Institutional Investor Clique And Analysts’ Earnings Forecast

Analysts’ earnings forecasting behavior is important to reduce information asymmetry and improve the efficiency of capital market,and the quality of their earnings forecasts is influenced by multiple factors including themselves and the target companies.As an important part of the capital market,institutional investors have developed rapidly in recent years,experiencing a process from weak and chaotic to strong and orderly,with professional and resource advantages coming to the fore.Due to their relatively weak ownership position,traditional institutional investors often express their views through collective exits or a combination of threats of exits and private communications.However,with the strengthening of their own power and the transformation of their investment philosophy,their participation in corporate governance has also changed from passive to active,exercising the right to vote and make proposals to participate in corporate decision-making.At the same time,the joint shareholding of several institutional investors makes it possible to establish network links between them,and the act of information sharing makes it possible to pass homogeneous or heterogeneous information to each other and reduce the unit cost of participation in corporate governance,while cooperation makes it possible to have a stronger voice and the ability to participate in corporate governance.In addition,institutional investors have the incentive and willingness to monitor the quality of information disclosure by management in order to reduce agency costs and improve their own returns,thus having an impact on analysts’ earnings forecasts.Previous studies on the impact of institutional investors on analysts’ earnings forecasts have tended to view institutional investors as an individual or aggregate,ignoring the interactive hookups within institutional investors or implicitly assuming the strong assumption that institutional investors necessarily reach unanimous action.Therefore,based on social network theory and game theory ideas,this paper uses the Louvain algorithm to construct community networks,and further uses the Bron-Kerbosch algorithm to approximate the highly aggregated institutional investor groups to empirically test the impact of institutional investor group holdings on analysts’ earnings forecasts.This paper concludes that the joint ownership ratio of institutional investors can significantly improve the quality of analysts’ earnings forecasts,while the endogeneity of the model is controlled by the instrumental variables method and Heckman two-stage regression,and the reliability of the findings is enhanced by a series of robustness tests.The influence mechanism test proves that the quality of information disclosure plays a mediating role.The analysis of the effects based on the nature of company ownership,industry characteristics,and external governance environment shows that the enhancement effect of institutional investor huddles on analysts’ earnings forecasts is more significant among private firms,traditional industries,and in the presence of lower audit quality and weaker external constraints.Unlike the assumption of independent homogeneity of investors,this paper expands the boundaries of the factors influencing analysts’ earnings forecasts and provides some theoretical and empirical basis for the cross-fertilization of network relations and financial research.