Publication Type

Master Thesis

Publication Date



Using data from Morningstar Principia CDs and employing standard methodologies, we examine the extent to which two mutual fund ratings: Morningstar star ratings and Morningstar stewardship grades can predict future fund performance. In particular, we investigate whether the combined predictive power of the two ratings exceeds that of a single rating. We decompose funds into various groups characterized by fund age and fund categories in order to address such issues as whether predictive performance is uniform across characteristic-based groups. Although our analysis shows that none of the ratings alone possesses strong predictive power, there is statistical evidence to support the notion that combined rating is superior to single rating in forecasting future returns. However, the evidence is not overwhelming enough to justify the efficacy of an investment strategy based primarily on Morningstar ratings. Besides studying predictability of ratings, we also construct a logistic regression model to seek potential determinants of the stewardship grades. We find that funds with good stewardship grades are generally those that incur low expenses, possess a large asset base and are managed by experienced fund managers. Finally, we investigate whether the two Morningstar ratings exhibit short-term persistence. Our findings indicate that the degree of persistence (as measured by the percentage of funds that retain their initial rating over a 12-month period) of the stewardship grades is much more pronounced than that of the star ratings.


mutual funds, performance persistence, portfolio performance, rating services, risk aversion

Degree Awarded

MSc in Finance


Portfolio and Security Analysis


GOH, Jeremy

Creative Commons License

Creative Commons Attribution-Noncommercial-No Derivative Works 4.0 License
This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 4.0 License.