On a Trajectory to Sustainable Investing: Is Sustainability and Proﬁtability Mutually Exclusive?

(1)

On a Trajectory to Sustainable Investing:

Is Sustainability and Profitability Mutually Exclusive? ^∗

by Simon Ginnerup Andersen

^†‡

May 17 2021

^§

Master Thesis

∗MSc Applied Economics & Finance

†Supervisor: Mads Stenbo Nielsen, Associate Professor, Department of Finance, CBS

‡Student ID: 103377

§Number of pages: 79.8 — STU count: 181,561

(2)

Abstract

It is a common belief that sustainability and profitability are mutually exclusive con- cepts. Traditionally, investment decisions have been based on a narrow set of criteria relating to the trade-off between risk and return. Nevertheless, a growing demand for ESG strategies has led to a proliferation in offerings centered around sustainable investment practices, and advocates of ESG investing argue that investors can ”do well by doing good.” Despite acknowledging sustainable finance as an emerging field that builds on a broader set of investment principles, the empirical evidence documented in the literature has been rather ambiguous with no consensus reached yet. More recent studies tend to favor a positive link between ESG and financial performance though. This thesis aims to provide an empirical assessment of the relationship between ESG metrics and corporate financial performance over the period from 2008 to 2020. Using annual ESG ratings from Refinitiv this paper constructs 10 portfolios based on positive screenings, which track the S&P500 and the STOXX600. Results suggest that low-rated portfolios tend to outperform high-rated portfolios for aggregated ESG scores as well as for individual pillar scores where financial relevance seems to be strongest for the governance component. Results are mixed during times of crisis, suggesting that ESG performance depends on the specific circumstances. Another interesting finding is obtained through Fama-French multi-factor regressions, showing how investors can yield higher returns by pursuing a long-short trading strategy that takes a long position in portfolios with poor ESG performance and a short position in portfolios with strong ESG performance, thus suggesting that doing the right thing comes with a price.

Keywords: ESG investing, sustainable finance, portfolio management, asset pricing, Fama-French factor regressions, S&P500, STOXX600

(3)

”There is one and only one social responsibility of business – to use its resources and engage in activities designed to increase its profits so long as it stays within the rules of the game, which is to say, engages in open and free competition without deception or fraud.”

(Milton Friedman, 1970)

(4)

1 Introduction

In today’s world of globalization and interdependence and in times of financial crisis, issues such as climate change, biodiversity, human rights, social inequality, economic inequality, business ethics and corporate governance are at the forefront of public and political attention.

How companies respond to these issues is becoming as important as traditional financial metrics when evaluating corporate performance, hence playing a more central role in investors’ decision- making efforts to identify long-term opportunities and risks for companies. In more recent years the concept of sustainable finance has emerged focusing on the allocation of capital flows towards projects that support the sustainable agenda. This gives market participants a choice to act in a responsible way and support positive change in day-to-day business and by engaging in sustainable capital allocation, investors can take part in redefining existing business standards and contribute toward a more sustainable future. According to the United Nations-backed Principles for Responsible Investment (PRI) responsible investing is an approach that aims to incorporate Environmental, Social and Governance (henceforth ESG) factors into investment decisions to better manage risk and generate sustainable long-term returns. The concept pro- gressed further with the notion of the triple bottom line where people, planet, and profit are considered equally important for the long-term success of society at large. This concept has gained traction over the years and has morphed into the ESG components that are now being integrated into most sustainable investment practices across all asset categories.

Sustainable investing has become a buzzword of the modern investment society with to- morrow’s investors being more committed to promote a sustainable transition. The growing investor interest in ESG factors rests on the notion that a sustainable strategy can affect the long-term performance of companies and should therefore be given appropriate consideration in investment decisions. Now the financial sector is also beginning to take an active position in trying to mobilize efforts through sustainable investment practices to contribute to a global improvement. Statistics seem to suggest that sustainable finance and investment is no longer just a niche practice (BCG, 2021). Sustainable investing has surged worldwide with sustainable assets having grown 34% since 2016 and today the total asset base is estimated to be worth more than$30 trillion according to the Global Sustainable Investment Alliance. Europe and the US are still the most dominant regions and account for the largest concentration of sustainable assets. The number of sustainable funds available to investors have increased exponentially and this trend is showing no signs of slowing in the near future. This provides investors with an opportunity to better align their portfolios with ESG criteria, facilitating a sustainable development. The recent COVID-19 pandemic has also revealed how sustainable investments are seen as a safe haven in times of crisis and volatile market conditions (P´astor & Vorsatz, 2020).

During the pandemic ETFs experienced large capital outflows due to investors relocating their money towards more resilient investments. According to Morningstar there was a total capital

(7)

outflow of USD 387.7 billion across all types of investment funds during the first quarter of 2020, whereas sustainable funds achieved capital inflows equal to USD 46.6 billion during the same period (Lodberg et al., 2020).

The relationship between ESG and financial performance is a hot topic among investment professionals, asset managers, and academic researchers. The long-standing question remains whether the integration of firms with high sustainability scores actually improves stock performance in terms of risk and return characteristics. The empirical evidence has been inconclusive and no consensus has yet been reached, leaving the investment community divided in two groups: those arguing that the incorporation of ESG elements will sacrifice returns and those believing that companies with strong ESG profiles will tend to outperform market benchmarks leveraging from high profitability and reduced systematic risk (Lee et al., 2020). Another group of investors may simply accept the cost of imposing restrictive ESG guidelines as the cost of doing the right thing. For those, leaving a better world for the next generation or improving living conditions for other people will lead to non-monetary benefits, which in itself is an incentive to pursue sustainable investment strategies.

The ESG performance is a concept that covers a wide spectrum of categories - including climate change, water management, carbon emission, protection of human- and labor rights, corporate ethics etc. While progress has been made to advance the use of ESG practices, the large number of contributions has instead generated the spread of a wide array of investment terminology, and disclosure frameworks which have caused methodological inconsistencies and lack of comparability for investors. This implies that portfolio formation, and thus empirical results to a large extent may be driven by choice of data provider (Boffo & Patalano, 2020).

Generally, scholars distinguish between negative and positive screening when forming portfolios.

Negative screening excludes all companies that are involved in controversial business areas considered unethical, such as alcohol, tobacco, weapon, fossil fuels, gambling etc. Rather than excluding certain companies, positive screening refers to a process where investors proactively select stocks based on positive ESG attributes. Despite the mixed findings of previous studies there seems to be an increasing awareness of how complex it is to measure all aspects of ESG performance in just a few numbers. The manifestation of ESG as an integral part of financial decision-making is mainly driven by a strong client demand, but the surge in academic research together with better availability of more sophisticated data from professional data providers – notably ESG rating agencies – have also played a central role in paving the way for ESG investing. This has made an assessment of ESG strategies more feasible and research on the subject will most likely continue in years to come.

