Previous research has indicated that momentum strategies earn abnormal returns, but most empirical studies that analyze global risk-adjusted performance have not considered a momentum factor. The authors examine the impact of index momentum factors on international (excluding US) and global equity funds. They demonstrate that momentum factors affect risk-adjusted performance, performance persistence, and analysis of luck versus skill.
What’s Inside?
The selection of proper risk factors is crucial in the measurement of mutual fund performance. The authors expand an index-based version of the Fama–French three-factor model (i.e., including beta, size, and value factors) by including both country and sector momentum factors. The inclusion of these factors results in a five-factor model, which, according to the authors, better explains variation in fund returns.
They compare one-, three-, and five-factor models for international and global risk-adjusted performance and discover that more than 50% of global and international equity funds show a significant exposure to at least one momentum factor. On average, after adjusting for risk, the funds underperform their corresponding benchmarks. In addition, the authors find that sorting funds according to their country momentum and sector momentum exposures reveals a positive relationship between these exposures and risk-adjusted performance.
How Is This Research Useful to Practitioners?
Investors pay high fees for active management, but because momentum factor investing does not require superior investment skill, it probably does not justify active fees. The authors find that in international equity funds, when country and sector momentum are added as risk factors in the five-factor model, the resulting alpha becomes negative or only slightly above zero, depending on the fund weighting.
Fund-of-funds managers would be interested in the authors’ test of the persistence of mutual fund performance. Remarkably, they find that if performance is measured on a three-factor model, investors achieve outperformance when their capital is allocated to the top decile of international funds. However, when sector and country momentum factors are included (i.e., the five-factor model), the alphas of the top-decile portfolios are lower and not significant.
The authors’ analysis of luck versus skill using the three-factor model suggests that among the top 10% of international funds, there are skilled managers at a 5% level. However, using the five-factor model, they find skilled managers among only the top 1%.
How Did the Authors Conduct This Research?
The fund data are from the CRSP Survivor-Bias-Free US Mutual Fund Database. Funds are screened on description (funds whose names include such terms as exchange-traded fund, or ETF; long–short; and alpha-only are eliminated), and funds with fewer than 24 consecutive months of performance are removed. The initial dataset includes 586 international (excluding US) and 209 global equity funds. The period evaluated is from January 1996 to December 2009.
The authors conduct tests on several index momentum strategies. Country and sector indices are ranked according to their 6- and 12-month performance and sorted into five country portfolios and three sector portfolios. The portfolios are held for performance periods of one and six months. The authors find that country strategies based on the past-12-month return and a holding period of 1 month yield the highest average return, at 1.10% per month (based on returns in local currency).
They also investigate the relationship between index momentum exposures and fund characteristics. They find that funds with high index momentum exposure are likely to be invested in small growth stocks. Additionally, funds in the top country momentum quintile or funds in the top sector momentum quintile appear to be more profitable than other funds; they typically exhibit higher alphas than those in the bottom quintiles. The top momentum quintiles also show high expense and turnover ratios.
Abstractor’s Viewpoint
The authors’ research is comprehensive, and it is difficult to do it justice in a summary. The authors include tables of summary statistics, which allow readers to review the statistical significance of the data and draw their own conclusions.
The research raises such questions as, What is alpha, and what is beta? The increasing popularity of investment strategies sold as “smart beta” has led many to view the capital asset pricing model version of beta as overly simplistic. Newly created investment strategies are attaching such risk factors as value stock exposure, small-cap exposure, and price momentum to smart beta. The authors clarify these return dimensions as additional risk factors and thoroughly analyze risk-adjusted performance. They have conducted interesting research that is definitely worth reading.