Monday, October 29, 2007

Countrywide Profits: Difficult to See

Countrywide's surprising announcement on Friday that they expect to be profitable in 4Q and CY 2008 pushed the stock market higher and allowed financial bonds to tighten modestly. I'm of two minds on this development. Its not that I can't imagine how Countrywide could turn things around, its just predicated on a couple very big ifs.

First, its not that hard to imagine a scenario where mortgage lending profitability improves suddenly. This probably sounds hard to swallow for many readers, so remember... I'm saying its possible. What mortgage lenders have going for them is improved margins and decreased competition. Lenders with capacity to lend can pick what loans they want to underwrite, and can name their rate. I think its also safe to say that mortgage lenders will assume weak home price appreciation (HPA) when making new loans, insist on larger down payments, and be more weary of any kind of adjustable rate loan.

All this adds up to 2H 2007 and 2008 vintage loans being much more profitable than 2005-1H2007 vintage loans, assuming both were merely held on balance sheet. In other words, investing in mortgage loans should become more profitable. It is also likely that the Fed will steepen the yield curve in 2008, which should decrease the cost of funds for most banks and other finance companies, and improve lending profitability.

Of course, Countrywide has been doing more mortgage originating and less mortgage investing in recent years. It is less clear that the origination for securitization game will improve in profitability. Securitizing through the GSEs will remain perfectly viable, but not necessarily more profitable. If mortgage underwriting spreads widen, its unclear how much of that would be realized by the underwriter and how much by the GSE in the case of an Agency securitization. We can all agree that Freddie Mac and Fannie Mae's guarantee is more valuable today than in 2006, and I expect both will command a higher price for that guarantee. Plus we know that fewer mortgage loans overall will close in coming years vs. recent past. So even if the profit per loan is marginally higher when securitizing via Fannie Mae or Freddie Mac, volume will be way down.

In non-agency space, there will be similar issues. Even assuming loan margins improve in 2008, I'd expect most of that margin will be passed onto the ultimate investor. Following with my market-power theory from above, while its true that mortgage lenders have more market power than before, investors willing to buy non-Agency MBS have the ultimate market power at the moment. And I expect that will continue for a while, especially in subprime. For many young investment analysts, the Great Subprime Collapse of 2007 represents their first serious bear market in any product. They will carry this their whole career. Don't believe me? Find an analyst whose career started in 1985 or so and ask him/her about program trading.

Anyway, so the result will be that the pool of potential investors in subprime securities will be permanently smaller for a long time going forward. So based on the laws of supply and demand, the spreads on subprime bonds will have to stay wide, even if there is universal agreement that newer vintages are of better quality. Even if various improvements are made to the subordination of subprime ABS, and even if the market starts believing in AAA-rated MBS again, I still think the overall spreads on subprime deals will stay much wider than was the case in 2006.

There is another issue, and one that hasn't gotten a lot of play in the media. The highest quality subprime loans which will be underwritten in the next couple years will be mostly refinancings of ARM loans. Several lenders, including WaMu and Countrywide, have announced programs to offer below market fixed-rate loans to subprime borrowers who won't be able to afford their reset. Far from being altruistic, this is a loan modification to avoid foreclosure disguised as a refinancing. Not that there's anything wrong with that. I mean, it would seem like the best move for the bank, similar to our discussion of loan mods the other day. On top of that, there is the FHA Secure as well as several state-run programs, which will similarly skim off the best subprime loans. It may seem strange to describe a de facto loan mod as a strong borrower, but based on how these programs are being marketed, it sounds like only stronger borrowers will qualify. In other words, borrowers whose only problem is the rate reset as opposed to borrowers in more dire straights.

So back to Countrywide. I don't think they are in a good position to take advantage of higher loan margins. I think actual banks with actual balance sheets and better access to emergency liquidity are in stronger position to realize new opportunities in mortgage lending. On the other hand, if we assume that Countrywide underwrites nothing other than easily securitized stuff, and has indeed written down all its assets (including both loans and servicing rights) to their true value, then there is no reason why they shouldn't be profitable to some degree in the near future.

