Master thesis risk and return of esg rated portfolios

--- Page 1 --- Hochschule für Wirtschaft und Recht Berlin Berlin School of Economics and Law Department 1 – Business and Economics Master of Science FACT - Finance, Accounting, Controlling & Taxation - SS 2024 MASTER THESIS RISK AND RETURN OF ESG-RATED PORTFOLIOS --- Page 2 --- by Tuong Khoa Nguyen, Nguyen Student ID No.: 77211946833 First supervisor: Prof. Dr. Natalie Packham Second supervisor: Prof. Dr. Hans Lieck Date of submission: July 19, 2024 Word count: Abstract With an increasing understanding of the possible influence on risk and return, environmental, social, and governance (ESG) aspects have been important concerns in the choice of investments in recent years. The goal of this thesis is to find out how risk-adjusted return performances of European listed companies from 2019 to 2023 vary with different level of ESG ratings using the portfolio approach method, which divides companies into five groups based on their ESG ratings. Performance models of the Fama-French three-factor and CAPM models are employed to test the portfolios'''' performance, as well as risk-adjusted return performance metrics such as Jensen''''s alpha, Treynor ratio, and Sharpe ratio. The findings reveal that portfolio risk-adjusted return performances vary with different levels of ESG ratings. It implies that, although companies with higher ESG ratings show less volatility than those with lower ESG ratings, better risk-adjusted returns are not always the outcome of higher ESG ratings. --- Page 3 --- Keywords: risk-adjusted returns, ESG ratings, factor model, portfolios Sworn declaration I hereby formally declare that I have written the submitted Master Thesis entirely by myself without anyone else’s assistance. Whereever I have drawn on literature or other sources, either in direct quotes, or in paraphrasing such material, I have given the reference to the original author or authors and to the source where it appeared. I am aware that the use of quotations, or of close paraphrasing, from books, magazines, newspapers, the internet or other sources, which are not marked as such, will be considered as an attempt at deception, and that the thesis will be graded with a fail. I have informed the examiners and the board of examiners in the case that I have submitted the dissertation, entirely or partly, for other purposes of examination. Berlin, 19.07.2024 Place, date Signature List of abbreviations List of figures Figure 1 Cumulative returns chart 16 List of tables Table 1. ESG ratings by year 12 --- Page 4 --- Table 2. Average industry ESG ratings and industry weights 12 Table 3. ESG ratings divided into quintile by year 14 Table 4. Quantity of companies in portfolios 14 Table 5. Descriptive statistic of five portfolios 24 Table 6. CAPM regression result 26 Table 7. Fama-French Three-Factor regression result 27 Table 8. Dummy variables regression result with A is the baseline portfolio 30 Table 9. Dummy variables regression result with B is the baseline portfolio 30 Table 10. Dummy variables regression result with C is the baseline portfolio 31 Table 11. Dummy variables regression result with D is the baseline portfolio 32 Table 12. Dummy variables regression result with E is the baseline portfolio 33 Introduction Background Environmental, social, and governance (ESG) aspects nowadays have become an integral part of investment strategies around the world. ESG factors are used to assess the sustainability of investments or assets and a company''''s performance in terms of environmental, social, and governance aspects (Townsend, 2020). While traditional investors aim to maximize shareholder profits, more and more investors nowadays incorporate sustainability into their investment strategies. The ESG Global Study (2023) report showed that the percentage of investors who consider ESG factors in their investment decisions has increased from 84% in 2021 to 90% in 2023. In 2023, the perceived demand for sustainable funds compared to non-sustainable funds in Europe increased by 56% (Statista, 2024a). According to the Global Sustainable Investment Review (2022), global investment in sustainable assets will reach $30.3 trillion by 2022, marking a substantial increase from $22.8 trillion in 2016. The value of assets in funds incorporating ESG criteria rose significantly from $5 billion in 2006 to $391 billion in 2021. This number had risen to $480 billion by November 2023 (Statista, 2024b). Furthermore, according to Lefkovitz (2023), 78% of sustainability indexes outperformed traditional indexes between 2018 and 2022, indicating the well-being of sustainable investments. Besides, there has been an increase in awareness of sustainability as well as the demand for organizations'''' ESG reporting in recent years. In 2000, the United Nations (UN) launched the Millennium Development Goals with eight goals with the target of tackling issues like health issues, hunger, poverty, illiteracy, gender inequality, and environmental harm by the year 2015 (United Nations, n.d.-a). In 2015, the Sustainable Development Goals were established to replace the Millennium Development Goals, which focused and expanded more on the pillars of human rights and equity (United Nations, n.d.-b). In the subsequent decade, additional principles and frameworks were established, offering further guidance on how companies can incorporate and report on ESG factors. The European Union (EU) introduced the new reporting regulation of the Corporate Sustainability Reporting Directive (CSRD), which became effective in January 2023 and requires listed and large companies to reveal details about social and environmental concerns (DG FISMA, n.d.-a). As sustainability reporting has emerged as a relatively new legal obligation, the practice of using established ESG ratings to support investment decisions for firms, portfolios, and funds has evolved gradually over the years, which has made investors concerned not only about financial performance but also about sustainability. Research from Amel-Zadeh & Serafeim (2018) found that investors aim for profitability and reduce risk by incorporating ESG into their investment decisions. The result reveals that only a small number of investors, approximately 32%, consider ethical motives as the main driver for them to implement ESG-based investment strategies, while there are more than 63% of investors who aim for good returns. They also revealed that 82% of investors perceive ESG as a crucial factor and consider it when making investment decisions. This leads to the question of whether investors can achieve high returns with low risks while also supporting a sustainable future and whether ESG performance has an impact on company returns. For this reason, understanding how ESG is interrelated with risk and return is not only interesting but relevant to most investors. Understanding of how portfolios with different ESG ratings have performed could be decisive for investment allocation in general and sustainability investment in particular. Research question --- Page 5 --- The purpose of this research is to determine whether there are differences between companies that are ranked as top ESG performers compared to those that are ranked as bottom ESG performers in terms of risk-adjusted returns. The thesis''''s research question is as follows: Do risk-adjusted return performances of portfolios vary with different levels of ESG ratings in the European market between 2019 and 2023? With the main goal of examining the risk-adjusted performances of portfolios with different levels of ESG ratings over a mid-term period, this study targets on the European market with the constituents of the STOXX Europe 600 Index in five years from 2019 to 2023. This thesis uses the portfolio approach as the methodology that sorts companies into quintiles based on their ESG ratings from the previous year. The ESG ratings used in this research were retrieved from Refinitiv Eikon. The performance analysis of these portfolios is tested using various performance models, which are Fama-French three-factor (FF3) and the CAPM models. In addition, Jensen''''s alpha, Treynor ratio, and Sharpe ratio are also determined as the risk-adjusted return metrics. Dummy variable regressions are further tested to access the differences between portfolios, allowing for the comparison of one portfolio''''s incremental performance relative to others. After the introduction part, the literature review section will introduce relevant theories and literature, followed by a proposal of hypotheses. The data collection, sample, and methodology will be presented in detail, followed by the study''''s empirical results. The results are then examined in connection with previous literature and theories to provide potential conclusions. In the end, the thesis concludes with a summary of key findings, limitations, and future research directions. Literature Review The literature review will cover a general view of the historical context and theoretical basis for developing the hypotheses for this study. It will start by giving a summary of socially responsible investing (SRI) and ESG. The link between market efficiency and ESG will then be investigated. To illustrate what has already been found, it will also go through empirical evidence from earlier studies on the subject, including results from studies with positive, neutral, and negative effects on risk-adjusted performance. Finally, it will introduce the developed hypotheses and research contributions. SRI and ESG investing The practice of SRI in the past is usually known as "ethical investment," in which investors take into account both financial aspects and social aspects (positive social outcomes) in their investment decisions (Cowton & Sandberg, 2012). Other notions, such as value-based investing, alternative investing, sustainable investing, community investing, green investing, and other terms, can be used interchangeably to replace this term (Townsend, 2020; Cowton & Sandberg, 2012). The definition of SRI is also defined by the European Sustainable Investment Forum (Eurosif), which indicates that it is an investing strategy taking into account environmental, social, and governance aspects when making portfolio decisions with the goal of generating long-term returns for investors (Eurosif, n.d.). In the 1960s and 1970s, SRI gained momentum through the movements for anti-war, environmental, and civil rights (Townsend, 2020). There was a lack of integration of social criteria into investing during this time. Friedman''''s shareholder theory dominated the idea that a company is responsible for maximizing profits for shareholders while still complying with legal and ethical standards (Friedman, 1970). Later, in the 1980s and 1990s, corporate social responsibility (CSR) and Freeman''''s stakeholder theory became well-known, challenging this view with the opposite opinion. Freeman''''s stakeholder theory emphasized the importance of stakeholders to the organization''''s existence and that companies should prioritize stakeholder values (Freeman, 2010). According to Cowton & Sandberg (2012), what distinguishes SRI from mainstream investments such as stocks, bonds, bank deposits, or property investing is that it considers ethical merit. Investors implement SRI through strategies of positive screening and negative screening. Negative screening excludes investments in portfolios or companies that violate ethical standards, while positive screening involves investing in those that actively engage in ethical business practices. Companies perceived as having negative SRI issues are those who have business relating to alcoholic beverages, tobacco, gambling, weapons, etc., and those perceived as having positive SRI issues are education, healthcare, pollution control, etc. (Cowton & Sandberg, 2012). ESG criteria evolved as an extension of SRI, becoming essential to investment strategies globally, with a focus on sustainability analysis (Townsend, 2020). ESG investing first appeared in the mid-2000s in Europe, driven by a combination of legal, environmental, and governance considerations (Townsend, 2020). In 2005, the term ESG was first used and gained popularity through the Who Cares Wins (2004) report, which has rapidly grown in the last two decades. The report broke down the ESG concept into three components: environmental, social, and governance. In 2006, the UN Global Compact and the UN Environment Program Finance Initiative introduced the Principles for Responsible Investment (PRI) (PRI, n.d.). The PRI framework consists of six core principles, providing guidelines for investors to help them integrate ESG into their choice of investments. It highlights that risk management and long-term sustainability can be enhanced through ESG investing. The launch of PRI and the rising accessibility of ESG data, along with other global initiatives such as the Global Reporting Initiative in 1997 and the UN Global Compact in 2000, have provided standard guidelines for reporting ESG impacts, shown the growth of CSR, and emphasized the importance of ESG issues on investment performance (UN Global Compact, n.d.).  Nowadays, ESG analysis is one of the three core principles of modern SRI; with the other two principles are values-based avoidance screening and corporate engagement (Townsend, 2020). ESG analysis focuses on assessing environmental, social, and governance factors to gauge its long-term sustainability and potential for financial success. ESG factors are used to evaluate the non-financial impacts of specific investments and companies (Bergman, Deckelbaum, & Karp, 2020). ESG can also serve as a risk management tool; changes in a company’s ESG characteristics may act as valuable financial indicators (Giese, Lee, Melas, Nagy, & Nishikawa, 2019). According to the Global Sustainable Investment Review (2022), the most popular sustainable investment strategy globally is corporate engagement and shareholder action, followed by ESG integration and negative screening as the second and third investment strategies, respectively.

