The Biggest Losers Redux: F&F Delenda Est
There is a viral rumor that Fannie and Freddie will unilaterally reduce the principal amounts owed on underwater mortgages.
This is generating a lot of criticism, and talk of “bailout.” Yes, there is probably a bailout element to it, and the political timing would be highly suspect. But to the extent these mortgages are underwater, F&F are eventually going to eat losses. Any such program, if rumor is fact, will not be entirely a bailout. The program would be a mixture of political payoff and recognition of reality.
But even the rumor and the credence many people give it emphasizes a point I’ve tried to make repeatedly about F&F. Namely, that those who strive mightily to absolve F&F of any responsibility for the housing boom and bust refuse to come to grips with the GSE’s massive losses. The losses already recognized are huge. The rumor is essentially an acknowledgement that the losses to come are, er, huger.
Those who defend F&F tiresomely point to statistics about the measured characteristics of F&F’s portfolio, e.g., the percentage of subprime and Alt-A. They argue that F&F’s participation in such lending was small, compared to non-GSEs.
But the ultimate measure of risk and credit quality is realized performance on the portfolio. F&F’s huge losses demonstrate that the credit quality of the portfolio was actually quite low, that the value of the portfolio was quite sensitive to housing prices and/or concentrated in overpriced regions, or both. Most likely, both. In other words, the huge losses say far more about the credit quality and housing price risk of the portfolio than analyses based on characteristics measured ex ante. Especially inasmuch as there is reason to believe that F&F fudged these measurements (and isn’t fudging one of Fannie Mae’s main activities?) in order to increase the amount of credit extended to riskier borrowers in the most bubbly and overpriced markets, while maintaining the pretense that it was continuing to maintain underwriting standards.
The proof of the pudding is in the eating. We are now eating F&F’s losses. They demonstrate, quite forcefully, that their brand of pudding was rotten. Going on and on about statistics allegedly demonstrating the quality of the ingredients doesn’t mean squat if the first bite makes you puke.
The rumor relates to the consequences of past actions. Sunk costs are sunk. What is a mistake is to ignore the causes of these past actions, and to repeat the mistakes going forward. Fannie Mae, however, is falling back into bad habits:
The Washington Independent has an excellent story today about an ongoing housing programme in the US that is almost breathtaking in its stupidity.
Known as “Affordable Advantage”, it involves a kind of partnership between Fannie Mae, the government-sponsored enterprise, and the housing finance agencies of individual states. Here’s the explanation from the Washington Independent:
Before the recession hit, these housing finance agencies, known as HFAs, issued tax-free bonds and used the funds on programs to encourage developers to build in underserved areas and to support single-family mortgages. When the financial crisis hit, private companies — leery of the collapsing housing bubble and freezing mortgage market — no longer wanted to buy the HFAs’ bonds.
Then Fannie Mae entered the scene, agreeing to buy these bonds under the assumption that the homeowners had strong credit histories. In exchange, the HFAs said they would buy back loans that went delinquent. And this is how the programme now works (also from the Washington Independent, emphasis ours):
Now, qualified homebuyers in the three states pioneering Affordable Advantage do not need to put down the 3.5 percent minimum down payment required by the Federal Housing Administration, or much of a down payment at all. They can get 100 percent financing — a loan as big as the purchase price of the house — for a 30-year, fixed-rate mortgage — a vanilla mortgage. The deal includes a program to help homebuyers if they become unemployed, lowered fees and there is no requirement that the homebuyer purchase mortgage insurance.
The only requirement is a $1,000 downpayment. Perhaps you can already see where this is going.
Yeah, we can see where it’s going because we’ve already been there. “Groundhog Day” is not a business plan.
But what’s the one thing that the “financial reform bill” didn’t do? That’s right: it didn’t do a damn thing about F&F.
So if the rumor about F&F’s treatment of existing mortgages turns out to be true, that will be less worrisome than the things that they continue to do and are likely continue to do going forward. The only way to prevent that going forward is to put these monstrosities down, with extreme prejudice.
