As much as I'd like to believe that CEOs are basically payed for chance, this research is deeply flawed. As in, it's literally trying to measure the (practicall

As much as I'd like to believe that CEOs are basically payed for chance, this research is deeply flawed. As in, it's literally trying to measure the (practically) immeasurable. More specifically, it's this assumption: > We model the impact of a manager on her firm’s performance by assuming that a manager who exerts effort and manifests her talent increase the firm’s expected return by the magnitude of her talent. that is simply not true. The performance of every CEO is amplified by the amount of people he's affecting. And that is made even worse by the fact that companies are competing against each other, meaning that usually it's not their total performance that matters, but the difference to the next best company. You end up with so much noise in your data that yes, the result looks random. But is it? We don't know. The only accurate measure of performance (that we know of) is performance itself.