Analyst: Commercial Banks Are Undervalued [Housing Tracker]

Subprime Banking Fallout 

An Acid-Tongued Maverick Keeps Bankers on the Edge of Their Seats. “Ladenburg Thalmann analyst Richard X. Bove, one of the most outspoken analysts tracking banks… remains negative on the prospects for investment banks like Goldman Sachs (GS) and Merrill Lynch (MER). But he has upgraded several commercial banks this year. He thinks banks are healthier because of the new capital they have raised and because investors have pushed their stocks too low. But, once again, he is pushing against the grain — and he has yet to be proved right: “The fact is that banks are too cheap.” (NY Times, Aug. 4^th)

Merrill's Confession May Mark the Bottom. “Merrill Lynch last week agreed to sell certain collateralized debt obligations that were notionally valued at $30.6 billion for $6.7B, or 22 cents on the dollar. There was an unmistakable whiff of desperation: Merrill will even provide 75% of the financing, and the firm will try and raise $8.5 billion by selling even more shares. Liquidating worthless securities on its book… puts immense pressure on Merrill’s peers to follow it into the confessional… Traders have begun bracing for a next big wave of markdowns… Long-exhausted financial investors hope this might usher in the final purge-and-cleanse phase necessary if the financial system is to heal.” (Barron’s, Aug. 4^th)

Ambac Terminates Citigroup CDO Deal. “Bond insurer Ambac Financial Group (ABK) paid $850 million to Citigroup (C) to terminate a guaranty contract tied to $1.4 billion in troubled mortgage-related securities… Bond insurer [stocks rose], as investors anticipated more cancellations of similar contracts… helping insurers clear out some of their losing positions… The CDO in question was backed by securities issued by other CDOs, and was highly likely to default. Ambac had previously agreed to cover interest and principal payments on the $1.4 billion security… Ambac [will] book a paper gain of $150M because it previously wrote down the contract's market value by $1B.” (WSJ, Aug. 2^nd)

Suntrust Buys Failed First Priority Bank. “SunTrust Banks (STI) acquired $214 million in insured deposits and $42 million in total assets from failed First Priority Bank, according to the FDIC. Atlanta’s largest bank, and one of the country’s 10 largest, will operate First Priority Bank’s six branches in the Bradenton market for 90 days, then close them. The branches will re-open Aug. 4 for customers with insured deposits. “This is an efficient way to add clients and deposits in an attractive long-term growth market,” said Barry Koling, SunTrust spokesman. Koling said the assets SunTrust acquired were primarily cash.” (Atlanta Business Chronicle, Aug. 1^st)

$625,000 House On A Street Wrecked By Subprime Loans? “In November, Wells Fargo issued a $289,275 mortgage for 920 W. Camile to an investor who had purchased the home at a foreclosure auction. In January, after the house was spruced up, Wells Fargo (WFC) issued a $500,000 mortgage to the new owners… At a time when America's biggest financial institutions are reporting billions of dollars in losses from bad bets on risky mortgages, why would a blue-chip bank like Wells Fargo extend so much credit on a street where comparable homes are selling for $300,000? Paul Leonard, director of the Center for Responsible Lending: "It suggests… that not all the problems have been wrung out of the mortgage market today.” (MSNBC, July 26^th)

 Seeking Alpha's Housing Tracker is a collection of housing-related excerpts from various sources, grouped by topic. Feel free to post any interesting links on the subject in the comments section below.

Get Seeking Alpha's housing market coverage by email -- it's free and takes only seconds to sign up.

Comments

[tagthatstock:]

lol,,,,,,, and yet they have huge negative eps when you look past the headline 'earnings per share' 'revenues (but no profits) growth',,,,,,,, its no wonder we are in this mess, the media,anals and the bankers cant do simple biz math.

2008 Aug 05 08:23 AM

[JasonC:]

The ones who can't do business math are the permabears and the end of the world trade herd.

You can't lose money as a bank with 5% spreads. Banking is all about the spread. They have losses now because they entered a ton of business at narrow spreads in the 2003-2006 period. But right now they can get all the funds they'd ever want at 3% or less, and they can lend them to A corporations at 7%, or government backed paper at 5.5%.

All you have to do to value them properly is look at their asset size and expect the ROA to revert to the long term mean in a year or three. Then for the weaker ones you have to leave some allowance for dilution in the meantime - present shareholders may not own the whole bank 3 years out. The more an individual bank has already raised and reserved, the less than forward dilution is going to be.