(8)

1.1 Research Question

This study sets out to investigate whether strong ESG characteristics can enhance corporate financial performance, and if so, how investors can set up investment strategies to exploit such market inefficiencies. The study is drawing on data from Refinitiv Datastream and is designed to examine ESG performance from 2008-2020 in a US and European context. The purpose of this research is to study the impact of incorporating ESG screening criteria on stock performance by emphasizing constituents from the S&P500 and STOXX600. It is key for investors to understand the linkage between ESG ratings and financial stock performance since it has become a screening criterion that is widely used by investment professionals, asset managers, and regular retail investors in selecting stocks and forming portfolios. Furthermore, it can provide useful insights as to whether it is possible for investors of all types to do good for society at large while also doing good financially. A positive relationship would likely cause more investors to steer money towards ESG-oriented assets which could potentially help to speed up the sustainable transition. Hence, this study aims to answer the following research question:

“To what extent does the incorporation of sustainable ESG metrics into global stock portfolios affect the risk-adjusted return and would it be possible for investors to come up with strategies to capitalize on such information?”

The research question above will guide the analysis, but in order to get a broader under- standing of the subject this study will moreover seek to explore underlying aspects of the topic.

This may render valuable insights to (i) how ESG performance differs across regions, (ii) how ESG portfolios perform in times of crisis, (iii) which mechanisms that drive sustainable performance - is it the E, S or G, and (iv) how performance is affected when adjusting for more conventional equity factors.

1.2 Motivation

There are several good reasons highlighting the relevance of this particular research area, of which some of them have already been briefly touched upon. While a lot of research on the subject has already been done, results have not always been consistent. In fact, results have often revealed to be highly dependent on what data type, choice of country, methodological terminology, and time period being examined, thus underlining the importance of renewing and reformulating what has already been done to strengthen the meaningfulness of ESG investing.

The term sustainability has become the number one buzzword and top priority on the global agenda. Politicians formulate policies and regulation to secure a more sustainable future, which has also led to a disruption of the global financial sector. To accommodate the increased demand for sustainable investments, a large number of funds aligning more properly with

(9)

sustainable beliefs have been launched by asset managers worldwide. This further highlights the importance of greater efforts toward transparency and consistency in order for investors to be able to navigate among the abundance of options and align them with financial materiality.

Another motivation of this study is to examine the most recent time period and apply some of the most sophisticated data material which has been developed over the last decade. This thesis contributes to the existing and mounting body of literature by deconstructing the ESG trinity to assess how each individual pillar influences financial performance, something which has only been scrutinized with scarcity. Additionally, this study includes evidence from the ongoing COVID-19 pandemic. Whereas numerous former studies have utilized the CAPM, Fama-French three-factor model and Carhart’s four-factor model, only a few studies have used the Fama-French five-factor model to test if ESG portfolios yield abnormal returns. This study applies both the Fama-French three-factor and five-factor model to nuance the analysis and see how robust results are to model specification. In sum, this highlights the relevance of ESG investing in a contemporary context and summarizes what motivates this thesis.

1.3 Delimitation

This paper has its theoretical roots in the literature on capital markets. It builds on the assumption that capital markets are efficient and stock prices reflect all information available at a given point in time. Historically, these theories have had great practical application, but the dynamic nature of today’s rapidly changing world calls for a modification of conventional theory. Hence, this paper applies traditional portfolio choice theory combined with the broader stakeholder perspective and other strands of theory that can provide useful insights.

The analysis of how ESG performance affects financial performance is centered around the US and Europe, thus being delineated from focusing on other regions. This may cause some degree of regional bias in the results, but the choice of US and Europe is rationalized by data quality being higher and more sophisticated in these regions. Nevertheless, the accessibility of ESG data in other markets have improved over the years and it would certainly be interesting to replicate the analysis to these markets, bearing in mind that data quality might still be somewhat lacking. Some studies examine ESG performance by looking at mutual funds. Yet, this approach has distinct weaknesses. This study does not cover ESG performance from the perspective of mutual funds because these depend on the timing ability and competences of the portfolio manager, making it difficult to separate the ability of fund managers from the actual performance of financial assets. Moreover, it is extremely difficult to examine the effect from individual screens because mutual funds often tend to employ multiple screens simultaneously and fail to disclose all the screens being used (Kempf & Osthoff, 2007). To overcome such problems, this study uses ESG screens at the company level. Neither does this study examine ESG performance by looking at accounting measures related to the operating profitability and

(10)

liquidity of a firm. This is definitely a highly relevant angle, but it has been considered beyond the scope of this paper to follow such path. Instead the focus of this thesis will be to assess the ability to earn abnormal returns from an investor’s perspective although these different angles tend to be correlated somehow. In addition, this study concentrates around the overall ESG rating together with the individual pillar scores. In other words, the impact from each of the 10 category variables that constitute the pillar scores are not covered. This could have been a way to break down ESG ratings even further to see the metrics being most relevant. Due to lack of availability within the field of ESG data, this paper is delimitated from using ESG ratings from multiple sources although this could be valuable and possibly uncover some of the discrepancies between different providers of ESG data.

1.4 Structure

In the course of this study, I will argue how ESG investing generally seems to underperform portfolios consisting of stocks with low ESG ratings over the period from 2008 to 2020. The methodological setup utilized render little support of ESG as an instrument to achieve superior risk-adjusted returns vis-´a-vis market returns. In this thesis it will also be argued how investors could possibly exploit significant relationships between ESG and financial performance to capitalize on strategies that steer capital into low-rated portfolios. The empirical rejection of ESG investing can partly be explained by the application of a rather narrow screening process. Hence, the results’ reliability is a matter of debate. The rest of this paper is divided into chapters and is structured as follows. First, Chapter 2 provides a historical overview of how sustainable investing has evolved over time. Chapter 3 offers an account of the underlying theoretical framework deemed relevant when studying the relationship between ESG and financial performance. Chapter 4 proceeds by screening existing literature for empirical evidence on ESG investing and sustainable finance. Chapter 5 subsequently outlines the methodological framework of this paper. Chapter 6 presents the empirical findings related to the risk-adjusted ESG performance. Chapter 7 discusses the results and elaborates on shortcomings while also suggesting areas for further research. Finally, Chapter 8 summarizes what has been done and concludes the thesis by reflecting upon the broader implications.

