Effect of Earnings Announcement on Share Price in Kenya: a Study of Nairobi Securities Exchange

Abstract

The objective of this study is to determine whether an effect exists on share prices due to the announcement of earnings by individual companies listed on the Nairobi Stock Exchange and the significance of these announcements on their aforementioned companies. An event study research design was selected since it explains how a given response variable, in this case- the share price is influenced by a set of explanatory variables, such as time or earnings announcement. A reasonable event window of 30 days had been selected and was expected to provide ample allocation for any expected reaction in share price as a result of an earnings announcement. The data used incorporated daily share price data from 65 listed companies in the Nairobi Securities Exchange. The scope of the data was 2 years (2014 and 2015). For data analysis, the market model was used. The response variable was the actual return on stock and this return was compared to the general return in the market. The abnormal return (AR) was thus a difference between the actual return and the market return. To model the actual return, a regression model was used, with the market return as the main explanatory variable. For model testing, the t-model was used. For additional model testing, the Chi-Square test was used and it had been selected due to its ease of implementation, especially for categorical data such as the ones used in this study (returns for stocks in different sectors). Through observation of abnormal returns, it was seen that the market is not strong form efficient and that due to relative delays in market adjustments, investors might exploit such “inefficiencies” in order to generate excess risk adjusted returns. This investigation showed that earnings announcements are a significant factor that influence share prices. This was shown through t-tests and chi square tests. In addition, it showed that the market is more sensitive to announcements of a negative nature (earnings decline or loss) as compared to positive announcements (earnings increase). This investigation also led to the discovery that effects of earnings announcements are observed even before the announcement date and they continue to be felt even after the announcement. It is very reactionary and this is seen where the market takes a relatively long period of time to adjust market prices back to their expected levels.