Literature Review & Empirical predictions
Most of the research done in prior pertains to analysts’ ability to recognize the mispriced stocks, focusing on the stock recommendation instead on the analysts’ valuations per se. For example, Womack (1996) documents substantial market reactions to the announcement of analysts’ recommendations, indicating investors believe that recommendations are valuation relevant. However, Bradshaw (2001) uses analysts’ earnings forecasts as an input to residual income model, and discovers that resulting valuations are inadequate to describe the affiliated stock recommendations even when there is a proof that these valuations recognize mispriced stocks (Frankel and Lee 1998). These findings indicate that personal valuations of the analysts differ from the findings of residual income model, but there is not much of empirical work examining the valuations that they do provide.
The investigation of the information content on target prices determined by the analysts fortify to extant research on the information substance of their other two signals: earnings forecasts and stock recommendations.
This research commonly finds substantial negative (positive) price reaction to recommendation downgrades (upgrades) (e.g., Womack , Stickel  and Elton, Gruber, and Grossman ) and that the future gross abnormal returns are predicted by the degree of consensus analyst stock recommendations (e.g., Barber, Lehavy, McNichols, and Trueman ).
Recommendations have also been exhibited to contain information that is normally squared to the information in other variables certified to have predictive power for stock returns (Krische and Lee ). Francis and Soffer  finds that each signal is informative in the existence of the other when they focussed on the relative informativeness of analysts' earnings forecast revisions and stock recommendations, while Markov , Bradshaw , and Stickel  examined the consistency between consensus earnings forecast revisions and consensus recommendation.
Mark T. Bradshaw (2002) investigated a random sample of analysts’ reports to determine (i) how often analysts validate their recommendations with target prices, and (ii) what do they use instead, when they do not use target prices. For instance, they randomly picked 103 companies across entire industries and obtained the latest analyst report that contains a recommendation. They identified the reasons for the recommendation after reading each report. The tendency to disclose a target price is greater for more favourable recommendation because analysts disclosed target prices in roughly two-thirds of the reports.
Bradshaw (2002) documented a correlation between price targets, earnings and recommendations based, which is measured on a sample of103 analyst reports. Brav and Lehavy (2001) used a large database of target prices and found a significant market reaction to price targets both unconditionally and conditional on contemporaneous recommendation and earnings forecast revisions.
Brav and Lehavy (2003), work is related to the empirical literature that investigates the properties of target prices. They used a large database of target prices over the period 1997–1999 and documented significant abnormal returns around target price revisions both unconditionally and conditionally on contemporaneous stock recommendations and earnings forecasts. They employed co-integration analysis to apprehend the long-term relation between these two measures, they estimate that the long-term mean target-price-to-current-price ratio is 1.28, i.e. target prices are 28% higher than current stock prices. Furthermore, they claimed that their analysis serves as an outset for further research on miscellaneous related questions, including the valuation models used to ascertain target prices.
Asquith et al. (2005) affirmed that in the cases of downgrade or recapitulation recommendation, there is amalgamation of important information with the change in target prices. Target price revisions enclose new information beyond earnings forecast revisions and stock recommendations, and the market reaction to a change in target price is stronger than to an equal percentage change in earnings forecast. Asquith et al. (2005) used 818 analyst reports over the period 1997–1999 to investigate whether there is a substantial relationship between the target price accuracy and the valuation methodology used to generate the target price and discovered no significant relationship.
Gleason, Johnson, and Li (2006) use an extensive database of analyst target price and earnings forecasts over the period 1997–2003. They follow Bradshaw (2004), as a part of their study by putting in analyst earnings forecasts to price-to-earnings-growth and residual income valuation models to formulate pseudo-target prices. They find little evidence that the pseudo target prices produced by the RIV model using specific earnings forecasts are related with more accurate target prices.
Bradshaw and Brown (2006) used a large-scale sample from the First Call database for the period 1997–2002 of analyst target price forecasts to analyze the accuracy of target prices. They showed that analysts do not present differential ability in deciding target prices and the market does not react differently to analysts with bad or good track records. They emphasized upon the importance of the difference between target price and current stock price as a determinant of target price accuracy. Furthermore, they argued that the analysts can show their optimism by issuing higher target prices since there are no formal rankings of analysts based on target prices.
Bonini, Zanneti, and Biachini (2006) reached to a similar conclusion by analyzing equity research reports for Italian listed firms published during 2000–2003.
M. Walker et al (2007) complemented the above studies as they reported updated evidence on the relation between target prices and current stock prices.
Previous empirical studies analysis is based on the data from periods when stock market indices extended to unprecedentedly high levels. The quality of their research and the practices of sell-side analysts during these periods have been heavily criticized, and since 2001 the industry has been undergoing a structural change. According to Nocera 2004, the only certainty right now maybe is that the old research model is dead.
