Contrarian and Momentum Investment Strategies
Disclaimer: This work has been submitted by a student. This is not an example of the work written by our professional academic writers. You can view samples of our professional work here.
Any opinions, findings, conclusions or recommendations expressed in this material are those of the authors and do not necessarily reflect the views of UK Essays.
Published: Thu, 15 Mar 2018
Literature Review of Contrarian Investment Strategies and Momentum Investment Strategies
2.1 Aim of Chapter
This chapter is designed for reviewing the existing academic papers relating to the well-recognized Investment strategies. To begin with, the two critical investment strategies, contrarian investment strategies and momentum investment strategies, are debated with supports from empirical evidences found in prior studies. Then, the analyses of empirical researches are reviewed separately for the US market, UK market and other international markets.
2.2 Momentum strategies and opposed theories
Recent studies have shed new light on the important area of financial study, Behavioral Finance, which in the past three decades becomes a popular subject among many people in professional occupations, i.e. psychologist, economist, analyst, etc. Behavioral finance is the field of study focused on examining the behavior of market and market participants. Several researches in this area point out numerous types of market anomalies which are remarkable for both professional and individual investors. One of the most notable issues is overreaction – the state when people incorporate the news or information and respond more than they should.
The term overreaction has some inferred meaning that has been debated among academics. Kahneman et al (1982) show that Bayes’ rule, the well-established norm which is used to address the correct reaction to new information, is not an appropriate way to characterize how investors respond to the new data. Their findings reveal that individual investors seem to overweight recent information and underweight base rate data. Other researcher, Grether (1980), observes the overreaction bias among professional security analysts and economic forecasters. He finds that this bias exist in professionals as same as in individuals. This has been supported by the evidence from Theory of Investment Value (Williams, 1956) that stock prices rely too much on current earnings and too little on long-term dividend payment.
Other academic researches try to examine the behavior of overreaction to information in securities trading. De Bondt and Thaler (1985) study the market efficiency and observe the effect of investors’ behavior to the stock prices. Regarding the belief of overreaction, the hypotheses of their empirical test lie on the question whether stock prices movements will track the extreme movements in the preceding period in the opposite direction, and that the more extreme the former path of stock movements, the greater will be for the latter. They test the relation between residual return behavior for the period of 72 months surrounding portfolios formation date, using monthly returns of stocks listed in New York Stock Exchange that have extreme capital gains or losses. Portfolios of winners and losers are formed regarding the rank of cumulative excess returns in the past 36 months, and cumulative average residual returns are computed for test period of subsequent 36 months. In their analysis, they find the results consistent with the profitability of contrarian investing, thus this violates weak form of market efficiency. Loser portfolios earn excess return over the market while winner portfolios underperform the market. Therefore, investment using contrarian strategies (buying past losers, selling past winners) generates profit by earning abnormal returns. In addition, their findings reveal the idea of overreaction effect is asymmetric; it has much impact on the losers than on winners. Findings also reveal that excess returns can be observed mostly in January which can be explained by tax-loss selling.
De Bondt and Thaler (1987) further evaluate the overreaction hypothesis by conducting researches regarding seasonality of the returns path of both winner and loser portfolios and seeking if the difference in CAPM-betas is the explanation of this anomaly. The first objective of the paper is to investigate whether high amount of excess returns in January is linked to the tax code or to seasonality in the relationship of risk-return. The authors employ same data set for the test and find that during the test period, both losers and winners earn excess returns mainly in January, but the amount of winner portfolios are smaller than those of loser portfolios. For long-term effect, the relationship is negative for losers; on the other hand, positive for winners. Further analysis of OLS regressions also show that both long-term and short-term reversals for losers may reflect pressure of tax-loss selling while only short-term reversals for winners are due to capital gain tax “lock-in” effect. Furthermore, in previous paper, the authors tried to explain the excess returns by difference in CAPM-betas and found that losers’ beta is relatively lower than winners’. They concluded that the market-adjusted excess returns were the estimates of the true risk-adjusted excess returns. However, many academics (Chan (1986, 1987) and Vermaelen and Verstringe (1986)) brought up the debate regarding the inappropriate procedure of estimating betas while betas vary with changes in market value. This paper corrects the weakness by constructing the (zero-investment) arbitrage portfolios that holding long position of losers and short position of winners (so called “contrarian strategies”) based on different market conditions. Finding shows positive beta in bull market condition, whereas it exhibits negative beta in bear market condition. In other words, it can be concluded that the arbitrage portfolios work well in both circumstances. Losers will gain more than winners in up market, while winners will lose more than losers in down market. The authors detect the reversal pattern in both winners and losers portfolios which is consistent with overreaction, and that earning reversal appears more for losers than for winners. Thus, this proves that contrarian investing generates abnormal returns for investors.
Other well-known papers of Jegadeesh (1990) and Lehmann (1990) also claim that there is strong evidences supporting the theory of shorter-term return reversals. Their studies show the existence of contrarian strategies in portfolios of stocks which determined by their returns in limited period of time (short-term i.e. weekly or monthly data). Despite the significant level of abnormal returns generated, these strategies rely on highly frequency of trading, which are transaction intensive and based on short-term price movements, it may conclude that the result shows more of short-term price pressure and a lack of liquidity in the market rather than overreaction.
