BI Norwegian Business School - campus Oslo GRA 19502 Master Thesis Component of continuous assessment: Thesis Master of Science Final master thesis – Counts 80% of total grade Can Sustainability Criteria Enhance Returns and Reduce Risk on Stocks? This thesis is a part of the MSc programme at BI Norwegian Business School.The school takes no responsibility for the methods used, results found and conclusions drawn. ID number: 0940426, 0944587 Start: 02.03.2017 09.00 Finish: 01.09.2017 12.00 GRA 19502 0940426 0944587 Abstract This thesis investigates whether sustainability criteria can be used to enhance return and reduce risk on stocks. This is done through conducting an empirical analysis on European stocks from 2007-2016, with the purpose of identifying a four-factor model that includes the sustainability score in addition to the three Fama and French factors. The methodology is based upon famous techniques to test asset pricing models, performing one two-pass regression inspired by Fama and Macbeth (1973) and one two-pass regression inspired by Fama and French (1992). The results show that the criteria can be used to obtain higher expected return, less volatility and less company-specific risk by investing in companies with better sustainability scores. 1 GRA 19502 0940426 0944587 Content Content ................................................................................................................................ 2 1.0 Introduction .................................................................................................................. 3 2.0 Responsible Investing ................................................................................................... 6 2.1 Community investment .......................................................................................................... 7 2.2 Shareholder Advocacy ........................................................................................................... 7 2.3 Screening ................................................................................................................................ 7 3.0 Related Literature ......................................................................................................... 9 3.1 Fama and MacBeth .............................................................................................................. 10 3.2 Fama and French: Three-factor Model ................................................................................. 11 4.0 Modern Portfolio Theory ............................................................................................ 12 4.1 Mean-variance Analysis ....................................................................................................... 12 4.2 Portfolio Evaluation ............................................................................................................. 13 4.3 The Link between MPT and ESG Investing ......................................................................... 13 5.0 Methodology ............................................................................................................... 14 5.1 Motivation and Limitations .................................................................................................. 14 5.1.1 Analysis on Portfolios .................................................................................................. 14 5.1.2 Analysis on Individual Securities ................................................................................. 14 5.2 Portfolio Analysis ................................................................................................................. 15 5.3 Two-pass Regression of Portfolios ....................................................................................... 15 5.3.1 First-stage Regression .................................................................................................. 15 5.3.2 Second-stage Regression .............................................................................................. 16 5.3.3 Adjusting Standard Errors ............................................................................................ 16 5.4 Two-pass Regression of Individual Stocks .......................................................................... 16 5.4.1 First-stage Regression .................................................................................................. 16 5.4.2 Second-stage Regression .............................................................................................. 17 6.0 Data ............................................................................................................................. 18 6.1 Data Description ................................................................................................................... 18 6.1.1 Return ........................................................................................................................... 18 6.1.2 Market Portfolio ........................................................................................................... 18 6.1.3 Risk-free Rate ............................................................................................................... 18 6.1.4 Size ............................................................................................................................... 18 6.1.5 Book-to-market ratio .................................................................................................... 19 6.2 Thomson Reuters ESG Rating ............................................................................................. 19 6.3 Preliminary Results .............................................................................................................. 20 6.3.1 Summary Statistics of Individual Stocks ...................................................................... 20 6.3.2 Summary Statistics of Portfolios .................................................................................. 21 7.0 Analysis ...................................................................................................................... 25 7.1 Two-pass Regression of Portfolios ....................................................................................... 25 7.1.1 First-pass Regression .................................................................................................... 25 7.1.2 Second-pass Regression ............................................................................................... 25 7.3 Two-pass Regression on Individual Securities ..................................................................... 27 7.3.1 First-pass Regression .................................................................................................... 