Contrarian investment strategies have become a dominant theme in finance especially since the performance of value stocks is not limited to the US market. In this study I will refer to value stock as those stocks that have relatively low price when compared with a fundamental or book value. On the contrary, glamour stocks are those with a relatively high price relative to their fundamental value. Therefore, contrarian investors would make a bet against the market. Researchers and academics believe that valuation remains the best arbiter of future returns and valuation opportunities are best presented in non-consensual ideas.
A contrarian philosophy is to invest against the herd, not for the sake of it, but to refute and reform the consensus. It needs to estimate the intrinsic value of a firm and purchase shares when their price is well below that value to enhance his/her returns and potentially decrease the risk exposure. Value strategies have proven to create the possibility for superior performance and past literature is in agreement that such performance is achievable however, the controversy lies in the reasons why such performance was achieved (Alan, et al., 2001). One the one hand, investors adopt the contrarian strategy to invest against those investors who extrapolate the past performance. On the other hand, some say that contrarian investments achieve higher returns because they are deemed to be riskier. Whichever way it is interpreted, contrarian investment strategies are one of the anomalies of the Efficient Market Hypothesis (EMH).
Efficient Market Hypothesis (EMH) is one of the most explored phenomenon in finance and investment. EMH states that the market has the ability to reflect the “true” prices of financial assets. In other words, when a new piece of information is released the market prices should reflect that information quickly and accurately (Lumby & Jones, 1999). Until the beginning of the twenty-first century, economists and the academia have accepted the EMH and agreed that market prices of financial assets had the ability to reflect all information available inclusive of new information without delay. Thus, neither fundamental analysis nor technical analysis would enable investors to achieve abnormal returns by holding a randomly selected portfolio of stock. However, in recent decades the academics began to dissect and dispute the EMH concluding that it is a much more complicated matter than they were led to believe by the standard finance theory. Particularly, they began to investigate the anomalies of EMH and discovered a different approach to finance theory such as the behavioral finance. Standard finance theory assumes that investors have little difficulty making financial decisions because they make those decisions based on prompt and accurate information, not influenced by their emotions. Behavioral finance however, examines the psychology of financial decision-making using mechanisms such as cognitive errors, overconfidence, role of biases in decision making, such as the rule of thumb when making certain investment decisions, the pain of regret and problems of self-control, against the standard finance theories such as capital asset pricing theory, option pricing and arbitrage theories (Statman, 1995).
The aim of this paper is to investigate mean reversion, that is, the investors’ ability to earn profits by investing in portfolio of assets that have been sorted using accounting ratios. To be more specific, the investors would buy stocks with low market prices relative to their fundamental values. Researchers assume that there is a negative serial correlation in market prices (Brouwer, et al., 1997) giving contrarian investors the ability to pursue the value premium by investing against the market trend.
Lakonishok, et al., 1994, Basu, 1977 and Fama and French 1992 have shown that stocks with higher earnings (dividends, book-to-value or other measures of value) to price ratios have earned higher returns than those with lower ratios, over the subsequent years. These results and other previous literature have raised the question: How efficient are financial markets?
Lumby and Jones, 1999, have identified three levels of efficient markets:
Weak form efficiency
Semi- strong efficiency
Strong form efficiency.
The idea that market prices incorporate all available information is no longer accepted. Fama, 1970 carried out tests to determine how efficient financial markets really are, against the three efficiency levels. Weak form efficiency implies that market prices reflect all market information and states that past return will have no effect on future returns making technical analysis a redundant tool for earning higher returns. Semi-strong efficiency implies that market prices reflect all publicly available information stating that a company’s stock price reflects all publicly available information making fundamental analysis a redundant tool since it cannot give any advantage to investors to find mispriced stocks. Strong form efficiency implies that market prices reflect all publicly and privately available information such as insider information that is not yet released by the company. This is an extreme case scenario which gives investors an exploitative power over the information regarding the market price formation.
Some academics assimilate the contrarian investment with the overreaction hypothesis claiming that people tend to overreact to good or bad performance of a company and argue that as a consequence of the overreaction of the market, value stocks significantly outperform glamour stock (Andrew and Craig, 1990). Further more recent research suggests that the effect of the overreaction hypothesis is more pronounced in some parts of the world than others. For example, Doan, et al., 2014, through their research on the Australian stock market have shown that, depending on the time horizon, the contrarian strategies prevails in the short-term and momentum strategies prevail in the medium- to long-term. Their study of the momentum and contrarian investment strategies in the Australian stock market found mixed findings of previous studies in the same market.
