An index fund that holds all available common stocks in proportion to their outstanding market values would provide the best possible tradeoff between expected return and risk for the investor who has no research advantage. A number of practical problems arise, however, when one attempts to translate this ideal into practice. The S&P 500, for example, is dominated by stocks of large American companies with a number of characteristics — industry representation, competitive position, financial strength and exposure to government regulation — that make them unrepresentative of U.S. common stocks in general. Differences are even more marked when one compares the S&P companies with foreign ones.

Whether an index fund can actually economize on transaction costs depends on its size, the frequency of additions and withdrawals and the techniques used by the manager to minimize these costs. If a very large index fund purchases 500 separate issues with the proceeds of each day’s dividends, it will raise its transaction costs excessively. Yet any attempt to economize by making purchases less frequently, accumulating dividends in the interim, will move the client away from his desired objective. Holding substantially fewer than 500 stocks, on the other hand, will add to risk without adding to return.

Even in the absence of special investment insights, a managed portfolio can compete effectively with index funds in coping with these problems. But if such insights are available, the managed portfolio can take advantage of the opportunities they present.