Government Intervention in Esg Investing: Financial Effects of France’S Energy Transition for Green Growth Act

GOVERNMENT INTERVENTION IN ESG INVESTING: FINANCIAL EFFECTS OF FRANCE’S ENERGY TRANSITION FOR GREEN GROWTH ACT

Elizabeth Ritger

An honors thesis submitted to the faculty of the Kenan-Flagler Business School at the University of North Carolina at Chapel Hill

Chapel Hill 2020

Approved by:

______(Dr. Riccardo Colacito, Ph.D.)

ABSTRACT

Elizabeth Ritger

Government intervention in ESG investing financial effects of France’s Energy Transition for Green Growth Act

(Under the direction of Dr. Riccardo Colacito, Ph.D.)

Environmental, social and governance (ESG) investing is gaining popularity in the finance sector by incorporating positive social impact and quantitative and qualitative risk factors. Investors using ESG strategies are assessing the impact of this shift on portfolio returns, and governments are regulating ESG data reporting. This study builds on prior research to (1) determine the impact of government reporting standards on ESG investment outcomes and long-term returns of portfolios and (2) address changes in the returns of ESG-focused investment strategies using the largest French companies before and after the implementation of the Energy Transition for Green Growth Act, a 2016

French law impacting ESG reporting standards. The results outline differences in returns between high and low-scoring ESG portfolios in France and changes in these portfolios before and after the regulation passed. The report concludes that France’s regulation increased average ESG scores but did not have a statistically significant impact on the average returns from ESG-focused portfolios.

ii AWKNOWLEDGEMENTS

I would like to thank the following people for their support throughout my honors thesis research and writing process. My project would not have been possible without your help along the way.

• Dr. Ric Colacito: Thank you for helping me with each step and teaching me so

much about research and data analysis.

• Dr. Kristin Wilson and Dr. Olga Hawn: Thank you for taking the time to serve on

my committee and offer valuable advice and feedback.

• Dr. Patricia Harms: Thank you for your constant encouragement and belief in my

success, even when I felt like quitting.

• Kim Miramontes and Erin Sanzone: Thank you for always making me smile when

work became overwhelming.

iii

TABLE OF CONTENTS

ABSTRACT ...... ii AWKNOWLEDGEMENTS ...... iii TABLE OF CONTENTS ...... iv LIST OF FIGURES ...... vi LIST OF TABLES ...... vii INTRODUCTION ...... 1 LITERATURE REVIEW ...... 4 Introduction ...... 4 Defining ESG investing ...... 6 Quantifying and standardizing ESG scores ...... 7 Government intervention in sustainability practices ...... 9 ESG’s Financial Impact ...... 12 Non-negative correlation ...... 14 Negative correlation ...... 15 Conclusion ...... 16 RESEARCH METHODOLOGY...... 18 Introduction ...... 18 Datasets ...... 18 Qualifying Subjects ...... 20 Time period ...... 21 Procedure overview ...... 21 Equations ...... 23 Research limitations ...... 24 RESEARCH RESULTS ...... 26 Introduction ...... 26 Section 1: SBF 120 Companies ...... 27 Overview ...... 27 Conclusion ...... 31

iv Section 2: CAC 40 Companies ...... 32 Overview ...... 32 Conclusion ...... 36 Section 3: CAC 40 Companies – Environmental ...... 37 Overview ...... 37 Conclusion ...... 41 FUTURE RESEARCH OPPORTUNITIES ...... 43 APPENDIX ...... i REFERENCES ...... iv

v LIST OF FIGURES

Figure 1: SBF 120 ESG Portfolio Performance (High-Scoring Minus Low-Scoring) ..... 28

Figure 2: SBF 120 High-Scoring ESG Portfolio Performance ...... 28

Figure 3: SBF 120 Low-Scoring ESG Portfolio Performance ...... 29

Figure 4: CAC 40 ESG Portfolio Performance (High-Scoring Minus Low-Scoring) ...... 33