Therefore this study aims to examine this phenomenon on the UK equity market and it will be interesting to see how contrarian investment strategies perform over the recent financial crisis. A collection of 113 stocks will be analyzed using fundamental analysis for a period of 10 years, 2000-2010, where the returns for subsequent five years will be monitored. Section 2 will look at the existing literature on contrarian investments and its causes. Section 3 will detail the methodology used to analyze and construct the portfolios. In this section I will provide additional data on which ratios will be used to differentiate between value and glamour stocks. Section 4 will present and interpret the results of the study. Section 5 will present the concluding remarks and recommendations for improvements.
2. LITERATURE REVIEW
2.1 BEHAVIORAL FINANCE
We currently find ourselves in an environment of low growth, which reduces the margin for investment error. In a high return environment mistakes are less costly because the returns can help overcome the frustration. Nowadays, the search for enhanced returns requires ever more discipline, conviction, ingenuity and courage. As humans we have to make many decisions as way of living and are faced with questions such as: Should I have a healthy lunch or a McDonald’s burger? How much should I tip my waiter? We make these decisions with ease by using a set of rules of thumb called “heuristic” rules that allow us to run our lives without which we would be paralyzed due to the volume of daily choices we would be faced. Behavioral finance is the study of those and other financial decision making biases that can be avoided, if we are aware what caused them. Below I have discussed some of the behavioral biases that affect the investment behavior.
Confident investors tend to overestimate their abilities when picking stocks and tend to overlook broader factors that influence the performance of their portfolio. Overconfidence is often linked with too much trading which could have a negative effect on their portfolio. Professors Brad Barber and Terry Odean analyzed the returns of US investors, differentiating between them as most active and least active traders and found that those investors with the lower trading activity achieve a higher portfolio return than those who were most active (Barber and Odean, 1999).
2.1.2 Representativeness heuristic
Kahneman and Tversky introduced representativeness heuristic as part of their research into cognitive error and it is used to analyze an individual’s behavior when he/ she is face with the probability of an event of uncertainty. The results of many experiments conducted by them have shown that individuals judged the probability of an event by its representativeness, i.e. event A is judged more probable than event B if event A is more representative than event B (Kahneman & Tversky, 1974).
One of the interesting experiment performed by Kahneman and Tversky is when a set of participants were given the following information about Linda; “she is single, outspoken, 31 years old and very bright. While studying philosophy, she was very concerned with issues of discrimination and social justice and was involved in antinuclear demonstrations”. The participants would then had to choose which one of the following event is most probable:
Linda is a bank teller (A), or
Linda is a bank teller and is active in the feminist movement (B).
The results have shown that 87% of the participants have chosen event B as the most probable event. This is a surprising result since the rules of probability state a combination of 2 events cannot be more probable than one simpler event. Therefore, this proves that the vast majority of the participants have used representativeness heuristic to judge Linda’s ability as a bank teller (Kahneman & Tversky, 1983).
Investing in the “best” business in the stock universe can still be a bad investment if you invested at the wrong price. Likewise, a “bad” business can become a profitable investment if it has been bought at a relatively low price. Fortunately for contrarian investors, markets are not always efficient and investors are not always rational resulting in mispriced businesses where contrarianism has its best effect.
Reversion is a property of the stock market that is not fully appreciated by al investors. A significant part of the literature based on financial markets suggests that investors tend to overreact to certain information. If a company announced a good performance, investors tend to be over optimistic in their forecasts and likewise, if the company has announced bad performance they tend to be over pessimistic (Andrew and Craig, 1980). However, following the market’s overreaction the price will tend to move towards its fundamental value suggesting that there is some of reversion in returns. This suggests that if stocks that have done well in the past does not mean they will continue to do well in the future and vice versa, stocks that have performed poorly in the past will tend to do better in the future. Therefore a successful implementation of the contrarian investment strategy implies investing in a portfolio of stocks that have been performing poorly in the past and their price is therefore below their fundamental value.
