Not logged in.

Contribution Details

Type Master's Thesis
Scope Discipline-based scholarship
Title Augmenting Factor Investment Strategies with ESG-Scores
Organization Unit
Authors
  • Aron Horvàth
Supervisors
  • Thorsten Hens
Language
  • English
Institution University of Zurich
Faculty Faculty of Business, Economics and Informatics
Number of Pages 86
Date October 2019
Zusammenfassung Responsible investments are one of the most popular topics currently discussed in finance. Yet, it is by no means a new field of study, as academic research has been exploring the connection between corporate social responsi- bility (CSR) and financial performance for over four decades. But the question of whether decision making based on ethical, societal and environmental factors can create economic benefits attracts financial institutions, asset managers and individual investors just the same. Still, there is much uncertainty concerning the financial impacts of decisions based on CSR. In many areas of CSR-related research, neither the practical nor the academic world could reach conclusive statements. While the general conclusion of most research regarding the link between CSR and company financial performance (CFP) is that there exists an at least modestly positive link between CSR and CFP (Friede et al. (2015)), the same link rarely applies on an aggregated portfolio level. Most strategies sorted on environmental, social governance criteria (ESG) have offered no performance increases over regular strategies, and sometimes suggested that investors even have to pay a price to be invested in more sustainable companies. Further, a risk-reduction effect of focusing on sustainable stocks could also not be observed (Brooks and Oikonomou (2018)). Previous research has offered multiple explanations for the lack of performance in SRI-portfolios. In a literature review, Friede et al. (2015) identify three main reasons. First, overlapping effects from market and non-market factors in a portfolio can cover a potentially existing ESG-alpha. Second, Derwall et al. (2011) find different performance effects from ESG-screens which can net out when combining them in portfolios. And third, portfolio studies often involve transaction costs which might mitigate those slight positive effects of CSR on the companies in the portfolio. Pizzutilo (2017) further extends the list by showing the possible negative effects of having an underdiversified portfolio because of exclusionary screens which also could net out any gains from the higher CSR-levels in the portfolio. To better understand these issues on the portfolio level, and to explore how to possibly better include ESG into an existing portfolio strategy, this paper will focus on the effects of ESG-sorting on elementary factor strategies, including size (SMB), value (HML), profitability (RMW), investment (CMA), momentum (MOM) and short term reversal (STREV). Trying to isolate effects in specific style-investments might give a better understanding which parts of a portfolio are affected in which way by an ESG-sort. As previous research has shown, better CSR has an influence on various performance metrics, which in turn often are used to sort into the factor strategies. ESG-sorting was performed based on the aggregated ESG-scores from Thomson Reuters (Refin-itiv). The scores included regular ESG, E, S, G as well as a Controversies score C and an aggregated ESGC score combining ESG and C. First, I established a baseline by looking at the effect of ESG-sorting on the whole market. My results stay in line with prior research. First of all, good-minus-bad (sustainable-minus-unsustainable) strategies performed on a global market showed almost no stat-istical significance, which means generally such a long-short portfolio cannot generate significant performance. Regardless of which ESG-dimension I sorted on, a bad portfolio usually performed slightly better than a good one, providing more proof that investors have to pay a price for being invested in more sustainable companies. Furthermore, in most cases even long-only strategies could not generate significant alpha over the market. The dataset covered over 3000 companies and during most of the time had over 1000 companies with available ESG-data. This provided a good opportunity to analyse regional differences as well. There were indeed large regional differences. In North America and the Asian/ Pacific region, sustainable portfolios were more likely to have average returns above those of the unsustainable ones, while in Europe the opposite was true. Also, it could be observed that the effects of sustainability sorting were much more pronounced for small companies and in those cases it was more likely that a sustain- able portfolio could create higher returns than the unsustainable. As the next step, I created the factor strategies in the same way as in prior related research Asness et al. (2019); Fama and French (2015); Jegadeesh (1990); Jegadeesh and Titman (1993) and sorted using the ESG-scores on top of the factor legs, thus separating every factor leg into sustainable and unsustainable. Unsurprisingly, most factors were affected differently by the inclusion of ESG. After seeing high variability in ESG-sorted long-short strategies, I looked at the different factor legs separately. Almost no ESG-sorted long leg outperformed the unsorted one but rather they underper-formed or were very similar to the regular legs. The only signifi- cantly better long leg was found in the sustainable STREV. Short legs varied more. Key findings were that a portfolio of unsustainable (CMA), momentum (MOM) and short term reversal (STREV). Trying to isolate effects in specific style-investments might give a better under- standing which parts of a portfolio are affected in which way by an ESG-sort. As previous research has shown, better CSR has an influence on various performance metrics, which in turn often are used to sort into the factor strategies.
Export BibTeX