Quality Investing – Industry Versus Academic Definitions – full version on SSRN – excerpt below
Georgi Kyosev
Erasmus University Rotterdam (EUR) – RSM Erasmus University; Robeco Asset Management
Matthias X. Hanauer
Robeco Asset Management – Quantitative Strategies; Technische Universität München (TUM)
Joop Huij
Erasmus University – Rotterdam School of Management; Robeco Quantitative Strategies; Erasmus University Rotterdam (EUR) – Erasmus Research Institute of Management (ERIM)
Simon D. Lansdorp
Robeco Quantitative Strategies; Erasmus University Rotterdam (EUR) – Erasmus School of Economics (ESE); Tinbergen Institute
June 13, 2016
Abstract:
In this study we provide an overview of common quality definitions that are currently used in the industry and those used in academic studies, and we outline the differences between these definitions. We show that there is a large dispersion in the definitions that are used for the quality factor with ‘industry’ definitions ranging from return-on-equity and profit margins to leverage and earnings variability, and ‘academic’ definitions such as operating accruals, net stock issues, and gross profitability. We document large performance differences between the different quality definitions. While ‘academic’ definitions for quality all seem to have significant predictive power for stock returns above and beyond common factors, we do not find significant predictive power for individual ‘industry’ definitions. Our results have important implications for the design of investment vehicles that provide investors exposure to the quality factor.
Quality Investing – Industry Versus Academic Definitions – Introduction
Perhaps one of the most significant developments in the asset management industry over the past decade is the rise of factor investing. An increasing number of investors have been adopting this approach to investing where investment portfolios are strategically allocated to specific segments of the market such as the small cap, value, and low-risk segments.1 In brief, the underlying rationale of the factor investing approach builds on the body of academic literature that shows that a significant portion of the value added of active management can be attributed to factors such as small cap, value, and momentum that have been documented by numerous academic studies (see for example the work of Carhart, 1997). By strategically allocating to these factors directly (instead of by selecting a fund manager that might allocate the investment portfolio to the factors) investors gain benefits such as increased transparency and control, and lower costs. The most common factors that are currently used in factor investing approaches are small cap, value, momentum, and low-risk. These factors have been reported by many academic studies (see, e.g., Fama and French, 1993, Jegadeesh and Titman, 1993, and Carhart, 1997). In addition, funds that provide investors exposure to these factors are currently readily available (with exception of the momentum factor that seems somewhat more challenging to implement because of the higher turnover that is associated with the factor).
A newcomer to the factor investing arena over the most recent years is the so-called quality factor. A notable observation regarding the quality factor is that there does not appear to be an unambiguous definition for quality. While different definitions are also used to measure value (e.g., book-to-price and earnings-to-price), momentum (e.g., 6-minus-1-month return and 12-minus-1-month return), and low-risk (e.g., 36-month volatility and 52-week market beta), the dispersion in definitions is substantially larger for quality.
Examples of anomaly variables from academic studies that are seen as quality indicators are accruals (Sloan, 1996), gross profitability (Novy-Marx, 2013), and low investments (Pontiff and Woodgate, 2008). Due to the lack of any market data in these characteristics they are generally referred to as quality variables. At the same time, the definitions that are often used for quality in the industry2 seem to be very different from the definitions that have been put forward in academic studies. For example we mainly find variables related to bottom-line profitability measures such as return-on-equity or derivations of it in index definitions (see, e.g., Piotroski, 2000, GMO, 2012, MSCI, 2013, and S&P, 2014). As a consequence, it is unclear if the quality returns that have been documented by academic studies can validly be used as expectation for quality funds that are currently offered to investors.
The contribution of this study to the literature is threefold: first, we provide an overview of common quality definitions that are currently used in the industry and those that have been used in academic studies, and we indicate the differences between these definitions. Second, we perform asset pricing tests to investigate which of the used definitions have predictive power for relative stock returns above and beyond common factors such as market beta, small cap, value, and momentum. Third, we analyse the robustness of the predictive power for an investable international setting as well as within a corporate bond universe. Existing academic studies investigating the quality-type factors have mainly been performed using broad U.S. equity data that can be dominated by microcaps.3 Furthermore, we are among the first to test the predictive power of quality outside the equity space4. Therefore, the out-of-sample evidence supporting the existence of a quality factor in international and investable universes as well as a different asset class is threefold and currently non-existent.
