In a penetrating review of Yukari Kane’s new book, Haunted Empire: Apple After Steve Jobs, the New York Times’ Farhad Manjoo points out all the ways in which Apple seems to be clicking along perfectly well without its legendary former CEO Steve Jobs. This is entirely consistent with what the empirical studies tell us about corporate leadership. Most studies have found that — contrary to the prevailing belief that CEOs control their companies’ fate — corporate leadership accounts for a relatively small percentage of the variance in a company’s performance. These studies grapple with a host of variables and there is a range of results, but the weight of the evidence is that on average, the identity of the particular CEO doesn’t matter very much. This is not to say that anyone can run Apple. These studies are comparing the performance of companies run by the type of CEOs boards typically hire; they don’t include the test case of randomly selected individuals with no business background being given the chance to run multinational corporations. They do indicate, though, that which person with the usual — very impressive — credentials, experience and talent runs the company may not matter very much. The most important drivers of a company’s success or failure are the external environment (both generally and within the industry) and the company’s internal resources (personnel, physical assets, intellectual property, culture, etc.). Leadership can matter in particular cases, but it does not matter very much on average and its impact is not predictable. All of which tends to undermine the view that companies should be paying CEOs huge sums because they create enormous value for shareholders. Companies create enormous value for shareholders; CEOs’ role in that creation is far less certain.