Introduction

Background

Environmental, social, and governance (ESG) aspects nowadays have become an integral part of investment strategies around the world ESG factors are used to assess the sustainability of investments or assets and a company's performance in terms of environmental, social, and governance aspects (Townsend, 2020) While traditional investors aim to maximize shareholder profits, more and more investors nowadays incorporate sustainability into their investment strategies The ESG Global Study (2023) report showed that the percentage of investors who consider ESG factors in their investment decisions has increased from 84% in 2021 to 90% in

2023 In 2023, the perceived demand for sustainable funds compared to non-sustainable funds in Europe increased by 56% (Statista, 2024a) According to the Global Sustainable Investment

Review (2022), global investment in sustainable assets will reach $30.3 trillion by 2022, marking a substantial increase from $22.8 trillion in 2016 The value of assets in funds incorporating ESG criteria rose significantly from $5 billion in 2006 to $391 billion in 2021 This number had risen to $480 billion by November 2023 (Statista, 2024b) Furthermore, according to Lefkovitz (2023),78% of sustainability indexes outperformed traditional indexes between 2018 and 2022,indicating the well-being of sustainable investments Besides, there has been an increase in awareness of sustainability as well as the demand for organizations' ESG reporting in recent years In 2000, the United Nations (UN) launched the Millennium Development Goals with eight goals with the target of tackling issues like health issues, hunger, poverty, illiteracy, gender inequality, and environmental harm by the year 2015 (United Nations, n.d.-a) In 2015, theSustainable Development Goals were established to replace the Millennium Development Goals,which focused and expanded more on the pillars of human rights and equity (United Nations,n.d.-b) In the subsequent decade, additional principles and frameworks were established,offering further guidance on how companies can incorporate and report on ESG factors TheEuropean Union (EU) introduced the new reporting regulation of the Corporate SustainabilityReporting Directive (CSRD), which became effective in January 2023 and requires listed and large companies to reveal details about social and environmental concerns (DG FISMA, n.d.-a).

As sustainability reporting has emerged as a relatively new legal obligation, the practice of using established ESG ratings to support investment decisions for firms, portfolios, and funds has evolved gradually over the years, which has made investors concerned not only about financial performance but also about sustainability Research from Amel-Zadeh & Serafeim (2018) found that investors aim for profitability and reduce risk by incorporating ESG into their investment decisions The result reveals that only a small number of investors, approximately 32%, consider ethical motives as the main driver for them to implement ESG-based investment strategies, while there are more than 63% of investors who aim for good returns They also revealed that 82% of investors perceive ESG as a crucial factor and consider it when making investment decisions

This leads to the question of whether investors can achieve high returns with low risks while also supporting a sustainable future and whether ESG performance has an impact on company returns For this reason, understanding how ESG is interrelated with risk and return is not only interesting but relevant to most investors Understanding of how portfolios with different ESG ratings have performed could be decisive for investment allocation in general and sustainability investment in particular.

Research question

The purpose of this research is to determine whether there are differences between companies that are ranked as top ESG performers compared to those that are ranked as bottom ESG performers in terms of risk-adjusted returns The thesis's research question is as follows:

Do risk-adjusted return performances of portfolios vary with different levels of ESG ratings in the European market between 2019 and 2023?

This study examines the risk-adjusted performance of portfolios with varying ESG ratings using the STOXX Europe 600 constituents over five years (2019-2023) Companies were categorized into quintiles based on their ESG ratings from the previous year, using Refinitiv Eikon data Performance was analyzed using Fama-French three-factor (FF3) and CAPM models, and risk-adjusted return metrics (Jensen's alpha, Treynor ratio, and Sharpe ratio) were calculated Dummy variable regressions were employed to evaluate the performance differences between portfolios and compare the incremental performance of one portfolio relative to others.

After the introduction part, the literature review section will introduce relevant theories and literature, followed by a proposal of hypotheses The data collection, sample, and methodology will be presented in detail, followed by the study's empirical results The results are then examined in connection with previous literature and theories to provide potential conclusions In the end, the thesis concludes with a summary of key findings, limitations, and future research directions.

Literature Review

SRI and ESG investing

Socially responsible investing (SRI) has evolved from "ethical investment," considering both financial and social factors, with interchangeable terms like value-based and sustainable investing As defined by Eurosif, SRI incorporates environmental, social, and governance (ESG) criteria in portfolio decisions for long-term returns The SRI movement emerged in the 1960s-1970s with social and environmental activism Friedman's shareholder theory once dominated the view that companies should prioritize shareholder profits, but CSR and Freeman's stakeholder theory later challenged this, emphasizing the importance of stakeholders in organizational success.

SRI (Socially Responsible Investment) differentiation from mainstream investments (stocks, bonds, bank deposits, property) lies in its consideration of ethical criteria SRI employs positive and negative screening strategies Positive screening invests in companies adhering to ethical practices, while negative screening excludes investments associated with unethical practices (e.g., alcohol, tobacco, gambling, weapons) Positive SRI issues include education, healthcare, and pollution control.

ESG criteria evolved as an extension of SRI, becoming essential to investment strategies globally, with a focus on sustainability analysis (Townsend, 2020) ESG investing first appeared in the mid-2000s in Europe, driven by a combination of legal, environmental, and governance considerations (Townsend, 2020) In 2005, the term ESG was first used and gained popularity through the Who Cares Wins (2004) report, which has rapidly grown in the last two decades The report broke down the ESG concept into three components: environmental, social, and governance In 2006, the UN Global Compact and the UN Environment Program FinanceInitiative introduced the Principles for Responsible Investment (PRI) (PRI, n.d.) The PRI framework consists of six core principles, providing guidelines for investors to help them integrate ESG into their choice of investments It highlights that risk management and long-term sustainability can be enhanced through ESG investing The launch of PRI and the rising accessibility of ESG data, along with other global initiatives such as the Global Reporting Initiative in 1997 and the UN Global Compact in 2000, have provided standard guidelines for reporting ESG impacts, shown the growth of CSR, and emphasized the importance of ESG issues on investment performance (UN Global Compact, n.d.)

Nowadays, ESG analysis is one of the three core principles of modern SRI; with the other two principles are values-based avoidance screening and corporate engagement (Townsend, 2020).ESG analysis focuses on assessing environmental, social, and governance factors to gauge its long-term sustainability and potential for financial success ESG factors are used to evaluate the non-financial impacts of specific investments and companies (Bergman, Deckelbaum, & Karp,2020) ESG can also serve as a risk management tool; changes in a company’s ESG characteristics may act as valuable financial indicators (Giese, Lee, Melas, Nagy, & Nishikawa,2019) According to the Global Sustainable Investment Review (2022), the most popular sustainable investment strategy globally is corporate engagement and shareholder action,followed by ESG integration and negative screening as the second and third investment strategies, respectively.

ESG and Market efficiency

The influence of ESG performance on a firm's performance is debatable in terms of expected returns and level of risk There are three types of risk exposure to a company: systematic risk,idiosyncratic risk, also called unsystematic risk, and total risk (Bouslah, Kryzanowski, & M’Zali,2018; Chollet & Sandwidi, 2018) Total risk comprises two components: systematic and idiosyncratic (unsystematic) risk (Sharpe, 1964) According to Markowitz's Modern PortfolioTheory (MPT), investors allocate investments that maximize the expected return with significantly reduced risk so that they can create an optimal portfolio and achieve optimal risk- adjusted returns (Markowitz, 1952) The theory suggests that the idiosyncratic risk associated with individual companies can be reduced by a well-diversified portfolio On the other hand, theEfficient Market Hypothesis (EMH) relates to the estimation of expected returns EMH, in its semi-strong or strong form, asserts that stock prices represent all information that is available to the public, suggesting that no investor can win the market by achieving returns that exceed the average (Lo, 2007) EMH and MPT are aligned on the assumption that the expected return of a portfolio depends on the degree of risk that investors accept to take in the context of an efficient market.

According to the implications of diversification from MPT, investors could hinder such optimization by imposing constraints, as a fully diversified portfolio cannot be obtained Le Sourd (2024) argued that the inclusion of ESG factors in portfolio allocation results in increased risk or reduced returns, as well as the appearance of opportunity cost Revelli & Viviani (2015) also claimed that the opportunity cost, or "diversification cost," associated with SRI prevents it from being an optimal investment strategy This is due to the fact that the exclusion of unethical investments reduces diversification and, as a result, risk-adjusted returns Chang & Witte (2010) also observed that ESG investing generally results in lower average returns and Sharpe ratios than unscreened investing, which supports the theoretical prediction of lower performance because of these constraints.

ESG information integrated into stock prices reflects the alignment with the EMH, indicating that ESG factors are fully incorporated into market valuations Research has consistently shown the positive impact of ESG practices on financial performance Companies with strong ESG practices have enhanced risk-adjusted returns, reduced risk, and improved financial performance (Ashwin Kumar et al., 2016; Giese et al., 2019) Moreover, higher ESG ratings often correlate with lower discount rates (Cornell & Damodaran, 2020), further supporting the integration of ESG factors into market valuations.

