Economists are always interested in both Market behaviour and the psychology of individuals while they make a decision. This study was undertaken to investigate that these phenomenon are related to each other by more than just appearance.
The term Overreaction carries with it an implicit comparison to some degree of reaction that is considered to be appropriate. In revising their beliefs, individuals tend to pay more importance to recent information and less weightage to prior data. There is also considerable evidence that the actual expectations of professional security analysts and economic forecasters display the same Overreaction bias.
In this report I looked at the UK stock Market and checked that weather or not does the UK stock market overreacts. Due to limited time and resources, the research was done on a very small scale. I randomly selected 6 companies from FTSE 100, viz-a-viz Burberry PLC, Glaxo Smith Kline PLC, Lloyds Banking PLC, J. Sainsbury PLC, Tesco PLC and Marks and Spencer PLC.
Overreaction in this paper was proved by showing that stocks which have underperformed the market over a period of time (3 years in the report) will outperform over a subsequent and similar time period.
There are 3 main objectives to the research:
To understand Overreaction Hypothesis
To study the various factors that lead to the phenomenon of overreaction in the stock market
To check weather UK stock data reflect overreaction hypothesis
The results show that previous losers have tended to subsequently outperform previous winners over the period November 2009 to November 2012, hence proving the UK Stock exchange does show Overreaction.
CHAPTER 1: INTRODUCTION
UK Stock Exchange
London Stock Exchange Logo.svg
The London Stock Exchange is a stock exchange located in the City of London in the United Kingdom. As of December 2011, the Exchange had a market capitalisation of US$3.266, making it the fourth-largest stock exchange in the world by this measurement and the largest in Europe.
Location London, England, United Kingdom
Founded 1801
Owner London Stock Exchange Group
Key people Christopher S. Gibson-Smith, (Chairman)
Xavier Rolet, (CEO)
Currency GBX
No. of listings 2,864 (as of December 2011)
Market Cap US$3.2 trillion (Dec 2011)
Volume US$1.7 trillion (Dec 2009)
Indexes FTSE 100 Index
FTSE 250 Index
FTSE 350 Index
FTSE Small Cap Index
FTSE All-Share Index
Website londonstockexchange.com
FTSE 100 Index
The FTSE 100 Index, also called FTSE 100, FTSE, or, informally, the "footsie" , is a share index of the 100 companies listed on the London Stock Exchange with the highest market capitalisation. It is one of the most widely used stock indices and is seen as a gauge of business prosperity. The index is maintained by the FTSE Group, a subsidiary of the London Stock Exchange Group. The index began on 3 January 1984 at the base level of 1000; the highest value reached to date is 6950.6, on 30 December 1999. After falling during the financial crisis of 2007-2010 to below 3500 in March 2009, the index recovered to a peak of 6091.33 on 8 February 2011, it's highest since mid-2008. It fell under the 5000 mark on the morning of 23 September 2011. As of 21 October 2012 it was at 5896.15.
Largest constituents of FTSE index
The following table lists the ten largest FTSE 100 companies measured by market capitalisation as of 9 March 2011.
Rank
Company
Sector
Market capitalisation (£ billion)
1
BHP Billiton
Mining
148
2
Royal Dutch Shell
Oil and gas
135
3
HSBC
Financial services
118
4
Vodafone Group
Telecommunications
93
5
BP
Oil and gas
91
6
Rio Tinto Group
Mining
86
7
GlaxoSmithKline
Pharmaceuticals
61
8
Unilever
Consumer goods
56
9
British American Tobacco
Tobacco
49
10
BG Group
Oil and gas
49
FTSE 100 index chart
File:FTSE 100 Index.png
Source: http://en.wikipedia.org/wiki/London_Stock_Exchange
Overreaction Hypothesis
There is now substantial evidence to suggest that asset returns are predictable over both long and short time horizons for both individual stocks and stockmarket indices. Whether or not such predictability reflects changes in expected returns is not at all clear (see Fama, 1991).
In this paper we investigate a potentially important aspect of long-run mesm reversion in stock prices known as the Overreaction Hypothesis:^ This hypothesis is based on US evidence presented by DeBondt and Thaler (1985 and 1987) which showed that stocks which experienced poor performance over the past three to five year period (losers) tend to outperform prior period winners over the following three to five years. Using UK data from 1955 to 1990 drawn from a random sample of up to 1000 stocks in any one year, we find that losers outperform previous winners over a two year period by a statistically significant 1.7% per annum. On further investigation we find that such overreation may in fact be a manifestation of the smcdl firm effect. This paper is constructed as follows: in the next section we review the relevant literature; in the third section we outline our methodology and data; in the fourth section we test the Overreaction Hypothesis using disaggregated stock data and discuss the results; in the fifth section we control for firm size and focus on possible seasonal differences in the returns of losers and winners: the final section concludes the paper.