(11)

MASTER THESIS - CAND.MERC 2 THE EMERGENCE OF SUSTAINABLE INVESTING

2 The Emergence of Sustainable Investing

The previous chapter has just shown how socially responsible investing (SRI) has experienced accelerated growth over the past decade. This reflects an increasing awareness by investors on social, environmental, ethical and corporate governance issues. Unlike conventional investment strategies, SRI is an investment process that integrates exactly these kind of issues into investment decision-making (Renneboog et al., 2008). Although sustainable metrics are now being used extensively and can easily be integrated into portfolio analysis, equity research, screening and quantitative analysis, they have not emerged out of the blue. It is not a new phenomenon – in fact ethical investment practices are as old as the concept of investing itself and can be traced back to ancient times. Ethical investing has roots in Jewish, Chris- tian, and Islamic traditions. For instance, in medieval Christian times ethical restrictions on loans and investments were in place, which were based on the Old Testament. According to the Koran, Islamic investors should refrain from investing in companies involved in pork production, pornography, gambling and interest-based financial institutions. Throughout history profits have been derived from trade with weapons and slaves, but in the 17th century Quakers refused to engage in sinful trade or profit from exploiting others any longer. A path that was later to be followed by the Methodist Church in UK, which would no longer invest in sinful companies involved in the production of alcohol, tobacco, weapons or gambling (Renneboog et al., 2008). By applying these negative screening mechanisms to the construction of investment portfolios, ethical codes and religious beliefs shaped the earliest instances of sustainable investing.

In contrast, the innovators of modern SRI put more emphasis on the distinctive personal ethical and social convictions of investors. Throughout the 1960s and 1970s increased activism around anti-war and anti-racism fueled public sentiment in favor of more SRI practices. In fact, public sentiment led to the establishment of the Pax World Fund in 1971 – the first sustainable mutual fund which was launched to oppose US involvement in Vietnam, the use of chemi- cal weapons and urge companies to adhere to higher standards of social and environmental responsibility (Liu, 2020). During the 1980s more legislation on corporate responsibility was introduced and increased environmental concern propelled the growth for sustainable investing. The racist system in South Africa became a focal point of protests by the anti-apartheid movement advocating for divestitures of South African assets and encouraged mutual funds to stop the allocation of capital into any South African firm or subsidiary. Furthermore, environmental disasters in the late 1980s increased awareness by investors of negative environmental consequences. The nuclear accident at the Chernobyl nuclear power plant in 1986 spread ra- dioactive materials across Europe causing a surge in the number of radiation-induced cancer deaths, while 11 million gallons of crude oil was spilled into the ecosystem of Alaska when the Exxon Valdez ran aground in 1989. In the face of such environmental disasters, activists

(12)

began to dissociate from nuclear and fossil fuel companies and coordinate efforts to speed up the sustainable transition (Renneboog et al., 2008). Through the 1990s awareness of global warming continued to grow and the Kyoto Protocol can be seen as a global call to action with a commitment from world leaders to set goals on addressing global warming and climate change.

The story of ESG investing dates back to 2004 when former UN Secretary General Kofi Annan reached out to more than 50 CEOs of major financial institutions, inviting them to par- ticipate in a joint initiative – the Global Compact initiative. The goal was to clarify how ESG could be incorporated into capital markets based on corporate citizenship and efforts guided by human rights, labor, environmental, and anti-corruption principles (Kell, 2018). In the following year The Global Compact launched a report entitled “Who Cares Wins”, suggesting that good corporate sustainability performance tends to coincide with good financial results.

The report moreover offered recommendations on how to incorporate ESG factors into asset management and other investment analyses. The report was essential for the design of the Principles for Responsible Investment that were launched at the New York Stock Exchange in 2006. The PRI contain six voluntary and aspirational principles (see appendix A1) that provide a set of guidelines on how to incorporate ESG issues into investment analysis. They have attracted a majority of the world’s asset managers and investment professionals, who have joined the international signatory base with the aim to develop a more sustainable financial system (UNPRI, 2021). Since then numerous initiatives have been introduced and the number of ESG funds available to investors has grown steadily. In 2015 the Paris Agreement was signed when world leaders came to consensus on how to tackle climate change and adapt to its effects.

Unlike SRI, ESG investing builds on the assumption that ESG components have financial relevance and are important to truly understand corporate purpose. Hence, ESG investing provides investors with the opportunity to vote with their money. Historically institutional investors have been reluctant to embrace the conceptualization of ESG arguing that their fidu- ciary duties are limited to shareholder maximization irrespective of environmental, social, or more general governance concerns (Kell, 2018). Nonetheless, sustainable investing gained im- petus towards the millennium and has continued to grow globally due to remarkable client demand and concentrated initiatives enacted by politicians and international organizations like the UN. Previously, the number of sustainable funds available to investors was rather limited, yet ESG investments have proliferated and today ESG funds have hit a record high of almost 4,000 sustainable funds in September 2020 (Morningstar, 2021; Janes, 2020). All this adds up to market participants having great expectations for public companies to be good stewards of the environment, to put emphasis on the well-being of all stakeholders, and to impose good governance mechanisms enhancing both transparency and accountability. To what extent finance professionals and asset managers actually integrate these ESG variables is another question. In

(13)

a survey, Eccles, Kastrapeli & Potter (2017) revealed that only 21% of institutional investors utilize a full ESG integration, whereas the vast majority of 47% mainly use exclusionary screening mechanisms to disregard worst-in-class performers. This obviously raises the question if ESG investing merely reflects pure marketing spin and false advertising. This will be discussed further later on.

Where does that leave us today? This historical overview of how sustainable investing has evolved clearly illustrates how investment decisions have changed from being guided purely by ethical norms and religious beliefs to reflect a broader and more nuanced perspective focusing on long-term performance and the footprint set on surrounding environments.

(14)

MASTER THESIS - CAND.MERC 3 THEORETICAL BACKGROUND

3 Theoretical Background

This chapter will offer a theoretical overview of relevant theories to better understand the economic mechanisms that can be useful in explaining the relationship between ESG and financial performance. This can provide valuable insights that can be used to discuss the results of this thesis. No universal finance theory has yet been finalized on the subject and therefore these theoretical strands, which utilize different approaches, can all provide useful insights on how the integration of sustainability and ESG metrics can impact financial performance.

