Look at the disclosed earnings per share of GE over Jack’s reign. Steady growth—10 to 15 percent per year—just as promised, year after year. I used to show GE’s earnings chart to Wall Street analysts that I trained on governance issues and asked them, “How many of you believe this chart? How many of you believe that a company in cyclical industries like medical devices, aircraft engines, and financial services can deliver such regular growth. Jack must be a genius. One industry surges ahead just as another one hits a bump–year after year.” No one ever raised their hands, including the analysts covering GE.

Immelt’s misfortune is being groomed as CEO by a master earnings manipulator, but becoming CEO in an age where some of those manipulations are no longer acceptable.

Why only some?

I define two classes of earnings management: (1) accounting manipulation and (2) short-term decisions. The former simply affects how numbers lay on the page; it has no effect on actual cash flow. The latter represents things like deferred investment or, in the case Elizabeth points out, ‘timely’ asset sales; these items actually current and future affect cash flow. I note that the first class of earnings management has gotten severely constrained by tighter accounting rules and oversight.

So, guess what managers bent on earnings management will resort to? A 2005 study by Graham, Campbell, and Rajgopal found that over 75% of top managers would sacrifice economic value via short-term decisions in order to “make their (promised) earnings numbers.” This second class of earnings management is not even on anyone’s radar. Most directors don’t know about this issue until I introduce them to it. Like Elizabeth, I’m appalled that companies do this stuff, but I’m more appalled that they do it so openly (like they used to get aware with accounting manipulation), and no one is calling them on it.