Analysts' Consensus and Target Price Accuracy: A Study in Latin America.

AutorAntonio, Rafael Moreira
CargoReport

Introduction

The information provided by analysts concerning bonds and securities are used by the market participants in their investment decisions. The literature of this field reports a significant market reaction to analysts' forecasts (e.g., Demiroglu & Ryngaert, 2010; Li & You, 2015). In this respect, results which indicate a significant association between the excess of returns regarding the recommendation issuance and the recommendation level are presented. Other studies in the field, such as the ones carried out by Barber, Lehavy, McNichols and Trueman (2001); Brav and Lehavy (2003); Stickel (1995); and Womack (1996), suggest that target prices are significantly associated with excess returns in the period of the event.

Analysts' decision to follow and assess a company is associated with the incentives related to the maximization of their reputation, the optimization of the reaction of the agents who use their reports (i.e. managers and investors), and also with the value maximization of their analyses, since the majority of the models take for granted that their payments are linked to one or more of these incentives. Therefore, analysts prefer to cover companies whose investors tend to react strongly to their forecasts, specially companies which present a greater volume of shares traded (Beyer, Cohen, Lys, & Walter, 2010).

Moreover, each forecast type (earnings forecast, share and target-price recommendations) is an enhancement to newsletters for the explanation of price movement (Beyer et al., 2010). However, Hansen (2015) highlights that empirical studies on how analysts create value by means of their activities, and their effects on company value, keep evolving.

In order to corroborate, Li and You (2015) note that analysts' forecasts add value to the companies analyzed, even though the way this value is created is not clear. Therefore, it is appropriate to discuss the relationship between analysts' consensus and market reaction.

On the other hand, Gul and Lundholm (1995) identified evidence that managers of financial resources tend to choose their portfolios based on the choices of other managers and financial market agents, since from this perspective analysts' forecasts tend follow the direction of others, thereby creating consensus in the estimates.

Nevertheless, Clement and Tse (2005) argue that the forecasts which originate from consensus present lower accuracy, and they also reflect less private information relevant to the analysts. For these authors, the behavior of tendency to consensus reflects concerns related not only to the analyst's career, but also to his or her abilities to produce idiosyncratic information, which may generate asymmetrical information for the market.

From the reputation and strategic forecasting theories, Huang, Krishnan, Shon and Zhou (2017) argue that there is a tradeoff between analysts' reputation and their strategic decisions when they make available the information they process. Thus, when making this information available, if their concern in maintaining their reputation is stronger than their strategic and competitive decisions, the forecast content may be degraded, and just the categorical and standard information is given to the market. On the other hand, Huang et al. (2017) argue that the forecasts adjusted by the consensus tendency present less bias than individual forecasts, a finding which contradicts Clement and Tse's (2005).

According to Yezegel (2015), analysts review their recommendations towards earnings surprise, which is measured based on their own estimates and consensus, since the relationship between pricing of earnings and responsiveness to the analysts' recommendations suggest that analysts contribute to market efficiency. Nevertheless, this author argues that the analysts lend greater weight to consensus expectations than to their own forecasts.

When considering target prices, Bilinski, Lyssimachou and Walker (2013) observed that, in studies on analysts' traits, reputation aspects and tendency to consensus also apply to the analyses regarding the accuracy of target prices. The authors highlight that analysts who produced better past forecasts for target prices, who have experience in forecasting, who follow a greater number of companies, who are experts in a country's market, and who are employed by large brokerage firms, tend to produce target prices with greater accuracy.

Several studies analyzed the relationship between accuracy and consensus, taking into account several signs embedded in the consensus (e.g., Clement & Tse, 2005; Huang, Krishnan, Shon, & Zhou, 2017). However, it is discussed in the literature that the target price presents relevant predictive and informational power (Bilinski, Lyssimachou, & Walker, 2013; Da & Schaumburg, 2011), which is used in this paper as a construct for analysis of the consensus effects in the market.

