The Gregarious Behavior Of Investors From Baltic Stock Markets

The main objective of this research is to investigate market participants’ gregarious behaviour in Baltic stock markets, namely Lithuania, Latvia and Estonia during the period January 2003-December 2013. The herding behaviour derives from the investors’ irrationality, who trade financial assets based on their positive expectations about prices future growth, generating manias among other market participants, thus ignoring the real return rates and the risk levels of their investments.The investors’ irrational behaviour is influenced by actions, feelings and impulses that are intertwined: mimicry, fear, trust, greed, optimism, pessimism, euphoria, panic. These features highlight an erroneous perception of investors in point of unsustained increase in prices, which has been generated by the stock prices deviations from their fundamental value.Under these premises, optimism, overreaction and speculative bubbles are appearing on the market and may constitute triggering factors of a financial crash.The probability of the occurrence of the speculative bubbles and financial crashes is influenced by the continuous entry on the market of new investors and less informed participants, which often act based on impulse, following a benchmark, without considering their own analysis and information that they hold. The existence of a “collective behaviour” of the investors, which is manifested by their tendency to imitate other market participants actions and to “follow the herd”, so ignoring their own beliefs, may increase market sensitivity to shocks and the probability of the occurrence of the systemic risk.In order to identify the investors’ herding behaviour, I have applied an adjusted CSSD model proposed by (Yao, Ma, Peng He, 2014), which implies the inclusion of two additional variables, the first one, to reduce the effect of multicollinearity and a second one, a lag term of the dependent variable, in order to improve the power of the model. Furthermore, I have constructed portfolios ranked according to daily market capitalization, by using quartile analysis, having as objective the identification of herding behaviour for small, medium and large companies. I have obtained mixed results. In the case of Lithuanian stock market, there is no statistical evidence in favour of herding behaviour of investors. Moreover, in the case of Latvia, the findings indicate a strong presence of herding behaviour in the case of large companies. In the case of Estonia, the empirical results highlighted the existence of herding behaviour for medium companies, which can be explained by the investors’ tendency to imitate the strategies of others market participants and to ignore their own beliefs.