« Harry Turtledove Has Another Fan | Main | The State of the Volunteer Army »

June 19, 2005

How Far Ahead Does the Stock Market Look? Five Years

The young and brilliant Stefano della Vigna (whom I never see because he's a fifth-floor person and I'm a sixth floor person) has what looks to me like a serious finance research hit--extra bases and RBIs:

Looking Long Term? Get Your Glasses - New York Times: [F]ew investors can focus on events more than five years ahead, even when those events are very predictable and almost certainly will have a big impact on a company's earnings. The study, 'Attention, Demographics and the Stock Market,' was conducted by Stefano della Vigna, an assistant professor of economics at the University of California at Berkeley, and Joshua M. Pollet, an assistant professor of finance at the University of Illinois at Urbana-Champaign. A copy is at http://www.nber.org/papers/w11211

The study is attracting much interest because researchers for many years couldn't figure out how to disentangle the attention span of investors from other factors that could also explain their behaviors. How, for example, to interpret why investors are unimpressed with a company's announcement of a new research and development effort that it says will lead to higher profits in 10 years?.... Della Vigna and Professor Pollet solved this problem by focusing on industries whose profitability depends heavily on the age distribution of the population. Bicycle makers are a good example... people in their late 30's or early 40's bought bicycles at nearly twice the rate of the overall population, in large part because they were buying them for their children. By contrast, consumers under 27 or over 55 bought bicycles at rates far below the national average.

The professors focused on two dozen age-sensitive industries - from toy and beer makers to nursing home operators and funeral homes - from January 1935 through December 2003. For each industry, they built a model of the year-by-year changes in demand caused by shifts in the age distribution of the population. On average, they found that when their model predicted a one percentage point increase in demand for an industry's goods or services, its profits that year were 5 to 10 percent higher.

If investors conformed to the completely rational, fully informed ideal described in an economics textbook, they would immediately take into account the long-term effects of a changing population. The stock prices of age-sensitive companies would thus be bid up or down soon after major changes in the country's birth rate, for example, even if the changes' effect on companies' earnings was not felt for several decades.

The professors found, however, that virtually no investors conformed to this ideal. Instead, the study found that the price of a stock began to change only about five years before shifts in age distribution started to have big effects on that company's earnings.... [O]ver the 68 years studied, competition did not eliminate the extra profitability in age-sensitive industries. One reason, the professors suspect, is the entry barriers to at least some of them.

Posted by DeLong at June 19, 2005 10:41 PM