27 7.3.2 Second-pass Regression ............................................................................................... 28 7.4 Discussion of Results ........................................................................................................... 30 7.4.1 Unifying Results ........................................................................................................... 30 7.4.2 Methodological Limitations ......................................................................................... 30 7.4.3 Practical Relations ........................................................................................................ 31 8.0 Conclusions ................................................................................................................ 32 9.0 Bibliography ............................................................................................................... 33 2 GRA 19502 0940426 0944587 1.0 Introduction Remember in 2015 when Volkswagen lost more than 20% shareholder value the week of an emission scandal?1 Or in 2010 when BP stock prices fell 55% after the Deepwater Horizon incident?2 Or in 2017 when a video footage of law enforcement forcible dragging a ticketed passenger from United Airlines plane went viral?3 The company stock price subsequently fell approximately 4%. Events as these have triggered critical questions on the relationship between sustainability and financial performance. The focus on sustainability has boomed over the last years. From 2014 to 2016 assets being professionally managed under sustainable strategies have increased by 25 percent (The Global Sustainable Investment Alliance, 2016). Some even claim sustainability to be one of the most significant trends in financial markets for decades (Clark et al., 2015). To facilitate for more capital flowing into a sustainable economy, the financial impact needs to be addressed. Previous research has failed to reach consensus on this link. According to Modern Portfolio Theory, imposing constraints on the investment universe will sacrifice diversification. This thesis’ contribution is to shed light on the link between sustainability and the financial performance with a focus of an investor who integrate the sustainability score of a stock into his investment decision analysis. This is in contrast to most previous research that has focused on how the average ESG (Environmental, Social and Governance) investment does. More precisely, the following research question and hypotheses have been chosen: 1 In September 2015, German car manufacturer Volkswagen admitted that 11 million of its vehicles were equipped with software that was used to cheat on emissions tests. 2 In April 2010, there was an explosion on the Deepwater Horizon rig caused by a blowout that killed 11 crew members. Two days later, Deepwater Horizon sank while the well was still active and caused the largest offshore oil spill in U.S. history. 3 In April 2017, a United Airlines passenger was forced to give up his seat due to an overbooked plane. The videos and footage of the scene show how he was dragged down the aisle by the arms and legs while other passengers shouted in protest. 3 GRA 19502 0940426 0944587 Can sustainability criteria enhance returns and reduce risk on stocks? 1. An investor can use the ESG rating to enhance return 2. An investor can use the ESG rating to reduce risk The analysis in this thesis is limited to the European market. Europe has the highest portion, 52.6%, of global sustainable investments assets in the world, and is considered a region with high living standard which has the right conditions for ESG policies (Global Sustainable Investment Alliance, 2016). According to the Country Sustainability Ranking as of October 2016, several European countries are considered to be among the top performers in the world (RobecoSam and Robeco, 2016). As a proxy for sustainability, the Thomson Reuters ESG score is used. It provides a reliable objective way to evaluate how investments are meeting ESG issues challenges, and can be downloaded from the database for investors to use. The analysis conducted builds upon two different two-pass regressions. Both regressions are using a four-factor model that includes an ESG term in addition to the three Fama and French factors. In the first analysis stocks are grouped into factor-mimicking portfolios based upon their ESG score, and subsequently the Fama and French factors. Then, these portfolios are used in a two-pass regression inspired by Fama and Macbeth (1973). The second method is a two-pass regression of each individual stock. The results achieved indicate a positive, significant relationship between ESG and return, and a negative relationship between ESG and risk. The practical implication of this is that an investor can benefit from adding the ESG score of a company to his investment analysis process. The remainder of this paper is composed as follows:  Chapter 2 provides background on material of responsible investing, by starting with a more broadly and historical perspective and narrowing it down to the sustainability term that will be applied in this thesis. Thereafter, general strategies of responsible investing will be introduced. 4 GRA 19502 0940426 0944587  Chapter 3 summarizes the core literature that exists on sustainability, and is followed by a description of Fama and Macbeth (1973) and Fama and French (1992).  Chapter 4 introduces the fundamental theory, Modern Portfolio Theory.  Chapter 5 contains a description of the methodology used.  Chapter 6 explains the data used in the empirical analysis, goes more into depth of the Thomson Reuters ESG rating, and introduce preliminary results.  Chapter 7 and 8 provide the analysis and conclusion. 