The relationship between investor sentiment and unit trust alpha on the JSE

Abstract

When examining the returns of a fund, there is a general consensus that risk adjusted measurements are standard tools to assess performance. Some popular examples would be the Sharpe and Treynor ratios. Alpha, which measures the difference between the expected return of a fund and the actual return based on a benchmark model, such as the Capital Asset Pricing Model (CAPM) is also used. The interpretation of a positive alpha is that a fund has performed better than the market. Traditional benchmark models such as the CAPM and its asset pricing variants are also used to understand the reason behind the occurrence of alpha. In classical asset-pricing models, only risk should be able to predict stock returns; however, numerous variables, without any clear association with risk, have been documented to predict stock returns. As such, factors that are non-traditional or not fully rational, such as investor sentiment which stem from behavioural finance, are usually not considered in these models. This study investigates whether there is a link between investor sentiment and unit trust alpha by using a macro-economic asset pricing factor model with two investor sentiment proxies (the Twitter Economic Uncertainty Index and Economic Policy Uncertainty Index) in South Africa between 2011 and 2020. The results show that investor sentiment is a statistically significant variable in an asset pricing model. The implication is that, after finding that investor sentiment is significant, a manager that fails to consider investor sentiment is likely to achieve inferior results to a manager that does.