The Trouble With Long-Term Investing

Justin Fox says the concept of “shareholder value” was always meant to embody a long-term concept:

The goal was to get corporate executives to pay less attention to accounting earnings and focus instead on economic earnings — which Alfred Rappaport, who taught at Northwestern University’s Kellogg School of Management, defined as anticipated cash flow discounted by the cost of capital. It was an argument for paying attention to what created value over time instead of stressing out about quarterly earnings. Which doesn’t sound dumb.

“I don’t know how many times I kept saying long term, long term, long term,” explains Rappaport, who is now 77 and living in semiretirement in Southern California, but still pens the occasional Harvard Business Review article and has a new book in the works. “To me, shareholder value was not about an immediate boost to stock price.”

Felix Salmon says the problem is that “we live in a world where the overwhelming majority of stock-market investors mark their holdings to market daily, even if their time horizon is measured in years . . . in order to believe that shareholder value makes sense, you also have to believe that if you buy a stock at $100 and it drops to $50, then you haven’t lost any money so long as you haven’t actually sold the stock.”


I think there’s actually a more fundamental problem here that goes back to what Keynes was talking about in Chapter 12 of the General Theory. The problem, to state it concisely, is that it doesn’t appear to be the case that “investing for the long term” is the best way to make money over the long term. “[H]e who tries to guess better than the crowd how the crowd will behave” will typically do better than the “skilled individual who, unperturbed by the prevailing pastime, continues to purchase investments on the best genuine long-term expectations he can frame.” The whole enterprise, recursive as it already is, becomes even more recursive when firm managers start focusing their activities on the determinants of stock prices since this turns out not to involve more guesses about how the crowd will guess the crowd will behave than it does assessments of the long-term future.