5 GRA 19502 0940426 0944587 2.0 Responsible Investing Responsible Investing or Social Responsible Investing (SRI) is a strategy which combines an investor’s intention to maximize both financial return and social return. This fast growing industry is particularly growing among among women and the millennial generation, two groups that are quickly becoming more influential investment decisions makers. A survey conducted by the Morgan Stanley Institute for Sustainable Investing (2015) found that (1) female investors are nearly twice as likely as male investors to consider ESG factors when making investment decisions, and (2) millennial investors are twice as likely to make sustainable investment decisions as other investors. It is useful to have a common understanding of the investment strategy that incorporate ethical conditions in order to optimize financial return. SRI has emerged in recent years as a dynamic and quickly growing segment of the financial services worldwide. Traditionally, SRI was about the alignment of investments and the values of the investor. Common themes that were inconsistent with the value of the SRI investors were typically gambling, tobacco, alcohol etc. Investors practiced this by avoiding investments in companies that offer such products. The asset managers easily implemented the exclusion strategy of such areas, but those investors with values concerning sustainability were missing a reliable basis for selection of stocks. Investors required more information about companies’ behavior related to ESG issues. Researchers addressed this by creating ESG evaluations, where the companies that do well on these evaluations indicate sustainable companies. Still, it is often difficult to classify an ESG issue as only an environmental, social or governance issue, as they are often interlinked. Even though investors use slightly different measures of ESG, some common examples are presented in the table below. 6 GRA 19502 0940426 0944587 Table 1 ESG Issues Retrieved from Hayat, U., & Orsagh, M. (2015). Environmental, Social, and Governance Issues in Investing: A Guide for Investment Professionals. Copyright by the CFA Institute. More broadly, sustainable and responsible investment is defined as an investment approach that incorporate the environmental, social and governance factors in the investment process. Within this context, there are three main strategies investors employ for responsible investing: community investment, shareholder advocacy and screening. This thesis will focus on screening. 2.1 Community investment Community investment is a way of sustainable investing. Investors allocate a percentage of their investment directly to Community Development Financial Institutions (CDFIs) to support economic development. Typically, they provide capital to low-income or disadvantaged communities. 2.2 Shareholder Advocacy Generally, stock ownership comes with rights, such as the right to vote in annual meetings. Shareholder advocacy describes the actions investors take by using their shares in companies to improve the environmental, social and governance practices. Other examples of shareholder advocacy are proxy voting, dialogues with corporate leaders and shareholder resolution. 2.3 Screening Screening is the practice of excluding or including companies from portfolios based on ethical criteria. Generally, investors seek to own profitable companies that make positive contribution to the society. There are several types of approaches for screening:  Norm-based screening is a strategy that involves assessing each company held in the investment portfolio against global norms, principals or 7 GRA 19502 0940426 0944587 standards such as environmental protection. The norms or principals are typically set out in international initiatives and guidelines such as OECD, UN Global Compact or other governmental or intergovernmental organizations, for example international labour organization (ILO).  Negative screening excludes companies from investors’ investment universe, due to the fact that these companies operate in industries that do not meet the ethical criterion of sustainable investment. Typically, companies are avoided due to their controversial business areas such as alcohol, tobacco or gambling. Negative screening may cause a reduction in investment opportunities since investors exclude companies, consequently, limiting diversification of risk. For example, Norway’s Governance Pension Fund excludes companies that base 30% or more of their activities on coal, and/or derive 30% of their revenues from coal.  Investors that practice positive screening include companies in their investment universe based on ESG performance. While negative screening will only reduce the investment universe, positive screening will lead to different optimal weights for each investment in the optimal portfolio. In other words, investors are facing three objectives: maximize financial return, minimize risk and maximize impact. The screening process can be very expensive for individuals, and as a consequence the demand for a reliable rating has soared. Two major agencies providing this rating are Thomson Reuters and Morningstar. Both Thomson Reuters and Morningstar provide positive screening based on a best-in-class approach. This approach is favoring investments with best practice amongst several sector peers, and is chosen as it will allow a sector balance within the investable universe. 8 GRA 19502 0940426 0944587 3.0 Related Literature In 2009, Hong and Kacperczyk found that “sin stocks” outperform market benchmark in the US. Sin stocks are stocks that promote vice, that is, alcohol, tobacco and gaming firms. They further argued that these stocks are neglected by investors because of social norms, and are undervalued. Yet, this research has been criticized as it compares sin stocks (which are not value-weighted) with a value- weighted benchmark. Since small capitalization (cap) stocks tend to outperform large cap stocks their findings might be biased. To cope with this, Lobe and Walkshäusl (2011) studied similar value-weighted sin stock and found that value- weighted portfolios do not significantly outperform their benchmarks. Still, there is a lack of applicability of earlier research since it relies on a different definition of sustainability. Research using ESG inclusion criteria is relatively new. An analysis concluded that 85% of the studies were focusing on one ESG dimension only (United Nations Environment Program Finance Initiative and Mercer Investment Consulting, 2007). Results have been mixed, but these studies are often criticized due to the interconnection of the three dimensions. Common findings of these studies are that companies with higher ESG scores are associated with less company-specific risk, lower cost of debt and higher credit ratings (Bauer et al., 2009; Bauer and Hann, 2011, Lee and Faff, 2009, cited in Hoepner, 2013). Examining several meta-studies and review papers, a general conclusion can be drawn that there is a positive correlation between sustainability and operational performance (Fulton et al. 2012, Hoepner and McMillian 2009, McWiliams et al. 2006, Salzmann 2005). Moreover, there seem to be an increase in the number of studies finding a positive link between ESG performance and financial performance. Eccles et al. (2014) found that “high” sustainability companies outperform “low” sustainability companies in the US in terms of stock market and operational performance. More specifically, they found that the annual abnormal performance is higher for the high sustainability group compared to the low sustainability group by 3.0% (significant at less than 5% level) on a value-weighted 9