This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.
A vast and extensive literature covers the quantitative (numerically measurable) information that affects stock market prices but little has been done to study the impact of qualitative information (e.g. managerial optimism and analyst recommendations). The substantial movements in stock prices do not always compliment the changes in the quantitative information. Thus, a better link is sought to be established by linking the stock price changes, quantitative variables and qualitative variables together (Shiller, 1981). Cutler, Poterba and Summers (1989) suggested that, “qualitative variables may help explain stock returns better.” Subsequent paragraphs discuss the studies and opinions that examine the effect of such qualitative information. The role of liquidity as a quantitative variable is also discussed.
Let us start with managerial optimism as the first variable affecting the performance of the ‘GC anomaly' affected firms. Managerial opinion such as earnings announcements and other voluntary disclosures in the quarterly and annual reports are instrumental in determining the daily stock market prices. In this part we will discuss the literature on managerial optimism (pessimism). This kind of information is often referred to as “soft information” and compliments the hard earnings news in understanding the market efficiency hypothesis. The soft information has a very important feature that it is not always true that it will give useful information (Crawford and Sobel, 1982 and Benabou and Laroque, 1992). But this type of information is used by every rational investor. The information is also more used by large investors as compared to the small investors. This is because the small investors absorb this type of information after it affects the market whereas the large investors use this information as soon as it is publically available.
Optimism as a subject of study for finance researchers came into existence because of two reasons. First, Managers underestimate uncertainty because they think they can control the firm's performance (March and Shapira, 1987). Second, Managers who are highly committed to their work because of their reputation, money at stake, and employment are more optimistic in their managerial behaviour (Gilson, 1989).
“Managerial soft information has two linguistic dimensions – net optimism and certainty. The net optimism dimension is associated with short-window announcement period returns”(Tetlock, Maytal Saar-Tsechansky & Sofus Macskass 2008). The post earnings announcement drift (the tendency of stocks to earn abnormally high returns following a favorable earnings announcement and to earn abnormally low returns following an unfavorable earnings announcement) is also justified by this dimension of the soft information.
The second dimension of soft information i.e., certainty is inversely related idiosyncratic volatility (the risk related to unique circumstances of a stock which can be eliminated easily through diversification) during the short window announcement interval” (Tetlock, Maytal Saar-Tsechansky & Sofus Macskass 2008). The soft information is negatively associated with variance resulting in greater levels of fluctuation in the stock prices post the release of such information. The certainty level of the soft information better captures or defines the nature of managerial information.
Managerial unexpected net optimism conveys such information which influences the hard earnings, released at the same point of time in the market. The certainty level in such managerial views is inversely proportional to the firm's abnormal returns. It has also been documented in the literature that the firm or managerial optimism is absorbed by different stocks differently (Elizabeth Demers and Clara Vega, 2008):
• Net optimism in high tech firms, high price-earning ratio firms, low quality of accounting data firms have a high level of net optimism.
• The firms which have larger analyst following and get a lot of media coverage also have a high level of net optimism.
• The changes in the prices of stocks due to change in the net optimism is quickly absorbed by the stock's turnover.
Therefore, it can be said that the demand for soft information in the market is increasing and is quickly and promptly appreciated by changes in the market prices. The studies like Tetlock, Saar-Tsechansky and Macskassy (2008), Engelberg (2007), Li, (2006) etc. suggest that the soft information like, linguistic managerial comments and media- expressed comments are used by investors and lead to stock price changes.
It has also been documented in the literature that the managerial net optimism not only affects the short window returns but also explains the post earnings announcement drift. Extending the work of Davis et al. (2007) and Demers & Vega (2008), it has been hypothesise that there is a partial and delayed reaction to the soft information of the managers. This links the post earnings announcement returns to the net optimism (unexpected). It has also been found that a larger reaction for soft information as compared to the hard earnings information in the recent past. The soft information has greater predictive powers for estimating the returns in the long run (Engelberg, 2007).
In terms of capital structure, Heaton (2002) advocated that optimistic managers are of the opinion that a project will yield more returns than it is expected to. Therefore, they think that the securities of their firm are undervalued in the market. That is why these managers try and communicate optimistic and positive information to the market, so that it incorporates such information in the otherwise undervalued stock prices. Heaton's results complimented the pecking order theory (prioritize everything) such that optimistic managers will be more inclined towards the pecking order type of behaviour. Also, managers who are more optimistic and overconfident will choose to raise capital through more of debt than other less optimistic or pessimistic managers.
