Last year the banks had an easy way to juice their profits. All they had to do was allocate a little less money to loan-loss reserves -- the money they set aside to cover bad debt. As the economy has improved and defaults have slowed, many decided they didn't need as much in reserve as they did in 2003, and presto, their earnings per share would rise a few cents.
But investors who assume the profits are humming and decide to buy bank stocks could be in for a shock. In 2005 many banks won't have this profit source. Some have already pared loan-loss reserves as much as they reasonably can, analysts say. "A lot of banks may do this from time to time to meet estimates," says Brian Shullaw, senior research analyst at SNL Financial in Charlottesville, Va.
The trouble with whittling away the reserves is that as banks write more loans, they will have to replenish the reserves. Plus, if credit conditions worsen as economic growth slows and interest rates rise, they will need to set aside even more, eating further into profits.
Do a little digging, and the current numbers don't look so great. Detroit's Comerica Inc. (CMA) had one of the largest drops in its loan-loss reserves relative to total assets, according to a study of large banks' fourth-quarter earnings done by SNL for BusinessWeek. Not only did Comerica fail to add money in the fourth quarter, it also extracted $21 million from the pot. That gave it an extra $98 million in income, or 57 cents a share, that it didn't have last year. The bank beat analysts' earnings estimates by 10 cents. Comerica Chief Credit Officer Dale Greene says muted loan growth, coupled with major improvement in credit quality, justify the move.
Others, such as Citigroup (C), garnered a few extra cents from replenishing reserves by a smaller amount than before. But it was enough to help them beat analysts' earnings estimates by a penny or two. Citi Chief Financial Officer Sallie L. Krawcheck said in a Jan. 20 conference call that the reserving process was done in mid-quarter based on a mathematical formula. She noted: "We as a company work very hard to systematize the process around rigorous analytics."
Of course, banks can't just shift funds around willy-nilly. Accounting rules dictate that they have to justify decreases in loan-loss allowances, for example by citing substantial improvement in credit trends. This past quarter, a bevy of bank earnings releases cited fewer nonperforming loans, improving asset quality, and a stronger underlying global economy as reasons for smaller loan-loss provisions. Bill Lewis, leader of the U.S. banking practice at PricewaterhouseCoopers, notes that subjectivity is often involved, but "most banks, in light of heightened regulatory scrutiny, are more precise in their estimation methodologies today than they have been in the past."
Maybe so, but even if the decreases in reserves are perfectly justifiable, there are still problems with this common industry practice. Besides cutting reserves to the core, banks "are increasing the cyclicality of earnings," says Richard Bove, a banking analyst at Punk, Ziegel & Co. "When bad times come, you know they are going to be increasing the size of the reserves." Already, Citi's Krawcheck has warned analysts not to expect substantial reductions in provisions in the future.
Tinkering with loan-loss reserves also distorts the true quality of bank earnings. A dollar of earnings in a quarter when the provisions are being reduced isn't as valuable as one when they're the same or rising. "Everybody can play the loan-loss game to varying degrees, and the market is generally going to see through it," says Craig Woker, a bank analyst at Morningstar Inc.
But that's not so easy for those investors who aren't well-versed in the nitty-gritty of bank bookkeeping. Bove thinks that bank reserve accounting needs a major overhaul. "In reality, it is really nothing more than a bunch of accountants flipping coins and playing games," he says. For now, investors should realize that the gravy train of 2004 may soon grind to a halt.
By Amey Stone