3.1 Shareholder Theory

Shareholder theory can be used to assess the linkage between ESG and profitability. The normative theory on business ethics was advanced by economist Milton Friedman (1970) who argues that the sole responsibility of firms is to maximize shareholder value. Management must ensure to conduct business in accordance with shareholder interest and allocate resources ef- ficiently to increase profits in a way that is compatible with ethical customs and the general rules of society. In other words, Friedman does not agree with the notion that business has any responsibility to engage in activities that serve the public interest alone. In fact, he claimed that ethical questions related to catchwords such as employment, discrimination and pollution should be taken care of by individuals and governmental institutions (Friedman, 1970). The shareholder perspective is based on the idea that management is appointed to safeguard the interest of shareholders, hence being both legally and morally obliged to do so. Once management start to incorporate ESG into investment decisions it allocates company resources to pursue investments that reflect some public perception of what good is and no longer acts as an agent of the shareholders. As a result, corporate executives arguably become employees of society instead of employees of the owners of the business (Friedman, 1970). Because the introduction of ESG initiatives will accommodate to the desires of stakeholders, advocates of the shareholder theory believe that it will decrease profits as a consequence of constraining the company’s ability to pursue optimal strategies that maximize shareholder value. Instead of charging higher prices to reflect the inclusion of ESG factors, it would be more efficient for businesses to set lower prices and let individuals make ethical contributions for themselves (Renneboog et al., 2008).

According to agent theory businesses (principals) delegate decision-making authority by hiring management (agents) to perform a service on their behalf (Tirole, 2001). The underlying problem is that principals and agents often have diverging interests (Fama & Jensen, 1983; Tirole, 2001). By making investments to enhance ESG it could for instance be argued that management sets aside shareholder interest and instead pursue their own self-interest to achieve personal gains at the expense of profit maximization. To avoid that management acts

(15)

opportunistically it is therefore crucial to align interests between shareholders and corporate executives. Designing appropriate compensation programs can be an effective mechanism to incentivize management to act in the best interest of shareholders, thereby aligning interests (Fortin et al., 2014).

Shareholder theory has been subject to a lot of criticism on the grounds that corporations fail to take into account other stakeholders when making decisions (Shank et al., 2005). Since managerial actions have impact on the welfare of other stakeholders, and not just shareholders, this should be reflected in managerial decisions. Based on Adam Smith’s ”invisible hand”

and the social welfare theorems, classical economics states that there is no conflict between maximizing shareholder value and doing good for society at large. This rationale stems from the idea that resource allocation is Pareto-optimal and social welfare is maximized when all companies maximize their own profits (Renneboog et al., 2008). However, an equilibrium where social welfare is optimized is not always consistent with the real world because assumptions underlying the welfare theorems do not always hold and stakeholders impose externalities on each other. An example could be when a business is upscaling production activities to maximize profits. As a result, the local environment will suffer from the increased pollution caused by the upscaling. Likewise, if a business engages in corruption and bribery it will impose negative externalities on society. Social welfare is the sum of the profits of all stakeholders and the examples above illustrate how social welfare is not always maximized (Jensen, 2001). Recent global challenges have highlighted the merits of a more stakeholder-oriented perspective and in essence, the firm should internalize these externalities on the various stakeholders. Hence, the next section will elaborate on stakeholder theory.

3.2 Stakeholder Theory

Stakeholder theory provides a broader view of corporate responsibility and stresses that the firm should maximize the sum of the various stakeholder’s surpluses, and not just shareholder value. If market participants consider a stakeholder-driven approach and proper ESG management to be linked to the long-term wellbeing of a corporation, it would be an inherent duty of corporate executives to give due regard to the interests of those groups (Freeman & McVea, 2001; Freeman, 1984; Jones, 1995; Walsh, 2005; Droste, 2020). ESG can be seen as a concept of stakeholder capitalism where relations between a corporation and its customers, suppliers, employees, investors, and communities are highly interconnected with frequent interactions to jointly create and trade value. According to Freeman (1984) it is paramount for management to know how the dynamics of these relationships work in order to be able to manage and shape these relationships to increase value-creation for all stakeholders. In situations with conflicting interests, it is also the responsibility of corporate executives to rethink problems and come up with solutions that address the interests of a broad group of stakeholders (Harrison et al., 2010).

(16)

Of course, it may sometimes be difficult to balance interests of a large group of stakeholders, but executives must figure out for themselves how to make trade-offs that are favorable for all parties (Freeman et al., 2007). The stakeholder perspective does not oppose shareholder’s ability to maximize wealth, but questions that this should be the most fundamental objective of business by recognizing a broader and more balanced set of objectives. The stakeholder view proposes a flexible framework that enables ESG issues to be taken into account without compromising a business’ ability to maximize shareholder wealth. Over the medium-to-long run, firms that take into account these ESG factors are more likely to avoid controversies and maintain good reputations, better retain customers and employees, and preserve high levels of shareholder trust during periods of uncertainty and transformation (Boffo & Patalano, 2020).

This can act as a catalyst for financial growth and social impact.

3.3 Discounted Cash Flow Models

Another strand of scholarly literature assesses the relationship between sustainability and financial performance using a “discounted cash flow” (DCF) approach to see how sustainability affects the value of a company when discounting the future expected cash flows (Giese et al., 2019). The idea is that sustainability and ESG matters relate to a company’s valuation through three different channels.

First, companies with strong ESG profiles are typically more competitive and profitable than their peers. One possible explanation could be more efficient allocation of resources, superior human capital development and a strong focus on strategy and innovation. In addition, high ESG-rated businesses tend to formulate long-term business plans and long-term incentive programs for senior management making ESG a good proxy for the quality of a company’s strategic and financial management. The second channel describes how a company manages its business and operational risks indicating that high ESG-rated companies often face lower idiosyncratic tail risk. The economic rationale being that companies with strong ESG characteristics have above-average risk control and compliance standards across the company and within their supply chain management. Due to better risk management, high ESG-rated companies suffer less frequently from incidents of fraud, embezzlement, corruption, and litigation, which will likely impact firm-value negatively (Godfrey et al., 2009; Jo & Na, 2012; Oikonomou et al., 2012). Finally, a company’s ESG score reflect the level of systematic risk. High ESG- rated companies are less vulnerable to systematic market shocks and thus demonstrate lower systematic risk. According to CAPM lower systematic risk corresponds to a lower cost of capital, which will lead to a higher valuation (Eccles et al., 2014; El Ghoul et al., 2011; Gregory et al., 2014). Furthermore, efficient incorporation of ESG metrics can improve companies’ appeal towards risk-averse and socially conscious investors who sidestep companies with poor ESG performance. The ability to attract a large investor base will trigger a higher valuation. To

(17)

sum up, strong ESG characteristics can be seen as a quality stamp enhancing a company’s competitive advantage and profitability, while lowering both idiosyncratic and systematic risk (Giese et al., 2019).

