| With the continuous progress of information technology,financial markets are more closely linked than ever before,market volatility has increased significantly and extreme phenomena have become increasingly frequent.Since the financial crisis in 2008,stock markets,bond markets,energy and commodity markets,and foreign exchange markets in major developed countries and regions around the world have all experienced roller coaster-like ups and downs.For Chinese investors,the 2008 crash is still fresh in their minds,and the violent A-shares in 2015 are even more vivid.In addition to financial markets,there have been Black Swans in international politics,with Brexit,Italy's failed constitutional referendum,the South Korean presidential scandal...In this uncertain era,people seem to have become accustomed to extreme events.What's more,when the Black Swan event changes from a once in a century event to a once in a few years event,the market is bound to readjust its expectations of future risk events and the distribution of returns,especially for extreme risks.This also urges the people to carry on the more thorough ponder and the research to the systematic tail risk.In essence,one of the reasons why the extreme risk known as the Black Swan is so frequent is that financial markets are highly information dependent.Traditional economics usually assumes that prices aggregate and reflect all information,and that the relationships and interactions between decision-makers are included in prices.This assumption is clearly quite different from the reality.First,information is often incomplete and asymmetric,making economic individuals only have limited capacity and limited rationality;Second,market participants are bound to observe and learn from each other and exchange information,which influences their information sets and trading decisions,ultimately affecting asset prices.Therefore,it is a new trend to analyze the interaction and influence between investors and explore the pricing mechanism through the mechanism of information diffusion,which can reveal the nature of exogenous price shocks.In recent years,the application of social network theory in capital market research has provided us with a powerful tool to analyze investor interaction and its influence.Institutional investor groups are more likely than individual investors to build relationships and share information.Scholars at home and abroad have conducted many in-depth studies on this issue,all of which have confirmed that information transmission has an important impact on investors' behavior,that is to say,we can not ignore the information network and information transmission mechanism behind the capital market when we explore its behavior characteristics.This paper aims to further study how the institutional investors' decision-making affects the systematic tail risk of stocks by means of the micro-mechanism of information network structure.Specifically,we would like to know whether the structural characteristics of information networks among institutional investors have an impact on the systematic tail risk of stocks.As mentioned above,extreme risk can wreak havoc on the liquidity,information transmission and pricing,investment and financing functions of the financial system.It is necessary to understand the logic of extreme risk,therefore,the use of institutional investor information network analysis of the formation of systemic tail risk is also of great theoretical and practical value.This article mainly includes seven chapters.The first chapter is the Introduction,it elaborates the research background and significance,research ideas and structure arrangement,as well as the main innovation points and shortcomings of this paper.The second chapter is Literature Review,in this part,we review and comb the literature on investor networks,fund flows(representing fund investors' behavior),Fund Management Initiatives(representing fund managers' behavior),and stock extreme risk.The third chapter is about the construction of investor network and the preliminary verification of its existence.First,it explains the construction method of fund network and its economic meaning.Then,it excludes the alternative explanation of non-information exchange channel through network comparison analysis.Finally,by observing the private information flow in the network through the fund's trading behavior in different information circles,the social attribute of the heavy position holding network is confirmed.The fourth and fifth chapters verify the hypothesis of "embeddedness" of economic behavior,in order to confirm whether fund market participants will include social network in their behavior decision-making information set.Chapter Four verifies the embeddedness of fund investors' behavior and examines the behavior of fund investors from the perspective of investor networks.This chapter explores the impact of the performance of other funds in the network on the fund flow of the Central Fund,that is,whether the performance of other funds linked to the fund will be included in their own evaluation system.Chapter Five verifies the embeddedness of fund manager's behavior and examines the fund manager's behavior from the perspective of investor networks.This chapter uses the average deviation of the fund's own transactions from those of other funds in its network as a measure of the fund's use of information in its network,this paper probes into the relationship between the degree of using the shared information in the fund network and the management ability of fund managers.That is,whether the fund manager with poorer managerial skill depends more on sharing information within the network.Consistent with the conclusions of existing literature,the results of the fourth and fifth chapters show that investor networks have a significant impact on the behavior of market participants,indicating that investor networks are real.On this basis,Chapter Six further confirms that the investor network still exists under extreme market conditions through the trading consistency of members in the fund network under extreme conditions,and also discusses the relationship between the structural features of the information network and the systematic tail risk.Chapter Seven is the Conclusion,which summarizes the conclusions of the full text,and gives the possible future research