| This dissertation aims to study the influencing factors and potential economic consequences of security analysts’ behavior and decisions,so as to promote the rational allocation of resources by security analysts in our country and improve the efficiency of resource allocation in the capital market.Security analysts are important information intermediaries and information providers in the capital market,and they have attracted widespread attention from scholars.They pay close attention to the predicted company,make corresponding investment analysis with through their profession and processing capabilities,and use certain valuation methods.Also,research reports which contain earning forecasts,stock and investment recommendations,provide investors with more valuable information,guide investors to make rational investment decisions,and promote stock prices return to its intrinsic value.Therefore,in theory,by issuing research reports,security analysts promote the transmission of listed company information,increase market information content,and to a certain extent help alleviate information asymmetry in the capital market,thereby promoting stock prices.The return of its intrinsic value guides the rational flow of resources and improves the efficiency of resource allocation in the capital market.However,in reality,security analysts may release biased research reports,which distort market prices and mislead investors’ investment decisions,and on the contrary increase market inefficiency.Security analysts’ decisions are affected by a variety of rational or irrational factors,and analysts’ predictive behavior or ratings may be inefficient（La Porta,1996）.Recent literature points out that analysts’ expectations of the company’s future are not rational,and that they have irrational deviations in their perceptions and decision-making.At the same time,they have examined the impact of these cognitive deviations on stock mispricing.The research of Cen et al.（2013）shows that when analysts make earnings forecasts,they refer to the industry median,and their forecasts are largely biased toward the industry median.Bordalo et al.（2018）found that analysts excessively extrapolate the company’s long-term earnings growth forecasts and overreact to past earnings growth.At the same time,Bouchaud et al.（2019）found that analysts’ expectations are sticky in the short term,and their continued response to the company’s profitability is insufficient.Hirshleifer et al.（2019）found that as the number of research reports released by analysts in a day increases,the accuracy of predictions decreases.As an important information intermediary and information provider in the capital market,can security analysts’ research promote the return of stock prices to their intrinsic value,guide the free flow of resources,and improve the efficiency of resource allocation in the capital market? This is a topic worth reexamining.Therefore,this paper firstly analyzes the influencing factors that affect the analyst’s decision;secondly,it studies the economic consequences of the analyst’s behavior and decision.Research on the influencing factors of analyst’s earnings behavior decision.This paper mainly includes two perspectives,one is a bounded rational perspective,and the other is a rational perspective.In the third chapter of this paper,based on the analyst’s bounded rationality,from the perspective of the reference effect of behavioral finance theory,the influence of the reference point effect on the analyst’s behavioral decision is studied.The reference point effect is a common perceptual characteristic of human beings.Just like the process of adapting to light,sound or temperature,people use the formed reference point to view and react to events.In fact,reference point dependence not only makes investors "cognitive stickiness"（such as investors accustomed to company dividends）,but may also cause a psychological deviation of "salience effect",that is,the difference between new information and cognitive reference points.Also,the strong contrast between the two may affect the perception of investors and thus their judgment.Use the earnings announcements and analyst earnings forecast data of Chinese A-share listed companies from 2002 to 2019 to test the effect of the reference point in the revision of analyst forecasts.The research found that:（1）The company’s previous fundamental information will become the analyst’s cognitive reference point for the current fundamental information.At the same time,the prominence of fundamental information changes will increase their perception of new information and make analysts better.It is possible to revise its expectations based on the latest SUE,and both star analyst and non-star analyst will rely on reference points to make decisions.（2）Sub-sample testing found that,regardless of whether the latest SUE is positive or negative,analysts will rely on the reference point of the previous period to make decisions.If the previous period is negative（positive）,the analyst is more likely to have unanticipated earnings.If it is positive（negative）,it will revise its forecast upward（downward）.Therefore,traditional theories such as information fluctuation,optimism bias,and conflict of interest cannot provide a unified explanation for the phenomena observed in this paper.As a rational market participant,analysts often cater to institutional investors in order to maximize their personal interests,which harms the interests of investors and creates conflicts of interest.In the fourth chapter of this paper,based on the rational perspective,the impact of the conflict of interest on analysts’ decisions is studied.Selecting all A-share listed companies in Shanghai and Shenzhen stock markets from2002 to 2019 as a sample,the impact of conflicts of interest from institutional investors on analysts’ earning forecasting behavior is explored from the perspective of earnings revision magnitude and the same direction revision probability.The study found that:（1）Analysts’ earnings forecast revisions are affected by the conflict of interest of institutional investors.Specifically,for companies with a higher institutional shareholding ratio,analysts revise their earnings forecast upwards by a greater extent.