In a parallel study, Cornell & Damodaran (2020) apply a DCF model framework to show how the relationship between sustainability and financial performance can be reinforced by mutual positive influence. If the degree of ESG can be used as a measure of business quality, then companies with a strong ESG focus should be able to achieve higher growth and improved profitability – thereby outperforming companies with low ESG rankings. All else being equal, this should enable companies with high ESG rankings to conduct better, more innovative, and more sustainable investments which will translate into recommenced growth and profitability (Cornell & Damodaran, 2020). The authors however regard the ESG hype with caution as they are not convinced that ESG initiatives will always promote shareholder wealth and research on the topic is rather ambiguous and inconclusive. The previous sections have now accounted for distinctive theoretical perspectives, each making contributions to the long-standing question of how ESG relates to financial performance. However, this thesis aims to construct ESG portfolios and assess the risk and return performance of these relative to market benchmarks.

Hence, subsequent sections will go into more depth with capital market theory.

3.4 Capital Market Theory

Capital market theory broadly defines an efficient marketplace where various entities trade financial securities with the aim to maximize risk-adjusted returns. The risk-return trade-off rests on the notion that stock prices fully reflect all information available and investors can therefore only achieve higher returns by accepting to take on more risk.

3.4.1 Efficient Market Hypothesis

The Efficient Market Hypothesis (EMH) was put forth by Eugene Fama (1970) stating that stock prices reflect all information, making it impossible for investors to beat the market and generate consistent alphas because stocks are already accurately priced and traded at fair value.

This makes it impossible for investors to exploit market inefficiencies by buying undervalued stocks and selling stocks with inflated prices. The economic rationale is that once information indicating that a stock is underpriced and hence provides an opportunity to derive profits, investors flock to purchase the stock and immediately bid up its price to a fair value, where only normal returns can be expected (Bodie et al., 2018). The only mechanism to obtain a higher return is through speculative investments posing substantially higher risk – hence the investor faces a risk-return trade-off.

(18)

The EHM has been commonly disputed by scholars pointing towards market anomalies that the EHM is struggling to explain and which directly contradict the foundation on which the EMH is derived. The hypothesis is a simplification of a highly complex reality with underlying dynamics that are difficult to capture in a theoretical model. To overcome this type of criticism, Fama (1970) proposed three versions of the EHM – weak, semi-strong, and strong form – all being varying degrees of the same theory. Theweak form insinuates that all data on historical prices are reflected in today’s stock prices and therefore investors cannot effectively utilize a trading strategy that predicts future stock prices. Besides past prices, the semi-strong form proclaims that current stock prices also incorporate all company-specific information available to the public. Subscribers to this version would argue that strong ESG characteristics would not help investors boost their returns and outperform the market benchmark since ESG data is publicly available and already accounted for in current stock prices. Only investors with access to insider information will be able to utilize excess returns. Finally, the strong form version of the EHM asserts that all information – both publicly available and information unknown to the public – is accurately priced into current stock prices and under no circumstances would investors be able to outperform the general market. Neither strong ESG profiles nor insider trading would then aid investors to obtain excess returns. However, this latter version of the EHM lacks empirical backing and can at best be used as a benchmark to test deviations from the market equilibrium (Fama, 1970; Fama, 1991). Despite ambiguity in the financial literature, investors continue trying to outperform the market by selecting the “right” stocks. Subsequent sections will elaborate on asset-pricing models trying to explain the variation in stock prices.

3.4.2 CAPM

The Capital Asset Pricing Model (CAPM) is a single-factor model that describes the relationship between systematic risk and expected return of risky assets. The CAPM was introduced in the 1960s by Sharpe (1964), Lintner (1965), and Mossin (1966). The model is built on the assumptions that (i) investors are rational mean-variance optimizers seeking the highest possible return for a given level of risk, (ii) investors have homogeneous market expectations, consistent with the assumption that all relevant information is publicly available, and (iii) all investors can borrow or lend at the risk-free rate allowing investors to also take short positions on traded securities (Bodie et al., 2018). These assumptions may be a rigorous simplification of reality, but the true market portfolio of all assets cannot be observed in the first place and therefore the EMH cannot really be tested directly. Many scholars have shown that it is beyond the ability of most investors – even professional ones – to generate returns that consistently outperform the market. Due to its central implication, that risk premia are proportional to exposure to systematic risk and independent of idiosyncratic risk, the CAPM remains widely accepted throughout the investment industry despite its empirical shortcomings (Bodie et al., 2018). The general CAPM was first expressed by Sharpe (1964):

(19)

E[Rit] =Rf t+βi(Rmt−Rf t), (3.1) whereE[Rit] is the expected return of assetiin periodt,Rf tis the risk-free rate in periodt,βi

measures the systematic risk of asseticompared to the market portfolio, andRmt is the return on the market portfolio. The CAPM implies that the expected return on any individual asset or portfolio is equal to the risk-free rate plus the product of the systematic risk component (beta) and the risk premium on the market portfolio. The beta coefficient measures the covariance of asset i with that of the market portfolio and is calculated as follows:

βi= Cov(Rit, Rmt)

σ_m² , (3.2)

whereCov(R_it, R_mt) is the covariance between assetiand the market, and σ_m² is the variance of the return on the market portfolio. So, the extent to which asseticovaries with the market will relate to the expected return of that asset. By definition the market beta will be equal to one. Assets that fluctuate more than the market – i.e. beta above one – will correspondingly face higher systematic risk exposure, which should be compensated for by yielding a higher expected return. Vice versa, assets being exposed to less systematic risk – i.e. a beta below one – will generate lower expected returns. The expected return-beta relationship can be portrayed graphically as the security market line (SML) in Figure 3.1 (Bodie et al., 2018).

Figure 3.1: The Security Market Line (SML)

(20)

As already touched upon in the previous section, the rationale underlying the CAPM is formed on the notion that capital markets are efficient and stocks are accurately priced. This implies that deviations from the CAPM would immediately be corrected for by the market, bringing stock prices back towards the SML equilibrium. Investors will relentlessly buy undervalued stocks, which will increase prices and adjust expected returns downwards. The same price mechanisms will cause investors to short sell overvalued stocks, which will drive prices downwards and boost expected returns. Empirically, this relationship is often tested through statistical analysis, regressing the excess return of a given asset on a market index that ade- quately represents the full market portfolio. In order for CAPM to hold, one should expect to observe values of alpha for any group of assets to cluster around zero since all variation in stock returns supposedly is captured by the systematic risk component (Bodie et al., 2018).

However, this rather simple model has often been criticized for not being able to capture all variations in stock returns. Hence, the next sections will provide an overview of asset-pricing models based on a multi-factor framework.