Figure 5: CAC 40 High-Scoring ESG Portfolio Performance ...... 33

Figure 6: CAC 40 Low-Scoring ESG Portfolio Performance ...... 34

Figure 7: CAC 40 Environmental Portfolio Performance (High-Scoring Minus Low-

Scoring) ...... 38

Figure 8: CAC 40 High-Scoring Environmental Portfolio Performance ...... 38

Figure 9: CAC 40 Low-Scoring Environmental Portfolio Performance ...... 39

vi LIST OF TABLES

Table 1: SBF 120 Portfolios Pre-Legislation Annualized Summary ...... 30

Table 2: SBF 120 Post-Legislation Data Annualized Summary ...... 30

Table 3: CAC 40 Pre-Legislation Annualized Summary ...... 35

Table 4: CAC 40 Post-Legislation Annualized Summary ...... 35

Table 5: CAC 40 Environmental Pre-Legislation Annualized Summary ...... 40

Table 6: CAC 40 Environmental Post-Legislation Annualized Summary ...... 40

Table 7: SBF 120 portfolio compositions ...... i

Table 8: CAC 40 portfolio compositions ...... ii

Table 9: CAC 40- Environmental portfolio compositions ...... ii

Table 10: Examples of ESG metrics per category ...... iii

vii

INTRODUCTION

Environmental, social and governance (ESG) focused investing is changing the way consumers and investors analyze companies and quantify impact. Driven by consumer demand for investment portfolios that provide both positive social impact and adequate financial performance (Bannier, Bofinger, & Rock, 2019), investment managers adopted ESG factors into their strategic practices. ESG scoring emerged to help investors analyze a company’s social impact. Data providers use indicators such as carbon emissions, diversity statistics and worker’s rights to produce these scores (Huber,

Comstock, & Davis Polk & Wardwell LLP, 2017). The subjective nature of ESG indicators results in a lack of standardized methodology and complicated variables that are difficult to uniformly measure. The introduction of ESG scores in investment decisions necessitates a regulation system.

The debate over the effectiveness of using ESG data as a driver of investment decisions increases demand for a relevant regulation system. While some researchers claim ESG investing could drive positive long-term financial performance (Friede, G.,

Busch, T., & Bassen, 2015), others are less optimistic about the use of ESG metrics as a factor in portfolios (Auer and Schuhmacher 2016). The debate over the effectiveness of

ESG investing prompted much of the research done in the industry; however, the lack of consistent ESG data increases the confusion as to the effectiveness of ESG strategies and

1 causing investors to struggle to produce the most financially and socially efficient outcomes.

As it becomes clear that current ESG data providers may front underlying inefficiencies, developed countries, particularly in the European Union and the United

States, are considering implementing government reporting standards for ESG metrics.

Regardless of its specific reporting standards, these laws raise awareness and scrutiny of

ESG behavior within companies. Many countries acknowledge the benefits of ESG reporting standards, but few governing bodies implemented these standards as thoroughly as France. France publicized the Energy Transition for Green Growth Act early in 2015 and the legislation gained momentum at the 2015 United Nations Climate Change conference, leading to global attention (Balibar 2017).

A notable and unique section of this law is Article 173-VI, a comply or explain government policy passed in 2015 as part of the Energy Transition for Green Growth

Act. This article of the law requires institutional investors to report specific ESG metrics of their portfolio management and validates them along set criteria (Evain, Cardona, &

Nicol, 2018). Article 173-VI is the first set of legislation to directly address ESG metrics within investing. By setting standards for investors to publicize their ESG activities,

Article 173-VI both increased awareness of ESG practices within French businesses and pressured French investors and companies to publicize and incorporate ESG metrics into regular business activities. A 2017 study of Article 173-VI reporting companies in France found that 91% of investors incorporated ESG factors into their investment strategies. Of these, 52% used France’s Socially Responsible Investment (SRI) label as an indicator while 22% used the Energy and Ecological Transition for the Climate (EETC) label

2 (Delérable, Gazzo, & Perez, 2017). These reports investigated the participation by ESG- focused companies in France following Article 173-VI’s implementation, but data is lacking on the correlation between these reporting standards and the financial performance of these investment strategies.