In the reality, markets are rarely perfectly efficient, despite the fact that EMH is the foundational financial theory that asserts market efficiency (Fama, 1991) Mixed empirical evidence shows that markets are often efficient, but not always due to market anomalies and behavioral factors(Naseer & Tariq, 2015) Cao, Titman, Zhan, & Zhang (2019) found that ESG factors can lead to abnormal returns and mispricing, suggesting the need for the markets to integrate ESG information effectively This study demonstrates that investor preferences for ESG can create opportunities for abnormal returns In contrast, research by Bofinger, Heyden, & Rock (2022) illustrated the potential for market inefficiencies by overvaluing stocks with high ESG ratings.They observe that improved ESG ratings lead to an increase in firm values relative to their actual value It suggests that higher ESG ratings lead to an increase in overvaluation and a decrease in undervaluation The cause of this mispricing is positive market sentiment towards ESG criteria,reflected in growing demand for sustainable investments Investors can therefore gain exceptional returns using ESG information, which goes against EMH, which presumes that all relevant information is already represented in stock prices.

Previous studies

Several studies have investigated the link between the ESG factor and portfolio performance. However, the results revealed mixed outcomes Some findings suggest a positive relationship, while others suggest a negative or neutral relationship Early research from Kempf & Osthoff (2007) and Statman & Glushkov (2009) proposed that portfolios with high ESG ratings tend to outperform those with low ratings Ashwin Kumar et al (2016) found that for most sectors, companies that incorporate the ESG factor have lower risk and higher returns than equivalent stocks Lee, Fan, & Wong (2021) and Teti, Dallocchio, & L'Erario (2023) are recent studies that also found investing in ESG can positively affect portfolio performance According to Teti et al. (2023), ESG may help manage risks by reducing portfolio volatility and downside risks In the opposite view, Luo (2022) conducted an analysis of the influence of ESG scores on UK stock returns, finding that companies with lower ESG scores generate higher returns, especially for less liquid securities Herzel, Nicolosi, & Starica (2011) also looked at the relationship when they used S&P 500 stocks from 1993 to 2008 to make optimal portfolios with limits They did this by using the mean-variance method of MPT and leaving out assets that did not meet social responsibility standards Research shows that sustainability screening has a slight negative affect on the Sharpe ratio, but it has a significant positive affect on market capitalization These studies present a negative correlation between high ESG ratings and superior portfolio returns Lee, Faff,

& Rekker (2013), on the other hand, did not observe any substantial variances in performance adjusted for risk between low and high corporate social performance (CSP)-constructed portfolios Furthermore, the differences between low and high CSP portfolios in terms of book- to-market ratio, performance, momentum characteristics, and size were not uniform The performance, momentum characteristics, and size of both high and low CSP portfolios were consistently the same, regardless of the method used to construct the portfolio.

The connection between ESG ratings and return performance is more complex, influenced by factors such as ESG criteria and the specific time frame Zehir & Aybars (2020) found that ESG- based portfolios on average did not outperform or underperform the market However, the lowest-performing ESG and governance portfolios actually outperformed the market Further investigations have looked into how ESG ratings impact industries and events like COVID 19. Díaz, Ibrushi, & Zhao (2021) discovered that, during the COVID-19 period, the social and environmental aspects of ESG were the drivers that affect sectors This emphasizes the importance of considering ESG factors in specific industries Prol & Kim's (2022) study revealed that portfolios optimized for high ESG result in lower returns and volatility than those optimized for low ESG They suggest that portfolios optimized for low ESG have a higher Sharpe ratio because there is not enough compensation from lower volatility for the lower return.

Historically, studies demonstrated a negative correlation between portfolio returns and ESG considerations However, recent research post-2018 indicates a shift towards positive or neutral relationships Wang, Guise, and Nagy's (2023) analysis from 2017-2021 supports this trend, showcasing the improved performance of ESG-integrated investments This evolution is attributed to investors' increasing awareness and acceptance of the role ESG plays in investment decisions.

The relationship between environmental, social, and governance (ESG) factors and risk-adjusted performance is a complex topic that has been the subject of numerous studies However, the findings of these studies have been mixed, with some showing a negative association between ESG and performance, while others have found either a positive or no significant relationship Further research is needed to better understand this relationship and determine whether there are any specific ESG factors that have a consistent impact on performance.

Hypotheses and Contribution

Different findings from the previously carried out literature review are discussed for both low and high ESG portfolios More investigation of this relationship with these theories is attempted in this thesis The hypotheses are derived from current research patterns that indicate higher ESG portfolios are outperforming lower ESG portfolios (Kumar, 2023).

To answer the research question, the hypotheses are formulated as follows:

H1: Portfolios of firms with higher ESG ratings demonstrate lower volatility than portfolios of firms with lower ESG ratings.

H2: Portfolios of firms with higher ESG ratings generate better risk-adjusted returns than portfolios of firms with lower ESG ratings

This research examines the risk-adjusted return performance of ESG-rated stock portfolios inEurope from 2019 to 2023 This thesis adds to the literature by taking a mid-term approach by examining a latest five-year period to analyze the link between ESG factors and financial performance, in contrast to other studies that focus on longer timeframes The chosen period includes significant regulatory advancements in ESG reporting and compliance, particularly the implementation of the EU Taxonomy in 2020 as part of the European Green Deal initiatives (DGFISMA, n.d.-b) Europe's advanced and consistent sustainability regulatory framework and best practices make it an ideal setting for this research Consequently, any impact of ESG factors on financial performance is likely to be more apparent in this context.

Data collection

ESG ratings data

The ESG ratings are an important tool for this research analysis, as they are necessary for constructing and analyzing the performance of portfolios with higher and lower ESG ratings. Thus, it is crucial to understand the data provider and methodology underlying the ratings.

Investors now utilize ESG scores or ratings as reliable instruments to obtain independent evaluations of the ESG performance of companies and to include ESG considerations in investment choices (Townsend, 2020) Most SRI research applies ESG measurements as a factor determining sustainability performance (Widyawati, 2020) There are many ESG rating companies that evaluate the effects and risks of investments; each one uses a different approach. Among the most well-liked suppliers are Sustainalytics, MSCI, Refinitiv, and S&P Global (Tayan, 2022) According to Tayan (2022), ESG ratings are intended to measure "ESG quality" as a reflection of a company's impact on stakeholders or as a measure of the impact of societal, environmental, and governance factors on a company's financial performance Different stakeholders have diverse interests and needs regarding ESG ratings, ranging from ethical concerns to financial risk assessment to reputational management.

The ESG ratings used in this thesis are retrieved from Refinitiv Eikon, previously known as Thomson Reuters, which was acquired by the London Stock Exchange Group (LSEG) in 2021. Being a top worldwide supplier of infrastructure and financial market data, it provides a large ESG database together with effective, data-based ESG ratings (Environmental, social and governance scores from LSEG, 2023) Of more than 630 ESG measures used by the evaluation system, 186 important ones were chosen for scoring because of their materiality and comparability Ten themes are formed out of these measurements, and these topics are further divided into three pillars: governance, social, and environmental LSEG employs a percentile rank scoring methodology to assess companies' ESG performance on a scale from 0 to 100.Ratings below 50 are considered relatively poor, while ratings above 50 are regarded as good.These ratings are then converted into letter grades ranging from D- to A+, where D indicatesESG laggards and A+ indicates ESG leaders This grading system offers a clear and interpretable assessment of a company's ESG performance in relation to its industry peers (Environmental, social and governance scores from LSEG, 2023).

Refinitiv updates ESG data weekly, aligning with corporate reporting patterns, fiscal year updates, and controversy events (Environmental, social and governance scores from LSEG, 2023) Despite the frequent updates, ESG ratings are issued annually and remain consistent throughout the year, often reflecting the company's performance from the previous year. Consequently, portfolios are revised annually using the ratings from the previous year (t-1) for the current year (t) For example, the ESG rating of 2018 will be used for portfolio construction in 2019 Group portfolios remain unchanged until new ratings are published.

Based on the 2022 study conducted by Capital Group, the majority of investors identified the lack of consistency among ESG rating providers as the primary challenge when integrating ESG data (ESG Global Study, 2022) Since each ESG rating provider employs unique methodologies,there is a lack of consistency among the ratings, resulting in significantly varied outcomes when these ratings are used to construct a portfolio (Li, 2020).

Financial data

In addition to ESG ratings, stock returns for companies in the STOXX Europe 600 index are obtained from Refinitiv Eikon Refinitiv Eikon provides a total return index for individual equities, incorporating reinvested dividends This study uses total return index as the total returns, assuming additional units are bought at the closing price excluding dividends They are collected on the last trading day of each month on a monthly basis, covering the period from

2019 to 2023 Furthermore, monthly market capitalization data for all companies is also collected from Refinitiv Eikon This data is used to create value-weighted portfolios based on each company's market size, with weights calculated using the firm's market capitalization for the respective month All data is collected in the Euro currency to make it comparable among different countries without changing the currency Given that the scope of the research is theEuropean market, the risk-free rate is determined by the one-month Euro Interbank Offered Rate(Euribor) The rate has been extracted from Refinitiv Eikon on the same dates as the returns.

Data sample

The STOXX Europe 600 index consists of 600 companies, but not all of them had complete ESG ratings for the period under examination As a result, companies without available ESG ratings were excluded from the portfolios during those periods Consequently, 66 companies lacked ESG ratings data and were eliminated from the sample.

534 companies The high level of aggregation in the portfolios and the small number of missing data points mean that these omissions do not pose a significant issue for the study As previously mentioned, ESG ratings were obtained from 2018 to 2022 for portfolio construction from 2019 to 2023, while other data, including total return index, market value, and the risk-free rate, were obtained from January 31, 2019, to December 31, 2023.

Table 1 ESG ratings by year

Year Obs Mean Median Min Max

Table 1 describes the sample size as well as mean, median, minimum and maximum ESG ratings of STOXX Europe 600 constituents (2018-2022)

Table 2 also details the number of industries included in the sample, following Díaz et al (2021). According to LSEG, the industries are categorized by Refinitiv Business Classifications, resulting in a total of 11 industry sectors listed in Table 2, along with their distribution (LSEG, n.d.) The table reveals that the industrial sector contains the highest number of companies, followed by the financial and consumer discretionary sectors The energy sector has the highest average ESG ratings, while the real estate and technology sectors have the lowest ESG ratings.

