Effect of presidential elections on investor herding behaviour in African stock markets

Purpose - The purpose of this paper is to investigate investor herding behaviour and the effect of presidential elections on investor herding behaviour in African stock markets at the sector level. Design/methodology/approach - The study segregates listed firms into financial, consumer goods, consumer services and basic materials sectors and uses the cross-sectional absolute deviation approach as a metric of detecting herding in each of the sectors. The authors extend the model to tease out the effect of presidential elections on investor herding behaviour. Findings - The study reveals that sectoral differences are fundamental to the evolution of herding. Herding is prominent in a financial services sector dominated by banks. The phenomenon also prevails in markets with smaller consumer goods and services sectors. A post-presidential election effect on investor herding is found for the consumer goods and services sectors of Ghana and a pre-presidential election effect is documented in Nigeria's consumer services sector. The authors conclude that post-presidential election effect is as a result of political connections whilst a pre-presidential election effect is attributable to political business cycles. Research limitations/implications - The study is based on four African countries due to data constraints. Nonetheless, the study is the first in Africa to the best of the authors' knowledge, and the results are very solid and have a lot of practical and policy implications. Practical implications - The study has implications for investors as it guides investment behaviour in pre- and post-presidential election periods. Originality/value - Past studies on investor herding behaviour in African stock markets have largely concentrated on the aggregate market. Knowledge on sectoral differences in investor herding is almost non-existent for African stock markets. Furthermore, premised on the fact that stock markets react to presidential elections, there is no known study that have attempted to examine the effect of presidential elections on investor herding behaviour. This paper contributes to the literature by providing evidence on sectoral differences in investor herding behaviour and the effect of presidential elections on sectoral herding behaviour.