Media Coverage and the Cross Section of Stock Returns A Probe into the JSE

Through reaching a wide-ranging population of investors, both institutional and individual, mass media coverage of stocks markets can alleviate financial information frictions and consequently affect the valuation of securities even when it does not present genuine news. The empirical objective of this research is to investigate this hypothesis by studying media reporting and changes in average stock returns. By constructing two portfolios of stocks divided into “stocks without media coverage” and “stocks with media coverage” an investigation can be carried to find out which portfolio outperforms the other and sometimes even after accounting for risk factors. Previous literature news media and the stock market has failed to address African financial markets including the Johannesburg stock exchange (JSE) market. The Johannesburg stock exchange is Africa’s oldest and largest stock market. An opportunity exists to replicate empirical work on news media reporting and changes in average returns in South Africa and Johannesburg stock exchange. The methodology employed in this study is adopted from the widespread research previously conducted in other more developed markets. Media coverage has been derived from the number of headline articles about a stock in a certain month in 23 influential South African print newspapers. Only headline articles are used to proxy for a stocks overall media attention. A systematic search of the LexisNexis database is carried out to find articles published in 23 major, influential newspapers in South Africa. The examination period is from 1 January 2013 to 31 December 2017 (a total of 7620 firm-month observations). The results indicate no statistically significant (at the 95% confidence level) outperformance of stocks without any news media reporting over stocks with news media reporting as found in more developed markets. Further analysis of data indicates that media reporting of the JSE stocks is surprisingly low and 99% of observations having only 6 headlines or less in the media. Therefore, about 1% of the observations are reported at least 7 times in the South African newspaper media