The analysis of managerial net optimism discussed in the subsequent sections of this paper is done with the help of Diction linguistic software used by earlier studies such as Davis et al (2008). The overall results of Davis et al (2008) reveal that the managers more often than not use optimistic and pessimistic opinions about the future firm performance, in the earnings press releases to facilitate changes in the stock prices. To measure the manager's optimism in the earnings press releases we use DICTION (textual-analysis software). The software counts the number of optimistic words such as words of praise, inspiration, and satisfaction and also the number of pessimistic words such as failure, denial, etc.). The software then calculates the net level of optimism or pessimism.
The level of optimism in managerial opinions does not actually justify the rational behaviour that a manager is expected to employ. It has been discussed in the previous literature that high level of optimism in managers opinions in the press releases is accompanied with high level of future return on investment and vice versa. Also, investors quickly respond to such optimistic and pessimistic opinions of the managers. The relationship between the optimistic and pessimistic information and the other quantitative information has also been established in the past. It has been found that optimistic and pessimistic information actually contains information which is synchronal to the other quantitative information without which it does not make any sense. Also, the linguistic style of managers provides information to the investors in addition to that provided by the linguistic style of the earnings press releases (Hoskin et al, 1986).
In the subsequent part of this paper, a clear relationship has been established between optimism and return generated by the ‘going concern anomaly' affected firms. The null hypothesis for this test is that the highly liquid stocks will generate low returns (in the case of going concern anomaly affected firms).
Let us now review the second variable of our research i.e. liquidity of a going concern. Keynes (1930) proposed that, “an asset is more liquid than another if it is more certainly realisable at short notice without loss” (Keynes, 1930, p. 45). The concept of liquidity originates from the phenomenon that every investor who buys a stock today, has an intention to sell it in the future. To do this he will have to incur a transaction cost. This gives rise to the concept of discounting. Thus, it is the number of transactions or trading days that actually determine the level of liquidity for a particular stock. But of course, it remains very difficult to establish a benchmark level of liquidity for any particular type of a stock. The relationship between liquidity and returns is studied and analysed in the later part of this paper, for more than 500 US stocks which are under the threat of Going Concern anomaly.
Bagehot (1971) and Glosten and Harris (1988) advocated that wrong selection of stocks is the primary reason of illiquidity in the markets. They justify the negative relationship between return and liquidity. The return-liquidity relationship for large hybrid market in US is very well documented in the literature. The obvious negative relationship has always been seen in the regularly traded stocks. This confirms the prevalence of the positive liquidity premium. On the contrary, not much is known about the small pure order driven markets.
The unique cross-sectional return-liquidity relationship proposes that the liquidity is related to the firm size and possesses better explanatory power than the beta and size justification given by Banz, 1981 and Fama ad French, 1982 (Amihud and Mendelson, 1986).
Another research paper, Brennan and Subrahmanyam (1996) documented the relationship between trading volume and stock price changes. This relationship was captured by a variable, λ (Kyle, 1985). A negative relationship was established between return and liquidity among the then time existing stocks. Easley et al. (2002) showed that, “a trade-based measure of information risk is positively related to returns using NYSE data.” The information risk spread was then shown to be positively related to spreads and negatively related to turnover, which suggests that it too is a proxy for liquidity” (Ben R. Marshall, 2004). Further, Jones (2001) and Avramov et al. (2002) proposed that if a firm accounts for liquidity then its asset prices better adjust to the dynamic market conditions and hence generate better looking market returns.
Jegadeesh and Titman established a link between momentum and value strategies through pre-trading volumes. Pre-trading volume also defines the “magnitude and persistence of price momentum.” The pre-trading volume also justifies the “intermediate-horizon under-reaction” and “long-horizon over-reaction” effects. Pre-trading volume estimates the magnitude and tenacity of future price momentum. It has been found that high volume traders (frequent traders) face faster momentum reversals and vice versa.
