August 21, 2009
High & Low Finance
By FLOYD NORRIS, New York Times
Banks are now losing money and going broke the old-fashioned way: They made loans that will never be repaid.
As the number of banks closed by the Federal Deposit Insurance Corporation has grown rapidly this year, it has become clear that most of them had nothing to do with the strange financial products that seemed to dominate the news when the big banks were nearing collapse and being bailed out by the government.
There were no C.D.O’s, or S.I.V.’s or AAA-rated “supersenior tranches” that turned out to have little value. Certainly there were no “C.D.O.-squareds.”
Staying away from strange securities has not made things better. Jim Wigand, the F.D.I.C.’s deputy director of resolutions and receiverships, says banks that are failing now are in worse shape — in terms of the amount of losses relative to the size of the banks — than the ones that collapsed during the last big wave of failures, from the savings and loan crisis.
The severity of the current string of bank failures shows that many of the proposed remedies batted about since the financial crisis erupted would have done nothing to stem this wave of closures. These banks did not get in over their heads with derivatives or hide their bad assets in off-balance sheet vehicles. Nor did their traders make bad bets; they generally had no traders. They did not make loans that they expected to sell quickly, so they had plenty of reason to care that the loans would be repaid.
What they did do is see loans go bad, in some cases with stunning rapidity, in volumes that they never thought possible.
The fact that so many loans are souring is a testament to how bad the recession, and the collapse in property prices, has been. But looking at some of the banks in detail shows that they were also victims of their own apparent success. Year after year, these banks grew and grew, and took more and more risks. Losses were minimal. Cautious bankers appeared to be missing opportunities.
As the great economist Hyman P. Minsky pointed out, stability eventually will be destabilizing. The absence of problems in the middle of this decade was taken as proof that nothing very bad was likely to happen. Any bank that did not lower its lending standards from 2005 through mid-2007 would have stopped growing, simply because its competitors were offering more and more generous terms.
Take the recent failure of Temecula Valley Bank, in Riverside County, Calif. For most of this decade, it grew rapidly. Deposits leapt by 50 percent a year, rising to $1.1 billion in 2007, from less than $100 million in 2001.
That growth was powered by construction loans, on which it suffered virtually no losses for many years. By 2005, loans to builders amounted to more than half its total loans — and to 450 percent of its capital.
Temecula appeared to be very well capitalized. But virtually all that capital vanished when the boom stopped.
When the F.D.I.C. stepped in last month, the bank had $1.5 billion in assets. The agency thinks it will lose about a quarter of that amount.
Across the country, at Security Bank of Bibb County, Ga., the story was remarkably similar. Its fast growth was powered by construction loans, although in this case the loans mostly financed commercial buildings, not houses. When those loans went bad, what had appeared to be a well-capitalized bank went under. The F.D.I.C. estimates its losses will be almost 30 percent of the bank’s $1.2 billion in assets.
In both of those cases, to get another bank to take over the failed bank, the F.D.I.C. had to agree to share future losses on most of the loans. That is one reason the agency’s estimates of its eventual losses could turn out to be wrong. In the best of all worlds, the loss estimates would be too high because the economy and property prices recover rapidly. But if the recovery is slow, the losses could grow.
In either case, the F.D.I.C. may soon need to seek more money to pay for failing banks. It could seek that cash from the Treasury, where it has a line of credit, or it could seek to raise the fees it charges banks.
So far this year, the F.D.I.C. has closed 77 banks, and there almost certainly will be more on Friday, the agency’s preferred day for bank closures. Last Friday there were five. Not since June 12 has there been a Friday without a bank closing. By contrast, there were three failures in 2007 and 25 in 2008.
Of the 77 failures in 2009, the F.D.I.C. could not even find a bank to acquire eight of them. Of the other 69, the agency signed loss-sharing agreements on 41.
By contrast, the agency found acquirers for all of the 25 failed banks in 2008, and had to sign loss-sharing agreements for just three of the banks.
“Loss-sharing” is something of a misnomer. In practice, the vast majority of the losses are borne by the F.D.I.C. Typically, it takes 80 percent of the losses up to a negotiated limit, and 95 percent of losses above that level.
Although the losses on current failures stem mostly from construction loans, it is possible that commercial real estate will be the next big problem area. Losses in that area were growing at the Temecula bank, although its portfolio was relatively small.
During the credit boom, loans on those properties became easier and easier to get, on more and more liberal terms. Unlike residential mortgages, commercial real estate loans typically must be refinanced every few years. With rents and values down in many areas, that will not be possible for a lot of buildings, and some owners are just walking away from their buildings.
Two years ago, when the subprime mortgage problems began to surface, Washington took great comfort from solid balance sheets, which regulators thought meant the banks could easily weather the problem.
Last year, we learned that the regulators, like the bankers, did not comprehend the risks of some of the exotic instruments dreamed up by financial engineers. This year we are learning that the regulators, like the bankers, also failed to understand the risks of the generous loans that the banks were making in the middle of this decade.