（2）Conflicts of interest from institutional investors will also have an impact on the probability of analysts’ earnings forecast revisions in the same direction.Further,the results show that the salience of company’s fundamental information interacts with conflicts of interest,which has asymmetrical effects on analysts’ earnings forecast revisions.Specifically,when the company’s latest fundamental information is positive,the prominence of the company’s fundamental information will increase the probability of analysts’ positive earnings forecast revision,and if the company’s latest fundamental information is negative,the higher the institutional shareholding will be,the lower the probability of analysts revising negative earnings forecasts,and the prominence of fundamental information does not enhance the conflict of interest.In short,conflicts of interest from institutions will not only cause analysts to increase the magnitude of positive earnings forecast revisions,but also reduce analysts’ response to negative company information.As an information intermediary in the market,whether security analysts can reduce information asymmetry in the capital market and improve the efficiency of the capital market depends on whether they have the ability to discover distorted information in the market.Undoubtedly,earnings management is a distortion of the company’s true performance information.Next,based on the perspective of earnings sensitivity,Chapter 5 of this paper judges the ability of Chinese security analysts to discriminate distorted information by examining the sensitivity of analysts’ earnings forecasting behavior to earnings management.Selecting all A-share listed companies in Shanghai and Shenzhen stock exchanges from 2002 to 2019 as the research sample,the results show:（1）Chinese security analysts have a certain ability to discriminate the company’s earnings management,however they will be affected by the concealment of earnings management.Specifically,security analysts have the ability to discriminate accrued earnings management based on changing the reflection and distribution of actual earnings in different accounting periods,but they do not have the certain recognition ability to change the company’s actual economic activities based on the more concealed real earnings management.（2）This paper considers the difference in the ability of star analysts and non-star analysts to identify earnings management.The results show that whether it is accrued earnings management or real earnings management,there is no significant difference in the recognition of earnings management between star analysts and non-star analysts’ earnings forecasts.（3）Institutional shareholding will lead to conflicts of interest,which will cause analysts to increase the magnitude of positive earnings forecast revisions.At the same time,institutional shareholding will also have an incentive effect,which can promote analysts’ recognition of calculated surplus and further expand institutional investment mechanism of the governing effect.Finally,asset prices are an important factor affecting the allocation of resources in the capital market.Reasonable asset prices can optimize the flow of resources and improve the efficiency of resource allocation in the capital market.To answer whether the analyst,as a relatively rational participant,promotes the return of stock prices to its intrinsic value,thereby increasing capital market efficiency,Chapter 6 of this paper investigates whether the analyst’s rating behavior takes advantage of the widespread anomaly in the capital market.The research found that:（1）First,examine the relationship between analyst forecasts and anomalous returns at the overall level.The20 anomalous factors used in this paper can provide predictable information about stock returns,but the consensus forecast by analysts is inconsistent with anomalous returns,indicating that analysts have not effectively used anomalous information when predicting ratings.Specifically,analysts tend to give more favorable ratings to overvalued stocks.However,these stocks often have significantly negative excess returns afterwards.（2）Further,examine the reasons why the analyst failed to effectively use the anomaly information.Aggregate the above 20 anomalous factors into a fundamental group and a non-fundamental group,and examine the relationship between the returns of the two sets of investment portfolios and analyst ratings.The study found that analysts can make better use of fundamental anomalies in the forecasting process,but ignore market anomalies.Specifically,analysts give lower ratings to stocks that are overvalued by fundamental groups,and the combination of ratings and the ordering of fundamental indicators has a significant monotonic decreasing relationship;analysts give higher ratings to overvalued stocks in the market group.There is a significant monotonic increasing relationship between the rating and the combination of the ranking of the market indicators.Further research shows that the results of this paper are still valid in the sub-samples with relatively complete pricing and trading mechanisms,strong analyst expertise,and high institutional shareholding.The above results all show that analysts may have a certain bias in their abilities.They are good at traditional fundamental analysis,but are lagging in response to market-level changes such as prices and transaction volumes.Limited attention may be the reason for their inefficient forecasting.