After 40 years or so, quantitative investing (learn more about quant investing at our Democratize Quant Conference) has evolved into a thriving practice. A major feature of the quantitative approach involves developing underlying numerical models and testing them on a historical (data) record and then forecasting where alpha may be embedded into the prices of a set of stocks. Whether you agree or disagree with this approach, it is difficult to deny that with the advanced state of data access and computational skill, “quants will win the day in ESG investing”. Such is the premise of this article and happily, it is accompanied by a compelling argument.
How does ESG challenge quantitative approaches?
What other approaches are used for constructing ESG portfolios?
Which are the realistic advantages of the quantitative approach relative to the others?
What are the Academic Insights?
The tidal wave that is ESG (environmental, social and corporate governance) presents the quantitative investing world with a number of new challenges. The acronym “ESG” has now become a portfolio goal within the constructs of modern portfolio theory (MPT) where the objective is the best tradeoff between risk and return. Individual ESG factors or fundamentals may or may not conflict with the objectives of enhancing alpha or in forming an efficient portfolio. For example, one issue that arises with ESG is that of the length of the time horizon. The identification of ESG factors that are enhancing or detracting depends on the length of the time horizon. For example, E&S features tend to add value over longer horizons, while governance has a short term horizon.
The authors construct an arbitrary set of 3 nonoverlapping categories to implementing an ESG portfolio:
Slicing or screening: Negative screening involves the screening out of specific sectors or industries and/or stocks based on environmental, social or governance (ESG) criteria; Positive screening involves including only sectors or stocks selected on the basis of “best-in-class” performance on ESG criteria.
Seasoning ESG integration involves the express incorporation of ESG criteria and practices into the financial analysis of stocks or industries. For example, sustainability or theme investing entails investment in areas such as clean energy, sustainable farming, sustainable green technology, green bonds or ETFs, and so on, and,
Solving explicitly for ESG factor exposures. The rigor of the third approach is on par with most quantitative methods. The authors argue that this approach allows for variability in preferences and. Investors have the choice to optimize the tradeoff between their ESG objectives and risk/return preferences. Although it sounds easy, the problem is not only generating alpha but maintaining target ESG exposures. Will the quantitative approach “be the future of formally integrating ESG with wealth creation”? The authors argue yes, and discuss a number of advantages unavailable to slicing, dicing or seasoning an ESG portfolio.
Better integration with MPT. Given the vast amount of “artillery” developed by quantitative researchers and practitioners, including various definitions of risk, pricing models, CAPM, factors, efficient markets, and so on, ESG may be seen as a simple extension whereby optimizing alpha becomes optimizing a combination of alpha plus ESG metrics. Furthermore, quant skills that combine knowledge of the ESG domain and the expansive nature of ESG data with computational and digital technology can be used to produce models that identify ESG features of stocks. See Exhibit 6 for an example
Better model building with larger and deeper datasets and systemizing into ESG domain knowledge. Everyone would agree that the quality, quantity and consistency of ESG data is problematic. From the quant point of view some of the deficiencies of ESG data, or lack of, can be addressed. For quality, outliers may be easily identified with statistical methods, missing data may be estimated by using other facts and information. For consistency and quality, ESG data gaps may be filled by analyzing nonstandard information using natural language processing to gauge various aspects of corporate governance. Conference calls, SEC reports, company-issued communications are examples of nonstandard data that can be mined for governance characteristics.
Better replication with digital estimation methods. As an example of the improved precision available with quantitative estimation methods, the authors report the results of an internal study design to replicate commercial ESG ratings. The predictive variables consisted of a linear combination of proprietary ESG factors that managed to produce statistically significant replications of MSCI ESG ratings up to 14 months out, over the period 2012 – 2020. Essentially, the quantitative approach was very successful at building an expert system that successfully mimic human decision-making. See Exhibit 6 below.
Why Does it Matter?
The contribution of this article is to point out the natural and complementary features of ESG investing and quantitative methods for building portfolios. The authors argue that expert systems can be built that mimic very effectively, human thinking but only with the current status of our ability to conduct digital estimations.
The Most Important Chart from the Paper
Sustainable (also known as environmental, social, and corporate governance [ESG]) investing is currently of intense interest in the investment world. In this article, the authors consider the salient challenges associated with ESG investing and how quantitative approaches may address them. Compared to fundamental methods of sustainable investing, the authors see quantitative methods as having several advantages: These methods can build on and extend the vast analytical toolbox of modern portfolio theory to incorporate investor preference in portfolio construction; they can leverage the recent data explosion to obtain insights on many intangible sustainability metrics; and they do not have the black box label. Instead, subjective judgment applied to building the quantitative system is essential. A thoughtful analytical system can be applied to a large universe of stocks, and quantitative methods may also be leveraged to predict popular ESG vendor ratings. Although these are the early days of quantitative sustainable investing, the authors believe these advantages will prove the quantitative method’s worth in sustainable investing.
Dr. Tommi Johnsen was a past Director of the Reiman School of Finance and a tenured faculty at the Daniels College of Business at the University of Denver. She has worked extensively as a consultant and investment advisor in the areas of quantitative methods and portfolio construction. She taught at the graduate and undergraduate level and published research in several areas: capital markets, portfolio management and performance analysis, financial applications of econometrics, and the analysis of equity securities. Her publications have appeared in numerous peer-reviewed journals.
Performance figures contained herein are hypothetical, unaudited and prepared by Alpha Architect, LLC; hypothetical results are intended for illustrative purposes only. Past performance is not indicative of future results, which may vary. There is a risk of substantial loss associated with trading stocks, commodities, futures, options and other financial instruments. Full disclosures here.