Mind Over Money: Cognitive Bias
11/15/2002 12:00 am EST
"There is no truth," said Gustave Flaubert. "There is only perception." Stocks swing from periods of undervaluation to overvaluation, and the extent of these changes goes far beyond what would be justified simply by changing fundamentals. What does change is perception. Here, we look at behavioral finance, which applies psychology to market analysis and tries to quanitfy investor perception.
"Cognitive scientists have devoted a great deal of effort toward understanding the complex processes that the human brain employs to make decisions," says Jim Oberweis, editor of The Oberweis Report. "Although many questions remain unanswered, research has demonstrated a few key patterns that often lead to poor decisions. In particular, studies have shown that people tend to consistently overestimate the probability of unlikely events if the memories associated with them are recent and vivid. For example, if you see a house burning or witness an automobile crash (or a stock market crash, for that matter), it increases your belief that such an accident will recur far more than it should based on statistics. Likewise, heavy media coverage of uncommon events also tends to upward bias our perceptions of frequency of occurrence.
"Consider the following question: Did violent crimes committed on school grounds in the U.S. increase or decrease during the 1990s? You might be surprised by the answer. In fact, the number of school-related student homicides remained stable throughout the 1990s at 30-35 per year, including 1998-99 (the year of the Columbine shootings). However, according to a Washington Post poll in 1999, 62% of people polled by the paper believed children were getting more violent and seven in ten thought a school shooting was "likely" to happen in their communities. We suspect that the vivid media coverage of recent events at the time, namely the Columbine shootings, skewed the perception of many of those interviewed.
"Consider how such a bias might influence beliefs about the stock market today. Virtually every investor in stocks experienced a remarkably painful period in the third quarter as the equity market continued to decline sharply. The S&P 500 declined 17.3% for the quarter (-28.2% for the first 9 months of the year); the Russell 2000 Index declined 21.4% for the quarter (-25.1% for the 9 months), its worst quarterly performance ever aside from the fourth quarter of 1987; and the Russell 2000 Growth Index, which is perhaps the best benchmark for our recommendations, declined by 21.5% for the quarter (-35.1% for the 9 months). Indeed, the third quarter of 2002 was one of the toughest quarters of the last 70 years for equity investors.
"According to an AP poll conducted Oct. 4-8, 2002, people were asked if they had $1,000 to spend whether they thought it would be a good idea to invest it in the stock market. The poll found that only 29% thought it was a good idea and 64% said it was a bad one. In contrast, a similar poll conducted in April of 1998 found that two-thirds of those surveyed thought it to be a good idea. Human judgments are strongly influenced by the recent past. Note that the 1998 poll was taken after several years of very strong market results. Notably, even though the market continued to rally initially, the S&P 500 fell about 20% over the next 4 years.
"Unfortunately, although the evidence is strong that recent results influence people’s judgments about the markets, empirical evidence suggests that there has been little or no predictive value of using a historical period’s return to forecast the future. In fact, if any conclusion can be learned from history, it is that major market corrections have proven to be exceptionally good times to buy stocks, as lower stock prices have often offered unusually attractive valuation opportunities.
"Few investors are able to objectively separate cognitive biases from their investment decisions. Right now, stock valuations relative to growth rates, within our universe of small-growth stocks, appear to be more attractive than at nearly any other time of the past decade. Despite the widely held perception in September that stocks would continue to go down after the third quarter, it has not turned out to be the case. The S&P 500 is up 10% so far in October, marking one of the best months in a long time. Stay tuned."
Professor Robert Shiller of Yale University found that at the peak of the Japanese stock market bubble, 14% of Japanese investors expected a crash. However, after the market did indeed crash, the number of investors anticipating another crash rose to 32%. People generally shift their opinion to be aligned with recent experience and extrapolate recent trends into the future. At the peak of the market bubble in 2000, many studies showed that investors strongly believed they would receive above average returns from their investments for years to come. Many planned their retirement on the expectation of 20% annual portfolio gains.
One of Schiller's theories is called "anchoring" which suggests that in the absence of better information, investors assume that current prices are "correct". Expectations are then anchored by recent experience, causing investors to focus on recent trends and to expect these trends to continue. Schiller believes this factor impacts over- and under-reactions by the marketplace. In other words, investors tend to become more optimistic as prices rise and more pessimistic as they fall. Augmenting this phenomenon is a theory called "conservatism" in which investors are slow to alter their opinions, despite new evidence that runs counter to their already-existing expectations.
The risk of these psychological processes is that extrapolating yesterday's market conditions into the future may leave you unprepared for a changing market environment. Today, many investors are conditioned by several years of poor performance in stocks, and they are now basing future investment decisions on that experience. Just as extrapolating the gains of the late 1990s proved disastrous for many investors, extrapolating the losses of the last three years into the future may prove as much a disservice. If your investment analysis leads you to conclude that the future for the stock market is bearish, so be it. Just don't reach that conclusion simply because your cognitive processes have conditioned you to that expectation.