The stock market’s reaction to contract announcements : an empirical study of companies in the maritime industry listed on Oslo stock exchange

Abstract

The research question of the thesis is “how does the stock market react to contract announcements by companies in the maritime industry listed on Oslo Stock Exchange?”. This is answered using the event study methodology as described by MacKinlay (1997). The sample consists of 208 contract announcements by 28 companies in the maritime industry listed on Oslo Stock Exchange from January 1, 2014, to December 31, 2017. The primary objective is to ascertain whether contract announcements lead to a cumulative average abnormal return that is significantly different from zero on the event day. In addition, the thesis will attempt identify the determinants of the stock market’s response, investigate if there are signs of information leakage prior to the announcement, and examine if there are any post-event stock price drifts. The analysis finds cumulative average abnormal returns on the event day ranging from 2.47 % to 2.56 % using four different normal performance models. They are all significant at a 1 % level. There are no significant effects in the pre-event day window or the post-event day window. Given that no evidence of information leakage or post-event stock price drift is found, the market appears to be efficient on the semi-strong form according to the efficient market hypothesis. The cross-sectional analysis finds that, everything else equal, the cumulative abnormal return on average increases between 7.02 and 8.22 percentage points when relative contract size increases by 1. This is intuitive as larger contracts relative to the size of the company are stronger signals of increased future earnings than smaller contracts. Furthermore, Tobin’s q is negative and significant as expected. This variable being negative is argued to be explained by the market being more surprised when low Tobin’s q firms announce contracts as it is expected that they are less able to extract value out of their assets. No other variables are found to be significant in explaining event day cumulative abnormal returns in the cross-sectional analysis. Lastly, the results obtained in the analysis appear to be robust to the choice of significance test, normal performance model, and outliers. In addition, the assumptions of OLS does not seem to be violated to the degree that the general inference is altered.