An empirical study on adjustment of stock prices to information implicit in stock splits: Evidence from Colombo Stock Exchange

Abstract

Announcements of stock split have been very common phenomena among firms with the implementation of the new Companies Act No. 07 of 2007. Stock splits are only cosmetic transactions that increase the number of shares outstanding while decreasing the share price. There are numerous theories to explain reasons for stock splits by corporations. Among them, most of the empirical studies, especially on developed markets, have been carried on for the optimal trading range hypothesis and signaling hypothesis. However, in case of Sri Lanka, only a few studies are found in literature in relation to this area, of which Gunnathilaka and Kongahawatte, (2011) and Hua and Ramesh (2013) are only the published studies available in the literature. They find that the market reacts positively to stock split announcements and the information implicit on the announcements are accurately and instantly incorporated into stock prices. Accordingly, this study intends to contribute to the existing literature by providing additional insights on information implicit on stock split announcements. Therefore, the objective of this study is to examine the impact of stock split announcement on stock prices and whether the prices are efficiently adjusted to such information on the CSE.

Data and Methodology

The sample period of this study is five years, from 1st January, 2009 to 31st December, 2013. A total of 61 stock splits announcements on the CSE were selected for the sample as shown in Table 1. This study is based on daily data obtained from the Data Library of CSE. The standard event study methodology as discussed in Brown and Warner (1980, 1985), and Campbell, Lo and MacKinlay (1997) is used to evaluate the reaction of stock prices to the announcement of stock splits and for assessing semi-strong form of market efficiency on the CSE. The event window consists of total of 3 days, the day before the event date, the day after the event date and the event day itself. A 41-day investigation window is defined for this study, which extends from day –20 through day +20 relative to the day of the stock split issue announcement (t =0). The estimation window includes 100 days before the investigation window.

Abnormal returns are determined for each sample firm event over the investigation window. The abnormal return is the different between actual return and the expected return. The expected returns are computed using Market Model of which the parameters are estimated in the estimation window and are assumed to be constant over the investigation window. The test statistics proposed by Brown and Warner (1985) are used to assess whether the Average Abnormal Returns (AARs) and the Cumulative Average Abnormal Returns (CAARs) of each day in the investigation window are statistically significant.