I concur with this Atlantic article that CEOs do not matter much and when they matter they destroy value more than create it. Moreover, the business press adulations of CEOs is much too short sighted. A CEO who turns around a company in 4-or-5 years has proven nothing much more than the fact that he is a turn around artist, not a leader who can guide the company through good and bad times. The tenure of a CEO matters before he is given all kinds of accolades. The yardstick to be used is a CEO who can successfully guide a company through at least 2 business cycles which means a tenure of at least 15-20 years. By this measure, neither Lee Iacoca, Jack Welch nor Lou Gerstner meet this criteria.
James March, a management professor at Stanford, goes so far as to say that in any well-run company that’s conscientious about grooming its managers, candidates for the top job are so similar in their education, skills, and psychology as to be virtually interchangeable. All that matters is that someone be in charge. “Management may be extremely difficult and important even though managers are indistinguishable,” he writes. “It is hard to tell the difference between two different light bulbs also; but if you take all the light bulbs away, it is difficult to read in the dark.”
And from Jeff Immelt, current CEO of GE:
“Not only could anyone have run GE in the 1990s, [a] dog could have run GE. A German shepherd could have run GE.”
A more nuanced view is here which tends toward the argument that CEOs do not matter much but they do matter. Thus the fact the CEOs should be rewarded based on performance is very much in conflict with the evidence.