Pretend you are the CEO of a very large multi-national corporation with a massive product line, horizontal and vertical integration in several class-leading categories, and have historically displayed a certain penchant for "earnings management".
If economic conditions were poor, would you like to clean some nasty things off your balance sheet and tell all the analysts: "It's the macro-environment, stupid", and that forecasts for business growth are "poor"? (all the while never mentioning how accurate your predictions of market conditions have been) Or, would you rather continue to report "bad" results when conditions were favorable?
Then, when (warning: silly market analogy alert) the brush fire allows new plants to grow and economic conditions improve, you can finally look like a rock star and cash out on all those stock options that have been 20 fathoms deep.
Given compensation structures, short term incentives, and the fact that this particular company (o.k., o.k., its GE if you have not figured it out by now) has not had a good record of soundly trouncing street estimates, the above seems like an awfully good strategy.