Capital ratios are not a good tool for identifying  problem banks in time to save them, according to a recent study by the   Federal   Reserve Bank of Boston.     
In their report, "The Use of Capital Ratios to Trigger Intervention in  Problem Banks: Too Little, Too Late," authors Joe Peek and Eric S.   Rosengren argued that capital ratios are a lagging indicator of trouble.   
  
Published in the September/October issue of New England Economic Review,  the authors studied failed New England savings and commercial banks between   1988 and 1994. They found that problem institutions often delayed reporting   poor capital ratios until after supervisory exams and waited to add to   loan-loss reserves until after a problem was identified.       
In reaction to the thrift crisis, Congress in 1991 required regulators  to segregate banks by capital. Under this "prompt corrective action"   scheme, the activities of banks with risk-based capital ratios below 6% or   leverage ratios less than 4% are restricted.     
  
But while federal law relies heavily on capital ratios, bank examiners  use far more data to identify problem performers early, the authors   concluded.   
Mr. Peek, an economics professor at Boston College, and Mr. Rosengren,  an economist at the Boston Fed, recommended regulators start basing   corrective action on the examiners' Camel rating system.   
"One possibility is to use Camel rating downgrades, rather than capital  ratios, as the trigger for prompt corrective action intervention," the   report noted.   
  
The authors also suggested that the prompt corrective action model could  be revised to raise the capital-ratio triggers. 
Robert Rowe, regulatory counsel for the Independent Bankers Association  of America, said banks would be wary about any subjective approach to   evaluating performance. Part of the examiners' evaluation includes   measuring management quality.     
Mr. Rowe said the industry prefers objective measures like capital  ratios. "Bankers are more comfortable when you add two and two and get   four," he said.   "When you start talking about sines and   cosines and tangents, they get a little more uncomfortable."       
Mr. Shea writes for the Medill News Service.