The Overreaction Hypothesis asserts that stocks which have underperformed the market over a period of time (often one to five years) will outperform the market over a subsequent and similar time period. This was first noted by * The authors are respectively from the Department of Economics, University of Reading, and the European Business Msinagement School, University of Wales. They would like to thank Padmini Kurukularaachi for her help in the preparation of this paper and the anonymous referee for helpful comments on an earlier draft of the paper. The financial support of the ESRC is gratefully acknowledged. (Paper received July 1993, revised and accepted May 1994) AddresB for correspondence: Andrew Clare, Department of Economics, University of Reading, Reading RG6 2AA, UK.
DeBondt and Thaler (1985) who found, using US data from 1926 to 1982, that previous losers outperformed previous winners by nearly 25% over the subsequent three year period.
Two explanations for this phenomenon have been proposed. The first is that the overreaction is a manifestation of the size effect, i.e. that losers tend to be small and that small firms outperform large firms. Zarowin (1990) and Chopra et al. (1992) investigate the impact of the size effect within the overreaction hypothesis using US data and find that adjustment for size does reduce the extra return available from losers. Indeed, Zarowin (1990) believes that all ofthe extra return can be explained by the size effect.^ DeBondt and Thaler (1985) did not believe that their results were due to the size effect, but they did not carry out the rigorous controls which Zarowin (1990) conducted, to prove this proposition.* The second explanation is that the reversal in returns reflects changes in equilibrium required returns which are not controlled for in DeBondt and Thaler (1985) - Chan (1988), Ball and Kothari (1989), and Chopra et al. (1992) all take this view. It is indeed likely that extreme changes in leverage will feed through to large changes in a firm's CAPM beta. Ball and Kothari (1989) find that the betas of extreme losers exceed the betas of extreme winners by 0.76 in the period following portfolio formation. Clearly such differences in betas, taken in conjunction with historical experience of risk premia can explain large differences in realised returns.
Although all the authors above use US data, there are some potentially important differences in their methodologies, particularly with respect to their definitions of winners and losers. DeBondt and Thaler (1985) defined winners and losers as being the best and worst 35 performing stocks over their monitoring period respectively. This number represented a relatively large proportion of stocks in 1926, but a far smaller percentage at the end of their sample period in 1982. Zarowin (1990) considers the top and bottom quintiles of stocks by performance; Kryzanowski and Zhang (1992) using Canadian stock data, look at extreme deciles of stocks; and Chopra et al. (1992) only consider the extreme 5% of performers. There is no obvious definition of 'extreme' winners and losers, but such differences could explain the contradictory research findings, particularly since all studies except for Kryzanowski and Zhang (1992), use basically the same US data. The findings of Chopra et al. (1992) that extra returns are available after controlling for firm size effects may well refiect the fact that the bottom 5% of stocks was used, compared to the bottom 20% used by Zarowin (1990). Kryzanowski and Zhang (1992) find that 'continuation', rather than overreaction is a feature of the Canadian equity market.
One feature common to the above studies is the existence of the January effect. Indeed this phenomenon is put forward as a possible explanation for the additional returns available from holding losers (Zarowin, 1990). Around 80% of the extra return available from investing in losers in the studies is available in January; however, since a high proportion of over the year returns is available in January this is not at all surprising and can in no way be considered an explanation of the Overreaction Hypothesis.
CHAPTER 2: LITERATURE REVIEW
"Does the stock market overreact?" , F.M. De Bondt and Richard Thaler, 1985
Research in experimental psychology has suggested that, in violation of Bayes' rule, most people 'Overreact' to unexpected and dramatic news events. The question then arises whether such behavious matter at the market level.
Consistent with the predictions of the overreaction hypothesis, portfolios of prior "losers" are found to outperform "winners". Thirty six months after portfolio formation, the losing stocks have earned about 25% more than the winners, even though the latter are significantly more risky.
Several aspects of the results remain without adequate explanation; most importantly, the large positive excess returns earned by the loser portfolios every January. Much to our surprise, the effect is observed aslast eas five years after portfolio formation.
"The Overreaction Hypothesis and the UK Stockmarket", Andrew Clare and Stephen Thomas, 1995
In this paper we tested the Overreaction Hypothesis using disaggregated UK stock price data. Our results show that previous losers have tended to subsequently outperform previous winners over the period 1955 to 1990, although the difference in performance is probably economically insignificant. We tested for asymmetric differences in post portfolio formation performance between large and small firms. We found that losers tended to be sm£ill and that the limited overreaction effects documented were probably due to the size effect. Furthermore, Ball and Kothari (1989) find that when annual return rather than monthly return data is used, support for the Overreaction Hypothesis becomes weaker; hence our results for the UK using monthly return data should be viewed with some caution.