[…] This post was mentioned on Twitter by Craig pirrong, Craig pirrong. Craig pirrong said: Updated my SWP blog post: ( https://streetwiseprofessor.com/?p=4141 ) […]
Pingback by Tweets that mention Streetwise Professor -- Topsy.com — August 6, 2010 @ 10:54 am
I realise that this is hopelessly naive but I’ve always wondered why someone (this is the naive part….who?) didn’t buy up all those CDOs at 10 cents on the dollar, amalgamate the tranches back into complete pools and then go and offer the mortgagees 50% reductions in amount owed?
The loss has already been acknowledged in the price of the bonds themselves. And buying at 10 cents and then settling for 50 cents looks like pretty good business to me.
Yes, yes, I know, wasn’t ever going to happen…..
Comment by Tim Worstall — August 7, 2010 @ 5:57 am
Tim–Back at the apex of the crisis, I proposed what I called the “Humpty Dumpty Solution”, which is based on the same intuition as yours: take the pieces and put them back together again, a la Humpty Dumpty. My proposal involved government coercion, alas, not something that made me comfortable, but it seemed better in comparison to many of the other proposals mooted at the time, eg, nationalizing banks, bailouts forever, etc. You can type “Humpty Dumpty” in the search bar to find the posts.
The main problems with a pure arbitrage type solution like you propose are (a) the information problem of knowing where all the pieces are, and (b) the severe asymmetric information problem, a problem that was particularly acute in the fall-winter 08-09. I concluded some sort of mandated reconstruction was required to overcome the coordination problem. In a nutshell, I proposed requiring all holders of CDOs, etc, exchange their securities for an interest in a corporation that would re-amalgamate all the underlying mortgages. Since the underlying mortgages would be easier to value, and the corporation’s equity would be a homogeneous claim on those underlying mortgages, I argued that the process would improve price discovery and liquidity, and reduce the liquidity discounts that were devastating balance sheets.
There would inevitably be distributive effects, as the determination of how many shares of the corporation a given CDO would command inevitably would be noisy given the state of the market and the complexity of the instruments. I argued, however, that any windfall losses that one party suffered would be considerably, and perhaps more than entirely, offset by the increase in value achieved by detoxifying these securities.
And yes, it was naive. I floated my suggestion to a friend, at the time a Fed governor, who sort of patted me on the head in a patronizing way.
Hi Professor,
You distort the facts by leaving out hard #’s behind Fannie’s paper losses. You consider them “huge” and a sign of Fannie’s true credit quality when you probably haven’t even bothered to look at there 10-Q or k’s. Using words like “recognized” and “realized” in the wrong context leads me to believe you also don’t know much about accounting.
Lets start with the facts.
Fannie has accumulated over $60 billion in credit loss provisions to use for charge-offs for bad mortgages. They were very conservative in their loss assumptions and very quick/aggressive to build loss provisions compared to other banks. Last 10-k, they mentioned they were basically done with building of these reserves.
The amount of losses that they realized in 2009 was roughly $33 billion. 2/3rds of losses came from the acquisition of mortgages from their trusts that was held off balance sheet. When you purchase credit impaired loans/HAMPloans to be modified from the trust, you buy them at PAR but you have to book them on your balance sheet and market value. This doesn’t make sense if you plan to hold these to maturity and the losses have not yet materialized/realized. Here’s the bigger kicker . . . they had to fund these purchases using the Senior Preferred 10% dividend. Ouch!
Because of accounting changes in 2010 that affected all banks, the $2 trillion MBS trusts came on Fannie’s balance sheet at PAR. One of the benefits of this was they could borrow from the debt market at cheaper rates to fund the purchase of credit impaired/HAMP loans. To date, they’ve realized about $11 billion in net charge offs. These charge offs are due to escalate in 2010, but they have built up more than enough reserves to handle these charge offs. Charge offs in 2010-11 reflect MBS losses in 2005 to 2008 vintage. 2009 forward have lower delinquencies than historical reflecting better underwriting standards throughout the industry.
Did you know that misrepresented mortgages that were sold to fannie, can be put back to the banks that sold fannie the mortgages? Since 2009, these forced loan buybacks have been in the billions with well over $30 billion worth of loans that have been brought to question by the FHFA.
