How much can we believe companies that report increased earnings over 20 quarters?

A new study, reported in the International Herald Tribune, suggests we should probably take it with a bag or three of salt.

The report bases its conclusions from a paper "Earnings Momentum and Earnings Management," which appeared in the spring 2007 issue of the Journal of Accounting, Auditing and Finance found that only a handful of companies should have had earnings-per-share growth for 20 consecutive quarters. But nearly 600 actually reported such strings of growth, something the researchers took to conclude constituted "prima facie evidence of earnings management."

One caveat: the researchers found no evidence of specific acts of earnings manipulation at any given company and not all companies with 20 consecutive quarters of increased earnings would have necessarily engaged in such manipulation. But then, the manipulation does not necessarily amount to fraud, it could just amount to not spending money on R&D or marketing in a quarter. The study also found that when there was a decline in earnings, the stock price usually fell by a bigger than usual amount as investors punished the company for letting them down.

What does all this point to? First, it shows the enormous amount of pressure that's placed on companies to maintain earnings. No matter what. And secondly, it tells us not to put too much of our money on companies that keep reporting strong earnings consistently, quarter after quarter.