Alphabet Bias and Trading Activity: A South African Perspective

Abstract

This study examines the relationship between alphabetic position and trading activity for shares on the Johannesburg Stock Exchange (JSE) to determine if alphabet bias exists on the JSE. The study employs share turnover and illiquidity measures as proxies for trading activity, whilst incorporating controls for firm-level characteristics including market capitalisation, share price, and analyst coverage. Panel regressions from January 2000 to December 2023 do not reveal a statistically significant alphabetical bias on the JSE, in large contrast to evidence from studies from developed markets. This discrepancy is likely due to the relatively small universe of shares on the JSE, which can contribute to lower search costs, thus reducing the need for simplified methods such as alphabetic ordering. The empirical results confirm that market capitalisation, share price, analyst coverage, book-to-market ratios, leverage, and systematic risk maintain significant relationships with trading activity, consistent with established theoretical frameworks. Multiple robustness tests incorporating alternative measurement approaches and sensitivity analyses validate these conclusions. The findings contribute to the behavioural finance literature in South Africa and show that fundamental firm characteristics rather than alphabetical bias drives trading behaviour on the JSE. Robustness tests, which include alternative measures and sensitivity tests confirm these findings. Further research is suggested in the study to examine similar patterns across other emerging markets can be found in order to for the findings to be generalised.