Fannie’s significant draws on the treasury can be explain as well. Lets start at the beginning of the conservatorship, FHFA ordered fannie to wipe out/write down to zero, roughly 42 billion in equity, most of it in the form tax deferred assets (Low Income Housing tax Credits) because the FHFA, taking direction from Paulsons treasury(my opionion), said Fannie wasn’t going to be profitable in the forseeable future. That forced $42 billion in senior preferred draws that amount to 4.2 billion/10% div they have to pay the treasury. They would’ve made a profit this quarter if it wasn’t for the dividend. They made a $1.5 billion draw in senior preferred this quarter to pay the $1.5 billion in dividend.
It’s also very amusing to see banks who have more exposure to subprime and alt a on their books start reversing their credit loss provisions to increase profits since the beginning of 2010. Fannie is built reserves into 2Q2010 and haven’t reversed any provisions into profit. Funny.
Here’s a rumor I heard. The private market wants fannie and freddie out of the secondary market to reprice risk and increase profits and make the secondary market less efficient. From Warren Buffett to Goldman Sachs, everyone stands to benefit from less government except the american homeowner. got to go. cutting it short.
Comment by Carl Marks — August 8, 2010 @ 8:44 am
@Carl Marks (aren’t we so clever). It is inadvisable to refer to “accounting” and the GSEs in the same sentence, except as part of a joke–preferably a sick one.
To wit, from the CBO’s recent analysis of the anticipated costs of dealing with F&F:
Heretofore, F&F have received $148 billion of Treasury support. This support reflects the writedowns F&F have made, but as CBO note, these writedowns are insufficient to reflect the true whole in their balance sheets.
To put that in perspective, this is more than double what support to AIG now totals (because AIG has repaid much of the original support), and AIG support is expected to fall further, whereas F&F’s is anticipated to rise inexorably. The last is demonstrated by the removal of the $200 billion/entity cap on assistance.
CBO attempted to determine the true magnitude of the budget impact of F&F, including the losses on subprime and Alt-A loans not held as securities, and hence (per the earlier quote) not accurate valued on F&F’s financial statements. CBO estimated that the 2009 budged impact for pre-existing commitments made prior to 2009 totaled $248 billion. Huge enough for you?
Even more disturbing, in some ways, is the fact that CBO estimates that F&F subsidized credit to the tune of $43 billion in ’09, and projects that it will subsidize credit to the tune of $98 billion 2010-2019.
The 2009-preexisting losses almost certainly reflect disproportionately the appalling quality of 2005-2006 vintage commitments, which are experiencing 4 pct (FHLMC) and 5 pct (FNMA) default rates, in contrast to pre-2005 rates in the .5-1 pct range.
The CBO estimates are at the low end of the range of some other estimates. EG, Barclays puts the loss at $500 billion, under more pessimistic assumptions about real estate prices.
As to the fact that banks are reducing loan loss reserves while F&F are increasing theirs, the point cuts exactly the opposite way as you suggest: it is a manifestation of F&F’s tell-the-truth-slowly strategy, and provides a further caution on the dangers of putting any faith in F&F financial disclosures. As if that was ever a good idea, esp. post-Franklin Raines.
Re your rumor: you’re kidding, right? Massive subsidies to “the american homeowner” are the problem, not the solution. And, if you haven’t noticed, a lot of American homeowners are American taxpayers, you have been stuck with the bill. No free lunch: is that news to you?
Getting subsidies out of the secondary market is a good thing. The subsidies distorted the pattern of investment, encouraged excessive leveraging, and, if you haven’t noticed, in the end didn’t do homeowners any favors. We need risk priced right, which is exactly what F&F didn’t do. The implicit government guarantee–made explicit–underpriced credit risk and wreaked havoc on the economy. Which is why Fannie and Freddie delenda est.
[…] and Thoma–deny, deny, deny that F&F were primarily culpable for the subprime crisis. As I said in an earlier post, Krugman et al evaluate the F&F pudding by claiming that the ingredients were A-OK: […]
Pingback by Streetwise Professor » Occupy Fannie and Freddie, Why Don’t You? — October 20, 2011 @ 12:06 pm