DeBondt and Thaler, early in1990s, investigated whether the stock market overreacts to information by looking at the behavior of security analysts, what they considered to be a “reasonable source of rationality”. Their conclusion was that we are humans which means that we are prone to make naïve forecasts which are too extreme and definitely not rational resulting in an anomalous market. Interestingly, they have found the same pattern in economists’ forecasts when analyzing exchange rates and macroeconomic variables and concluded that “overreaction can pervade even the most professional of predictions”. Perhaps they are correct and the recent financial crises of 2008 has proven that fear makes the best of us.
This study is based on the Lakonishok, Shleifer and Vishny’s paper “Contrarian investment, extrapolation and risk” from 1994 and it is built on the belief that investors to not fully appreciate reversion. They classified stocks into value and glamour stocks based on financial ratios such as book to market ratio (B/M), cash flow to price ratio (CF/P), earnings to price ratios (E/P) and growth in sales (GS). They gathered stock traded on either New York Stock Exchange (NYSE) or the American Stock Exchange (AMEX). In order to build their portfolios, at the end of April in any given year they ranked the stocks according to a fundamental ratios and then divided the stocks into 10 portfolios where Portfolio 1 was made up of stocks with the lowest value of the ratio and Portfolio 10 comprised the stocks with the highest value of the ratio. Over the next five years they have monitored the performance of each portfolio.
They have found that value stocks (for example, Portfolio 10 out of 10) have outperformed the glamour stocks, even in the first year after the portfolio formation. The value stocks realized an astonishing average return of 19.8% compared with 9.3% of the glamour stocks. By adopting the buy-and-hold strategy for the five subsequent years, the value portfolio yielded 146.2% whereas the glamour portfolio yielded only 56%. An important fact that they outlined through their paper is that the gap between returns of value and glamour stocks increases as the years increase. Thus, contrarian investment is a profitable strategy in the long-term.
Furthermore, they investigated if these results are affect by the size of the companies and if some ratios were more important than others. Not surprisingly, value stock still outperformed the glamour stocks but interestingly they have found that GS and CF/P were the main drivers of the results contradict with Fama & French, (1992) who stated that equilibrium returns are best defined by the B/M, E/P and size and suggested that these ratios can be identifies as measures of risk.
These results were reproduced by both David (1994) based on a sample of large US companies over a 30 year period and Chan, Hamao and Lakonishok (1991) investigating the Japanese stock market. Extensive research produced by Dreman, 1997, and Graham and Dodd, 1934 have proven that value strategies have outperformed the glamour strategies and hence beat the market. They used the same methodology and applied fundamental analysis to identify value stocks from glamour stock. Their findings accentuates the expectation that value stock outperform glamour stock, particularly those stock with higher earning-to-price ratios are likely to earn higher returns ((Basu, 1977), (Chan, et al., 1991) and (Fama & French, 1992)).
The question what we should ask ourselves is: How likely are the excess returns of value stocks to persist in the long term?
Fama and French, (1990) attempted to explain the outperformance tendency of value strategies by adjusting their asset pricing model to include size and value as factors of market risk in the capital asset pricing model (CAPM). The famous three-factor model reflects that the value and small cap stocks outperform the market. Debating the three-factor model, Gregory, et al., (2001), through a comprehensive classification of stocks: one-way categorization as used by Fama and French in the three-factor model and two-way categorization based on both past performance and future expected performance. For the period of January 1975 to December 1998, using the one-way classification of stocks, they have found the three-factor model broadly explained the differences in returns of value and glamour stocks. However, the two-way classification portfolios shown that value stocks significantly outperformed glamour stocks in the UK and these results are robust when controlling for book-to-market and size factors. These results contained differences that were not accounted by the three-factor model, which is nothing more but a rational risk-pricing model.
Antoniu, Galariotus and Spyrou, (2006), produced a study on the London Stock Exchange to identify whether short-term contrarian profits can be achieve and investigated what are the sources of such profits. Based on the Fama and French (1996) three-factor model they have decomposed the profits according to these factors and the results have shown that in the UK contrarian investment strategies outperformed the market and most importantly these results are robust even when taking into account the risk, seasonality and market frictions characteristics of the portfolios. Their conclusion was that investors overreact to firm specific information.
These finding are parallel to Lakonishok, et al., (1994) who have shown that glamour stocks underperform value stocks when a book-to-market strategy, as well as other financial ratios, is used. Most importantly in their research, they provide evidence to other critics of contrarian investments such as “the size effect”, the claim that value portfolios are riskier and optimism and pessimism of investors.
