Businessweek has a great cover story on accounting estimates and how they affect stock returns.
The problem with today's fuzzy earnings numbers is not accrual accounting itself. It's that investors, analysts, and money managers are having an increasingly hard time figuring out what judgments companies make to come up with those accruals, or estimates. The scandals at Enron, WorldCom, Adelphia Communications (ADELQ ), and other companies are forceful reminders that investors could lose billions by not paying attention to how companies arrive at their earnings. The hazards were underscored again Sept. 22 when mortgage-finance giant Fannie Mae said its primary regulator had found that it had made accounting adjustments to dress-up its earnings and, in at least one case, achieve bonus compensation targets. The company said it is cooperating with government investigators. The broader concern is that corporate financial statements are often incomplete, inconsistent, or just plain unclear, making it a nightmare to sort out fact from fantasy. Says Trevor S. Harris, chief accounting analyst at Morgan Stanley (MWD ): "The financial reporting system is completely broken."
Indeed, today's financial reports are more difficult to understand than ever. They're riddled with jargon that's hard to fathom and numbers that don't track. They're muddled, with inconsistent categories, vague entries, and hidden adjustments that disguise how much various estimates change a company's earnings from quarter to quarter, says Donn Vickrey, a former accounting professor and co-founder of Camelback Research Alliance Inc., a Scottsdale (Ariz.) firm hired by institutional investors to detect inflated earnings.
This why you will never see a Businesspundit post about some company making a surprise earnings announcement. I don't trust what most companies put out as earnings. Mrs. Businesspundit has been an auditor for too long and I have heard too many stories about screwed up financials.
I read a book a long time ago called "The Quality of Earnings" which I would highly recommend. I think the author used to put out a newsletter where he analyzed the earnings quality at various companies. If any of you know of newsletters like this, please leave some information about them in the comments.