The momentum effect in the Tunisian stock market: Risk hypothesis vs. underreaction hypothesis

Abstract

The purpose of this paper is to examine two controversial explanations for the momentum in the Tunisian stock market: the risk hypothesis and the underreaction hypothesis. To address the risk issue, the five-factor model of Fama and French (2015) was used to estimate the momentum profits. We found a strong evidence of risk-adjusted momentum profits indicating that the risk cannot explain the momentum effect. To test the underreaction hypothesis, an event study was performed to track the market reaction to the information content of earnings before, on and after earnings announcement. We found that good earnings news are followed by positive abnormal returns; while bad earnings news are followed by negative abnormal returns over 12 months after the announcement date. Consistent with the underreaction hypothesis, these findings indicate that the market slowly adjusts in the same direction to the unexpected earnings. To control for this effect, we extended the five-factor model to include a factor based on unexpected earnings. We found that the momentum profits are captured by a zero-investment portfolio that is short on the portfolio with the lowest unexpected earnings and long on the portfolio with the highest unexpected earnings.