Therefore, considering the economic and reputation incentives that lead analysts to be inclined toward consensus, this paper intends to expand the discussion of the consequences of the effect of consensus in the market, discussing specifically the effects on target prices. To this end, we seek to answer the following research question: When there is greater consensus among analysts on share target prices, is the forecast error smaller?

A major contribution of this paper evidences that analysts' consensus is an informative item for investors and users of this information. In addition, the results indicate that the greater the effectiveness of government, the greater the mean accuracy of target price estimates issued by analysts.

Another aspect that reinforces the contribution of this study concerns the breadth of the countries (Latin America) and the period (2010 to 2017) that we analyze. It is important to emphasize that investing in stocks is an alternative for investors who often do not understand the stock market, and follow analysts' forecasts.

The work is organized in 6 sections. Following this introduction is a literature review of the studies on the consensus among analysts, and their forecasts and their effects on the stock market. Third section presents the justification of the hypotheses, while Fourth section describes the methodology aspects and the research tools which were used. Fifth section is a discussion and analyses of the results found, while Sixth section offers final considerations of the research and of the whole work developed. A bibliographic reference is found at the end of the paper.

Literature Review

Consensus

The consensus tendency refers to the tendency of different agents, who make individual decisions, to act similarly at the same time. Theoretically, two main reasons for such behavior are presented: the individuals tend towards consensus or act in a similar way due to the fact that they have correlated information, or because they are subject to an incentive structure which encourages imitation (Clement & Tse, 2005; Jegadeesh & Kim, 2010; Trueman, 1994).

Gul and Lundholm (1995) highlight that, in several situations, economic agents tend to underpin their decisions with the decision of other agents. The authors present a model in which the agents choose a share and the time to act, demonstrating that their decisions cluster according to the information distribution among themselves. Thus, the passage of time allows the first agent to anticipate some information from the second one, based on previous experiences.

When considering the context of the analysts' actions, Guttman (2010) states that when obtaining different private or relatively similar information in terms of accuracy (both initially and over time), analysts issue their forecast simultaneously, or at the same moment they would have issued them if other analysts had not been present, broadening the discussion of tendency to consensus outside the time factor.

In these terms, aspects such as incentive structure and reputation tend to promote consensus based on the nature of their career and compensation system (Clement & Tse, 2005; Hong, Kubik, & Solomon, 2000; Scharfstein & Stein, 1990). Considering that the analysts' actions are reflected in the market behavior, analyses which assess the market interpretation of the relationship with consensus becomes relevant (Jegadeesh & Kim, 2010; Yezegel, 2015).

However, the literature has provided different results about the effects of consensus in the market. First, Clement and Tse (2005) argue that forecasts which do not follow the consensus are more complete and provide more relevant and complete information. The authors point out that the tendency to provide independent earnings increases when analysts are characterized by having experience providing forecasts with accuracy, by the size of the brokerage firm, by the analyst's experience, and by the reduction of the number of sectors the analysts cover, which corroborates the theory which relates independent forecasts and issues of career and ability.

For these authors, monitoring of the consensus by analysts can reduce the information transmitted by their individual forecasts in cases where the use of private information is not totally applied when carrying out or reviewing forecasts, as they aim to just be close to the mean average.

On the other hand, Huang et al. (2017), with a database of analysts' forecasts from 1990 to 2010, argue that when adding multiple signs, the consensus tends to be more informative than the sum of individual parts. The authors observed that 60% of the analysts of their sample tend to act towards the prevailing consensus. This tendency is directed by economic factors.

Jegadeesh and Kim (2010) observed that the market reactions to analysts' recommendations are stronger when such recommendations move away from the consensus of share recommendations than when they move towards it. The authors point out that even analysts from the most reputable brokerage firms also present...

Para continuar a ler

PEÇA SUA AVALIAÇÃO

VLEX uses login cookies to provide you with a better browsing experience. If you click on 'Accept' or continue browsing this site we consider that you accept our cookie policy. ACCEPT