Previous research e.g., Datar, Naik and Radcliffe (1998) mentions that there is an inverse relationship between volume and future returns in the long run. The Charles, Lee and Bhaskaran (2000) study stated the opposite and contradicted the liquidity volume and return relationship given above. The study also states that the excess returns to volume-based strategies are due to trading liquidities, e.g., firms with which have a higher current volume when compared to the volume 4 years ago, generate significantly lower returns in the future and vice versa. Also, the low (high) trading volume stocks possess the characteristics of value (glamour) investing (derived from the concept of informative and speculative investment strategies). Consequently, the stocks which possess low (high) trading volume are linked to bad (good) current operating performance, high (low) book-to-market ratios, less (more) analyst followings and low (high) stock returns over the previous 5-year period.
Also, the analysts predict low (high) long-term earnings growth for low (high) volume stocks but good (bad) future operating performance. Apparently, the market has been surprised by the consistent high (low) earnings of low (high) trading volume firms.
Further, two papers Campbell et al. (1993) and Blume et al. (1994) provide valuable features of a security by analysing its pre-trading volume. Blume et al. in particular uses the past prices and past volumes of a stock to describe it. The only drawback of these studies is that they don't specify the nature (feature) of the information it reveals. Further, Conrad et al. (1994) reveal that the price reversal phenomenon exists only for the stocks which are heavily traded (at weekly intervals) while the stocks which are traded less frequently display return continuation. Datar et al. (1998) found an in direct relationship between stock turnover and stock returns. This finding was later used by researchers to examine the relation between trading volume and past price momentum.
Further, it has been empirically found that all the measures or sources of equity and total share turnover are immensely correlated. Also, when establishing a multivariate regression relationship using the two variables, these predict future market returns very accurately (Malcom Baker & Jeremy C. Stein, 2003). Baker and Wurgler (2000) unveil a uniform pattern in the average data, i.e. if the equity issuance is high in the market on an average then the market as a whole underperforms in the subsequent year. This is because of the long term fundamental values and gains that the firm managers anticipate. All this information is expected to be absorbed by the stock prices at regular intervals of time. A few important results (patterns) which clearly establish a relationship between stock turnover, equity issue and returns are given below (Baker and Wurgler (2000) :
• There is a very high correlation between equity issue (external share) and stock turnover. The correlation coefficient between the two is very high i.e. 0.64. Further this correlation remains unaltered by past period returns and price-dividend ratio.
• Stock turnover and equity issue play a very important role when regressed together leading to a boost in the overall predictive power of future returns.
• The predictive ability of stock turnover is very substantial in overall economic terms. At the same time, the role of error term in estimation cannot be ignored.
According to Amihud and Mendelson (1986), illiquidity of a stock can be calculated by the estimating the cost which an investor incurs to immediately dispose off his/her current holding. The cost originates because of the reason that the investor will either trade at a favourable or an unfavourable price which is called bid or ask price. Therefore, a true measure of illiquidity of a stock is the spread (trade-off) between the bid and ask (offer) price. Further, this spread is defined as the total of the buying premium (included in the offer price or ask price) and the selling concession (excluded from the bid price). The stocks with higher spreads have much higher expected returns compared to the ones which have a lower spread. “Clientele effect” clearly states that the investors who have a longer holding period choose stocks which have higher spreads and vice versa. This study further gives the importance of “securities market microstructure” for the estimation of asset returns. The financial policies that increase the liquidity of a stock also reduce the opportunity cost of capital and leads to overall improvements and efficiency in the trading process of the firm.
The relationship established by Anihud & Mendelson (1986) was contradicted by Jacoby et al. (2000). This paper established a positive convex relation between the spreads and the future asset returns as against the concave relation given by Amihud & Mendelson (1986). The Capital Asset Pricing Model (CAPM) given by Sharpe (1964), Lintner (1965) and Mossin (1966) has been challenged time to time by various other researches such as Fama and French (1992), Amihud and Mendelson (1986). Fama and French (1992) was later challenged by Datar et al. (1998) who showed that trading ‘volume' is superior than ‘size of the firm' in terms of explaining the future returns. Brennan and Subrahmanyam (1996) used estimated variable and fixed, transaction and proportional costs to establish a concave relation among variable costs and premiums and a convex relation among variable costs and premiums. Brennan et al. (1998) also tested if various firm characteristics; along with market liquidity (calculated through trading volume) has strong explanatory powers to justify the future returns. This study also found an inverse relation between trading volume and stock returns for NYSE stocks. It was further proposed that the CAPM model is incomplete without taking into account the effect of the bid-ask spread i.e. the liquidity costs. The overall risk of the firm and liquidity exist simultaneously. Therefore, these should always be accounted together (inseparably) for calculating future risk-adjusted returns.