As compared to the research of Asquith et al. (2005) and Bradshaw and Brown (2006) previous research sample contains a large number of reports with more cautious target price forecasts and reports with neutral/negative recommendations. Might be the reason of difference is either due to institutional differences between Wall Street and the City of London (Breton and Taffler 2001) or mirror overall changes in the sell-side analyst’s practices. Any effects of these fundamental changes on the value of analysts’ research output are partially captured by the study. By using a sample of 490 equity research reports issued over the period July 2002–June 2004 to analyze the accuracy of target prices opinion on earnings multiples and DCF models, they provided supplementary empirical evidence on the grounds of target price accuracy.
For the purpose of the research there is a need to examine the properties of analysts' target prices. The analysis is beginning with an investigation of market reaction to target price announcements. As we know, target prices provide investors with the analyst’s certain determinable statement about firms' expected returns therefore it is expected to observe substantial average market reaction around such target price announcements. In addition, it is expected that market reactions around target price revisions to increase in the timeliness of the revision, since extensive downward (upward) revisions in target prices are inferable to represent less (more) favorable news.
Target prices, however, are issued in conjunction with earnings forecasts and stock recommendations. Therefore, the analysis is extended to determine whether target price are incrementally informative. Given the discreteness of stock recommendations, even in its presence target prices are expected to be informative. Target prices are expected to convey information to market participants above and beyond that in the earnings revision, since earnings forecasts are but one input into the target price calculation.
Even though it’s quite reasonable to expect target prices to be informative relative to earnings forecasts and stock recommendations, the inverse is not indisputable.
If the information in the earnings forecasts is already reflected in the target price, then earnings forecasts together with other determinants such as payout ratios and discount rates are inputs to the analyst valuation model. Likewise, if preparing in advance a target price into a stock recommendation is redundant in informative terms, then such a target price is an output of a valuation model that throws back the analyst's beliefs in the basic value of the firm. Investors can instantaneously infer the recommendation by equating/comparing the target price to the current stock price. This view conjectures that both earnings forecast revisions and recommendations should be uninformative in the existence of target price revisions.
There is reasonable evidence to believe that stock recommendations and earning forecast revisions are informative when target prices exist. Firstly, for a given target price, stock price dynamism that increase the significance of mispricing, but in the analyst opinion it does not necessitate a target price revision which might lead the analyst to keep his target price and revise his recommendation. For instance, a percentage decrease in the stock prices which is currently rated a Buy, by an analyst, might increase the stock appeal and subsequently might even be upgraded to a Strong Buy keeping the same target price, when revealed to the investor. Secondly, additional information on the level of confidence regarding target price is provided by recommendations, in case of a given target to market price ratio. For instance, if a target price exceeds the current stock price by say 15%, it would be considered more credible if it is tagged as a Strong Buy instead of being a Buy. Thus, while on one hand target price reflects the mean of the analysts’ posterior belief about a firms value, on the other hand, recommendations provide additional information about the dispersion of these beliefs. Also in the presence of target prices, earning forecasts revisions might prove to be informative. This kind of situations occurs when an earnings forecast is considered to be informative by the investor, but is inconsistent with the way the analyst incorporated the earnings information into his pricing model.
Thus empirical tests would be conducted to check these competing views on the information role of stock recommendations and earnings forecasts in the presence of target prices.
There are other potential explanations as well such as one provided by a Goldman Sachs analyst, explaining that precise measure of firm mispricing relative to current market price is provided by target prices and additional mispricing information regarding a cross-section of similar stocks is conveyed through recommendations. For instance, a recommendation regarding the measure of intra- industry mispricing might be provided by an analyst, if certain investors are constrained to hold a certain subset of the market, such as of a given industry.
Furthermore, the co-movement of the stock and target prices would be examined. These are modeled as co-integrated systems since both of these variables are linked to the underlying firm fundamental value. The main objective of using the cointegration framework is to estimate the mean of this ratio and to enquire whether its magnitude is consistent with the predictions of some commonly used asset- pricing models.
Also the discussion would include if target prices are forward looking or not, and if yes to what extent.
A thorough in depth study is to be made regarding the price system’s reaction to deviations from its long-term equilibrium ratio and quantify the speed and magnitude of adjustment of the two series towards this long- term relationship. The basic goal is to examine which price series corrects to the long-term equilibrium once the price system has been shocked away from it.
Need help with your literature review?
Our qualified researchers are here to help. Click on the button below to find out more:
In addition to the example literature review above we also have a range of free study materials to help you with your own dissertation: