I recently discovered a paper by famous economist Robert Shiller which had I find amazing findings in terms of how we perceive the future and how it ends up being. I sent an email to Nassim N. Taleb to take his opinion in it, and he replied telling me that it was mentioned in his book Fooled by Randomness. I was quite surprised with his reply, especially after I had read Fooled by Randomness inside out, but I guess the reason the paper’s results didn’t stick in my mind is that the graph that is essential to the paper is not shown in Taleb’s book.
For starters: one of the most widely used models to value a stock price is Gordon’s Growth Model which states that stock prices should be equal to the present value of all future dividends earned by the stock. This is quite intuitive, as otherwise, why would you pay more to own a piece of paper (share or stock) which earns you less than the present value of all future dividends on it?
Shiller then had the following brilliant idea: if stock prices should be equal to the present value of all future dividends earned by owning the stock, then let’s take a look at how these present values of actual future dividends compare with… the actual future itself.
The results are shown in the following graph:
The red curve, P*, is the present discounted value of actual subsequent real dividends, and the blue curve is the value of the S&P.
What do you think? Does the blue line look like a close approximation of the red line it was trying to forecast?
Shiller’s paper title is “Do Stock Prices Move Too Much to be Justified by Subsequent Changes in Dividends?” I guess by now the answer should be pretty clear.
I guess this marvelous finding has more than one interpretation. Philosophically, one might think about the following question: Is the future more random than what we think it is going to be, partly because of the fact that we overreact to our expectations of it?
Shiller’s paper can be and can be found here: http://www.math.mcmaster.ca/~grasselli/Shiller81.pdf