Felix Salmon

The WSJ has a fabulous "Heard on the Street" column

today, worrying about all these losses being announced by big universal banks:

Citi, UBS, Deutsche. Are they worried that the losses are so big they will damage

the banks’ franchise? No: the worry is rather that the banks might be overstating

their losses:

While there is no evidence to suggest the banks are doing anything wrong,

numerous academic studies show that discretionary valuations related to possible

losses, which are often based on estimates, are an area of financial statements

that has in the past been manipulated by executives to manage earnings.

Doubtless there is some of this going on: if you’re going to write off, say,

$2.5 billion, you might as well make it a round $3 billion (that seems to be

the standard write-off these days), thereby giving yourself a bit of cushion

against further losses and possibly a nice bit of extra upside with which to

pad future profits.

On the other hand, all of these numbers are by their very nature inexact at

best, and I’m sure that most shareholders would prefer, in the present environment,

that these banks err on the side of caution.

Of course, the banks have to maintain the fiction that all these numbers are

entirely objective and accurate to umpteen decimal places:

Deutsche Bank said it "continues to apply accounting and valuation principles

consistently with prior periods." A UBS spokesman said the bank’s markdowns

"are appropriate and follow established accounting principles and industry

standards." A Citigroup spokesman said the bank "has taken impairments

in the third quarter based on a rigorous process applying appropriate accounting

principles."

I somehow doubt that anybody outside these banks considers the quality

of their earnings reports to be particularly consistent or rigorous, so I’m

not entirely sure who the banks think they’re kidding. What would happen if

a bank decided to be honest, and simply said something like this?

"We have a lot of very hard-to-value securities on our books, we know

they’ve fallen in value, but we can’t say how far with any exactitude, so

we’re going to take a large $3 billion write-down which should be more than

enough to cover any losses associated with the debt portfolio."

I have a suspicion that the sun would still rise the following morning. Although

you can never be sure.

DealBook

asks today whether Toronto-Dominion Bank is overpaying for Commerce Bank.

They’re both customer-serviced, says RBC analyst Gerard Cassidy, but they have

very different profit margins:

The difference, Mr. Cassidy told The Times, is that Toronto-Dominion, which

operates in a much less competitive market, pampers its customers at a much

lower cost. The Canadian bank, Mr. Cassidy calculates, spends about 50 cents

for every dollar it generates. That amount is 74 cents at Commerce.

It’s true that Commerce’s home base of New York and New Jersey is less competitive

than, say, Toronto. But I’m not entirely sure what Cassidy is talking about

when he refers to dollars "generated". If he means loan income, then

he’s not really talking about the cost of pampering customers, he’s talking

instead about ability to generate loans. Here’s what Murray wrote in the comments

to my last blog entry:

CBH is great at attracting depositors. But it’s terrible at putting those

assets to work: it hates lending – at 36% its loans to deposit ratio is among

the lowest (TD’s is 65%). So it put majority of deposits in a securities portfolio

that it’s been losing money on here and there for the last 2 years. Not clever.

In other words, if Commerce’s costs remain the same, but it increases its loan-to-deposit

ratio from 36% to 54%, then – presto – its cost-to-income ratio

will fall from 74% to Toronto-Dominion’s 50%. And if TD can loan out the same

proportion of Commerce’s deposits that it does at the rest of its banks, then

Commerce’s cost-to-income ratio would fall (assuming costs stay the same) all

the way to 41%.

Now, it’s certainly possible that there’s some deep connection between Commerce

Bank’s customer service, on the one hand, and the fact that it makes very few

loans, on the other. Maybe it’s good at being nice to its customers because

it owes its customers money, rather than the other way around.

But banks buy and sell loans to and from each other every day: if Commerce

didn’t want to lend money to its customers, it could just buy loans from the

bank next door instead, and let that bank have the bad customer relationship.

Loan-to-deposit ratios are easy to increase.

The $8.5 billion acquisition

of Commerce Bancorp by Canada’s Toronto-Dominion Bank is being greeted with

something of a shrug this morning. When Commerce CEO Vernon Hill was ousted

in June, it seemed only a matter of time before the company was sold –

but analysts were expecting the franchise to get something between $45 and $50

per share, rather than the $42.37 at which this offer is valued.

There’s lots of talk

this morning about how greenbacks are cheap for Canadians these days, of course.

But Commerce’s assets are all in US dollars, and they’re clearly worth less

– in US dollars – now than they were before the liquidity crunch

of July and August.

So let it just be noted that TD is paying more than 2.8 times book value for

Commerce Bank, which is a very high multiple for a retail bank heading into

an economic downturn. The multiple comes not from enormous growth prospects:

if anything, Commerce’s home market around New York and New Jersey is overbanked.

Rather, it’s a function of the rapid consolidation of the US banking system.

Mid-sized banks are extremely attractive, these days, to big banks which are

finding it hard to grow organically and therefore feel that they need to grow

by acquisition. In other words, they’re strategic, not financial, assets –

and get priced accordingly.

So the price that TD is paying might not be much of a premium to where Commerce’s

shares have been trading. But those shares have been pricing in a possible acquisition

since June.

