* While Paul Samuelson concedes that perhaps there are managers who can outperform market consistently, he questions why academic investigators have been unable to find any evidence of their existence. Professor Grubel suggests that answer may lie in Peter Principle. When a portfolio manager has investment success with a particular portfolio, it often presents him with opportunities to manage different and larger portfolios. So long as he continues to succeed, he will continue moving from one portfolio to another, larger portfolio. Ultimately, he will confront a portfolio so large that his talents are taxed to limit; having reached his level of incompetence, he will no longer be able to achieve above-average returns. a world in which recognition for success comes rapidly, portfolio managers rapidly reach their levels of incompetence and tend to remain there for some time. Thus, at any given moment, only a small proportion of all portfolio managers will be earning consistently above-average returns. And they will move through successful portfolios of their careers so quickly that their existence will not really be visible to tests of academics. If this hypothesis is correct, a successful portfolio manager should produce longer runs of above-average returns if he persists in managing small portfolios than if he moves on to larger portfolios. More generally, because successful portfolio manager tends to change jobs frequently, any proper test of his existence requires analysis of performance history of manager, rather than performance history of portfolio. O T HE efficient market hypothesis, together with empirical tests establishing its validity, represents one of most important achievements in finance during 1960s. The theory has received widespread public attention and has undoubtedly influenced behavior of many investors. fact, evidence suggests that investors who believe in efficient market hypothesis are less willing to follow professional investment advice, avoid turnover of their holdings and entrust less of their wealth to paid managers of mutual funds and other imperfectly diversified portfolios.' Analysts and portfolio managers, having seen great prestige and incomes they enjoyed during 1960s fall dramatically during 1970s (though principally for other reasons), have engaged in many debates about merit of efficient market hypothesis with its proponents from academic world. The academics tended to win these arguments easily by pointing to strong evidence that mutual funds on average earned a rate of return that was only equal to that of market as a whole, and that individual funds traced through time did not succeed in earning above-average rates of return for more than a very few years. Against these statistics, members of investment community can offer only casual evidence and a belief that some professionals, who are better analysts than others, produce above-average yields for portfolios they manage. This view has been spelled out by Arthur Zeikel: In any group of people . . . some will exhibit capacity to recognize and use information faster than others, and that capacity puts them at a competitive advantage. Professional portfolio managers are paid to exhibit that capacity on behalf of clients and customers. It follows that the Herbert Grubel is Professor of Economics at Simon Fraser University, Burnaby, Canada. He thanks Stephen Easton, John Herzog and his former colleagues at Finance Department of University of Pennsylvania, Pao Cheng, Arthur Zeikel and Paul Samuelson for their helpful comments. 1. Footnotes appear at end of article.