Some argue that smaller companies face a larger possibility of bankruptcy and therefore are part of a less competitive market requiring a higher return to compensate investors for taking on the additional risk. Lakonishok, et al., (1994) based on market capitalization selected the top 50% of the companies in their portfolio and evaluated the portfolio. Not surprisingly they have found the same results as before and hence, disproving the size effect.
Another factor of increased profitability of contrarian investment strategies is inadequate treatment of risk. In order to address this issue, Lakonishok, et al., (1994), analyze the difference in returns of value stocks and returns of glamour stocks and evaluate how the strategy will perform over time. They have found out of the 22 years examined only in 5 years the glamour stocks outperformed value stocks, out of which 4 have incremental differences and only in 1979 the difference was more significant, 16.8%. Furthermore, when analyzing the broader European market, Brouwer, et al., (1997) have found similar results as Lakonishok, et al., (1994) where the variance of stocks explained only a small part of the return differences between value and glamour stocks.
Optimism and pessimism is the result of forecasts made without a full appreciation of reversion. Using the two-way classification of stocks they separated those stocks that appear to be in favor of the market defined by growth of sales (GS) and cash flow to price (CF/P) ratio. In the pre-portfolio formation period glamour stocks had a lower CF/P ratio of 0.08 compared to value stocks CF/P ratio of 0.279 and outperformed the value stocks with returns rising by 139% for glamour stock and only by 22.5% for value stocks, a clear indication that the market preferred the glamour stocks and had more optimistic expectations. In the contrary in the post-portfolio formation period the glamour stocks had shown a much less impressive growth in fundamentals’ value. Particularly, the cash-flow performance of glamour stocks is lower than expected prior to portfolio formation even though over the 5-year period it reached a value almost double that of value stocks. The growth of sales performance of glamour stocks mimics a similar trait with pre-formation values being much higher than post-formation values outlining the consequences of investors being too extreme in their predictions.
2.4 MARKET EFFICIENCY
One of the most influential theories that the financial markets gave build upon is the Efficient Market Hypothesis (EMH). EMH states that the market has the ability to reflect the “true” value of stocks incorporating all available information and when a new piece of information is released, the market will quickly and accurately reflect it through its listed prices. Thus, fundamental analysis and technical analysis would prove useless in helping investors achieve abnormal returns. However the value of a company, the stock price, is calculated based on existing information about the company as well as future information such as investment opportunities that the company holds (Miller and Modigliani, 1961). Therefore, if the valuation of a stock in based on future interpretation how can it be classified as efficient?
A different approach to EMH is the “random walk” approach. The random walk approach implies that stock prices follow a random walk suggesting that subsequent prices of stocks are unaffected by previous stock prices (Malkiel, 1991). The logic behind this idea is that stock prices immediately reflect new information and that tomorrow’s price should then be independent from today’s price. However, tomorrow’s information is uncertain and unknown therefore it cannot be predicted, thus it is random. By following this logic the prices of these stock will also be random because they are based on future, unknown, information, hence the term “random walk”. Malkiel, in his 1991 paper, has shown that markets cannot be perfectly efficient otherwise there will be no incentive for fundamental analysts to dig through the market and find mispriced stocks if information is quickly reflected in stock prices . Exemptions from the rule of efficient markets and examples of irrational behavior are the crash of 1987 and the bubbles of 1999.
Lumby and Jones, 1999, have identified three levels of efficient markets:
Weak form efficiency
Semi- strong efficiency
Strong form efficiency.
Weak form efficiency suggests that market prices reflect all historical price movements of the underlying security. Fama 1970 has run tests on the weak form efficiency on a selection of stocks and have found some evidence of random share price movements. Therefore an attempt by an investor to use technical analysis to identify stock that would generate abnormal return would prove pointless. Nevertheless, a recent paper that examined the efficient market hypothesis in European markets has found that particularly France and UK reject the EMH because of the presence of mean reversion in the data, particularly strong in recent years (Borgess, 2010).
A less restraining form of efficiency is the semi-strong approach where all publicly available information is reflected in the market prices. Since all publicly released information of a company is incorporated in its stock price, fundamental analysis would not give investors the opportunity to gain higher returns by holding a randomly selected portfolio of stocks. Fama, et al., (1969) provided evidence in favor of semi-efficient markets by looking at the movement in stock prices before and after the company has made a bonus issue announcement. They have chosen a 30-month period both before and after the announcement to investigate whether a substantial return could be won. They have found that the share price increased 30 months prior to the announcement, but there was no change in the share prices after the announcement. One explanation that the authors have identified is changes in dividend level. A positive announcement of a company’s performance will lead to an increase in the level of dividends because the company is perceived more profitable and more likely to increase shareholders wealth.