On the opposite side of academic literatures, contrarian strategies have been debated against the early papers of the market efficiency which focused on relative strength strategies of buying past winners and selling past losers.
Levy (1967) mentioned in his paper that there are sufficient evidences that a trading rule of buying stocks with current prices relatively higher than their past 72-weeks average creates significant abnormal returns. Jensen and Bennington (1970) continue their research based on Levy (1967) focusing on longer term of observation and find that Levy’s trading rule does not outperform a buy and hold strategy and his strategy also reflect a selection bias in the process of choosing stocks.
Many literatures try to examine whether practitioners use relative strength strategies as one of their stock selection criteria rather than using contrarian investing. One of them is paper of Grinblatt and Titman (1989, 1993). They investigate the strategies of mutual funds and find that their investment is mostly based on buying stocks that performed well in the previous quarter. This suggests that relative strength strategies may generate abnormal returns.
More recent author, Burton (2003), studies the momentum strategies and finds that momentum investing produces positive relative returns only in some periods, such as in late 1990s. On the contrary, this strategy generates high negative relative returns during the year 2000. As a result, no specific investment strategy is confirmed to produce excess returns across times.
Literatures regarding momentum investing (relative strength strategies) will be analyzed thoroughly later in the next section of this chapter according to the source of paper and the data used in those methodologies.
2.3 Review of researches regarding profitability of momentum strategies in the US market
Jegadeesh and Titman (1993) believe that if investors are prone to overreaction or underreaction bias, then investment strategies that rely on selecting stocks due to their past returns performance will generate profits. In other words, the profitability from employing relative strength trading strategies will undermine the stock market efficiency theory, where investors cannot generate any excess returns above the markets. The authors observed data of NYSE and AMEX stocks from 1965 to 1989, and formed ten equally-weighted portfolios based on J-month/ K-month strategy of the 3 to 12 months holding periods. They conduct the methodology of calculating portfolio returns using overlapping holding periods. Thus, the total of 16 strategies have been conducted for the normal circumstance where there is no time lag between portfolio formation period and the holding period, while another 16 strategies have been developed for the methods that skip a week between portfolio formation period and the holding period in order to avoid bid-ask spread, price pressure, and lagged reaction effects which might affect the results. The results show that almost all returns are statistically significant, except for 3×3 strategy in the normal circumstance. The most profitable zero-cost strategy belongs to 12×3 strategy which gives the monthly return of 1.31%. Overall, the relative strength strategies are profitable in the observed period with slightly higher results from strategies using 1-week lag between formation periods and holding periods. The tests also indicate that predicted changes in stock prices may not be persistent and that sample stocks which are employed with the relative strength strategies earn negative excess returns starting from 12 months after the portfolio formation, which can be continued up to 31 months time. In addition, the bias in market expectation is observed during the analyses. It seems that winners portfolios earn consistently higher return than those of losers portfolios around the quarterly announcement of earnings that are made in the first 7 months around the portfolio formation date, while losers portfolios generate significant higher returns afterwards. Thus, this proves that former returns are positive and latter returns are negative for investment using relative strengths strategies. Moreover, the authors use one specific strategy, 6×6 strategy that does not skip a week, to be the representative of other strategies and conduct further analyses in details. They decompose the sources of profitability of relative strength strategies and find no evidence of systematic risk in trading or lead-lag effect from delayed stock price reactions to common factors. However, these evidences support the idea that profitability might come from the delayed price reactions to firm-specific information.
Despite the fact that this research provides strong evidences of market inefficiency, there are lots of debates questioning if the profits from relative strength strategies are either the compensation for the risk, or the consequence of data mining.
One side of the literature believes that the anomaly regarding the profitability of momentum investing is due to behavioral biases of investors. They are convinced that abnormal returns occurred because there is a delay in overreaction to information among investors. Thus, this drives up the prices of winners and draw down the prices of losers exceeding the limit of their long-term value. Besides, these behavioral models also predict that there should be excess returns of losers over winners when returns revert to their fundamental values in the subsequent periods. This perspective is supported by the papers from Barberis, Shleifer, and Vishny (1998), which have mentioned in details earlier in the previous chapter of this dissertation, Daniel, Hirshleifer, and Subrahmanyam (1998), as well as the paper of Hong and Stein (1999).
The other side of literature captures the weakness of this statement and argues that it is not sensible to reject the model that investors are rational. They propose the theory that momentum profits exist due to compensation for risk.
Conrad and Kaul (1998)
Jegadeesh and Titman (2001) carry out the same methodology from their former research with the new sample group of data to prove these criticisms. They employ additional nine years of data from the period between 1990 and 1998, and run out-of-sample tests.
To examine the conflicting arguments of two
Jegadeesh and Titman (2001)
2.4 Review of researches concerning profitability of momentum strategies in the UK market
2.5 Review of recently published empirical findings in international context
Cite This Work
To export a reference to this article please select a referencing stye below: