There isn’t a general consensus on what really causes momentum. On the one hand others argue using behavioural theories that state that momentum is a result of “naïve investors with biased expectations” Hvidkjaer (2006) and on the other hand, others argue that momentum is a product of the “rational response that individuals have to real market constraints” (Scowcroft & Sefton, 2005).
Random walk vs trend following
Momentum relies heavily on the autocorrelation between returns which assumes that there will be a trend continuation. However, according to Alexander (1961); Cootner (1964); Samuelson (1965) market price movement is random and additionally, market movements are difficult to predict (Lo & MacKinlay). Roberts (1959); Wärneryd (2001) argue that people, despite prices following a random walk, will always convince themselves that there is a pattern that has a predictive value. As such, people have certain psychological mechanisms of dealing with the random nature of prices (Zielonka, 2004). Anchoring, herding, representativeness and misperception of regression to the mean are some discussed by Zielonka (2004).
Anchoring is a cognitive bias where individuals place too much weight to the first piece of information that they get to make decisions. Individuals have an initial value that they have, new information is used to adjust that initial value to the real one. Research has shown that this process of adjustment is rarely completed, affording too much weight to the initial value (Goldberg & von Nitzsch, 2001; Slovic & Lichtenstein, 1971; Tversky & Kahneman, 1971; Zielonka, 2004). In terms of prices, the investor puts more weight on the latest or the most outstanding stock price, which becomes the anchor in making future stock predictions (Zielonka, 2004).
Herd behaviour is the predisposition of people to follow the actions of a much larger group in making individual judgement (Asch, 1952). Individuals will cease to rely on their judgement if the greater population of market participants are not similar to theirs. They assume that the general consensus is better than their own judgement (Zielonka, 2004).
An investor assumes the probability of the stock going in a certain direction by judging it relatively to its performance in the past (Shefrin, 2002). A derivative of representativeness is gamblers fallacy, that is, investors believe that there will be a return reversal and if it’s a return continuation then the behavioural bias is known as antiregressive (Wagenaar, 1988) .
Misperception of regression to the mean
Representativeness also alludes to how individuals do not acknowledge fully the degree to which there is a regression to the mean (Shefrin, 2002). For example, statistically, exceptionally large values in a chance process will be trailed by values that are not as large (Tversky & Kahneman, 1974). Due to this misperception, investors tend to forecast a trend continuation (Zielonka, 2004).
Underreaction and overreaction a result of the above?
These psychological biases lead to investors making cognitive errors. For example, due to representativeness analysts tend to overreact, they are more confident of the winning stocks continuing on a winning streak and the losing stocks continuing on a losing trend (Shefrin, 2002). Anchoring, according to Tversky and Kahneman (1974), means that an individual does not adjust adequately, in due time, the new information that is received and this leads to an initial under reaction.
There are a number of explanations as to why the there is a post-earnings-announcement drift. One of the theories is that the price responds to new information in a rather slow manner. This could be due to traders not being able to fully digest and incorporate this new information into their trades. Or, secondly, the costs of trading on this new information immediately actually exceeds the earnings achieved for a certain level of trades (Bernard & Thomas, 1989)
In other models prices initially underreact to information and that leads prices adjusting slowly to new information (Hong, Lim & Stein, 2000) and this is what leads to momentum. Evidence from the under reaction proves that over a time period of say 1-12 months, security prices underreact to news. News, as a result, will slowly be incorporated into prices, which is likely to reveal positive autocorrelations over this time period. In other words the latest positive news has the power to predict positive returns in the future. On the other hand, overreaction evidence shows that over longer periods of time such as 3-5 years, security prices will overreact to constant patterns of news pointing towards the same direction. In this way securities that have had a long-term record of positive news have a tendency to become overpriced and have low average returns afterwards. Therefore securities with a series of good performance, however measured, obtain exceptionally high valuations which on average, return to the mean (Barberis, Shleifer & Vishny, 1998).
Stocks that have information slowly disseminated will show greater momentum, in this same vein it seems the bigger the firm the faster information is diffused and vice versa. Firms that also have smaller analyst coverage, ceteris paribus, have information that moves slower amongst investors and react much slower to bad news than good news (Hong et al., 2000),
The return continuation postulated by Jegadeesh and Titman (1993) cannot be explained by the three factor model created by Fama and French (1996) which in fact predicts that there should be a return reversal for future returns and not a return continuation. This, they went to explain, could be a result of the behavioural biases that investors have. That is, investors have a tendency to underreact to the short term information in the past and this leads to the return continuation and an overreaction leads to the return reversal in the longer term, amongst other things.