3.4.3 The Arbitrage Pricing Theory

The explanatory power of single-factor models, such as the simple CAPM, has often been challenged for not being able to explain all variations in stock returns. As an alternative, the Arbitrage Pricing Theory (APT) was developed by Stephen Ross (1976), a multi-factor model that aims to advance the legitimacy of single-factor models by explicitly encompassing various macroeconomic risk factors that will allow for a more accurate prediction of stock price variation. The APT relies on three key assumptions; (i) asset returns can be explained by a factor model, (ii) there is a sufficient number of securities to diversify away idiosyncratic risk, and (iii) well-functioning capital markets do not consent to the persistence of arbitrage opportunities (Bodie et al., 2018). Whereas the CAPM assumes capital markets to be perfectly efficient, the APT assumes windows of arbitrage to exist. Arbitrage opportunities are present when investors can capitalize on securities being mispriced and construct a zero-net-investment portfolio that will yield a certain and risk-free profit. However, this window of opportunity is temporary because high trading volumes will cause a price correction until such arbitrage is precluded. Hence, the APT does not entail the same restrictive propositions as the CAPM.

Nevertheless, the extent to which the no-arbitrage condition is satisfied is important for the functioning of capital markets and also tells us something about investors’ opportunities to exploit market inefficiencies. If for instance, firm-specific ESG data is not immediately reflected in market prices, the no-arbitrage condition does not hold, and investors can earn risk-free profits by utilizing these temporary inefficiencies before markets restore to equilibrium and satisfy the no-arbitrage condition. The APT by Ross (1976) is a good starting point for forming more multifaceted models that accommodate more factors.

(21)

3.4.4 Multi-Factor Models

As mentioned earlier more factors can be included to better account for variations in stock returns, and one suggestion is to use variables that empirically have predicted average returns well and may thus reflect the various risk premia (Bodie et al., 2018). One example is the three-factor model by Fama & French (1996) that includes additional risk factors to mimic the risks not accounted for in more simple models. In addition to general market exposure, the authors also look at how sensitive the return of an asset is to the SML and HML risk factors.

The model has come to dominate much of the empirical research conducted on security returns and is expressed as follows:

R_it−R_{f t}=α_i+β₁(R_mt−R_{f t}) +β₂SM B_t+β₃HM L_t+_it (3.3) SMB is a shortening for “Small Minus Big” and measures the excess return of a portfolio of small-cap stocks relative to a portfolio of large-cap stocks. Fama & French (1996) find that small-cap stocks tend to outperform large-cap stocks. Similarly, Banz (1981) examines the empirical relationship between the return and the total market value of NYSE common stocks and finds that smaller firms on average tend to have higher risk adjusted returns than larger firms. HML is an abbreviation for “High Minus Low” and measures the excess return of a portfolio of stocks with a high book-to-market ratio relative to a portfolio of stocks with a low book-to-market ratio. Fama & French (1993) suggest that companies trading at lower P/E multiples tend to generate excess returns, or equivalently that value-stocks tend to outperform growth-stocks. The authors advocate that the additional factors more accurately capture risk and therefore better explain stock returns. Whether these particular risk factors, or more generally, how risk factors should be identified in the first place, has been subject to a lot of ongoing debate in the academic field, as the findings of Fama and French either point to market anomalies or suggest that risk factors determined based on past evidence are correlated with other factors that are difficult to specify (Bodie et al., 2018).

The paper by Titman & Jegadeesh (1993) documented that there was a premium for investing in high momentum stocks, implying that strategies which buy stocks that have performed well in the past and sell stocks that have performed poorly in the past, generate significant positive returns over 3-to 12-month holding periods. Carhart (1997) finds similar results where funds with high returns in the past also yield higher average expected returns the following year, yet this pattern seems to fade away thereafter. In his paper, he extends the three-factor model to include the momentum factor for asset pricing of stocks, so it becomes:

R_it−R_{f t}=α_i+β₁(R_mt−R_{f t}) +β₂SM B_t+β₃HM L_t+β₄M OM_t+_it (3.4)

(22)

Carhart (1997) argues that this will enhance the explanatory power of asset pricing models in explaining variations in stock returns. Nonetheless, the four-factor model does not add quality investment factors, which the three-factor model had been criticized for leaving out. To overcome such criticism, Fama & French (2015) extended their three-factor model by adding another two risk components – namely RMW and CMA – and now propose a five-factor model directed to capture size, value, profitability, and investment patterns in average stock returns on top of the equity risk premium on the market portfolio.

R_it−R_{f t}=α_i+β₁(R_mt−R_{f t}) +β₂SM B_t+β₃HM L_t+β₄RM W_t+β₅CM A_t+_it (3.5) RMW is a shortening for “Robust Minus Weak” and measures the excess return of a portfolio of stocks with robust earnings relative to a portfolio of stocks with weak earnings. Fama &

French (2015) find that this risk factor measures the profitability of a company and suggest that companies with high-quality earnings tend to outperform companies with earnings of a weaker quality. CMA stands for “Conservative Minus Aggressive” and measures the excess return of a portfolio of stocks with conservative capital investments compared to a portfolio of stocks with aggressive capital investments. The evidence seems to suggest that the former tends to outperform the latter. The inclusion of these two additional risk factors enhance the theoretical foundation on which excess stock returns can be examined. As argued by Roll and Ross (1980) it is difficult to figure out which factors to include and this discussion will likely remain a hot topic for years to come. This section has now gone through some of the most acknowledged asset pricing models, which will be drawn upon later in the analysis.

3.4.5 Portfolio Choice Theory

Harry Markowitz (1952) pioneered modern portfolio theory by describing how risk-averse investors can construct portfolios to maximize expected return based on a given level of market risk. Investors make investment decisions based on a mean-variance analysis enabling them to form optimal portfolios that weigh risk against the expected return. Investors are risk-averse and prefer less risky portfolios to riskier alternatives for a given level of expected return, or equivalently, portfolios with a higher expected return for a given level of risk. A portfolio is said to be mean-variance efficient when there is no alternative portfolio with a higher expected return given the level of risk. Rational investors are assumed to be mean-variance optimizers meaning that they will always choose a portfolio that is mean-variance efficient (Bodie et al., 2018). The efficient frontier is the graphical representation of a set of portfolios that maximize expected return for each level of portfolio risk – i.e. mean-variance efficient portfolios. Mean- variance efficient portfolios are found on the upward-sloping part of the efficient frontier, which can be seen in Figure 3.2. The two-fund separation theorem states that any portfolio on the efficient frontier can be generated by holding a combination of the minimum-variance portfolio

