Low-volatility strategies have quickly gained attention in the investment community. The authors investigate a heuristic-based design that leads to a portfolio with a superior Sharpe ratio. The back-tested methodology and results are very robust.
Many traditional low-volatility strategies, when optimized, create a bias toward smaller-cap stocks and overconcentration in a small number of sectors and/or countries. The authors provide a construction methodology that overcomes these problems and leads to a practical and investor-friendly portfolio with a superior Sharpe ratio. The portfolio is constructed by combining the return-enhanced engine of the fundamentals-based methodology with a low-volatility design that allocates to low-beta stocks.
Investors want low-volatility strategies, but many existing strategies are problematic. They require estimates of the covariance matrix and typically have a small-capitalization tilt. Such strategies incur large turnover. The authors provide a construction methodology that overcomes the problems in existing methods. Their approach has many advantages: The turnover rate is lower, the level of transparency is higher, and the resulting portfolio has broader representation.
Comparing the performance of their fundamentals-weighted, low-volatility portfolios with the relevant benchmark portfolios, the authors find that their portfolios significantly outperformed. The results are consistent in each of the three regions. Overall, relative to the traditional optimization-based, minimum-volatility portfolio, the fundamentals-weighted, low-volatility strategy provides comparable or slightly enhanced performance and considerable improvement in turnover rate and investment capacity with low volatility.
The authors back test their strategy with the United States, developed markets excluding the United States, and emerging markets. They select the largest 1,000 stocks from each of the three regions, based on company fundamental weight. The fundamental weight is calculated by equally weighting firm size–related company accounting matrices: cash flow, book value, sales, and dividends. In this way, the authors ensure liquidity and low transaction costs. The data include 45 years of history for the United States, 25 years for the developed markets excluding the United States, and 13 years for the emerging markets.
To eliminate the small-cap bias, the authors take the weight generated by a traditional optimizer methodology and multiply it by the market cap or the fundamental weight of the company. To remedy the bias related to overconcentration in a sector, they sort the 1,000 largest stocks by their betas within the sectors. They select the lowest 30% of stocks in each sector to be the final constituents of that sector. They then repeat the above procedure but substitute country/region for sectors to get a country/region-neutral portfolio.
The authors break down their analysis into growth and slowdown macroeconomic phases and find that their strategy outperforms the benchmarks across the three regions regardless of growth or slowdown phase. The level of return differences over the two phases is distinct. Their strategy provides similar or better performance than core equity indices during normal market conditions. But they find that during the market crash, their strategy reduced risk considerably while generating a significant return enhancement.
The fundamentals-weighted, low-volatility strategy described by the authors is straightforward and transparent. They use a wide variety of benchmarks to test their results and conduct back tests of their strategy in three very representative regions. They also conduct the tests over different macroeconomic cycles. The methodology and the results are very robust.
Summarized by Vipul K. Bansal, CFA. Copyright 2013, CFA Institute. Reproduced and republished from the CFA Digest with permission from CFA Institute. All rights reserved. Article originally published in the Journal of Index Investing, Spring 2013, Vol. 3, No. 4: 8-22.