Recent studies on the behavior of earnings growth over time raise doubt about the ability of past growth to explain differences in price-earnings ratios. Either future growth is difficult to predict, or investors are basing their predictions on information other than past growth.

Grouping common stocks into portfolios on the basis of price-earnings ratios, the authors find that the initial P/E differences among the portfolios persist up to 14 years. Growth appears to explain little of the persisting P/E differences, however. Price-earnings ratios correlate negatively with earnings growth in the year of the portfolio’s formation, but positively with earnings growth in the subsequent year, suggesting that investors are forecasting only short-lived earnings distortions.

Nor does risk supply the explanation for these differences. Although price-earnings ratios can vary either positively or negatively with market risk, depending on the market conditions in a given year, market risk is of little assistance in explaining the observed persistence in price-earnings ratios over periods longer than two or three years.

The authors conclude that the most likely explanation of the evident persistence in price-earnings ratios is not growth or risk, but differences in accounting method.