But what about the “old” losses we haven’t recognized yet? Don’t they become “new” when we get around to lifting up the rug?
Mere scaling arguments show that the assertion in the headline below can’t possibly be right. Over the last two years (one fifth the period for this latest forecast) the GSEs in conservatorship have already lost three times that estimate. In the seven days from the 6th to the 15th of this month foreign central banks dumped an amount of their Agency Debt holdings larger than that maximum estimated loss amount for the period starting two weeks from today and running for forty fiscal quarters. Fannie & Freddie have guarantee exposure to about $5.6 trillion ($5,600 billion) of existing mortgages. Wouldn’t that imply over 99-cents to the dollar recovery in the next 10 years for that principal and interest representing half the nation’s real estate book? [UPDATE: maybe not — see Skymutt2’s insightful comment below]
On Wednesday CalculatedRisk blog offered, “I think there are two key problems for the housing market: 1) the excess supply of existing housing units, and 2) negative equity.” Surely that combination will blow the F&F losses past the limit in just a few quarters. After all, if an objective outlook could be as rosy as Reuters’ headline why is there essentially no private mortgage market today?
Reuters (9/16 ’10): “Fannie, Freddie could cost govt $53 billion through 2020”
Mortgage finance giants Fannie Mae (FNMA.OB) and Freddie Mac (FMCC.OB) could cost the government $53 billion through 2020 or save the government as much as $44 billion, depending on the accounting principles used, the non-partisan Congressional Budget Office said on Thursday.
The Reuters report is missing the word “new,” which features very prominently in the CBO’s Director’s Blog Post on this issue and the image of their letter to Rep. Frank (PDF) that they have on their site.
Here’s a relevant snippet from the blog:
In its August 2010 baseline projections, CBO included an estimated $53 billion in costs for new mortgage guarantees that Fannie Mae and Freddie Mac will make over the 2011–2020 period.
However, the last paragraph in the blog contains this rather amazing text:
The Administration takes a different approach to showing the impact of Fannie Mae and Freddie Mac on the federal budget. It treats the GSEs as separate from the federal government, and records the cash transactions between those two GSEs and the federal government. That is, it shows the payments the government makes to those entities when it purchases preferred stock, less the dividends Fannie Mae and Freddie Mac pay to the government. CBO estimates that those cash transactions will result in net receipts to the government of $8 billion over the 2011–2020 period, reflecting additional costs for more cash infusions from the Treasury in the near term (2011 and 2012) and dividend payments from the GSEs to the Treasury that will exceed cash infusions in subsequent years. That budgetary treatment, however, does not reflect the governmental nature of the GSEs’ activities, nor does it capture the full cost of the risks associated with them.
I’m wondering if that last bit is the weasel signaling that they are ignoring the $5.6T piece. Now in the letter we have the language:
Therefore, CBO includes the cost of the entities’ mortgage guarantees and portfolio investments in its baseline projections of the federal budget. That is, the mortgages owned or guaranteed by Fannie Mae and Freddie Mac are treated as loans or loan guarantees of the federal government. For the entities’ new guarantee commitments and portfolio purchases, CBO projects budget outlays equal to the estimated subsidy inherent in the commitments at the time they are made.
So is the absurdly low figure for future losses to Fannie and Freddie per se based on the assumption that all the past losses to the Enterprises and losses subject to the new US implicit guarantee on future losses have already been socialized and thus aren’t losses of the GSEs?