Again, the subsequent part of this paper performs various tests to check whether liquidity has any hold on the returns generated by a stock. The null hypothesis for this test would be that liquidity does not affect returns.
It is now time to introduce the third variable of our research i.e. the role of media coverage in affecting a firm's performance and ultimately its going concern status. The emergence of various fraudulent cases like Enron, Adelphia, Tyco, and WorldCom in the corporate scenario have lead mass media to take the responsibility and ensure that corporate reputation plays an equally important role in a firm's success as actual market performance does. Media is very capable of guiding the investors to think in the way they want them to (Craig Eugene Carroll, 2004). The importance of such media coverage can be understood from the fact that the number of business magazines and TV networks have doubled in the past ten years (The New York Times, 2000). Also, the subject is important because of the fact that the investors learn almost everything about a company's operations through media or press. Ball-Rokeach, 1989 gave five reasons that encourage the media to show interest in the corporate sector:
• The media is concerned about its own reputation as well. Therefore, the reputation and image of the corporate entities has consequences for the media itself.
• The media rely on large organizations to get “information subsidies” (Gandy, 1982). The firms with substantial corporate reputations (favourable or unfavourable) are used by the media as source of information.
• The performance of the media is directly related to the performance of the corporate entities in most of the cases (Swisher & Reese, 1992).
• The media exercises its rights and responsibilities to form an opinion of a corporate entity and help the legal system by defining the possible scandalous and wrong intentions of such concerns.
• The democratic society permits the media to express its views about the corporate scenario and create awareness among everyone associated with it.
Klapper advocated that the exposure to media is more of reinforcement to the already built-up reputation of the firm rather than complete new formation of opinions for the same. He furthered the research by saying that even though media influences the general public opinion it is very hard to say that it holds a direct relationship with building their opinions entirely on the news and information that media provides them. Also, Lazarsfeld's concluded that media does not play a very vital role in the formation of public opinion (Gitlin,1981).
The media and press mentions have some indirect effects on public opinions and stock performance. With the introduction of “cognitive psychology” (Severin and Tankard, 2001), media has to redefine its purpose of providing information to the public. The role has now been shifted to influencing the perception of public rather than influencing their attitudes.
The analyst recommendations generate a lot of response from the investors. The first time going concerns do not have any certain background; therefore, there is a huge divergence of opinion regarding these stocks when they are issued for the first time in the market. Deepika Bagchee (2009) hypothesised that the investors generally shift their expectations of a particular stock (upward or downward), depending upon the analyst recommendations. Such downward (upward) shift leads to lower (higher) future returns than the non IPO firms. The investors in first time going concerns alter their expectations downwards by considering the analyst recommendations (sell-side). There is also evidence that when the analysts adjust their optimistic ratings downwards, the investors not necessarily alter their own ratings of the stock. The investors only adjust their ratings and views about a particular stock only after four to five continuous bad ratings given by the analysts.
Mikhail et al. (2007) found that the difference in the reaction of large and small investors to the analyst recommendations. Large investors base their trading decisions by considering the earnings forecast and analyst recommendations simultaneously whereas small investors base their decisions on the occurrence of such recommendations.
Stickel (1995) and Womack (1996) found a convincing negative (positive) price reaction to downward (upward) adjustment to the analyst recommendations. Also, it is documented that the positive recommendations are less credible because of their not reporting negatively and being optimistic.
Bhattacharya (2001) and Bhattacharya et al. (2004) analysed the response of large and small investors to the analyst recommendations. He found that the small traders are directly related to the random walk hypothesis (returns are unpredictable) and less related to the analyst recommendations. Whereas the large investors trading decisions are uncorrelated to the random walk hypothesis and more directly related to the analyst recommendations. This is probably due to the reason that small investors are unable to process the market related information and often fail to incorporate it into their trading activities. On the other hand, the large investors possess greater knowledge of the market and hence are better able to comprehend the situation and such recommendations.
The null hypothesis, press mentions do not affect the stock returns (‘going concern anomaly' affected firms) is tested and the results for the same are discussed in the subsequent part of this paper.