retained portfolios. Also notable was the composition of the guaranteed loans resulting in losses. While "nontraditional" loans accounted for a "disproportionate share" of the credit losses, losses on "conventional" loans were nonetheless significant, accounting for roughly one-third of the roughly $166 billion in capital reductions.
These facts have a number of implications to the restructuring debate. Stronger credit standards alone cannot be counted upon to eliminate risks to the guaranty business; in addition to the above-noted losses on conventional loans, the post-2006 experience demonstrated that all loan products are vulnerable to home price declines. Therefore, taxpayers will continue to be exposed to potential losses if the government backing is maintained unless borrower LTVs and Enterprise capital levels are significantly increased. Moreover, benefits that result from immediately liquidating the GSEs' $1.6 trillion or so retained portfolios will not be worth the potential disruptions to the MBS market. A potential solution might be to allow the portfolios to run off over time, with only loans ineligible for securitization purchased through the cash window added to the portfolio.
Most importantly, proposals to privatize the GSEs beg the question of whether the mortgage market can operate without government involvement. I think the prospects are doubtful; given the significant obstacles facing the revival of the private-label market, mortgage financing will require a strong government presence for the foreseeable future.
Attempts to restructure the GSEs should also take the nature of the MBS markets into account. While not technically "assets" of the GSEs, the agency passthrough market is highly valuable to their operations. The market facilitates a variety of different types of transactions; it integrates the separate issuance and secondary trading functions, it allows for trading of both generic securities (i.e., TBAs) and specified pools, and it creates an efficient funding mechanism through the trading of dollar rolls.
The efficiency of this market is not something to be taken for granted; consider the many failed attempts that the Securities Industry and Financial Markets Association and its predecessors have made to create a TBA market for ARMs. The problem is that even minor changes to the GSEs will likely disrupt the passthrough market to some unknowable degree. This could result in significantly higher mortgage rates and put the already fragile housing recovery at risk.
Unfortunately for would-be reformers, the environment was much more conducive to fundamental reform of the GSEs when issuers had non-agency securitization options at their disposal. Given the housing market's dependency on GSE-backed financing, the efforts to reform Freddie and Fannie run the serious risk of inadvertently weakening the housing market. The need to preserve the continuity of the TBA markets will limit the options available for fundamentally restructuring the GSEs.
Bill Berliner is a mortgage and capital markets consultant based in Southern California. His Web site is www.berlinerconsulting.net.