Table 2 Average industry ESG ratings and industry weights

Table 2 shows the average industry ESG ratings and industry weights for STOXX Europe 600 constituents

Methodology

Portfolio construction

A portfolio approach is used to examine the connection between ESG ratings and risk and return. Companies are assigned to five different portfolios, labeled A, B, C, D, and E, according to their ESG ratings This outcomes in a total of 25 sub-portfolios being evaluated Portfolio A includes companies with the lowest ESG ratings, while Portfolio E includes those with the highest ratings.

As ESG ratings change annually, the companies in each portfolio are reallocated each year This sorting allows for the assessment of whether higher-rated companies perform significantly differently from lower-rated companies The companies are ranked and divided into quintile portfolios, each representing 20% of the data, as follows:

Top Quintile (Quintile A): the top 20% of companies with the highest ESG ratings

Bottom Quintile (Quintile E): the bottom 20% of companies with the lowest ESG ratings Middle Quintiles (Quintiles B, C, and D): the companies with intermediate ESG ratings

Table 3 ESG ratings divided into quintile by year

Table 3 illustrates the ESG rating ranked and divided into quintiles by year (2019-2023)

The rules stated above indicate that companies will be ranked relative to other companies The quintile rules have been chosen to ensure that the number of companies in each portfolio remains approximately the same each year This also prevents any portfolio from benefiting from an added diversification effect.

Table 4 Quantity of companies in portfolios

Table 4 illustrates the number of companies assigned to each portfolio, based on ESG ratings (2019-2023)

After assigning firms to their respective portfolios for the analysis period, weights should be allocated to calculate the portfolio return and risk Following Díaz et al (2021) and Teti et al.(2022), all portfolios are constructed as value-weighted portfolios in this thesis Value-weighted portfolios are selected because of the significant role that the value-weighted market portfolio plays in asset pricing, as demonstrated by the CAPM proposed by Sharpe (1964) Value- weighted returns are computed by dividing each stock's market value by the total of the portfolio’s market value.

Risk-adjusted return measurements

Common measurement metrics for comparing risk-adjusted returns include the Treynor ratio, Jensen's alpha, and Sharpe ratio Besides them, alpha from factor models, which account for risk explained by the included factors, can serve as a metric for risk-adjusted returns (Elton, Gruber,

A risk-adjusted return is a method of evaluating an investment's return by accounting for the associated risk and seeking to minimize risk while achieving the same or greater returns (AshwinKumar et al., 2016) The firm's stock volatility serves as an indicator of total risk, which is quantified by calculating the annualized standard deviation of stock returns over a 12-month period (Bouslah et al., 2018; Chollet & Sandwidi, 2018) On the other hand, systematic risk is determined by a company's level of sensitivity to changes in market returns This refers to the risk that is caused by a stock's reaction to general market movements that affect all securities(Sharpe, 1964) Systematic risks are measured by a firm’s beta based on the standard CAPM model (Sassen, Hinze, & Hardeck, 2016) Idiosyncratic risk arises from firm-specific factors and represents the residual risk that cannot be determined by fluctuations in the average market portfolio returns MPT suggests that through diversification in a well-constructed portfolio, idiosyncratic risk can be eliminated, and investors are solely compensated for bearing systematic risk (Markowitz, 1952) This implies that only systematic risk is relevant for asset pricing in perfect markets (Amit & Wernerfelt, 1990).

The Sharpe Ratio, originally known as the reward-to-variability ratio, was introduced by Sharpe in 1964 and is a widely used tool in investment analysis According to Sharpe (1964), it is a method that evaluates the portfolio’s risk-adjusted return by considering its excess return above the risk-free rate and its volatility In accordance with Díaz et al (2021), the Sharpe Ratio will be used in this investigation with the purpose of evaluating the risk-adjusted return of various portfolios.

Where R f denotes the risk-free rate, R p denotes the return of portfolio, and σ p denotes the standard deviation of portfolio.

The Sharpe ratio measures how well the return on an investment portfolio compensates for the risk assumed A greater Sharpe ratio suggests that an investment has obtained a stronger risk- adjusted return.; therefore, it is a frequently employed metric by investors when comparing investment (Schmid & Schmidt, 2010).

The Sharpe ratio in this study is computed following portfolio returns and volatility annualization Annualized volatility, which refers to the measure of the investment’s risk over a one-year period, is calculated as the square root of 12 monthly periods in a year The annualization is computed using this formula following Zehir & Aybars (2020):

Where R a denotes the annualized return of portfolio, R p denotes the return of portfolio, σ a denotes the annualized volatility and σ p denotes the monthly standard deviation of portfolio.

The Treynor ratio, originally known as the reward-to-volatility ratio and developed by Treynor in

1965, is another measure of risk-adjusted return similar to the Sharpe ratio (Treynor, 1965) The The Treynor ratio is computed by dividing the portfolio's excess return over the risk-free rate by its beta, which measures its exposure to systematic risk Beta indicates how much the volatility of a stock or fund correlates with the overall market What distinguishes the difference between the the Treynor ratio and Sharpe ratio is the measure of risk they use The Sharpe ratio takes into account the total risk, whereas the Treynor ratio only examines the systematic risk.

Where R f denotes the risk-free rate, R p denotes the portfolio’s return and β p denotes the beta of portfolio.

Beta is calculated by dividing the covariance of the market's return with the portfolio's return by the market return's variance. β p =C ov(R p , R m ) σ m 2

Where C ov ( R p , R m ) denotes the covariance of the market's return with portfolio's return and σ m 2 denotes the market return's variance.

Jensen’s alpha (hereafter alpha), first presented by Jensen in 1968, is a risk-adjusted performance measure where alpha evaluates whether the achieved portfolio return exceeds the expected return predicted by a performance benchmark model, thereby indicating if an investor has been properly compensated for the risk undertaken (Jensen, 1968).

Alpha represents the excess return attributable to the investor's asset composition, not market movements A positive alpha means that the portfolio achieved better performance than the market, generating higher returns than predicted by CAPM Conversely, a negative alpha signifies that the portfolio did not achieve the required return If the stocks in the portfolio are fairly priced, the portfolio return will match the value predicted by CAPM, resulting in an alpha of zero, suggesting market efficiency In an efficient market where prices entirely incorporate all information, achieving abnormal returns through stock choosing would be unlikely Instead, a diversified portfolio and passive investing strategy would be preferable (Fama & French, 1996). However, the perfect market efficiency is rarely observed in the reality. α=R p −[ R f + β p ( R m −R f ) ]

Performance benchmark models differentiate between two elements of total risk: systematic risk,which is rewarded with higher expected returns, and non-systematic risk, which is not compensated as it is diversifiable Since actual returns often deviate from expected returns and benchmarks are imperfect, alpha becomes relevant Essentially, alpha is the intercept in the model, measuring the additional return beyond what is expected from an asset pricing model likeCAPM Alpha can also be applied to multi-factor models, adjusting the equation accordingly, to test whether an investor has managed to beat the market by generating a positive alpha.

Regression analysis

A market approach to assess return performance using asset pricing models is used in this study.Through time-series regression, alpha is used to access and measure the performance of the constructed portfolios with Fama-French three-factor and CAPM models Specifically, the study assesses whether alpha significantly differs between high and low ESG-rating portfolios.

A linear link between the movements of the market and the returns of an asset or a portfolio is demonstrated by the CAPM, which was established by Mossin (1966), Lintner (1965), and Sharpe (1964) CAPM illustrates that the expected return on a portfolio or asset is determined solely by its systematic, or non-diversifiable, risk This risk is measured by the beta factor, which is dividing the covariance of the market's return with the portfolio's return by the market return's variance Beta is a metric that quantifies the extent to which an asset is volatile in relation to the market It provides an indication of the amount of risk that the asset contributes to a diversified portfolio When an asset has a beta that is more than one, it indicates that it is more volatile than the market As a result, it requires a higher return in order to compensate for the higher risk. High-beta assets correlate more with the market, reducing diversification benefits and adding more risk.

CAPM regression model is tested in this thesis as follow:

R pt −R ft =α+β p ( R mt −R ft ) + ϵ pt

The excess return of a portfolio (R pt −R ft ) can be explained by its abnormal return (α), its sensitivity to market risk premium (β p ), and market excess return (R mt −¿ R ft ) The abnormal return captures returns not predicted by the model, while beta represents the portfolio's volatility relative to the market Market excess return quantifies the additional return from investing in the market over the risk-free rate The error term (ϵ pt ) reflects unexplained variations at a given time.

The Capital Asset Pricing Model (CAPM) employs the market portfolio, encompassing all risky assets equally, to represent the entire investable landscape By utilizing beta, the risk-free rate, and the anticipated market return, CAPM calculates an asset's expected return However, it assumes rational investor decision-making, negligible transaction costs, uniform knowledge and expectations, and public trading of all assets Rational investors, as Szylar (2013) asserts, seek optimally diversified portfolios and expect compensation only for risk that cannot be diversified.

4.3.2 Fama-French Three-Factor Model

Factor models provide a framework for understanding risk and return in financial markets, identifying the factors that drive asset prices and enabling investors to assess risks and opportunities associated with different investments The single-factor CAPM model, which attributes return solely to the market portfolio, has proven insufficient for explaining asset returns Fama and French (1993) introduced a multi-factor model that considers additional factors to better explain asset returns.

Fama and French developed the three-factor model, extending CAPM by supplementing two extra factors: size and value Their research identified these characteristics, alongside the market factor, as better explanations for return variability The size premium (SMB: Small-Minus-Big) captures the tendency of smaller stocks to outperform larger ones Size, in this context, refers to market capitalization and is calculated as the return difference between small and large stock portfolios The SMB factor measures the extra return investors can expect from holding a portfolio of smaller stocks compared to a portfolio of larger stocks The value premium (HML: High-Minus-Low) captures the value effect, where value stocks (higher book-to-market ratios) outperform growth stocks (lower book-to-market ratios) This factor is determined by subtracting the returns of portfolios of high- and low-book-to-market ratio companies The HML factor measures the additional return investors can expect from holding a portfolio of value stocks compared to a portfolio of growth stocks.

