Why do you think you have to invest in the stock market when prices have skyrocketed?      

The momentum bias

Suppose — and the truth is not far off — that real estate prices had increased by 15% in 2004, 2005, and 2006. Do you think that the probability is high that they will post the same hike in 2007?  Now suppose that the advance had been only 3% per year over the past three years. Is it still just as probable, in your opinion, that 2007 will finish with an increase of 15%?

Investors often think that what has occurred in the recent past can reoccur in the near future with a probability very much higher than is realistic. This behavior may originate in the failure to understand that when a random phenomenon is stable, it is necessary to observe it over a long period of time in order to obtain an accurate picture of its probability distribution.  A practical consequence is that the tendency to give excessive weight to recent information distorts expectations regarding return on investments. It seems that individual investors are optimistic when the markets are bullish and pessimistic when they are bearish.De Bondt (1993)[i] used a weekly survey conducted by the American Association of Individual Investors (AAII) during the period from 1987 to1992 to assess the predictions of investors. The results demonstrate that their predictions for the movements in the Dow Jones index for the following six months are directly tied to the performance of the index during the week prior to the survey. The balance of opinion (optimist-pessimist) widens on average by 1.3 points for each percentage point the Dow gains during the week preceding the survey. More generally, the survey shows that the feelings of investors about the index depend on the market performance over the previous six months.

In the same vein, a study by theGalluppolling institute, ordered monthly by UBS PaineWebber, shows that the expectations of individual American investors over the years 1999-2002 followed the fall of the markets. As the American market was falling, the forecasts of the investors for the following year were being revised downward as can be seen in Figure 1.1.

Figure 1.1: Annual forecasts of American Individual Investors

Year

Annual change in the

S&P 500

Forecasts of investors for the following year

(on Dec. 31)

1999

19%

15.3%

2000

-10%

10.5%

2001

-13%

8%

2002

-23%

5.9%

Source: UBS PaineWebber/Gallup Survey of Optimism of Individual Investors.

Conclusion 

Certainly the momentum bias arises as much from ignorance of the elementary laws of probability as from lack of information on historic returns. Increases of 15% in real estate prices or of 30% in the stock market are extreme phenomena that are much less probable than more modest changes conforming to historical averages. Betting on them is like betting on snow inBeijingin October. It is possible, but surely not very probable, even when July, August, and September have been quite cool.

Source: Mickaël Mangot, 50 Little Experiments for investors,Wiley Publisher, 2009

Endnotes

[i] De Bondt, W., “Betting on Trends: Intuitive Forecasts of Financial Risk and Return,” International Journal of Forecasting, 9(3), (1993): 335-371.