Bridge over ocean
1 February 2016 CFA Institute Journal Review

The “Science” and “Art” of High Quality Investing (Digest Summary)

  1. Stuart Fujiyama, CFA

The well-regarded tradition of investing in “quality” companies dates back to Benjamin Graham in the 1930s. Nevertheless, academics and investors have yet to agree on a clear definition of quality. The authors propose an approach to quality that combines “scientific” financial statement and market performance measures with more “artistic” corporate culture and environmental, social, and governance metrics.

What’s Inside?

The academic finance and accounting discipline has produced extensive literature on quality investing. The research has examined backward-looking corporate financial statement–based metrics, such as gross profitability and return on invested capital (ROIC). Meanwhile, a more recent but rapidly growing body of work is now being produced that focuses on forward-looking nonfinancial reporting, including environmental, social, and governance (ESG) metrics. The authors attempt to combine financial and nonfinancial quality metrics, placing an emphasis on their persistence, in an effort to identify companies that are likely to continue delivering high-quality performance over the long term.

How Is This Research Useful to Practitioners?

The authors find that, within the CAPM framework, no strategy driven by financial metrics generates statistically significant alpha at the 0.05 level, but one value metric–based strategy does—the price-to-earnings ratio (P/E). Within the Fama–French three-factor framework, the ROIC-driven strategy generates “somewhat significant” (p = 0.08) alpha, whereas the P/E-driven strategy again delivers significant (p = 0.03) alpha. The authors hypothesize that the explanatory value of financial metric–driven strategies has largely been arbitraged away by the increasing prevalence of data analytics.

They examine the three-factor alphas generated by metric persistence cohort portfolios and find that ROIC is close to statistical significance in its performance, with p = 0.08 for the one-year cohort and p = 0.10 for the three-year cohort. This result is the strongest showing among the quality metrics; consequently, the authors select ROIC as their key financial quality metric.

For the value metrics, the authors find that P/E is statistically significant, at least for the one-, two-, and three-year cohorts. The p-value equals 0.07 for the four-year cohort and is far in excess of acceptable limits for the five-year cohort.

The authors consider 12 nonfinancial indicators of quality—including those representing ESG, corporate culture, and sustainability—and find that governance (proxied by the Bloomberg governance score) has the most significant (p = 0.01) correlation with ROIC for 2014; consequently, the authors select governance as their key nonfinancial quality metric.

Finally, the authors find that the “sweet spot,” or the point at which high-quality companies tend to remain high quality, appears to be after the third year for ROIC and after the fourth year for governance. They use the intersection of the three-year persistence cohort for ROIC and the four-year persistence cohort for governance to create a list of 14 potential high-quality companies.

How Did the Authors Conduct This Research?

The authors begin by examining financial quality metrics, extending the work of Novy-Marx (working paper 2012), who used data from CRSP and Compustat. The authors, using annual data presumably from the same sources, identify 1,036 of the largest US equity stocks by market capitalization as of 2015 and examine their returns from 2000 to 2013.

They form seven long-only portfolios based on the top 30th percentile of companies in terms of five of the financial quality metrics examined by Novy-Marx and two traditional value metrics—P/E and book to price. The authors rebalance the portfolios annually and assess their returns using both a standard CAPM alpha approach and a Fama–French three-factor alpha approach.

For each of the seven financial metrics, the authors create cohort portfolios based on persistence—the ability to remain in the top 30th percentile—for at least the last one year and exactly the last two, three, four, or five years. They calculate the CAPM and three-factor alphas for the resulting five cohort portfolios for each of the seven metrics.

To examine the persistence of governance alongside the persistence of ROIC, the authors create five cohort portfolios for each metric based on the ability to remain in the top 30th percentile for at least the last one, two, three, four, or five years. The same sample of approximately 1,000 stocks is used to examine the persistence of ROIC, but because of data constraints, the S&P 500 Index is used to examine the persistence of governance.

Abstractor’s Viewpoint

To their credit, the authors use a cohort approach to discover more persistent patterns rather than simply focusing on short-term (less than two years) results. In the footnotes, the authors acknowledge that significance is usually assumed at the 0.05 level. But they select ROIC as their key financial quality metric based primarily on somewhat significant (0.05 < p ≤ 0.10) results. Explicitly and consistently setting significance at the 0.10 level might help to justify their decision, but I would prefer to see significance set at 0.05 (or even 0.01) and any results with higher p-values deemed not significant.

We’re using cookies, but you can turn them off in Privacy Settings.  Otherwise, you are agreeing to our use of cookies.  Accepting cookies does not mean that we are collecting personal data. Learn more in our Privacy Policy.