Fama-French three-factor regression model is tested in this as follow:

R pt −R ft =α+β pM ( R mt −R ft ) + β pSMB SMB t +β pHML HML t +ϵ pt

Where R pt −R ft denotes excess return of portfolio p for time (month) t, α denotes abnormal return unexplained by the model, β pM denotes the beta of portfolio p, representing portfolio's sensitivity to the market risk premium, (R mt −¿ R ft ) denotes is the excess return of value- weighted market portfolio for period (month) t, representing the additional return expected from investing in the market above the risk-free rate, SMB and HML represent the size premium and value premium at time (month) t, β pSMB and β pHML denote the portfolio p ' s sensitivity to the size factor (SMB) and value factor (HML) and ϵ pt denotes the error term at time (month) t.

Fama & French (1993)’s methodology was used to construct the SMB and HML factor All stocks are ranked by market capitalization The median of market capitalization is used as a cutoff point to divide the stock sample into two groups: small (S) and big (B) The stocks are continuously ranked by book-to-market ratio within each group The book-to-market ratio is obtained from the ratio 1/price-to-book ratio, where the price-to-book data is downloaded from Refinitiv Eikon The 30 th and 70 th percentiles are used as a cutoff point to divide the stocks into three groups: low (L), medium (M), and high (H) Six value-weighted portfolios are then created at the intersection of the book-to-market ratio group and the size group, which are big-high (BH), big-medium (BM), big-low (BL), small-big (SH), small-medium (SM), and small-low (SL). HML (High Minus Low) factors are calculated monthly by calculating the difference between the average monthly return of portfolios with high and low book-to-market ratios Similarly, SMB (Small Minus Big) factors are calculated monthly by subtracting the average monthly return of large portfolios from the average monthly return of small portfolios.

Multi-factor models, like CAPM, use beta to quantify risk However, time-series regression is used to validate these models and incorporate other elements in order to better understand asset price behavior (Fama & French 2015) While CAPM relies just on beta to assess risk, Fama &French (1993) propose additional criteria for more exact return projections Fama and French(1996) also highlight that the small enterprises and equities with a higher ratio of book-to- market, for example, tend to outperform due to market inefficiencies Research by Lee et al.(2021) demonstrates that larger enterprises are often found in high ESG portfolios, while smaller firms are less prevalent Tang (2023) argues that the Fama-French three-factor model is more effective in illustrating the variation in stock returns compared to the CAPM model.

Validity of the analysis

To determine the relationship between independent variables (factors) and the portfolio's return (dependent variable), ordinary least squares (OLS) regression is employed Microsoft Excel and STATA are utilized for the calculations and regressions, which are based on data collection and regression model evaluation This process aims to estimate the parameters that uncover the link between the factors and the portfolio's return over a specified time frame (e.g., 60 months).

2019 to 2023), time-series regression is used The equations stated in 4.3 are applied, respectively However, time-series data often exhibit common issues such as stationarity, autocorrelation, and heteroskedasticity The regressions are then tested with those issues to ensure the validity of the result.

To check the data for stationary, an augmented Dickey-Fuller (ADF) test was run and performed (Said & Dickey, 1984) A time series is considered stationary if its statistical properties remain constant over time Non-stationary series can produce misleading regression results The result rejected the null hypothesis that series have a unit root and demonstrated that all data were stationary.

To check the data for heteroskedasticity and autocorrelation, the White (1980) test and the Breusch-Godfrey test (Breusch, 1978; Godfrey, 1978), were run and performed, respectively. Time-series data frequently shows autocorrelation, where residuals are correlated with their past values, which violates the OLS assumption of independent errors Additionally, time-series data may display heteroskedasticity, meaning the variance of the errors changes over time.

The analysis revealed significant heteroskedasticity and serial correlation issues within the data Portfolios C and D exhibited both issues for both CAPM and FF3 regressions Portfolio E presented heteroskedasticity for CAPM regression and autocorrelation for FF3 regression Notably, Portfolio B solely exhibited autocorrelation for both regression models, whereas Portfolio A showcased neither heteroskedasticity nor autocorrelation To resolve these concerns, robust standard errors and Newey & West corrections were applied to rectify the errors.

(1987) standard errors, respectively After correcting these tests, the data can be used as a valid consideration All the results are presented in Appendix.

Analysis and Results

Descriptive Analysis

The figure of cumulative returns is provided for better understanding the fluctuations in the market following the figure below:

Figure 1 compares the performance of five different portfolios (A, B, C, D, E) against the STOXX 600 index from 2019 to 2023.

It is notable that all portfolios show an upward trend over the period, indicating a positive growth Portfolio A consistently achieves the highest cumulative returns, making it the best performer overall Portfolios B, C, and D also perform well, with C and D showing more stability compared to B Portfolio E, while having the lowest cumulative returns among the portfolios, offers a steadier growth pattern All portfolios significantly outperform the STOXX Europe 600 index, highlighting the effectiveness of these investment strategies compared to the broader market index This figure also reflects that portfolio A generates a higher return but is more volatile than portfolio E, which generates a lower return but is more stable.

More details on the portfolio performance metrics of five portfolios over the five-year period, including their returns and volatility, cumulative and annualized performance, risk-adjusted performance, distribution, and market characteristics, are presented in the table below:

Table 5 Descriptive statistic of five portfolios

Table 5 presents the descriptive statistic of five portfolios over the five-year period ((2019-2023)

From the table, it can be seen that portfolio A leads with the highest mean return at 1.70%, followed closely by portfolios D at 1.64% and B at 1.63% Portfolio E has the lowest mean return, at 1.14%, indicating generally lower performance compared to the others Portfolio E has the least severe minimum return, with the lowest minimum return at -10.55% and the highest maximum return at 16.87%, suggesting a relatively lower risk of experiencing extreme losses and the best potential for high gains Portfolio C has the lowest minimum return, at -14.29%, indicating the largest loss in the period Portfolio C also has the highest beta at 1.04, suggesting it is more sensitive to market changes compared to the other portfolios.

Portfolio D stands out with the best risk-adjusted performance, as reflected by the highest Sharpe ratio of 1.04 and Jensen's alpha of 1.70% This means that, considering its beta, portfolio D outperforms its expected return and offers the best returns relative to its risk level Portfolio A has the highest volatility of 17.24%, even though it has the highest cumulative return of 155.31% and the highest annualized return of 20.62% At 0.17, the Treynor ratio of portfolios A and D indicates that they yield the highest return in relation to the systematic risk.

In contrast, Portfolio E offers the poorest returns, with a 13.41% annualized return and a total return of 87.59% With a Treynor ratio of 0.11 and a beta of 0.87, Portfolio E offers the lowest return with respect to systematic risk and is less exposed to market volatility Nevertheless, it has the lowest annualized volatility of 14.85 and the lowest standard deviation of 4.32% Despite its low volatility, portfolio E has an annualized return that is insufficient to compensate for the volatility As a result, it has the lowest Sharpe ratio among all portfolios at 0.63.

Portfolio E holds the highest market capitalization ($84,241 million) and book-to-market ratio (0.83), indicating a portfolio of larger, value-oriented companies Conversely, Portfolio A possesses the lowest market cap ($7,596 million) and book-to-market ratio (0.58), suggesting a portfolio composed of smaller, growth-oriented companies.

Distribution characteristics show varied skewness and kurtosis across the portfolios All portfolios have negative skewness except for portfolio E, which has a positive skewness of 0.14, indicating it has a tail on the right side of the distribution This also suggests it has more frequent small gains and fewer extreme losses The other portfolios have negative skewness, indicating a higher probability of experiencing larger losses compared to gains Additionally, Portfolio E has the highest kurtosis at 5.36, suggesting it has a higher likelihood of extreme values compared to Portfolio A, which has the lowest kurtosis at 2.79.

Overall, each portfolio exhibits distinct characteristics for different investors Portfolio D is optimal for those seeking strong risk-adjusted returns, with the highest Jensen's alpha and Sharpe ratio On the other hand, portfolio A is for those willing to take on higher risk for higher returns,with the highest cumulative and annualized return but also the highest volatility Finally,portfolio E for those prioritizing stability and value investments has the lowest annualized volatility and standard deviation but also the lowest returns.

Regression result: CAPM

Table 6 summarizes the results of CAPM regression analysis for five portfolios (A, B, C, D, E) The coefficients for the market factor, alpha value, and adjusted R-squared values are provided, along with their statistical significance indicated by asterisks and t-values in parentheses (*, ** and *** denotes significance level of 5%, 1%, and 0.1%, respectively) The number of observations (N = 60) indicates that the regression analysis for each portfolio was based on 60 monthly returns, as the data is monthly over a 5-year period (2019- 2023).

About the market sensitivity, at the 0.1% level, all coefficients for Mkt-rf are statistically significant Portfolio A exhibits the highest sensitivity to market movements, with a coefficient of 0.996 The coefficient close to 1 suggests that Portfolio A moves almost in tandem with the with a coefficient of 0.935 Portfolio D, with a coefficient of 0.856, and Portfolio E, with the lowest market sensitivity at 0.800, show progressively less sensitivity to market fluctuations The returns of portfolio E are less influenced by the market compared to the others.

About the alpha, at the 0.1% level, all coefficients are also statistically significant All five portfolios demonstrate significant positive alpha values, indicating outperformance relative to their expected returns based on market risk Portfolio D leads with the highest alpha of 0.0170, suggesting it generates an excess return of 1.70% per period Portfolio B follows with an alpha of 0.0167, while portfolio C shows an alpha of 0.0146 Portfolio A and portfolio E have the lowest market sensitivity at 0.0100 and 0.0122, meaning that they outperform the expected return by 1% and 1.22% per period, respectively.