The strong from of market efficiency is the most extreme because it claims that market prices incorporate all publicly available information as well private insider information that has not yet been released to the public. One of the many concerns is that insider information is illegal to hold and use when trading and it is difficult to obtain, otherwise it will not be insider information. It is extremely difficult to test the validity of this form of efficiency and as Fama,1970 suggested, at it best it should be used as a benchmark to market efficiency deviation.
3 DATA AND METHODOLOGY
3.1 Data and Sample Selection
The sample period covered in this study is from January 2002 to December 2012. The empirical analysis uses annual accounting data from the BloombergTM database (Bloomberg, 2018) and monthly return data extracted from Yahoo FinanceTM. In order to add value to existing literature, I have selected an up to date data sample with the latest year possible being 2012 because the portfolios’ return are monitored and calculated over a 5-year horizon. The stocks were selected from the FTSE 350 stock index because it comprises of both large and small companies. I excluded companies that belong from the financial sector such as banks because of the complicated nature of their balance sheet and additional risks that they are exposed to and I have also excluded companies that hold very small market share since they will not be viable investment opportunities for large investors (Doeswijk, 1997). Therefore, portfolios of stocks are formed annually and for each year, 100 stocks are analyzed and grouped according to their fundamental values. In order to avoid the rise of ex-post selection bias, companies that seem to be underperforming were not discard from the sample (Banz and Breen, 1986). The formation of the portfolios is made at the end of December of the previous year, for example a portfolio formed in 2002 begins with data from the end of December of 2001.
The fundamental values used in this study are book-to-market and cash flow-to-market ratios. Firms are assessed and ranked according to these values in order to form the portfolios. In order to avoid negative ratios and distortion in the results, firms with negative values in the formation year were replaced with firms that displayed positive values. Hence, why 115 stocks were obtained from the UK stock index even though only 100 stocks were used in each year. In order to calculate the fundamental ratios balance sheet data for the companies was taken from the Bloomberg database (Bloomberg, 2018) and it was used to compute percentage of the book value of equity form their book-to-market ratios. In doing so, “total equity” is used as a proxy for book value of equity . From the cash flow statements generated by BloombergTM Database (Bloomberg, 2018) I extracted the cash flow from operating activities and calculated the cash flow-to-market ratio, which is normally symbolized by the cash generated by the company’s operations, net income plus depreciation and amortization. Market capitalization of the company is used as a proxy of market value of equity and is computed by multiplying the total number of common shares outstanding with the corresponding share price. However, this value was also taken from the BloombergTM Database (Bloomberg, 2018).
For each portfolio, I compute the return in each of the five years following the formation of the portfolio, i.e. the per year return, the average annual return and the cumulative five year return using annual compounding. Therefore data for the market prices and returns had to be obtained and estimated from 2002 to 2016 even though the sample period ends in 2012. In order to calculate the returns and to capture a better picture of the company’s return, the adjusted closing prices were obtained from Yahoo FinanceTM for all 115 companies. This is particularly important because the adjusted closing prices comprise changes in stock prices, dividends payments, etc. prior to the next day’s opening price.
To summarize the accounting variables used in this study are book-to-market value of equity (B/M) and cash flow to market ratio (CF/M). The following section will provide more details on how portfolios and returns are constructed.
3.2 Portfolio Formation
Following the Lakonishok, et al., (1994) definition of value and glamour stocks, this study will consider value stock be to those that have market value lower than its fundamental value and glamour stocks those that have a relative high market value relative to its fundamental value. Therefore, value stocks will be those with higher B/M and CF/M ratios whereas glamour stock will be defined by the lower B/M and CF/M ratios.
The first step in constructing the portfolios is to rank the portfolios in each of the formation years based on the computed B/M and CF/M ratios, where rank 1 is the highest ratio and rank 100 represents the lowest ratio. As the equations below outline, I have calculated the B/M ratio and CF/M ratios by dividing the total and cash flow to market capitalization respectively. As mentioned above, any stocks that experienced negative fundamental values were excluded from the sample and replace with stock that had positive values for that formation year to avoid distortion of the results. By the end of this exercise we should have a redefined list of stock presented in a descending order of the accounting ratio.