The system isn't going to work without profitable banks, and it is going to work. Spreads cure everything, the solution is already baked in. They just take time, as the higher earnings from them chomped their way through past mistakes.

The error the bears all make is to focus on the balance sheet only, or at most the current earnings without averaging them over full cycles. When they aren't just chasing the headlines. Commercial banks are franchises with funding cost advantages that always come back. They are just cyclical is all. People who can't be bothered to take a 7 year average of ROAs shouldn't touch cyclicals, but anybody who can add can see the banks were sells at 2006 prices but are buys at these prices.

2008 Aug 05 09:34 AM

[BobNyr:]

Banks with real estate exposure are in big trouble. Here is why... I was looking at buying condos in DC recently and made several offers on "short sales". This is where the real estate is listed for less than what the bank is owed for the loan that was made to the original owner. In each case, the bank would not respond or delayed the response. In other words, they did not want to sell the property even though nobody was paying the mortgage and the property was going to seed. I got the impression that banks are intentionally delaying selling these properties because when they do sell them, they need to recognize the actual "hit" on each such property. These hits will then flow through into the income statement and affect the earnings. I think banks are intentionally delaying such transactions so that the effect on reported earnings is smoothed out instead of taking a large hit at once. When these short sales eventually occur, their prices will then affect real estate appraisals and this will have a reduction effect on real estate prices. My suspicion is that this delaying tactic will go on until the end of the year so that banks can then report losses in the following year instead of the current year.... It is all about the "reported earnings" and the bonuses that the executives of the banks get based on the reported earnings.

2008 Aug 05 11:35 AM

[JasonC:]

Um, everyone knows that banks will lose money on some mortgages. The question is, can anyone do math?

No, they aren't nefariously delaying loss recognition, they just don't have the processing capacity to deal with so many loan workouts at once. That is why they are hiring everyone they can and paying them to perform workouts for them.

The rate of loans going into default on prime credits has increased to all of 2.5%. The spread on prime credits is 2.5% over their cost of capital. If they recovered nothing they'd still break even with spread that wide. They aren't recovering nothing.

The problems all stem from (1) prime mortgages written at tiny spreads and the rates lock in via derivatives at the time or (2) deadbeat mortgages with default rates well into double digits - 10-15% on alt-As and 35% on junk paper - where the spreads didn't cover that rate of default.

How much does a bank lose on a default? They expected something like 20% counting workout costs, but the reality these days is more like 50%, and if costs are high enough or resale hard enough, it can go still higher. But there is some recovery, of course. They get houses that aren't worth zero.

The math of it is, spread times 1 minus default rate (that is the profitable part) minus default rate times (100 minus recovery rate) - that being the loss part. With the spread part being per year and the rest causing loan exit and so being one-off.

Take prime mortgages at a 3% spread (e.g. 3% funding cost via CDs or whatever, and 6% mortgage rate), and a 2.5% failure rate (triple the long run average, but seen right now). They earn on the good ones .975 * 3% = 2.925% of total book size. Suppose the losses on prime loans run 50% when they default, then that is 1.25% of loan book, and the net is positive 1.675%. Push the recovery rate down to 30% and the losses rise to 1.75% of loan book, and they are still ahead 1.175%. Prime loans even in distress conditions, pay. The distress conditions cause low short rates on the funding side and thus keep the spread wide, and that is more than enough to cover even steep losses on the modest portion of prime loans that default.

But consider a subprime with a spread 3% higher (higher loan rate) thus 6% overall, but with a 35% default rate. The good portion of the subprime book earns 0.65 * 6% or 3.9% of total amount written. But the bad portion at 50% recoveries costs 17.5%. That still won't lead to 80% losses, only 13.6% losses. If the recoveries are only 30% it leads to 20% losses on the written book. The much steeper writedowns being taken on such loans reflect the banks conservatively projecting that they are going to be that bad every year over their expected lives of five years or so. By which time 90% of the loans will have defaulted and the book will have evaporated.

Loans to deadbeats don't pay, even at moderately higher spreads. They made them in the expectation that the losses on non-performing loans would be moderate, moderate enough that the spreads on the performing portion could cover them. That was false.

But it won't make loans on prime credits losers at wide spreads. It isn't the losses on non performers that matter, it is the spread on performers and how many perform.

FWIW...

2008 Aug 05 12:59 PM