(23)

Figure 3.2: The Efficient Frontier

Source: Munk (2017)

and the maximum-slope portfolio (Munk, 2017). The minimum-variance portfolio refers to the portfolio of risky assets that yields the lowest possible variance that can be attained for a given expected return of a portfolio. Risky assets are assets that bear an element of risk and is hence not risk-free. The variance in stock returns is exposed to both systematic risk and the firm- specific risk inherent in a particular investment due to its unique characteristics. However, the idiosyncratic risk component can be diversified away by adding uncorrelated risky assets to an investor’s portfolio. The fundamental purpose of pooling risky assets is to minimize the overall risk profile of the portfolio without compromising the level of expected return. The maximum- slope portfolio refers to the combination of risky assets that results in the steepest slope (Munk, 2017). Furthermore, the theorem tells us that the asset allocation decision can be separated into two steps. First, the optimal risky portfolio must be determined. For all investors, regard- less of risk aversion, the optimal risky portfolio is the tangency portfolio, which maximizes the average excess return on a portfolio over the risk-free rate divided by the standard deviation of the portfolio’s excess return (Sharpe ratio). In the second step, investors must decide upon the mix of the tangency portfolio and the risk-free asset, which will depend on the investor’s attitude towards risk (Kim & Boyd, 2008; Bodie et al., 2018). Hence, the efficient frontier can be derived by any combination of the minimum-variance and the maximum-slope portfolio.

(24)

To put this in context, it is noteworthy to look at how the efficient frontier is affected once we start to impose constraints. One example could for instance be not to let investors engage in any short selling of assets. From Figure 3.2 it is evident how such a constraint would cause the efficient frontier to shrink. The efficient frontier with the no short selling constraint is denoted

“frontier, risky – no short” (Munk, 2017). Once we start to impose portfolio constraints, it is no longer possible to construct portfolios that perfectly utilize the covariance matrix between risky assets, and therefore the risk-return trade-off of the constrained frontier will only be sub- optimal. Portfolios subject to one or more constraints will at best remain unchanged, but it is more likely that they will have lower expected returns for any given level of risk, whereby investors suffer a utility loss and are left worse off. Hence, rational investors with the objective to optimize mean-variance portfolios would always prefer a portfolio on the unconstrained efficient frontier since this would typically generate a more lucrative risk-return trade-off – at least it will not be harmful to the investor. The exact same rationale applies when constructing portfolios based on ESG data. Making positive and/or negative ESG screenings is no different from the no short selling constraint in terms of how it will affect the efficient frontier.

This chapter has now offered an account of different theoretical perspectives that each has something to say with regards to how sustainability measures and ESG proxies relate to the financial performance of a company. Moreover, this chapter has provided a detailed overview of some of the more theoretical models that can be used to explain why or why not high ESG-rated companies yield abnormal returns.

(25)

MASTER THESIS - CAND.MERC 4 LITERATURE REVIEW

4 Literature Review

During the course of past decades, a vast number of economic and financial scholars have insistently attempted to pinpoint whether high-performing ESG stocks deliver higher expected returns. Most studies have looked at how ESG information is captured in quantitative asset pricing models, and what (if any) alpha generating properties ESG data contains after con- trolling for common risk factors (Breedt et al., 2019). The results found in the literature is somewhat ambigious and despite being widely discussed, no consensus among scholars has yet been reached. This chapter briefly reviews a selection of academic literature and provide insights to the state-of-the-art empirical evidence on ESG as a means to higher risk-adjusted returns.

First, it may be worthwhile to look for possible explanations to the ambiguity in the existing literature. One explanation for the inconsistency in the literature stems from the absence of a uniform perception of what is meant by ”good” and ”bad” and how each ESG property should be measured. This problem has intensified in line with the increasing number of rating agencies. Li & Polychronopoulus (2020) identified 70 different data providers of ESG ratings by the end of 2019. This number does not even include the numerous investment banks, political institutions and research organizations that conduct ESG-related research used to produce customized ratings. Fish, Kim & Venkatraman (2019) document more than 600 different ESG ratings in 2018. The rating agencies not only differ in how they measure various ESG criteria, but also with respect to what criteria are deemed worthy of measurement. This implies that empirical results of the potential financial benefits of ESG can really differ, depending on which ESG methodologies being used. There is a substantial number of scholars documenting such divergence of ESG ratings, which clearly contributes to the flawing of the theoretical debate (Berg et al., 2019; Chatterji et al., 2016; Dortfleitner et al., 2015; Semenova & Hassel, 2015). It is therefore crucial to understand the underlying methodologies and results should be interpreted with caution.

4.1 Empirical Evidence on Neglected Stocks

One strand of literature describes how stocks with high ESG scores might exhibit underper- formance compared to low ESG stocks due to certain demand effects. This is in line with the argument put forward by Merton (1987) who points out that when certain stocks are neglected by a large group of investors, they may end up becoming undervalued. It could for instance be low ESG stocks being ignored by a vast majority of investors for not sufficiently living up to sustainability standards. Perhaps a little counterintuitive, these undervalued stocks will yield higher returns relative to high ESG stocks due to the market mispricing. This would not change even if the undervaluation of low ESG stocks is permanent, because a low stock price reflects a

(26)

high dividend/price ratio that will drive up returns (Hvidkjær, 2017). In addition, so-called sin stocks often shunned by ESG investors, have incentives to follow very conservative accounting procedures because these sin stock industries¹ fall under considerable scrutiny from regulators (Berman, 2002; Hong & Kaperczyk, 2009). This contradicts the hypothesis that companies with high ESG scores outperform the low-rated counterparts. In fact, this market anomaly of neglected stock outperformance seems to be well documented among scholars.

Hong & Kacperczyk (2009) is arguably the most prominent and most cited paper on the return effects of negative screening. By analyzing US stocks, they find that the sin stocks (in their paper defined as tobacco, alcohol and gambling firms) are held by relatively few institutional investors and followed less by financial analysts relative to a control group of stocks (Hvidkjær, 2017). Using Merton’s (1987) argument that stocks neglected by a large segment of investors will tend to have depressed prices, hence higher future returns, the authors find sin stock outperformance of 3-4% per year relative to control groups. They use a long sample period spanning from 1926-2006 and run standard Fama-French factor regressions to arrive at these results. Fabozzi, Ma & Oliphant (2008) use a one-factor model to analyze the returns of 267 stocks from alcohol, tobacco, biotech, defense and adult entertainment industries across 21 countries and find that sin stocks exhibit abnormal returns during the sample period 1970- 2007. Trinks & Scholtens (2017) reach similar results showing that sin stocks exhibit high returns during 1991-2012 in several international markets based on a large sample consisting of 1634 stocks across 94 countries. Instead of excluding particular industries, the authors select at the individual stock level. Hoepner & Schopohl (2016) conducted a very interesting and revealing research study of exclusionary screenings, where they use the CAPM to examine whether portfolios of excluded companies generate abnormal returns relative to their respective benchmark indices. More specifically, they look at the performance of stocks excluded from the Swedish AP-funds and the Norwegian Government Pension Fund-Global during the period 2001-2015. They also follow a norm-based screening process rather than a sector-based. Their results suggest that these portfolios of excluded stocks generate alphas that are statistically significant.

