Momentum Anomaly in Emerging Markets

The medium-term momentum effect is one of the most intriguing anomalies in modern financial theory. Unlike other pricing anomalies such as short-term and long-term reversal effects, medium-term momentum is the most difficult to explain using rational asset pricing models. According to Nobel laureates Fama and French, the main drawback of their three-factor risk model is its inability to describe the pattern of medium-term momentum. This study attempted to answer the question of differences in momentum returns in emerging markets through the lens of momentum crashes and to draw a parallel between the market-specific level of investor risk aversion and the characteristics of momentum crashes. Momentum crashes are a promising explanation for excess momentum returns, whereby significant negative WML portfolio returns occur during market recovery periods following a correction, as the portfolio of losers recovers faster than the portfolio of winners. This study hypothesizes that one of the reasons for the asymmetric growth of losers after a market correction may be a distorted perception of expected returns by investors during market stress, the degree of which is influenced by the aversion risk parameter, which is unique for investors in each market. To test this hypothesis, using the Python programming language, momentum strategies were constructed with a robust 6-1-6 specification for 18 selected emerging markets in accordance with the Jegadeesh and Titman methodology. In addition, the CAPM model was applied to momentum returns to clean up market risk. As a result, it turned out that the profitability of the strategy differs significantly in the studied markets, and the residual alphas are statistically significant in several of them, which makes momentum an anomaly. Further, the low profitability of the WML portfolio was confirmed in the phase of market recovery after the correction, which is most favorable for the occurrence of momentum crashes. From the data obtained, it became clear that it is during this phase that the largest share of all the most significant negative momentum returns occurs. In the final part, an attempt was made to describe the differences in the characteristics of momentum crashes through the attitude to risk specific to investors in each country. For this purpose, two indexes of aversion risk were used. As the main measure, we used the regression coefficients calculated on the basis of the global aversion risk index applied to the market indices of each of the markets. The aversion risk index of sociologist Geert Hofstede was used as an alternative indicator. As a result, countries were sorted based on investor’s attitude to risk. The resulting classification showed the presence of a consistent pattern between the attitude of local investors to risk, the amplitude of momentum crashes, and the total return on the WML portfolio. Thus, the markets in which investors are more risk-tolerant are characterized by lower losses of the WML portfolio in the phase of market recovery and higher overall profitability of the momentum strategy. This observation allows us to better understand both the differences in momentum between the markets and the very nature of the momentum effect.