Perspectives from ISB

This article was first published in the Free Press Journal on April 16, 2012

http://epaper.fpj.co.in/Details.aspx?id=17080&boxid=3452315

Refuting one of the most famous theories in Finance, the Efficient Market Hypothesis (EMH), Momentum and Overreaction are two phenomena experienced in different degrees in different markets across the globe.

Momentum refers to a phenomenon wherein the past winners continue to be winners and past losers continue to be losers in the short run, typically three months to one year period. Overreaction or reversal anomaly is that the losers in the past outperform the past winners and past winners turn into losers in the long-term, a period of three to five years.

Studies in the developed markets like the US, UK, Australia and Japan, and developing nations like Turkey and Brazil, have shown that due to the presence of either momentum or overreaction, or both phenomena, significant abnormal returns can be earned by the investors using simple strategies. The support for momentum is weaker in the emerging countries than for the developed countries. Also, Asian countries are found to exhibit weaker momentum than the European and American countries.

While both momentum and overreaction can be attributed to factors like size, risk, macro-economy, data mining biases, liquidity, etc. none of them are conclusive. Another strand of literature relies of the investor psychology to explain the two trends.

Investor Psychology

Psychological characteristics of investors might explain the reasons that could be behind over-reaction or momentum. Indian investors being more emotional in nature, psychological aspects may be very important in explaining the market trends.

One of the most popular investor characteristic is “overconfidence”. Overconfidence makes the investors overreact to any news. If there is good news about a stock, the investors will drive the prices up by buying more, and will continue to keep buying for some time due to overconfidence. According to this logic, the momentum will be higher in Bull markets.

On the other hand, momentum may also be a symptom of underreaction. That is, prices adjust too slowly to news. The underreaction of stock prices due to news (for example, earnings announcements) may cause the momentum, since a slow diffusion of information among investors could make the path to the ‘correct’ value of the stock longer than expected. But, for longer periods, an overreaction of stock prices may occur due to extrapolation of a series of good or bad news, especially if investors are overconfident.

Underreaction could also be a result of either the conservative nature of the investors or lack of confidence. The conservatism bias suggests that individuals underweight new information in updating their expectations. If investors act in this way, prices will tend to slowly adjust to information, but once the information is fully incorporated in prices, there is no further predictability in stock returns. Investors, who lack confidence and hesitate in making a decision, would also underreact to information, causing the prices to take longer to reach their correct value, hence exhibiting momentum.

It is also found that bad news generally have the effect of making the prices more volatile than good news. Also, people act faster on good news than on bad news, as they are averse to losses. They do not mind realizing profits, but dislike realizing losses, hence keep postponing them. Both small as well as professional traders have been found to hold their losing portfolios longer than their winning ones. This could be interpreted as negative feedback strategy with respect to past returns, which could lead to reversals in prices.

Studies have pointed towards other factors like self-attribution self-deceptions, emotion-based judgments, framing effects, and mental accounting, to explain the momentum and overreaction trends.

In a study which was published in the International Research Journal of Finance and Economics, my peers Chakrapani Chaturvedula and Nikhil Rastogi from IMT Hyderabad and I studied 156 months NSE listed stocks data for indications of momentum and overreaction effects.

Momentum in NSE, India

We find that momentum strategy of buying the winners and selling the losers results in significant positive returns for the interval of 3 months for all categories of stocks, low cap, mid cap and large cap (Table 1). However when we take the higher intervals like 6 months and 12 months, there is no momentum in small cap and medium cap stocks. But, for large cap stocks, it persists for all the intervals up till 12months. This result is surprising since these stocks are tracked more by the analyst and so information should be quickly incorporated into the prices resulting in no momentum for this category of stocks. The result is also in contrast with many previous literatures, which points towards existence of momentum in small cap, rather than large cap stocks.

Table 1: Momentum after accounting for size effect

Overreaction in NSE, India

Table 2 shows the results for Overreaction after accounting for size. We find no evidence of over-reaction for the small and the large cap stocks while we consistently find it in the case of mid cap stocks. Our results for the large cap stocks are not at all surprising as the large cap stocks are widely tracked by analysts and any new information would get disseminated very fast. However, the absence of over-reaction in small cap stocks is very surprising because small cap stocks, by virtue of being less traded and having slower dissemination of information than large or medium cap stocks, was expected to show signs of over-reaction.

Table 2: Overreaction after accounting for size effect

The evidence for overreaction is present only in mid-cap stocks. This is supported by the under-reaction hypothesis. What we can say is that in India, stocks under-react to new information initially, thereby exhibiting momentum in the short run. But, mid-cap stocks over-react in the long run.

Disclaimer: The views are those of the author and not necessarily of the organisation in which she works.