The big IF in the previous paragraph is the true value of their assets. By that I mean not the market value, but what those assets really turn out to be worth. We are living in a world where determining the value of mortgage assets is extremely difficult. We know that there are assets currently priced at 50 cents on the dollar which will eventually pay off in full. And we know there are assets similarly priced which will turn out to be worthless. If Countrywide has written down all their risky assets to x cents on the dollar, and on average, that's what those assets ventually pay out, then everything will be fine. If they realize x-y, then it may be several quarters before they're back in the black.

Finally, the thing that I don't like about Countrywide is their access to non-market capital. A bank can go to the Fed or to the Home Loan Banks and get emergency capital. So if WaMu or Fifth Third or US Bank or some other large retail bank were to have sudden trouble with securitization, they'd have options. Countrywide Bank is too small to consider it a realistic option to fund Countrywide's overall operation, and as we saw in August, Countrywide is subject to liquidity problems.

In terms of the market overall, both in credit and stocks, you'd hope that Countrywide delivers on their promise. If not, I think there will be a very negative reaction in both markets.

Disclosure: I own no Countrywide securities. I do own Washington Mutual and Wachovia bonds in client accounts, as well as various state housing agency debt.


Anonymous said...

Countrywide has the retail reach and existing client base, and Bank of America has the financial strength. With B of A just announcing the end of their wholesale lending division, the writing's on the wall for this merger to happen. Not to mention that B of A has an option to buy 20% of CFC already.

Anonymous said...

the merge is bound to happen. 1.2 billion in losses and countrywide's stock still goes up. that's ludicrous

Annie and Lyon Jewett said...

"Lenders with capacity to lend can pick what loans they want to underwrite, and can name their rate."

True,they will pick the loans with the highest margins. As they should.
However, I have to agree with the other comments that a merger will happen. CFC's reputation is poor but they have a massive book which B of A will aggressively profit on.

Accrued Interest said...

I'd say that the merger potential helps credit investors more than stock investors. You'd think BAC is only interested in the merger at a distressed share price. It isn't like they'll be taken out at $25.

Anonymous said...

You folks are forgetting something significant. How in the world can any lender issue new or refinance existing mortgages in markets where home price appreciation is negative? Think about it. Who in the world (buyer or lender) would take on such a commitment while home prices have further to fall?

Now it's a different game for home loans where prices are rock bottom. But there ain't that many of those opportunities around just yet and the current REO and auction markets don't look too promising.

Accrued Interest said...

I don't think negative HPA is a deal killer. They just have to assume neg HPA into the loan risk. Its really not unlike assuming a certain inflation rate, i.e., that the money the borrower is paying back over time is worth less.

What that means is every one will have to put 20% down to cover likely negative HPA for the first couple years.

Original Gangster said...

For the merger Bulls, please read:

Furthermore, Bank of America is subject to customary standstill restrictions prohibiting it from, among other things, acquiring beneficial ownership of additional voting securities of the Company, making any proposal to acquire the Company or otherwise seeking to influence control of the Company, in each case other than transactions in the ordinary course of Bank of America’s financial services business and not involving control of the Company. The Company is required to offer Bank of America the right to match the terms of any third party proposal entertained by the Company regarding a sale of the Company. Bank of America’s standstill and match rights continue until Bank of America no longer owns Company common stock and/or Convertible Preferred Securities, calculated on an as-converted basis, equal to at least five percent of the Company’s common stock then outstanding.

Anonymous said...

At 4:36 Accrued Interest wrote:
What that means is every one will have to put 20% down to cover likely negative HPA for the first couple years.

Hey, AI, that's pretty funny.

Eh, what's that? You aren't kidding?

Come on AI. Folks here in Florida are already walking away in droves from properties on which they've placed 20% deposits. Assuming the purchasers have perfect credit, could they even get mortgages at prime rates on those properties if they were so inclined?

I'm curious how this played out in LA during the 1990 real estate crunch. I understand that housing prices fell for years then even though they were only inflated by 10-20%, much lower than the current bubble.

