Testing Semi-Strong Market Efficiency in Relation to COVID-19 Announcements: An Event Study of Finland

Objectives The main objective of this study was to examine the response of the Finnish market to COVID-19 announcements. Thus, Hypothesis 1 was formulated as follows - The response of the Finnish market to COVID-19 announcements was consistent with the semi-strong form of the Efficient Market Hypothesis. Additionally, four auxiliary hypotheses were proposed. Hypothesis 2 – The response of the Finnish market to COVID-19 announcements became less severe with each subsequent COVID-19 wave. Hypothesis 3 – Finnish market reactions to COVID-19 announcements varied across sectors. Hypothesis 4 – The announcement of the start of a restriction period has a negative impact on the Finnish market. Hypothesis 5 – The announcement of the end of a restriction period has a positive impact on the Finnish market. To assess the hypotheses three waves of the pandemic were investigated. Summary To test the above, six Event Studies were conducted, with a sample of 139 companies publicly traded on the NASDAQ Helsinki stock exchange. The six events of interest were all announcements regarding COVID-19 restriction periods issued by the Finnish Government. The firms were divided into 9 sectors using the Global Industry Classification Standard taxonomy. Conclusions The reaction of the Finnish market to Event 1 was not consistent with the Efficient Market Hypothesis as defined by Fama (1963) and Samuelson (1965). Thus, Hypothesis 1 has been rejected. Hypothesis 2 has been supported, as the market returned to displaying characteristics of semi-strong form efficiency after the first COVID-19 wave. It was observed that reactions varied across different sectors of the Finnish economy, hence Hypothesis 3 has been supported. Hypothesis 4 has been confirmed by observing that the announcement of the start of the first COVID-19 State of Emergency had a negative impact on the Finnish market. No positive market reactions to announcements of restriction periods ends were observed. Hence, Hypothesis 5 has been rejected