Testing for Herding in the Finnish Equity Market: A CSSD and CSAD Approach During Periods of Uncertainty

Behavioural finance is a theoretical framework at the intersection of finance and psychology. It describes the irrational behaviour of individuals within an economic context. The basis of behavioural finance is to understand individual financial decision-making. Sometimes however, investors can behave irrationally as a group. This phenomenon is described as herding. This thesis tests whether the Finnish equity markets exhibit herding behaviour during the sample period of 2019 – 2024. The research is twofold. Firstly, it focuses on the full sample of 2019 – 2024 to test whether the Finnish equity markets exhibit herding behaviour in general. Secondly, it splits the time period into three sub-periods to focus on specific periods that contain events that exhibit market stress. Notably, the COVID-19 pandemic and the escalation of the war in Ukraine during early 2022. The empirical testing methodology used is a combination of two cross-sectional dispersion models: the cross-sectional standard deviation and the cross-sectional absolute deviation models. The data used for the tests is composed of the daily returns of 18 companies within Nasdaq Helsinki, and the OMXH25 index is used as a proxy for the market. A total of 1508 observations are included within the period. The empirical results of the cross-sectional standard deviation and the cross-sectional absolute deviation models suggest anti-herding behaviour within the Finnish equity market. Return dispersion tends to widen during periods of large market movements, contradicting with the convergence pattern of herding. However, these findings should not be interpreted as definitive evidence that herding does not happen within the Finnish equity market. Both models are based on theoretical assumptions of CAPM and market efficiency, suffering from the joint hypothesis problem. Thus, they may fail to capture varying types of herding in different market environments. The presence of multiple market stress periods during the sample period applies stress to the markets, which could amplify herding. This herding might not be fully captured by the cross-sectional standard deviation and cross-sectional absolute deviation approaches.