This In Practice piece gives a practitioner’s perspective on the article “Global Equity Country Allocation: An Application of Factor Investing,” by Timotheos Angelidis and Nikolaos Tessaromatis, published in the Fourth Quarter 2017 issue of the Financial Analysts Journal.
What’s the Investment Issue?
Factor investing—the investing technique that creates portfolios using such factor characteristics as momentum, low risk, and small stock size—has undergone significant advances in recent years but is generally undertaken only at the domestic stock level rather than the country market level.
The authors test whether global factor portfolios formed at the country level, using country indexes instead of factors represented in individual stocks, perform as well. Why? Because factor portfolios face major implementation challenges, and country portfolios—which are based on widely available indexes—are easier to manage and more liquid and have higher capacities and lower transaction costs.How Do the Authors Tackle the Issue?
First, the authors create a series of global portfolios using country-focused exchange-traded funds (ETFs) and futures. In each portfolio, they create one of four global equity factors widely accepted as having outperformed the market-cap-weighted index over the long term. These factors are value, small capitalisation, high momentum, and low risk.
As well as choosing countries with the highest factor exposure to create each factor, the authors apply market-cap-weighting, equal-weighting, minimum-variance, inverse-variance, and maximum-diversification methodologies to each portfolio. Market-cap weighting means companies with the largest market capitalisation have proportionally more representation in a portfolio, whereas equal weighting gives the same importance to each stock in the portfolio. Minimum variance is designed to lower volatility across the portfolio, while inverse variance assigns higher weights to individual low-volatility stocks. Maximum diversification aims to create a portfolio that is as diversified as possible without seeking to maximise returns.
This variety of methodologies is used in case one or more are found to be particularly effective in maximising risk-adjusted returns.
Second, the number of countries used to create global factor portfolios is expanded beyond that used in earlier research to include emerging markets. The total number of markets included in the article is 44, with the factor portfolios spanning the period from 1980 to 2015. The idea behind including emerging markets is to increase the opportunity set and to see whether doing so improves risk–return performance.
Third, the authors combine single-factor portfolios to create a global multifactor portfolio, which should perform more consistently across different market cycles. It should also limit portfolio turnover compared with managing four separate portfolios.
Fourth, the authors compare global, stock-based factor portfolios with country-based portfolios to see whether they perform similarly.
In addition, to reflect investors’ practical needs, the authors construct global multifactor portfolios using a 2% tracking error constraint to the world stock index and consider the effect of this constraint on performance.What Are the Findings?
A number of findings are likely to be of interest to investors seeking to build a globally diversified portfolio using factor premiums.
The first is that each of the four factor portfolios displays significantly better returns than the world cap-weighted index. The outperformance ranges from 2.48% for a low-beta portfolio constructed using minimum variance to 8.42% for a high-momentum, minimum-variance portfolio.
Second, adding emerging markets increases the Sharpe ratio (which measures excess risk–return versus the risk-free rate) of all the single-factor portfolios. The Sharpe ratios of portfolios that include emerging markets are, on average, 15% higher than the Sharpe ratios of portfolios based solely on developed markets.
Third, the Sharpe ratio of global multifactor portfolios is significantly higher than that of the widely used world cap-weighted index.
Fourth, country-based portfolios have higher returns than, and similar Sharpe ratios to, both stock-based factor portfolios and the MSCI factor-based indexes. An equally weighted global multifactor country-based portfolio, for instance, outperforms the MSCI combined multifactor portfolio by 2.85%.
Finally, ensuring that the country-based portfolios closely track the cap-weighted index impairs performance because of the extra costs involved.What Are the Implications for Investors and Investment Professionals?
Country-based factor portfolios—tilted toward value, low risk, value, and momentum and implemented through ETFs and index futures—offer a superior alternative to stock-based factor investing and an improvement on other traditional strategies.
A factor-based portfolio built from country exposures outperforms the world cap-weighted stock market. The outperformance is sizable even after allowing for transaction costs, various portfolio construction methods, and tracking error constraints.
In addition, country-based factor portfolios provide risk and return performance similar to or better than that of stock-based factor portfolios.
With greater capacity, more liquidity, reduced transaction charges, and greater ease of management, factor-based portfolios are easier to construct and so could become cheaper for investors. If their cost is reduced toward the level of passive cap-weighted index products, it is possible that factor-based investments could one day match flows into traditional passive funds.
This shift, in turn, may lead to more investment firms entering the market and increased research and innovation in factor investing. A niche part of the investment industry could thus move into the mainstream.