This factor, negative HPA over a multi-year time frame with expected price drops >20% in some major markets (the Case/Shiller indexes have all the gory details), is what makes this cycle unique.

Eyal Bar said...

So based on the laws of supply and demand, the spreads on subprime bonds will have to stay wide

Well I don't know where this comes from but I absolutely DO NOT believe that supply and demand set prices. I believe that prices are set by Mr. Market's perception of value, and SOMETIMES on top of that you have supply and demand accentuate a trend, such as a very tight supply of homes in early 2005 because people were falling over themselves to buy homes. So.... ya...

The second thing I have a beef with is your claim that CFC does not have access to emreg. capital and is "too small". I think you should re-examine what took place 2 months ago with the whol BOA discount window almost-northern-rock thingy. I mean I don't claim to be an expert on this either.

Third thing: there are two worthwhile comments from Tanta on CR. (a) CFC's servicing arm is too big to fail (b) like you said, CFC is being left over with a commodity business of underwriting and originating and funding, so ROIC should face headwinds.

Anonymous said...

tddg, you're right about Paul Krugman. He's truly gone off the deep end. Here's what he said this week...

""For one thing, there isn’t actually any such thing as Islamofascism — it’s not an ideology; it’s a figment of the neocon imagination.""


""Just to be clear, Al Qaeda is a real threat, and so is the Iranian nuclear program. But neither of these threats frightens me as much as fear itself.""

Oh boy. This is what comes from supposed intellectual elites these days. Sigh.

Anonymous said...

" We know that there are assets currently priced at 50 cents on the dollar which will eventually pay off in full. "

AI : How did you come to this conclusion? Could you please describe even one that you actually believe will pay off in full?

The cause of this bubble was the frequency of giving out loans to people who had no chance of repaying.. you can't simply fit those loans to a classical historical repayment model to do more risk analysis.. because those loans don't fit classic historical risk spreads at loan origination!

I firmly believe that there are more assets priced at 50 cents on the dollar that will turn out to be worth zero than there are which are worth the entire dollar. Under 5% will pay out fully, and I really think it's intellectually disingenuous to even consider those as a viable topic.

Remember the other problem.. OA and neg-A loans from banks that are claiming (as part of this year's income) the difference between the teaser and the 30-yr fixed's monthly.. income they will never see, and will have to restate negatively after the home goes thru foreclosure.

Accrued Interest said...

Anon @8:59:

I don't know exactly how the LA cruch played out, but I feel confident that mortgage loans were available.

Look, weak or negative HPA obviously increases the risk to lenders. But mortgage loans are still available, just under more stringent terms. I know for a fact that anyone with a 700+ FICO and 20% to put down can get a loan. Mortgage production isn't zero.

Accrued Interest said...


Maybe my statement in the post was too strong. I think the odds of default are pretty low for either CFC or any of the large banks, but the scenario seems more likely with CFC.

Accrued Interest said...

Dave: I just wish Krugman would go back to being this left of center economist focused on international issues. Now he's a left of center blowhard who sometimes writes about economics.

Accrued Interest said...

Anon @12:13:

I understand that is much sexier to claim that every ABS security should really be worth zero, but that's just not reality.

If you spend time playing with the actual cash flow models (not the Monte Carlo but a model showing who gets cash when based on default experience) its honestly not hard to get to a place where more senior tranches pay off in full.

The problem is that its common that a small differences in defaults cause a "cliff diver" effect. This is where the security goes from paying off in full to paying off very little given a relatively small change in defaults.

Now, let's assume that subprime foreclosure levels come in at 20% for the 2006 and 2007 vintage. BTW, that's double the previous high in FC rates. If we assume that 20% is randomly distributed among existing MBS securities, then there will be bonds where the default levels are only 5% or so. Then there will be others where FC is 50%. The senior bonds will survive at 5% in most cases.

Now of course the 20% won't be randomly distributed. Some underwriters were more aggressive than others and we'd therefore expect a higher rate of default with more aggressive originators.