"Do Markets overreact: International Evidence", Ahmet Baytas and Nusret Cakici, 1999
In this paper, using the Conrad and Kaul's methodology we test for the overreaction hypothesis - which maintains that stock prices systematically overshoot and therefore their reversal can be predicted from past performance - in seven industrialized countries. Consistent with the findings of Conrad and Kaul, we see no evidence of overreaction in the US. However, returns to long-term contrarian strategies in other countries seem to be generally significant. Moreover, we find that in the majority of the countries, while returns to arbitrage portfolios based on price are higher than those based on size, the latter generally outperform the winner-loser arbitrage portfolios.
"Further Evidence On Investor Overreaction and Stock Market Seasonality", F.M. De Bondt and Richard Thaler, 1987
In a previous paper, we found systematic price reversals for stocks that experience extreme long-term gains or losses: Past losers significantly outperform past winners. We interpreted this finding as consistent with the behavioral hypothesis of investor overreaction. In this follow-up paper, additional evidence is reported that supports the overreaction hypothesis and that is inconsistent with two alternative hypotheses based on firm size and differences in risk, as measured by CAPM-betas. The seasonal pattern of returns is also examined. Excess returns in January are related to both short-term and long-term past performance, as well as to the previous year market return.
"Size, Seasonality, and Stock Market Overreaction" , Paul Zarowin,1990
Recent research finds that the prior period's worst stock return performers (losers) outperform the prior period's best return performers (winners) in the subsequent period. This potential violation of the efficient markets hypothesis is labeled the "overreaction" phenomenon. This paper shows that the tendency for losers to outperform winners is not due to investor overreaction, but to the tendency for losers to be smaller-sized firms than winners. When losers are compared to winners of equal size, there is little evidence of any return discrepancy, and in periods when winners are smaller than losers, winners outperform losers.
CHAPTER 3: RESEARCH METHODOLOGY
Purpose of Study
The main objective of the research conducted here was:
To understand Overreaction Hypothesis
To study overreaction hypothesis, that was proposed for the first time by F.M. DeBondt and Richard Thaler in detail. Understand what the author had done and tried to prove. Also take into account that the research conducted was for the US stock exchange data from 1926 to 1982. So the data available for the research was not digital.
Further reading about what are the other various theories about overreaction hypothesis. Checking what others researcher studied on the same topic. What are their findings.
To study the various factors that lead to the phenomenon of overreaction in the stock market.
What do the researches think is the other reasons behind overreaction besides market sentiments and human psychology. There can be factors such as time period or size of the firm etc. that the researchers might not have taken into account while the research. Understanding how these can affect the results and how others have researched and used that data to either support or decline the base research of overreaction hypothesis.
To check weather UK stock data reflect overreaction hypothesis
To check this, stock exchange data had to be analysed. I mainly followed the research methodology adopted by Andrew Clare and Stephen Thomas in their paper "Overreaction Hypothesis and the UK Stockmarket". Due to the time constraints and limitation of resources and source of data, I could not conduct the research as vast has the original.
This research was conducted to check that does overreaction hypothesis exists on a very small scale as well.
Research Design
For my research, I randomly picked out 6 Public Limited Companies(PLC) from FTSE 100.
These companies were:
Burberry
Glaxo Kline Smith
Tesco
Sainsbury
Llyods Banking
Marks and Spencer
The monthly opening and closing data for the last 36 month was recorded and analysed. For the analysis Total Return on stock is required.
Total Return on Stock = (P1 - P0) + D
P0
Where
P1 = Closing Value
P0 = Opening Value
D = Dividend
The data was collected and organised in MS Excel.
Source of the data was: http://www.finance.yahoo.co.uk
All the data collected was totally secondary and the sample size was 6. Time period of review was 36 months(3years). The data collected was empirically analysed through pre-set formulae, given and done by the DeBondt and Thaler (Debondt & Thaler, 1985).
Procedure
Once data collection and organisation was done, next part was to understand what all operation and empirical formulae had to be applied to get to a decisive conclusion for the research.
According to (Thomas & Clare, 1995; DeBondt & Thaler, 1987) the first step after data collection was to calculate total return on stock for every stock. Using the above mentioned formula I calculated the Return on each stock (marked as the green column in the above tables).