These results highlight that all portfolios are significantly influenced by the market while also providing positive excess returns, with portfolio D being the standout performer in terms of alpha and portfolio A in terms of the highest market sensitivity The adjusted R-squared values are over 80% for all portfolios, showing that the market factor explains a substantial portion of the variance in returns Portfolio A has the highest explanatory power, followed closely by portfolios C and D, with portfolio E having the lowest, yet still strong, explanatory power.

Regression result: Fama-French Three-Factor

Table 7 Fama-French Three-Factor regression result

Table 7 summarizes the results of Fama-French Three-Factor regression analysis for five portfolios (A, B, C,

D, E) The coefficients for alpha value, the value, size and market factor, and adjusted R-squared values are provided, along with their statistical significance indicated by asterisks and t-values in parentheses (*, ** and

*** denotes significance level of 5%, 1%, and 0.1%, respectively) The number of observations (N = 60) indicates that the regression analysis for each portfolio was based on 60 monthly returns, as the data is monthly over a 5-year period (2019-2023)

About the market sensitivity, at the 0.1% level, all coefficients for Mkt-rf are statistically significant Portfolios A and B exhibit the highest market sensitivity, at 1.030 and 1.036, respectively, indicating a strong sensitivity to market movements With coefficients around 1.0, they move closely with the market They are followed by portfolio C, which, with a coefficient of 0.914, shows significant but slightly lower market sensitivity Portfolio D, with a coefficient of 0.807, and Portfolio E, with a coefficient of 0.858, demonstrate lower market sensitivities compared to the other portfolios.

Regarding the size factor (SMB), the coefficient for portfolio A of 0.573 indicates a positive and significant relationship with the size factor This suggests that smaller firms are positively contributing to its returns In contrast, Portfolio E shows a significant negative relationship with the size factor, as evidenced by its coefficient of -0.515, suggesting larger firms contribute positively to its returns Portfolios B, C, and D do not exhibit significant size factor impacts, with Portfolio B showing a non-significant coefficient of -0.0541 and Portfolios C and D having non- significant coefficients of 0.319 and 0.312, respectively, indicating a negligible impact of the size factor on the returns of portfolios B, C, and D.

For the value factor (HML), portfolio A has a significant negative coefficient of -0.213, indicating that growth firms, rather than value firms, positively influence its returns Similarly, portfolio B has a significant negative coefficient of -0.175, showing a preference for growth firms In contrast, portfolio E benefits from the value factor with a significant positive coefficient of 0.144, suggesting that value firms enhance their returns Portfolios C and D do not show significant value factor impacts, with portfolio C having a non-significant coefficient of 0.0684 and portfolio D having a non-significant coefficient of 0.0892, indicating a negligible impact of the value factor on the returns of portfolios C and D.

All portfolios exhibit significant alpha values, indicating consistent returns above their risk-adjusted expectations Portfolio E reports the highest alpha (0.0173), translating to an excess return of 1.73% per period Portfolio A demonstrates the lowest alpha (0.00434), outperforming by 0.434% per period Portfolios D, B, and C also display excess returns, amounting to 1.16%, 1.58%, and 1.33% per period, respectively.

The results of the FF3 regression analysis include more of the SMB and HML factors compared to CAPM regression, which helps provide more insights into the risk drivers of portfolio’s return.The regression reveals that market movements significantly influence all portfolios' returns.SMB and HML, on another hand, influence varies among portfolios Portfolio A benefits from smaller firms and growth firms and is the most sensitive to the market, while portfolio E benefits from larger firms and value firms Portfolios C and D both have no significant impact from the size and value factors Portfolio B also has no significant impact from the size factor, but it has a growth impact similar to portfolio A All portfolios have significant alphas, suggesting outperformance from expected returns Portfolio E has the highest alpha, indicating the highest excess return relative to expected returns, as well as the highest explanatory power.

Dummy variables regression

CAPM and FF3 regression results have illustrated the return performance of ESG-rated portfolios influenced by market factors, SMB, and HML factors However, the regressions are performed independently, so the differences between portfolios cannot be accessed effectively. Thus, this section will use dummy variables in regression analysis to access the differences between portfolios.

By including dummy variables, the incremental impact of being in other portfolios can be compared to a baseline portfolio, which enables for a comparison of the performance of one portfolio relative to others In addition, including dummy variables in a single regression model enables us to examine how market factors, SMB factors, and HML factors differentially affect each portfolio.

Table 8 Dummy variables regression result with A is the baseline portfolio

Alpha Mkt-Rf SMB HML

Table 8 summarizes the results of regression analysis using dummy variables with A is the baseline portfolio for five portfolios (A, B, C, D, E) The coefficients for alpha value, the value, size and market factor, and adjusted R-squared values are provided, along with their statistical significance indicated by asterisks and t- values in parentheses (*, ** and *** denotes significance level of 5%, 1%, and 0.1%, respectively)

Portfolios D and E exhibit lower sensitivity to market returns than portfolio A, while portfolios B and C have no significant correlation Portfolio E demonstrates the strongest influence on value firms compared to portfolio A Notably, all portfolios yield positive and substantial alpha values, surpassing portfolio A with excess returns of 1.05%, 0.895%, 1.16%, and 1.18% per period Among them, portfolio E boasts the highest excess return, while portfolio B remains distinct, lacking significant relationships with any of the examined factors.

Table 9 Dummy variables regression result with B is the baseline portfolio

Alpha Mkt-Rf SMB HML

Table 9 summarizes the results of regression analysis using dummy variables with B is the baseline portfolio adjusted R-squared values are provided, along with their statistical significance indicated by asterisks and t- values in parentheses (*, ** and *** denotes significance level of 5%, 1%, and 0.1%, respectively)

Portfolios D and E are less sensitive to market returns compared to portfolio B, while portfolios

A and C are not significantly correlated to market returns The size factor does not significantly impact all portfolio returns compared to Portfolio B Value firms positively affect the returns of portfolios C, D, and E compared to portfolio B Portfolio A has no significant effect on value firms The alpha values show that only portfolio A underperforms portfolio B, with excess returns of 1.05% per period.

Table 10 Dummy variables regression result with C is the baseline portfolio

Alpha Mkt-Rf SMB HML

Table 10 summarizes the results of regression analysis using dummy variables with C is the baseline portfolio for five portfolios (A, B, C, D, E) The coefficients for alpha value, the value, size and market factor, and adjusted R-squared values are provided, along with their statistical significance indicated by asterisks and t- values in parentheses (*, ** and *** denotes significance level of 5%, 1%, and 0.1%, respectively)

Portfolio E is less sensitive to market returns compared to portfolio C, while portfolios A, B, and

C are not significantly correlated to market returns Portfolio C has no significant SMB andHML coefficients, suggesting size and value factors do not significantly impact it Portfolio E has a significant SMB coefficient of -0.614, suggesting larger firms contribute positively to its returns Portfolios A and B have significant HML coefficients of -0.281 and -0.173, respectively,indicating growth firms positively affect their returns Portfolio A underperforms portfolio C with excess returns of 0.895% per period, while portfolios B and D do not show significant relationships with any factor.

Table 11 Dummy variables regression result with D is the baseline portfolio

Alpha Mkt-Rf SMB HML

Table 11 summarizes the results of regression analysis using dummy variables with D is the baseline portfolio for five portfolios (A, B, C, D, E) The coefficients for alpha value, the value, size and market factor, and adjusted R-squared values are provided, along with their statistical significance indicated by asterisks and t- values in parentheses (*, ** and *** denotes significance level of 5%, 1%, and 0.1%, respectively)

Portfolios A and B are less sensitive to market returns compared to portfolio D, while portfolios

C and E are not significantly related The size and value factors do not significantly impact portfolio D, while larger firms contribute positively to portfolio E's returns compared to portfolio

D Growth firms positively affect the returns of A and B Portfolio A underperforms portfolio D with excess returns of 1.16% per period, while portfolio C does not show significant relationships with any factors, indicating it behaves differently compared to portfolio D.

Table 12 Dummy variables regression result with E is the baseline portfolio

Alpha Mkt-Rf SMB HML

Table 12 summarizes the results of regression analysis using dummy variables with E is the baseline portfolio for five portfolios (A, B, C, D, E) The coefficients for alpha value, the value, size and market factor, and adjusted R-squared values are provided, along with their statistical significance indicated by asterisks and t- values in parentheses (*, ** and *** denotes significance level of 5%, 1%, and 0.1%, respectively)

Portfolios A, B, and C are more sensitive to market returns compared to portfolio E, while portfolio D is not significantly related Smaller firms contribute positively to the returns of portfolios A, C, and D compared to portfolio E Growth firms also positively affect the returns of these portfolios However, only portfolio A has significant alphas, indicating that it underperforms portfolio E with excess returns of 1.18%.

In summary, dummy variable regressions access the differences between portfolios, allowing for the comparison of one portfolio's incremental performance relative to others It can be concluded from the regressions that portfolios with lower ESG ratings exhibit greater market sensitivity than those with higher ESG ratings, which applied to all top, middle, and bottom portfolios Five regressions indicate that portfolio A underperforms the other four portfolios B, C, D, and E in terms of returns, indicating the lowest excess return, as demonstrated by CAPM and FF3 regression Compared to middle portfolios B, C, and D, smaller firms are observed to affect portfolio A’s return, and bigger firms are observed to affect portfolio E’s return Portfolio B includes more growth companies than portfolios C and D; however, the incremental impact on the value factor is not defined between portfolios C and D The incremental impact on the size factor is also not defined between the middle portfolios B, C, and D, which indicates that firm size behaves differently among them These analyses clearly demonstrate the incremental differences in returns, market sensitivity, size factor, and value factor between the levels of portfolios, but the results of the middle portfolios are mixed and do not lead to a conclusive conclusion.

Discussion

Interpretation of results

When comparing the portfolio of firms with the highest ESG ratings, portfolio E, and the portfolio of firms with the lowest ESG ratings, portfolio A, the results provide strong evidence that supports the first hypothesis H1, which states that "portfolios of firms with higher ESG ratings demonstrate lower volatility than portfolios of firms with lower ESG ratings." As presented in the descriptive statistics part, portfolio E shows the lowest total risk and systematic risk in terms of beta and annualized volatility Portfolio A, on the other hand, has the highest cumulative and annualized returns but also the highest volatility, indicating higher risk compared to portfolio E The trend over time also suggests that portfolio E is more stable than portfolio A.

