The results in the paper are based on a particular set of assumptions about 1) the length of the horizon, 2) the discount rate, 3) the growth rate of dividends, and 4) the PE ratio at the end of the horizon. It is easy to modify the assumptions in the paper and compute alternative earnings growth rates. This can be done for any time period. The paper used data as of June 30, 1999, but more recent data can be used. If you click Compute growth rate of earnings, you can make your own assumptions and see what they imply. The default values that are used for these computations are based on data as of September 4, 2003, but you can change these values if you wish. The goal is to find a set of assumptions that seems sensible and that does not imply an unrealistically high ratio of profits to GDP in the future. Can you find such a set? The paper essentially argued that such a set did not exist for the June 30, 1999, data.

It is also possible to use the stock-price formula to compute what the S&P stock price should currently be for alternative assumptions about the future growth rate of earnings. If you click Compute S&P stock price, you can do this. Again, the default values are based on data as of September 4, 2003, but you can change these if you wish. You should compare your computed stock price with the current S&P stock price in deciding whether the current S&P stock price is too high or too low. Again, can you find a sensible set of assumptions that justifies the current S&P stock price and that does not imply an unrealistically high ratio of profits to GDP in the future?

You should read the section of the paper mentioned above before running the experiments. The experiments are not necessarily self explanatory.