Last night I talked to Chris Coulthrust (not Colthrust) of Applied Cognetics,

about the NetBank

implosion. Understandably, for someone who’s just lost access to what he

thought was money safely in the bank, he’s not a happy bunny. And equally understandably

he was a bit concerned about my blog entry, because it was long on the throwaway

snark and a bit short on facts.

So the first thing to clear up is that Coulthrust is not some kind of subprime

spiv who took his morally-dubious profits and put them in NetBank. He did work

at a subprime lender in the late 1990s, long before the subprime bubble, but

he was not a founder of the company, which went bust, and in fact he left that

job with substantial debts. After many years at various technology startups,

he eventually founded Applied

Cognetics, which does have one subprime-related product, but which is making

most of its revenues from other sources.

Coulthrust also helped me find all manner of interesting information on the

FDIC website. NetBank’s total deposits came to $2.3 billion, which is less than

its total assets of $2.5 billion. That should come as some reassurance to Coulthurst

and other uninsured depositors, although of course the assets are long-dated

and illiquid and might take some time to monetize.

The scope of

the uninsured-depositor problem is not huge: NetBank had approximately $109

million in 1,500 deposit accounts that exceeded the federal deposit insurance

limit. Of that $109 million, 50% is being paid out in an immediate payment by

the FDIC (exactly what "immediate" means in this case is not entirely

clear), leaving just under $55 million in unreachable deposits in total. The

FDIC is also spending $110 million of its own money, from the Deposit Insurance

Fund, to help keep most depositors whole, and of course ING is paying $14 million

of its own money in order to get NetBank’s deposits. So by the time

all’s said and done, the losses to NetBank depositors are likey to be small.

Should NetBank’s depositors, like Coulthrust, have suspected that something

nasty was going down? If they were paying close attention, then yes. NetBank’s

share price was in the pennies for most of this year, which meant that the equity

cushion protecting depositors was very small indeed, and their only real hope

was that NetBank would be acquired by EverBank. So when EverBank abandoned

the proposed acquisition on September 17, the gig was up.

But the fact is that with $100,000 of FDIC insurance, most Americans quite

reasonably don’t spend much time worrying about how safe their banks are. And

so it’s understandable that a small and fast-growing company like Applied Cognetics,

whose accounts had much less than $100,000 in them most of the time, might not

consider a large windfall in terms of revenues to be cause for concern.

And while financial-market professionals are comfortable thinking of bank deposits

as one type of banking-industry liability, most individuals don’t consider themselves

to be lending money to the bank when they make a deposit. Rather, they think

of themselves to be simply placing their money in the bank for safekeeping and

easy access. Applied Cognetics had $100,000 in a NetBank checking account which

paid zero interest.

Interestingly, the reason that Applied Cognetics had all of its money in NetBank

to begin with was basically due to the fact that Commerce Bank, in Brooklyn,

made it so difficult for Coulthrust to open a business account there that he

eventually gave up and decided to do everything online instead. He says he went

back to Commerce Bank four times, with all the different bits of paperwork they

were asking for, but that they were never satisfied, and seemed more interested

in asking none-too-intelligent questions about his business model instead. My

guess is that they were stealth-underwriting a loan he hadn’t even asked for,

although it’s also entirely possible that they were simply incompetent.

Finally, it’s probably just as well that the legendary Robin Kelton died in

April. Kelton was a financial-industry giant, who founded not only the investment

bank Fox-Pitt Kelton but also the rating agency IBCA (now part of Fitch), and

the insurance rating agency ISI (now part of S&P). Kelton was on the board

of NetBank from day one, and he would have been devastated to see its depositors

out of pocket.

It hasn’t got a lot of headlines, but the US has now officially suffered its

biggest bank

failure since 1993. NetBank had $2.5 billion in assets, and somehow contrived

to lose more than $200 million in 2006. Deposits were insured up to the FDIC

ceiling of $100,000, and ING Direct is buying $1.5 billion in deposits for $14

million, which works out at about $1,000 $135 per new customer.

Of course, there’s always going to be someone with more than $100,000 in deposits.

But in

this case it’s hard to feel a huge amount of sympathy:

Applied Cognetics, a software development and online marketing firm based

in Brooklyn, N.Y., has about $1 million in deposits in NetBank…

NetBank customers with accounts exceeding the FDIC limit will become creditors

in NetBank’s receivership, the FDIC said Friday…

Applied Cognetics bills itself as a one-stop shop for online lead generation

and Internet development.

Colthrust and his team – who formed the company after a subprime mortgage

lender where they worked was sold – have built the 10-employee company’s sales

to more than $10 million since it was founded in 2000. Ironically, given that

lenders’ catering to subprime borrowers have led to a spike in home loan defaults,

subprime lead generation software is among the services Applied Cognetics

sells.

So Colthrust started a subprime lender, sold it, and then used the proceeds

to start a company finding leads for other subprime lenders. He also put all

his money in NetBank, which promptly went belly-up in the subprime meltdown.

There’s some kind of poetic justice there, I think.