In comparison to the middle portfolios B, C, and D, they also exhibit the same pattern: those with higher ESG ratings have higher annualized volatility and beta, except for portfolio C, which has slightly higher annualized volatility and beta than portfolio B but is considered not significant.

Contrarily, the study's analysis casts doubt on the second hypothesis (H2), which posited that portfolios with higher ESG ratings outperform those with lower ratings on a risk-adjusted basis Although CAPM and FF3 analyses indicate that portfolio E (with the highest ESG ratings) has a higher alpha than portfolio A (with the lowest ESG ratings), suggesting a superior risk-adjusted return, dummy variable analysis further corroborates this finding However, the Sharpe and Treynor ratios suggest a more nuanced conclusion.

A generates a higher return than portfolio E in terms of total risk and systematic risk Portfolio E,although having the lowest volatility, has an annualized return that is insufficient to compensate for the volatility As a result, it provides the lowest Sharpe ratio and Treynor ratio This, in contrast, indicates that portfolio A provides higher returns in comparison to the level of risk undertaken Therefore, the evidence suggests that while higher ESG ratings are associated with lower volatility and strong alpha, they do not necessarily lead to better risk-adjusted returns when considering other performance metrics like Sharpe and Treynor ratios.

Analyses from middle portfolios B, C, and D also provide inconclusive evidence that portfolios of firms with higher ESG ratings generate better risk-adjusted returns Dummy variable regressions do not show a defined conclusion about the incremental differences among middle portfolios The middle portfolio D, which comprises firms with higher ESG ratings than portfolio

Portfolio D exhibits exceptional risk-adjusted returns, with the highest Sharpe, Treynor ratios, and Jensen's alpha under CAPM analysis Furthermore, under FF3 analysis, Portfolio D outperforms Portfolio A and only falls behind Portfolio E in alpha Dummy variable analysis confirms this superiority Despite its lower ESG ratings compared to Portfolio E, Portfolio D finds the ideal balance between risk and return, making it the best choice for investors seeking strong risk-adjusted performance.

Despite inconclusive evidence linking ESG ratings to risk-adjusted performance, this study suggests that incorporating ESG factors into investment selection can yield abnormal returns This indicates that investing broadly in ESG-rated companies may offer benefits, even if targeting specific ratings may not These findings align with previous research and suggest ESG characteristics may be associated with higher valuations, supporting the notion that SRI-focused firms experience enhanced financial performance and risk mitigation.

All regressions show that the SMB and HML factors are significant for both top and bottom portfolios, with the exception of middle portfolios This indicates that, in the European market, firm size and firm value affect abnormal returns, with low ESG ratings being more common among smaller and growing companies and high ESG ratings being more common among bigger and more valuable companies For middle-class portfolios, firm size and firm value are not the primary drivers of returns The initial investigation of the portfolios' characteristics showed that the high-rated portfolio companies are frequently bigger and more valuable, while the low-rate portfolio companies are smaller and growing companies The high ESG portfolios have a generally larger average market capitalization per portfolio over the five years examined This tendency suggests that larger firms are more competent at sustainability, which is probably because they have more resources to allocate to ESG activities and are under more pressure to perform better because of their increased impact These findings are consistent with studies by Lee et al (2021) High ESG-rated portfolios have a lower market beta and have shown more resilience in times of economic slump, like the COVID-19 pandemic Rising demand and regulatory pressure have led to an increase in SRI, which in turn has driven capital inflows into assets with higher ESG ratings.

Furthermore, according to the Efficient Market Hypothesis (EMH), all assets are priced fairly by in the market, taking into account all available information, and adding ESG ratings should not lead to abnormal returns However, market inefficiencies, including information asymmetry,transaction costs, and irrational behavior, enable investors to locate and trade mispriced assets,resulting in abnormal returns The significant positive alphas discovered in this study suggest that markets are not completely efficient and that ESG information can be exploited to generate abnormal returns.

Limitations of the research

The applied research methodology and data gathering influence the generalizability of the results; hence, it is critical to address the limitations of the study As mentioned above, each ESG rating provider employs unique methodologies, resulting in a lack of consistency among the ratings This leads to significantly varied outcomes when using these ratings to construct a portfolio (Li, 2020) This study uses ESG ratings from Refinitiv Eikon, so the results could be significantly different from other studies that use different ESG rating providers This study's findings consider a five-year timeframe rather than a longer time span, which may present the possibility of additional factors influencing the outcomes Neglected factors that influence outcomes may result in inaccurate conclusions being reached It was also necessary to select an asset universe that had a sufficient amount of ESG rating coverage in order to obtain significant analysis and results The European market is investigated using STOXX Europe 600 to mitigate the data loss of companies lacking ratings Nevertheless, the study focuses on large-capitalization businesses and uses a small sample of marketplaces This suggests that the outcomes may vary among firms with small and medium capitalization.

Furthermore, this study only uses aggregated ESG ratings without further separation into environmental, social, and governance ratings, which does not provide a deeper analysis of the three pillars The sample period is set to include returns from 2019 to 2023, which encompassed the potential noise from the COVID-19 pandemic that began in 2020 and thus did not fully account for all the influencing factors It is highly probable that the results of an out-of-sample test would yield alternative conclusions.

Conclusion

Recommendations for future research

This paper explores the connection between ESG and risk-adjusted performance The findings suggest that investors can generate abnormal returns by investing according to companies' ESG scores However, the study cannot definitively explain the driving factors behind these abnormal returns The study suggests further investigation into the underlying drivers behind the abnormal returns, such as whether they are due to temporary market mispricing, additional risk factors, or a sign of higher quality The characteristics of the firms could also be investigated and controlled to determine whether the abnormal returns are still significant.

Another recommendation is that portfolio performance could be compared by utilizing ESG data from multiple providers The inconsistency in scoring methodologies used by rating agencies leads to low correlations between companies' final ESG ratings studying proposed strategies across various scores provided by different agencies to determine the best performance. Nevertheless, research on the subject will frequently yield inconsistent results with no clear direction of influence until a standardized framework for ESG rating is established In addition, future research could use alternative factor models, such as the Fama-French five-factor model, to gain a more comprehensive understanding of the ESG and risk-adjusted performance relationship.

Amel-Zadeh, A., & Serafeim, G (2018) Why and how investors use ESG information: Evidence from a global survey Financial analysts journal, 74(3), 87-103 DOI:10.2139/ssrn.2925310

Amit, R., & Wernerfelt, B (1990) Why do firms reduce business risk? Academy of

Ashwin Kumar, N C., Smith, C., Badis, L., Wang, N., Ambrosy, P., & Tavares, R (2016) ESG factors and risk-adjusted performance: a new quantitative model Journal of sustainable finance

Bergman, M S., Deckelbaum, A J., & Karp, B S (2020, August 1) Introduction to ESG The Harvard Law School Forum on Corporate Governance Retrieved from https://corpgov.law.harvard.edu/2020/08/01/introduction-to-esg/

Bofinger, Y., Heyden, K J., & Rock, B (2022) Corporate social responsibility and market efficiency: Evidence from ESG and misvaluation measures Journal of Banking & Finance, 134,

Bouslah, K., Kryzanowski, L., & M’zali, B (2018) Social performance and firm risk: Impact of the financial crisis Journal of Business Ethics, 149, 643-669 DOI: 10.1007/s10551-016-3017-x Breusch, T S (1978) Testing for autocorrelation in dynamic linear models Australian economic papers, 17(31) DOI: 10.1111/j.1467-8454.1978.tb00635.x

Cao, J., Titman, S., Zhan, X., & Zhang, W E (2019) ESG preference and market efficiency: Evidence from mispricing and institutional trading SSRN Electronic Journal Retrieved from: https://gssf.ch/wp-content/uploads/2018/12/ESG-Preference-and-Market-Efficiency.pdf

Chang, C E., & Doug Witte, H (2010) Performance evaluation of US socially responsible mutual funds: Revisiting doing good and doing well American Journal of Business, 25(1), 9-24.DOI: 10.1108/19355181201000001

Chollet, P., & Sandwidi, B W (2018) CSR engagement and financial risk: A virtuous circle? International evidence Global Finance Journal, 38, 65-81 DOI: 10.1016/j.gfj.2018.03.004 Cornell, B., & Damodaran, A (2020) Valuing ESG: Doing good or sounding good? NYU Stern

School of Business DOI: 10.2139/ssrn.3557432

Cowton, C., & Sandberg, J (2012) Socially responsible investment In R Chadwick (Ed.), Encyclopedia of Applied Ethics (2 ed., pp 142-151) Academic Press Inc DOI:10.1016/B978-0-12-373932-2.00086-7

Díaz, V., Ibrushi, D., & Zhao, J (2021) Reconsidering systematic factors during the COVID-19 pandemic–The rising importance of ESG Finance Research Letters, 38, 101870 DOI: 10.1016/j.frl.2020.101870

DG FISMA (n.d.-a) Corporate sustainability reporting European Commission Retrieved April

18, 2024, from https://finance.ec.europa.eu/capital-markets-union-and-financial-markets/ company-reporting-and-auditing/company-reporting/corporate-sustainability-reporting_en

DG FISMA (n.d.-b) EU taxonomy for sustainable activities European Commission Retrieved July 11, 2024, from https://finance.ec.europa.eu/sustainable-finance/tools-and-standards/eu- taxonomy-sustainable-activities_en

Elton, E J., Gruber, M J., & Blake, C R (1996) The persistence of risk-adjusted mutual fund performance Journal of business, 133-157 Retrieved from http://www.jstor.org/stable/2353461

Environmental, social and governance scores from LSEG (2023, December) LSEG Retrieved from https://www.lseg.com/content/dam/data-analytics/en_us/documents/methodology/lseg-esg- scores-methodology.pdf?esg=telekom+austria

ESG Global Study (2022) Capital Group Retrieved from https://www.capitalgroup.com/advisor/pdf/shareholder/ITGEOT-028-658081.pdf

ESG Global Study (2023) Capital Group Retrieved from https://www.capitalgroup.com/advisor/pdf/shareholder/ITGEOT-073-1043294.pdf

Eurosif (n.d.) Sustainable investment Eurosif Retrieved July 11, 2024, from https://www.eurosif.org/sustainable-investment/

Fama, E F (1991) Efficient capital markets: II The journal of finance, 46(5), 1575-1617 DOI: 10.1111/j.1540-6261.1991.tb04636.x

Fama, E F., & French, K R (1993) Common risk factors in the returns on stocks and bonds Journal of financial economics, 33(1), 3-56 DOI: 10.1016/0304-405X(93)90023-5

Fama, E F., & French, K R (1996) Multifactor explanations of asset pricing anomalies The journal of finance, 51(1), 55-84 DOI: 10.1111/j.1540-6261.1996.tb05202.x

Fama, E F., & French, K R (2015) A five-factor asset pricing model Journal of financial economics, 116(1), 1-22 DOI: 10.1016/j.jfineco.2014.10.010

Freeman, R E (2010) Strategic management: A stakeholder approach Cambridge university press.

