Safe-Haven Effectiveness of Cryptocurrency: Evidence from Stock Markets of COVID-19 Worst-Hit African Countries

Abstract

The study assessed the hedge or safe-haven property of five cryptocurrencies for stocks of three COVID-19 worst-hit African countries. We address two main concerns bordering on the predictive capacity of African stocks for cryptocurrency returns and the safe-haven property that cryptocurrencies could offer to African stocks. A distributed lag model, with explicitly incorporated salient statistical features, was adopted based on its efficient management of parameter proliferation and estimation biases. We ascertained the model’s in-sample predictability and evaluate its out-of-sample forecasts performance in comparison with the historical average model, using Clark and West statistics. While African stocks significantly predicted cryptocurrency returns, the cryptocurrency-stocks nexus revealed the diversifier and safe-haven property of cryptocurrencies for African stocks in periods of normalcy and crisis/pandemic, respectively. Our predictive model outperformed the historical average model in the out-of-sample. Our results may be sensitive to cryptocurrency-stocks nexus and sample periods but not the out-of-sample forecast horizons.