Rosner: GSEs Probably Need $100 Billion; Freddie Mac Reported to Be Meeting Treasury

Warning: Josh Rosner is controversial in some circles. He co-authored a simply terrific paper with Joe Mason, “Where Did the Risk Go? How Misapplied Bond Ratings Cause Mortgage Backed Securities and Collateralized Debt Obligation Market Disruptions.” If you haven’t read it, do so now. If you don’t have the time, we summarized it here.

However, the esteemed Tanta is not a fan. She believes that he has gotten a tad too aggressive in tooting his own horn and taking provocative positions to get media attention (hope that accurately sums up her views).

Nevertheless, Rosner provides the logic behind his back of the envelope calculation on how much capital would be required to shore up Fannie and Freddie and it doesn’t sound unreasonable.

Some readers have howled that estimates like Rosner’s are off base, that they are not at risk of making good on their debts. They seem to miss quite a few issues:

1. The GSEs are not subject to SEC financial reporting standards. Jonathan Weil of Bloomberg has argued that if they had to mark to market and report their deferred taxes like banks do, they’d both have negative net worth.

2. Banks and insurers are required to hold equity for a reason. Shit happens, and risk is a many-headed beast. For instance, the GSEs have massive derivatives exposures and have had serious losses on them in the past.

3. Mere solvency is not an appropriate standard for the GSEs. In their role of providing/guaranteeing mortgages, they have to be able to fund at the very best rates. That means they need to be an AAA credit, or at worst, a very strong AA. Mere solvency is BBB, perhaps even BB or B. Simply being solvent doesn’t cut it.

The notion that Freddie and Fannie are undercapitalized in the absence of goverment support is hardly controversial. Jim Hamilton discussed that issue, and the potential for the markets to demand that the government make good on the implicit guarantee, at the Fed’s Jackson Hole conference last August. Nouriel Roubini predicted the insolvency of the GSEs two years ago and in a recent piece, argued that in a worst-case scenario, the GSEs would need $200 to $300 billion.

The other piece of the equation that some commentators are missing is that the debt markets are driving this crisis of confidence. The rise in agency spreads is approaching the February/March crisis levels. Worse, then there were no fancy Fed facilities in place (save the TAF); the March creations were aimed at shoring up the agency market, and that worked only for a short while. We have been told that despite very clear language in Freddie and Fannie offering documents disavowing a Federal guarantee, that major foreign investors have been told that the officialdom will stand behind the paper (yes, this is a rumor, but from a credible source).

So the issue is a very simple one. Fundamentally, this isn’t about GSE solvency. Despite all the legalese otherwise, GSE paper’s rates ALWAYS presupposed a Federal guarantee. Those tight spreads made no sense based on their stand-alone financials. Now that guarantee is being tested, and the government has blinked. Any wonder the market is spooked and effectively forcing the Feds to recant?

What makes the problem pressing is that the GSEs have large, continuing funding needs (the Journal mentioned today, as we did yesterday, that they have over $225 billion to refinance by the end of September) and depend on foreign participation. Recent reports indicate that foreign buyers have bought less at the last two auctions and central banks have also been net sellers of GSE paper.

Our view has been that GSE auctions will not fail (“everything can be solved by price”) and thus were surprised when bond market participant John Jansen (who does not have a dog in this fight) said:

Notwithstanding the relative success of the Freddie sale the agency market is still a very troubled venue. One analyst notes that central bank demand for the sector has diminished significantly since June. He said the change in appetite could not be narrowed down by geography or by the asset size of the institution.

There are some outright sellers, some who just do not add or do not replace as paper rolls off and there is a group of first time buyers. On balance, however, central banks are buying considerably less of this paper.

Some are troubled by the recent statements of Secretary Paulson that he is not eager to use his new powers. Some have extrapolated from his statements that he is only prepared to exercise his powers in an emergency. What constitutes an emergency?

Suppose we walk in tomorrow and Freddie or FNMA can not get rolled over in the discount note market. Treasury exercises its powers and the taxpayers have an ownership interest in the GSEs.

So people in the marker see failure to roll paper as a possibility. That is grim indeed. However, sentiment is volatile on the upside as well. Jansen noted today that a story at Bloomberg nixing the idea of a Treasury takeover led to a considerable tightening in CDS spreads, which should by arbitrage provide relief to funding costs.

But even if there are no failed sale a further increase in spreads (or mere persistence at recent levels) might have repercussions elsewhere (in new mortgages, if nothing else, and the foreign buyer repudiation of formerly-seen-as-impeccable US paper could have other weird knock-on effects).

And it appears that the folks at Freddie, despite Daniel Mudd’s insistence that things are fine, are meeting with the Treasury to discuss contingency plans. From the Telegraph:

Senior Freddie Mac executives are reported to be meeting with the Treasury today, potentially to get some clarity about how the government will support the company if necessary, having been forced to pay a record premium for a $3bn debt issue on Tuesday.

While there has been little commentary on these meetings, the press has taken note that the Treasury has backed off of its “we don’t need to do anything” stance. From the Financial Times:

Although it was granted new powers to extend its credit lines to Fannie and Freddie and invest in their equity last month, the Treasury has been adamant it does not expect to have to make use of the new authority.

But on Wednesday, a Treasury spokeswoman declined to repeat that assurance. Instead, she said Treasury was “vigilantly” monitoring market developments and was “focused on efforts that will encourage market stability, mortgage availability and protecting the taxpayer”.

As the FT correctly noted, this change in tone is far short of a commitment to act.

So the pressure on Paulson to Do Something is considerable (although the easy way out is to put in big orders at the auctions…..a simple finesse, although it would probably be roundly criticized). But the Journal reports that the Treasury is studying more aggressive options:

For Treasury, which is mulling how to structure a rescue of the two mortgage companies, if necessary, the ability of the pair to keep turning over their debt is paramount….Treasury is developing a series of options that it could use to shore up the companies if a market seizure forced them to intervene….

Among the options at Treasury’s disposal is to make a large capital injection that would essentially result in a takeover by the U.S. government. Such a move wouldn’t be undertaken lightly, these people said, and would be done as part of a broader plan to remake the companies. Among the issues being debated is whether to oust management and whether to treat both companies equally or devise a rescue for one and not the other.

Finally, to Rosner’s quick and dirty analysis in the Telegraph:

Josh Rosner, of consulting and research house Graham-Fisher, argues that the minimum the pair needs to raise is at least $40bn, but he believes the realistic number is more like $100bn.

His estimates are based on the roughly $3,500bn in guaranteed books of business the two firms hold, and are based on past peak losses of approximately 1.3pc in the 2002-03 housing downturn.

However, the current housing crisis is far worse than it was six years ago and the pair’s problems could be as deep as Mr Rosner suggest.

Update 8/21, 9:45 PM: Reader William R reminded us that Standard & Poor’s has arrived at a far greater cost estimate than Rosner’s:

Although few are predicting an imminent need for a bailout [of Fannie and Freddie] just yet, credit rating agency Standard & Poor’s recently placed an estimated price tag on this worst case scenario — $420 billion to $1.1 trillion of taxpayer’s money…

S&P added that saving Fannie (FNM) and Freddie (FRE, Fortune 500) might cost so much that the federal government’s AAA credit rating, the top possible rating, might even be at risk. If that was lost, then all federal government borrowing would become more expensive.