direction in this field.Based on the above analysis,the main conclusions are as follows:(1)There is private information exchange among the funds in the network.The private information exchange can improve each other's performance,which reflects the cooperation between the funds(complementary features).However,information exchanges can also lead to consistency in fund transactions within the same network,making funds within the same network more vulnerable to common non-systemic shocks,and similar risk characteristics will increase the competition among funds(alternative characteristics).Thus,the performance of other funds in the network has two opposite externalities to the fund flow of the central fund,that is,“the complementary effect” and “the substitution effect”,and the final total effect depends on the relative strength of complementary effect and substitution effect.Further analysis shows that when the complementary characteristics are stronger,the sensitivity of the fund's fund flow to the average performance of funds in the network increases,while the sensitivity to the fund's own performance decreases.In summary,the performance distribution of other funds in the fund's network will also become an important reference for fund investors,which will significantly affect the fund's fund flow.(2)The performance of funds with low usage of the information shared in fund network is significantly better than that of funds with high information usage in the fund network.The higher excess returns of these funds are mainly due to their excellent stock selection ability,which bear more heterogeneous risks,but do not increase the overall risk level.The fund manager turnover data indicates that the fund's use of information in the network is more related to the fund manager's characteristics than to the fund's characteristics.The degree of information use in the network directly reflects the private information content of the fund,so it is more likely to be related to the ability of the fund manager.The above results show that fund managers use less information in the network because they have more valuable private information,which reflects the personal management ability of fund managers.In other words,the less skilled fund managers will rely more on shared information within the network.(3)Funds in the same network still informationally connected even when the market is in extreme decline.This makes them exhibit strong behavioral convergence whose extent influenced by the specific information delivery mechanisms.Further analysis suggests that this enhanced behavior convergence which is the reason of “collective stampede” will influence the systematic tail risk of a stock.Specifically,we find that stocks with lower network density or higher network concentration demonstrate higher systematic tail risk.Our explanation for this is that members of a dense network are more closely connected,communicate more frequently and smoothly,and have more coordinated interests.In the event of a crisis,they can communicate in a timely manner,reducing the consistency of sales among themselves,thereby reducing stampedes,to avoid hurting each other.When the network concentration degree is large,the information comes from the fund which is in the center position(is in the information superiority),the other funds which connects with it take the following strategy,therefore this kind of communication is often one-way.When the market extreme risk occurs,the central fund in the network with a high degree of network concentration will first notice because of some advantages of its own information acquisition,and then quickly flee.Then the information is passed down,and the funds that receive the information are sold in turn,racing to get out,sometimes superimposed with the spread of panic,finally leading to a severe stampede,causing extreme declines in asset prices.These conclusions hold true even when controlling for stock characteristics and public information.This means that the mechanism of posterior information transfer between investors plays a decisive role in this process.It is precisely because of the consistency of information acquisition and information response of investors in a crisis that the systematic tail risk is generated and facilitated,and the black swan event is more likely to be triggered.On the whole,the core contributions of this paper are as follows: On the one hand,the research on social network in finance mainly focuses on how economic individual chooses to establish social network and the influence of social network structure on economic individual behavior.Because of the constraints of conventional social network construction(based on background information of economic individuals),few literatures can touch on the relationship between social network structure and asset pricing.On the other hand,some existing literatures ignore the query about the social attribute of the heavy position holding network and try to use the heavy position holding network directly to analyze the relationship between the investor network structure and the asset pricing,which leads to the lack of theoretical foundation and its conclusions are unconvincing.In the face of the separation of the two types of literature,this paper innovatively relaxes the way of social contact,thus circumventing the constraints of the background network.Then,we prove the existence of the fund's heavy holdings network and the embeddedness of the market participants' behavior in a normative way,thus building a bridge between the above two kinds of literature,which help us successfully to introduce social network into asset pricing research systematically.At the same time,this research is not limited to a certain kind of social connection,the inclusive social network construction way enables us to make the best use of the existing research sample,therefore,the conclusion of this paper has more extensive applicability and guiding significance.To some extent,the research of this paper expands the application of social network theory in capital market.At the same time,the results of this paper will help to deepen the understanding of the formation of extreme markets and information transmission,and provide useful reference for the construction and implementation of regulatory policies and market stability rules. |
PDF Full Text Request