（3）Finally,according to the abnormal signal and the analyst’s rating level,this paper carries out an independent double ranking of all stocks.It is found that when the analyst’s rating is inconsistent with the abnormality,the abnormality will produce greater excess returns.That is to say,analysts’ biased recommendations may cause market friction and hinder effective correction of mispricing.The theoretical significance of this paper is as follows.First of all,this paper expands the research of reference point theory from the perspective of prominent effect,and verifies that analysts have reference point dependence in the cognitive decision-making process through the revision of analyst earnings forecast.Secondly,this paper expands the influence of the interaction mechanism of rational factors（conflicts of interest）and bounded rational factors（prominence caused by reference point dependence）on analyst decision-making behavior.Then,the results of this paper also provide a new interpretation path for corporate managers to influence the market through earnings management.Managers may,in order to maximize their own interests（or cooperate with the reduction of holdings）,utilize earnings management,especially real earnings management behavior,resulting in contrasting unexpected earnings,affecting the decision-making behavior of analysts,thereby affecting the stock market.Finally,this paper examines the value of analyst behavior and its impact on market efficiency from the perspective of the capital market anomaly,thus enriches existing research on the economic consequences of analyst behavior.This paper has strong practical significance.First,the results of this paper show that the behavior of analysts is affected by conflicts of interest.Regulatory authorities should strengthen the education of analysts’ professional ethics,improve the analyst’s principle of independence and integrity in the practice process,and further improve relevant laws and regulations so that analysts can truly and effectively play the role of information intermediary and ultimately promote the improvement of securities pricing efficiency,and the healthy and stable development of the capital market.Second,Chinese security analysts have a certain ability to discern the company’s earnings management,but they will be affected by the concealment of earnings management.Specifically,security analysts have the ability to discriminate accrued earnings management based on simple changes in the reflection and distribution of earnings in different accounting periods,but they do not have the ability to identify real earnings management based on the greater concealment of the company’s actual economic activities.This paper believes that security analysts should further improve their analysis level,not limit to the discovery of financial statement information,and need to broaden the dimensions of analysis.Third,this paper finds that there is no significant difference between star analysts and non-star analysts in their ability to identify earnings management.The results of this paper indicate that the selection mechanism of star analysts needs to be further regulated to better play the role of star analysts in stimulating.Fourth,the analyst group of this paper should improve its own research level,enhance the ability to detect distortions in the capital market,and ultimately promote the improvement of securities pricing efficiency and the healthy and stable development of the capital market.The innovations of this paper are mainly reflected in the following aspects.First,this paper expands the research of reference point theory from the perspective of highlighting effects,and verifies that analysts have reference point dependence in the cognitive decision-making process by revising analyst earnings forecasts.Second,this paper enriches the research on the economic consequences of reference point theory,revealing that reference point dependence does not necessarily lead to negative results.When the agent’s attention and information processing ability are limited,it is not necessarily irrational,or it will definitely lead to systemic bias.Third,this paper explores the impact of conflicts of interest on analysts’ decisions from two perspectives:the magnitude of analyst earnings forecast revision and the same-directional earnings forecast revision probability.It enriches the existing research on analysts’ earnings forecast revision behavior decision-making.Fourth,this paper expands the influence of the interaction mechanism of rational factors（conflicts of interest）and bounded rational factors（prominence caused by reference point dependence）on analyst decision-making behavior.Fifth,this paper studies for the first time the ability of analysts’ earnings forecast revision behavior to discriminate the company’s earnings management.It is found that analysts have a certain ability to discriminate earnings management,but they will be affected by the concealment of earnings management.Sixth,this paper studies the difference between star analysts and non-star analysts from the perspective of identifying differences in earnings management capabilities.Seventh,the results of this paper also provide a new interpretation path for corporate managers to influence the market through earnings management.Managers may,in order to maximize their own interests,utilize earnings management,especially real earnings management behavior,resulting in contrasts between before and after unexpected earnings,affecting the decision-making behavior of analysts,thereby affecting the stock market.Eighth,this paper examines the value of analyst behavior and its impact on market efficiency from the perspective of capital market anomalies.It enriches the existing paths for analysts’ behavior to influence market efficiency.Finally,this paper classifies different capital market anomalies and finds that analysts have different responses to different types of anomalies.And based on the theoretical framework of behavioral finance with limited attention,it is believed that analysts will fall into the predicament of "taking care of one and losing the other" when using anomalous information. |
PDF Full Text Request