Friedman, M (1970) The Social Responsibility of Business is to Increase its Profits New York

Times Magazine, September, 13, 122-126 Retrieved from https://www.nytimes.com/1970/09/13/ archives/a-friedman-doctrine-the-social-responsibility-of-business-is-to.html

Giese, G., Lee, L E., Melas, D., Nagy, Z., & Nishikawa, L (2019) Foundations of ESG investing: How ESG affects equity valuation, risk, and performance The Journal of Portfolio

Global Sustainable Investment Review (2022) Global Sustainable Investment Alliance.

Retrieved from https://www.gsi-alliance.org/wp-content/uploads/2023/11/GSIA-Report-2022- FINAL-Compressed.pdf

Godfrey, L G (1978) Testing against general autoregressive and moving average error models when the regressors include lagged dependent variables Econometrica: Journal of the

Econometric Society, 1293-1301 DOI: doi.org/10.2307/1913829

Herzel, S., Nicolosi, M., & Starica, C (2011) The cost of sustainability on optimal portfolio choices European Journal of Finance, 1, 1-17 Retrieved from https://core.ac.uk/download/pdf/ 6538183.pdf

Jensen, M C (1968) The performance of mutual funds in the period 1945-1964 The Journal of finance, 23(2), 389-416 DOI: 10.1111/j.1540-6261.1968.tb00815.x

Kempf, A., & Osthoff, P (2007) The effect of socially responsible investing on portfolio performance European financial management, 13(5), 908-922 DOI: 10.1111/j.1468- 036X.2007.00402.x

Kumar, S (2023) Exploratory review of esg factor attribution to the portfolio return in Fama- French factor model framework Academy of Marketing Studies Journal, 27(3) Retrieved from https://www.abacademies.org/articles/exploratory-review-of-esg-factor-attribution-to-the- portfolio-return-in-famafrench-factor-model-framework.pdf

Le Sourd, V (2024, March 20) Does ESG investing improve risk-adjusted performance?. EDHEC Risk Climate Impact Institute Retrieved from https://climateimpact.edhec.edu/does- esg-investing-improve-risk-adjusted-performance

Lee, D D., Faff, R W., & Rekker, S A (2013) Do high and low‐ranked sustainability stocks perform differently? International Journal of Accounting & Information Management, 21(2), 116-132 DOI: 10.1108/18347641311312267

Lee, D D., Fan, J H., & Wong, V S (2021) No more excuses! Performance of ESG‐ integrated portfolios in Australia Accounting & Finance, 61, 2407-2450 DOI: 10.1111/acfi.12670

Lefkovitz, D (2023, February 1) In a Period of Poor Performance for Sustainable Investments,

Gender Equality and Renewable Energy Were Bright Spots Morningstar Indexes Retrieved from https://indexes.morningstar.com/insights/analysis/blt2dc45febbf0bad6b/2022-morningstar- indexes-esg-risk-return-analysis

Li, F (2020, January) What a Difference an ESG Ratings Provider Makes! Research Affiliates. Retrieved from https://www.researchaffiliates.com/content/dam/ra/publications/pdf/770-what-a- difference-an-esg-ratings-provider-makes.pdf

Lintner, J (1965) Security prices, risk, and maximal gains from diversification The journal of finance, 20(4), 587-615 DOI: 10.1111/j.1540-6261.1965.tb02930.x

Lo, A W (2007) Efficient markets hypothesis, The New Palgrave: A Dictionary of Economics, 2nd edition Retrieved from https://ssrn.com/abstract1509

LSEG (n.d.) TRBC Sector Classification Refinitiv. https://www.lseg.com/en/data-analytics/financial-data/indices/trbc-business-classification

Luo, D (2022) ESG, liquidity, and stock returns Journal of International Financial Markets,

Institutions and Money, 78, 101526 DOI: 10.1016/j.intfin.2022.101526

Markowitz, H (1952) Portfolio Selection The Journal of Finance, 7(1), 77–91 DOI: 10.2307/2975974

Mossin, J (1966) Equilibrium in a capital asset market Econometrica: Journal of the econometric society, 768-783 DOI: 10.2307/1910098

Naseer, M., & Bin Tariq, D Y (2015) The efficient market hypothesis: A critical review of the literature The IUP journal of financial risk management, 12(4), 48-63 Retrieved from: https://ssrn.com/abstract'14844

Newey, W K., & West, K D (1987) A Simple, Positive Semi-Definite, Heteroskedasticity and Autocorrelation Consistent Covariance Matrix Econometrica, 55(3), 703–708 DOI: 10.2307/1913610

PRI (n.d.) About the PRI PRI Retrieved July 11, 2024, from https://www.unpri.org/about-us/about-the-pri

Prol, J L., & Kim, K (2022) Risk-return performance of optimized ESG equity portfolios in the NYSE Finance Research Letters, 50, 103312 DOI: 10.1016/j.frl.2022.103312

Revelli, C., & Viviani, J L (2015) Financial performance of socially responsible investing (SRI): what have we learned? A meta‐analysis Business Ethics: A European Review, 24(2), 158-185 DOI: 10.1111/beer.12076

Said, S E., & Dickey, D A (1984) Testing for unit roots in autoregressive-moving average models of unknown order Biometrika, 71(3), 599-607 DOI: 10.2307/2336570

Sassen, R., Hinze, A K., & Hardeck, I (2016) Impact of ESG factors on firm risk in Europe Journal of business economics, 86, 867-904 DOI: 10.1007/s11573-016-0819-3

Schmid, F., & Schmidt, R (2010) Statistical Inference for Sharpe Ratio Interest rate models, asset allocation and quantitative techniques for central banks and sovereign wealth funds, 337-

Sharpe, W F (1964) Capital asset prices: A theory of market equilibrium under conditions of risk The journal of finance, 19(3), 425-442 DOI: 10.1111/j.1540-6261.1964.tb02865.x

Statista (2024a, March 7) Perceived demand for sustainable funds 2023 in Europe and North

America Statista Retrieved from https://www.statista.com/statistics/1454696/perceived- demand-for-sustainable-funds-europe-north-america/

Statista (2024b, February 28) Global ESG ETF assets 2006-2023 Statista Retrieved from https://www.statista.com/statistics/1297487/assets-of-esg-etfs-worldwide/

Statman, M., & Glushkov, D (2009) The wages of social responsibility Financial Analysts

Szylar, C (2013) Handbook of market risk John Wiley & Sons.

Tang, Z (2023) Risk-Return Analysis of Equity Portfolios: Comparison Between CAPM and Fama-French Three Factor Model In 3rd International Conference on Economic Development and Business Culture (ICEDBC 2023) (pp 227-237) Atlantis Press DOI: 10.2991/978-94-6463-

Tayan, B (2022, August 24) ESG Ratings: A Compass without Direction The Harvard Law

School Forum on Corporate Governance Retrieved from https://corpgov.law.harvard.edu/2022/08/24/esg-ratings-a-compass-without-direction/

Teti, E., Dallocchio, M., & L'Erario, G (2023) The impact of ESG tilting on the performance of stock portfolios in times of crisis Finance Research Letters, 52, 103522 DOI: 10.1016/j.frl.2022.103522

Townsend, B (2020) From SRI to ESG: The origins of socially responsible and sustainable investing The Journal of Impact and ESG Investing, 1(1), 10-25 DOI: 10.3905/jesg.2020.1.1.010

Treynor, J L (1965) How to Rate Management of Investment Funds Harvard Business Review, 63-75.

UN Global Compact (n.d.) Uniting business for a better world United Nations Global Compact Retrieved July 11, 2024, from https://unglobalcompact.org/take-action/20th- anniversary-campaign

United Nations (n.d.-a) UN Documentation: Development - 2000-2015, Millennium

Development Goals Research Guides at United Nations - Dag Hammarskjửld Library Retrieved

July 11, 2024, from https://research.un.org/en/docs/dev/2000-2015

United Nations (n.d.-b) UN Documentation: Development - 2016-2030, Sustainable

Development Goals (n.d.) Research Guides at United Nations - Dag Hammarskjửld Library.

Retrieved July 11, 2024, from https://research.un.org/en/docs/dev/2016-2030

Wang, X., Guise, G., & Nagy, Z (2023, March 7) ESG Factor Returns: 2022 in Review - MSCI.

MSCI Retrieved from https://www.msci.com/www/blog-posts/esg-factor-returns-2022-in/03701563813

White, H (1980) A heteroskedasticity-consistent covariance matrix estimator and a direct test for heteroskedasticity Econometrica: journal of the Econometric Society, 817-838 DOI: 10.2307/1912934

Who Cares Wins (2004) The Global Compact Retrieved from https://www.unepfi.org/fileadmin/events/2004/stocks/who_cares_wins_global_compact_2004.pd f

Widyawati, L (2020) A systematic literature review of socially responsible investment and environmental social governance metrics Business Strategy and the Environment, 29(2), 619-

Zehir, E., & Aybars, A (2020) Is there any effect of ESG scores on portfolio performance?Evidence from Europe and Turkey Journal of Capital Markets Studies, 4(2), 129-143 DOI:10.1108/JCMS-09-2020-0034