Next part was to calculate the Market adjusted return of each stock (Uit). Market adjusted return is calculated as follows:
Uit = Rit - Rmi
T = 1….n
n = 1…12, or n = 1…24, or n = 1…36
Where Rit is the return on the stock, and Rmi is the return of the market.
Once we have Uit for all the months for all the individual stocks, we can add the uit for 12/24/36 months for each stock and then calculate the average.
This average is used to form the portfolio. Portfolio formation is done by putting the averages in descending order and choosing the Winners and Losers.
The test which we perform on the average portfolio is based upon forming a 'difference' portfolio, where Average return on portfolio Winners (Rpw) subtracted from where Average return on portfolio Losers (Rpl). If the return on this portfolio (Rot) is insignificantly different from zero than we can reject the simple Overreaction Hypothesis, assuming that differences in transactions costs between winners _and losers do not influence Rpi; however, a significant positive value for /?£>,, could be taken as confirmation ofthe US results.
We use the formula:
Rdt = Rpl - Rpw = alpha + e
t = 1…n
Where alpha is a constant and e is a white noise error term. The subscript t represents the period after portfolio formation, where n is equal to 12, 24 and 36 for the one, two and three year overreaction tests respectively.
A simple t-test on the significance of the constant aj tells us whether there is a difference in the means of the winner and loser stocks (again notwithstanding differences in transactions costs). A significant and positive value for ct^ can be seen as confirmation of the overreaction hypothesis, i.e. the average performance of the losers is greater than the average performance of the winners.
Hence for the T-Test we take the hypothesis as:
H0: No Overreaction Exists in the stock market in the selected sample size
H1: Overreaction does exist in the stock market for the selected sample size
Also Annual Return is displayed just for reference purposes, so that the readers can get an idea about the market in the year of review.
Annual Return = [(Ending Value/ Beginning Value)^(1/t)] - 1
CHAPTER 4: DATA ANALYSIS AND FINDINGS
Below is the table of the analysed and computed data.
Significant alpha value clearly indicates that overreaction does exist in the UK stock market. And the increasing alpha value from 1st year to 3rd year also shows that Losers have outperformed the Winners as time span of review increases.
Hence proving that there is significant evidence to reject the NULL hypothesis in favour of alternate hypothesis.
1st Year (N=12)
2nd Year (N=24)
3rd Year (N=36)
Return on Losers
-0.01269
-0.01216
-0.00964
Return on Winners
0.02004
0.00233
0.00736
alpha (Value)
0.00852
0.01560
0.06788
alpha (%)
0.85%
1.56%
6.79%
Market Annual Return Percentage
0.35%
0.04%
0.12%
I also checked the tabulated the winners and losers for all three years, since the sample size was small. This was done just to see which all stocks actually showed growth in terms of return and actually over took the existing winners.
The table given below shows the winner and loser portfolio in each year
1st Year
2nd Year
3rd Year
Winners
Losers
Winners
Losers
Winners
Losers
Marks & Spencer
Burberry
Burberry
Marks & Spencer
Burberry
Glaxo Smith Kline
Llyods Banking
Glaxo Smith Kline
Glaxo Smith Kline
Tesco
Marks & Spencer
Tesco
J. Sainsbury
Tesco
J. Sainsbury
Lyods Banking
J. Sainsbury
Llyods Banking
Some findings:
Only J. Sainsbury has been the consistent winner throughout the three years.
Burberry had over taken and became the top winner in the second year itself.
Tesco has been consistently in the loser portfolio for the review period.
Galaxo Smith Kline has been fluctuating from Losers to Winner to Loser again.
Marks and Spencer has been fluctuating from Winners to Losers to Winners again.
CHAPTER 5: CONCLUSION
Conclusion
In this paper we tested the Overreaction Hypothesis using disaggregated UK stock price data. Even though the research was on a very small scale, my results show that previous losers have tended to subsequently outperform previous winners over the period November 2009 to November 2012, hence proving the UK Stockexchange does show Overreaction.
We found that when annual return rather than monthly return data is used, support for the Overreaction Hypothesis becomes weaker; hence our results for the UK using monthly return data should be viewed with some caution.
Limitations
The research's main limitations were:
Time constraint: Project had to be complete within a months' time, hence a vast study for the UK stock market couldn't be conducted.
Resources: The research was self-funded so resources were also a constraint.
Secondary Data: All the data that was collected was secondary, this research doesn't need primary data collection.
Size of the Firm: Due to the time constraint and to reduce the complexity of the research, size of the firm and other factors that might affect the overreaction hypothesis were not accounted for.
Due to the above mentioned the limitations of the project, the research had to be down sized from the original research conducted by (Debondt & Thaler, 1985) and the period under review was shortened to 3 years.