B/M=(Total Equity )/(Market Capitalisation) Equation (1)
CF/M=(Cash Flow )/(Market Capitalisation) Equation (2)
The second step in constructing the portfolios involves calculating the returns and segmenting the ranked list of stock into deciles. Holding period return (HPR) is an important aspect of this step. HPR is defined as the total return received from holding an asset over a period of time. It is normally expressed a percentage and in annualized terms. This study analyzes the buy-and-hold strategy for the stocks and therefore, HPR are calculated for the 5 subsequent years. The equation (3) used is outlined below and for the purpose of this study t will take the value of the holding periods 1, 2, 3, 4, and 5 years. The HPR values will then be used to compute the cumulative returns for each of the stock in every formation year. For example, for the formation year 2004, the X stock would consist of its HPR for 2004, 2005, 2006, 2007 and 2008, representing the HPR gained from holding the X stock from 2004.
HPR=( Ptn-Pt)/Pt Equation (3)
The third step is segmenting the portfolios into deciles. The stocks ranked based on their fundamental values are grouped into portfolios of 10 stocks each. Portfolio 1 will therefore comprise of the first 10 highest values of subsequent returns, portfolio 2 will comprise of the following 10 subsequent returns and so on until the last 10 stock remain to form portfolio 10, the lowest ranked values. The average returns for each of the ten stocks are calculated reducing the portfolio size to a set of 5 returns for each subsequent year.
In the fourth and final step we need to commute the average returns (AR) and the cumulative returns (CR) for each of the portfolios. The AR is the mathematical average of per-year returns for a period of time and is calculated by dividing the return in a given year by the number of years that stock has been held for. The tables below numbered (1) and (2) show these returns as R1, R2, R3, R4 and R5, where the numbers represent the holding periods. Cumulative returns (CR) shows the total return what was gained from buying and holding investment. For example, CR5 represents the total return gained by holding an investment from the formation year until year 5.
Contrarian investment strategies are successful only if the value and glamour stocks are correctly defined. Therefore, based on the contrarian models and the information presented in the results tables below, we can classify portfolio 1 as the extreme value portfolio because it comprises the highest values of B/M and CF/M ratios and portfolio 10 as the extreme glamour portfolio because of the lowest values of B/M and CF/M ratios.
The study has followed the one-way classification of stocks implemented by Lakonishok, et al., (1994), because it has been adopted by many others, particularly Gregory, et al., (2001), and creates the opportunity to compare these results and see how they fit in with the consensus.
Tables 1 and 2 below outline the results found in this study of UK stock market. We will not analyze these returns by looking at each of the strategies separately.
4.1 Book-to-Market Classification
Table 1 shows the results for the one-way classification on the basis of BM. For each decile portfolio the return in each of the five years following formation is computed, as well as the average returns and the cumulative 5-year returns. The table clearly outlines that for the chosen sample value stocks outperformed the glamour stock, even in the first year after formation. These finding are in line with Lakonishok, et al., (1994), that this study has followed closely, but also similar with the results of Gregory et al., (2001). On average over the post-formation period the glamour stock have an annual average return of 15 percent and the value stocks reported an annual average return of 19 percent, resulting in a difference of 4 percent per year similar to Gregory, et al., (2001) who has found an difference in annual return of 6.34 percent. Furthermore, if the portfolios are held for the whole 5-year period the cumulative results show a difference of 40 percent between value and glamour stocks. The results are in line with both Lakonishok et al., (1994) and Gregory, et al., (2001), however, it is worth mentioning that the results reported by both Gregory, et al., (2001) and Lakonishok et al., (1994) show a much more pronounced effect of contrarian investment reporting a difference between value and glamour stocks of 78.83% and 161% respectively. The results based on the book-to-market ratio therefore reveal that there are significant value effects for the UK over this period and there is an average value premium of over 4 percent per annum.
We will now look at how the portfolios performed when we have classified them based in the cash flow-to-market ratio.