4.2 Empirical Evidence in Favor of ESG

The majority of academic research however seems to subscribe to the notion that some form of ESG integration has a positive impact on corporate financial performance (Nagy et al., 2016;

Friede et al., 2015). In general, the main argument in favor of ESG outperformance is that the stock market underreacts to ESG information. This implies that positive ESG information is not fully reflected in stock prices, which leave companies with strong ESG characteristics undervalued and investors can obtain abnormal returns by investing in these stocks (Hvid-

1Companies in alcohol, tobacco, weapon, fossil fuels, gambling, and pornography industries

(27)

kjær, 2017). The underreaction hypothesis is plausible and supported by evidence from Ball &

Brown (1968), Bernard & Thomas (1989), and Jegadeesh & Titman (1993). Moreover, ESG information is mainly categorized as an intangible asset, which does not appear directly on the balance sheet. Hence, it may be less salient to investors, explaining why the valuation of such intangibles is typically more uncertain than for tangibles. This strengthens the propositions of the underreaction hypothesis. Another common argument for ESG outperformance is that demand for ESG stocks have been driven up as a result of these stocks becoming more and more popular. That is, a growing demand for a particular set of stocks can push up the prices of those stocks, even without any new fundamental information about the value of these stocks (Hvidkjær, 2017). There seems to be a bulk of literature suggesting that investors can capitalize on pursuing ESG investment strategies.

Using data from KLD (now MSCI), Kempf & Osthoff (2007) set up long-short value- weighted portfolios of stocks from the S&P500 and DS400 in the period 1992-2004. Applying a four-factor asset pricing model, they find annual alphas that are significantly positive of around 5%. Based on data spanning from 1992-2007, Statman & Glushkov (2009) reach similar infer- ences. Borgers et al. (2013) report that the ESG outperformance does not seem to continue when extending the sample period of Kempf & Osthoff. They prolong the sample period to 1991-2009 and use a diverse set of selection criteria to construct long-short portfolios with some being weighted according to value and others being equally weighted. Their findings are robust to changes in the ESG measure and show four-factor alphas that are significantly positive until 2004 after which they become statistically insignificant centering around zero. Likewise, Halbritter & Dorfleitner (2015) perform an analysis that confirms the findings of Borgers et al.

The authors moreover examine stock returns based on different data providers. In their sample, data provided by ASSET4 spans from 2003-2012, whereas data from Bloomberg is included in the period 2006-2012. For all data providers, they find alphas that were generally insignificant and thus consistent with the MSCI results for the same period. However, to test the robustness of the findings, the authors also conducted Fama-MacBeth cross-sectional regressions, which showed results that were somewhat different. While alphas based on MSCI data were still insignificant, the alphas based on ESG data from both ASSET4 and Bloomberg were highly significant and positive. Once again this highlights the importance of interpreting results with care since results may differ depending on the underlying ESG methodology. Larsen (2016) presents a strong positive correlation between ESG scores from MSCI and realized returns during 2012-2016. On top of this, he finds that high ESG-rated stocks tend to demonstrate lower standard deviations in the return. These findings are confirmed by Lodberg, Nielsen & Zobbe (2020) who examine the relationship between ESG performance and financial performance over the period 2008-2020. Besides showing higher expected returns and lower standard deviations, they also suggest that sustainable portfolios are more resilient in times of crisis.

(28)

4.3 Investor Return when Decomposing ESG

When it comes to ESG performance, investors typically assume that the G-pillar is more robust and more important than the E and the S counterparts. Derwall, Guenster, Bauer &

Koedijk (2005) analyze the stock returns based on an eco-efficient investment strategy over the period 1995-2003. Their results suggest that more eco-efficient companies exhibit higher stock returns than their less eco-efficient equivalents. However, their results show lack of robustness, which is not very surprising given the short sample period. Specifically, the long-short alphas from the one-factor model are insignificant, whereas four-factor alphas are significantly positive at a 5% level. In a more recent study, the same authors apply the Tobin’s Q ratio to find a correlation between the level of eco-efficiency and the operating performance and equity valuation of a corporation (Guenster et al., 2010). Thus, the results can be used to explain the outperformance documented in their 2005 study – either eco-efficient companies were undervalued, or they became overvalued during the period of study. Giese et al. (2020) finds that the answer to whether governance is more important depends on the time horizon. Over a time-horizon of a one-year period it turned out that the G-component showed the strongest financial result.

However, the authors also looked at longer periods of time. Using a 13-year period to back-test the performance of the individual ESG pillars they find much more balanced results. In fact, all three pillars have showed outperformance when comparing the top quintile to the bottom quintile of each component during this 13-year study period. They explain this by pointing to two types of ESG key risk. The first type is referred to as event-driven risks, which are key risks that describe the likelihood of companies suffering in severe incidents such as a fraud cases or oil spills, which can hit the stock price in the short run. The second type of risk, which materialize over longer periods of time are called erosion risks because they can lead to erosion of the stock price over many years. The authors find that the G-score has a higher proportion of event-driven key risk issues, explaining why governance shows stronger financial relevance in the short run. Similarly, they find that E and S-scores have a higher proportion of erosion key factors explaining why these are more long-term in their financial relevance.

There are also scholars in the literature suggesting a positive relationship between the social pillar and the financial performance of a company. Edmans (2011) convincingly reports that firms with high employee satisfaction yield higher future stock returns. By using a value- weighted portfolio of the 100 best companies to work for in the US, he is able to show four- factor alphas of 3.5% from 1984-2009, which is 2.1% greater than industry benchmarks. Results remain robust when outliers are removed, and particular firm characteristics and different weighting methodologies are controlled for. Edman rationalizes his findings by pointing to the fact that markets fail to fully absorb the intangible ESG information. He moreover expects this mispricing to be corrected for once the intangible information becomes more obvious for market participants and the effects start to materialize through higher earnings.

On a Trajectory to Sustainable Investing: Is Sustainability and Proﬁtability Mutually Exclusive?