In some instances the prices of securities will overreact to news and for a while after that the prices will continue to overreact over a period of time De Long, Shleifer, Summers and Waldmann (1990).
Risk theories behind momentum
Although Conrad and Kaul (1998) have put forward for consideration a risk-based analysis of momentum Hong et al. (2000) express that there is not much that supports the risk based theories, even though their risk-based analysis is logical. Hung and Glascock (2008) show that risk cannot explain momentum as higher returns did not coincide with higher risk in their research and (Fama & French, 1996) were unable to capture the momentum effect in their three factor model.
Despite the lack of evidence to support technical analysis in the prediction of future prices, momentum is not an exception, it remains a popular strategy and this may be attributed to the behaviour of individuals. Its popularity may be a result of the cognitive biases mentioned above. Furthermore, the predictions made from analysing stock charts correspond with the behavioural finance heuristics (Zielonka, 2004). Judgemental heuristics and biases are pervasive even among the professionals (Stephan & Kiell, 2000).
Alexander, S.S. 1961. Price movements in speculative markets: Trends or random walks. Industrial Management Review (pre-1986), 2(2):7.
Asch, S. 1952. Social Psychology. 1952. Englewood Cliffs, Nueva Jersey, Prentice-Hall.
Bernard, V.L. & Thomas, J.K. 1989. Post-Earnings-Announcement Drift: Delayed Price Response or Risk Premium? Journal of Accounting Research, 27:1-36.
Conrad, J. & Kaul, G. 1998. An anatomy of trading strategies. Review of Financial studies, 11(3):489-519.
Cootner, P.H. 1964. The random character of stock market prices.
De Long, J.B., Shleifer, A., Summers, L.H. & Waldmann, R.J. 1990. Noise trader risk in financial markets. Journal of political Economy:703-738.
Fama, E.F. & French, K.R. 1996. Multifactor explanations of asset pricing anomalies. The journal of finance, 51(1):55-84.
Goldberg, J. & von Nitzsch, R. 2001. Behavioral finance (A. Morris, Trans.) Chichester ENG. [Online] Available from.
Hong, H., Lim, T. & Stein, J.C. 2000. Bad News Travels Slowly: Size, Analyst Coverage, and the Profitability of Momentum Strategies. The Journal of Finance, 55(1):265-295.
Hung, S.-Y. & Glascock, J. 2008. Momentum Profitability and Market Trend: Evidence from REITs. The Journal of Real Estate Finance and Economics, 37(1):51-69.
Hvidkjaer, S. 2006. A trade-based analysis of momentum. Review of Financial Studies, 19(2):457-491.
Jegadeesh, N. & Titman, S. 1993. Returns to buying winners and selling losers: Implications for stock market efficiency. The Journal of finance, 48(1):65-91.
Lo, A.W. & MacKinlay, A. A Non-Random Walk Down Wall Street. 1999. [Online] Available from.
Roberts, H.V. 1959. Stock‐Market “Patterns” And Financial Analysis: Methodological Suggestions. The Journal of Finance, 14(1):1-10.
Samuelson, P.A. 1965. Proof that properly anticipated prices fluctuate randomly. Industrial management review, 6(2):41-49.
Scowcroft, A. & Sefton, J. 2005. Understanding momentum. Financial Analysts Journal, 61(2):64-82.
Shefrin, H. 2002. Beyond greed and fear: Understanding behavioral finance and the psychology of investing. Oxford University Press.
Slovic, P. & Lichtenstein, S. 1971. Comparison of Bayesian and regression approaches to the study of information processing in judgment. Organizational behavior and human performance, 6(6):649-744.
Stephan, E. & Kiell, G. 2000. Decision processes in professional investors: Does expertise moderate judgmental biases.416-420.
Tversky, A. & Kahneman, D. 1971. Belief in the law of small numbers. Psychological bulletin, 76(2):105.
Tversky, A. & Kahneman, D. 1974. Judgment under uncertainty: Heuristics and biases. science, 185(4157):1124-1131.
Wagenaar, W.A. 1988. Paradoxes of gambling behaviour. Lawrence Erlbaum Associates, Inc.
Wärneryd, K.E. 2001. Stock-market psychology: How people value and trade stocks. Edward Elgar.
Zielonka, P. 2004. Technical analysis as the representation of typical cognitive biases. International Review of Financial Analysis, 13(2):217-225.