4.2 Cash Flow-to-Market Classification
Table 2 shows the results for the one-way classification on the basis of CF/M. The results show once again that by employing a buy-and-hold strategy of value stocks you will be able to profit from much larger returns. Particularly, the average yearly returns mimic the results of the strategy when book-to-market ratio is used, projecting a difference of approximately 4 percent per year. These results correspond with the findings of Gregory, et al., (2001) who reported a difference in the average annual return of 6.42 percent. Lakonishok, et al., (1994) have identified an even larger difference between value and glamour returns of 10.8 percent. On average, by holding the portfolios over the 5-year period, value stocks reported a cumulative return of 179 percent whereas the glamour stocks achieved only 119 percent, resulting in a difference of 60 percent. When categorizing stocks based on CF/M we cumulative returns are much higher than those when employing B/M categorization, approximately 20 percent difference. Once again, the results reported by both studies have found more pronounced results where Gregory, et al., (2001) found a difference of 72.60 percent in cumulative returns and Lakonishok, et al., (1994) have found 95.1 percent. Sorting based on CF/M thus appear to produce somewhat bigger differenced in returns than when sorting is based on B/M ratio implying that better value strategies arise when determining more directly market expectation of future growth.
In this section, the results have confirmed the results of previous studies and proved that value strategies based on classification of firms by a single fundamental ratio produce large results over the 10 – year period 2002 – 2012. Thus, there is scope for contrarian investment. The paper confirmed that by choosing value stocks, those that seem to be out of favor with the market and display a lower market price than their fundamental value, achieve much larger returns than their counterparts whose market price is much higher relative to their fundamental value. However, it begs the question, if those fundamental values clearly indicate that those stock are overpriced, why do investors chose them? One potential answer is that fundamental ratios such as the book to market (B/M) ratio when computed fails to take into account factors such as intangible assets (research and development) and growth opportunities Lakonishok, et al., (1994). A good example is Facebook. For the first 5 years once listed Facebook operating with a negative balance sheet and were not able to produce any profit, yet its market price was relatively high. This is because when investors compute the company valuations they took into account growth opportunities that the B/M ratio does not compute (Miller and Modigliani, (1961)).
Overall the findings of this paper outlines that market continues to misprice stocks and in line with existing literature investor do not always make rational judgments of market expectations. This contradicts with the consensus and questions the efficient market hypothesis validity that markets reflect all available information regarding the underling company. If this is correct the portfolios should not generate excess returns and fundamental analysis should not add value when constructing the portfolios. However, as the findings suggest there is scope for contrarian investing.
While this study has been conducted on an up to date sample, it did not cover all the fundamental ratios that were used in the Lakonishok, et al., (1994) and Gregory, et al., (2001) to demonstrate whether the results persist when a two-way classification of stocks is applied. Intriguingly, when Gregory, et al., (2001) used the two-way classification of stocks they have found that the results support contrarian investment, reporting an astonishing difference in cumulative return for the 5-year holding period of 141.47 percent between value and glamour stocks. Lakonishok, et al., (1994) has found a relatively high value of 100 percent difference in the cumulative returns. Whilst the findings in this paper correspond with their findings when a one-way classification of stock is applied, it would have been interesting to see whether the value of contrarian investing based on this sample is as strong as their findings given a more recent data sample.
When selecting stocks, a random sample of companies was chosen however; market capitalization was monitored to avoid choosing only large cap organizations from the index. Having said that, this study did not take into account the size-adjusted returns to test whether the sixe of the firms had any influence on the returns generated by value and glamour stocks. Gregory, et al., (2001) has found that when adjusting for size, value stock persistently outperform glamour stock, however there is a slight reduction on the difference. Moreover when, Lakonishok, et al., (1994) took into account the size of the firms they have found that in the first 2 years after the formation, glamour portfolios reported negative returns while value portfolio achieved high positive returns. Again, this leave room for improvement in this study and could introduced the size-adjusted returns to identify the behavior of value and glamour stocks.
A further improvement to the study is the introduction of earnings to price ratio and growth of sales to analyse returns or adopting the Miller and Modigliani, (1961) company valuation approach to estimate each company valuation and then construct the portfolios on these valuations. It provides insight into how different the market price is from the valuation based on fundamental values. However, this is a question left for another time.
This study, following the methodology of Lakonishok, et al., (1994), has proved that there are significant value effects in returns for the UK stock market by using two different measures of value: book-to-market and cash flow-to-market. The paper provided evidence that for the period of January 2002 to December 2012 value stocks have consistently outperformed the glamour stocks and these results correspond with findings of Lakonishok, Shleifer and Vishny (1994) when analyzing the US stock market.