Addressing the impact of policy implementation, my research explores the following question: is there a difference in the financial performance of ESG-focused public companies in French markets before and after Article 173-VI was passed?

Research on this topic can guide policy in countries seeking to implement similar ESG reporting controls by evaluating the financial performance of companies in France before and after policy implementation. This research can be expanded to aid governments in producing the most efficient and effective legislation for encouraging investment in ESG- focused companies.

LITERATURE REVIEW

Introduction

Environmental, social and governance (ESG) focused investing is disrupting the financial industry through its cascading effects on management and investment decisions1. ESG investing, a strategic investment approach combining traditional financial performance and non-financial data, is the financial industry’s response to increased consumer demand for positive social impact in developed markets. The majority of millennials, the generation representing a new wave of investors, are concerned with their social impact (Klober, 2015) and the concept of ESG-based portfolios offers investors both desired positive social effects and adequate financial performance to sustain such investments. These strategies are rapidly gaining popularity, with socially responsible investments (SRIs) making up approximately a quarter of US investments and half of European investments in 2019 (Bannier, Bofinger, & Rock,

2019). In addition to conventional tracking of financial performance to establish investment strategy, ESG investors use ESG scores as a factor to construct portfolios for their clients. These scores are assigned to individual companies and industries, using data

1 ESG investing is a term increasing in popularity within financial literature. It is often interchangeable with the terms socially responsible investing (SRI) and impact investing and will be treated as such in this paper (Boerner, 2012)

4 such as carbon emissions, diversity and worker’s rights to rate potential attractiveness

(Huber, Comstock, & Davis Polk & Wardwell LLP, 2017)2.

While certain ESG scoring factors such as diversity statistics and employee pay are evaluated on a standardized basis, many require analysis subject to interpretation by individual investment professionals or firms. Research within ESG is in the relatively early stages, leading to emerging errors in research, lack of standardized methodology and complicated variables that are difficult, if not impossible, to uniformly measure.

The introduction of ESG scores in investment decisions necessitates a regulation system. As it becomes clear that a seemingly harmless trend may front underlying inefficiencies, developed countries, particularly in the European Union and the United

States, consider implementing government reporting standards to standardize factors affecting ESG scores. The European Union is leading this trend with the United Nations’

Global Compact. This law encourages, but does not require, government-regulated reporting by public companies and provides access to uniform ESG data sources (Ortas,

Garayar & Álvarez, 2015).

France leads ESG regulation trends through Article 173-VI, a unique comply or explain government policy passed in 2015 as part of the Energy Transition for Green

Growth law (Giamporcaro, Gond, & O’Sullivan, 2019). This law builds upon the UN’s

Global Compact with additional reporting standards for public French companies and investors. Article 173-VI, among the first of its kind to establish a standardized reporting

2 These three categories of data points are universally recognized and included across ESG leaders, however there are many factors that vary and will be discussed in later sections (Huber, Comstock, & Davis Polk & Wardwell LLP, 2017).

5 system, is designed to encourage reporting of ESG practices by public companies and implementation of ESG investing by investment institutions.

Defining ESG investing

Once a niche subject area for socially conscious extremists, ESG investing is quickly moving into mainstream space, consequently altering key drivers among participants (Jemel-Fornetty, H., Louche, C., & Bourghelle, D., 2011). According to research done in 2011, this shift primarily indicates that, compared to pioneer movements, ESG strategies today revolve around business outcomes as opposed to ethical considerations. Investors are not solely focused on the positive social outcomes of their portfolios, but on maximizing the cross-section between customer demand and financial return. Qualitative research on the goals of ESG-focused companies concludes that this shift from ethics-based practices to a focus on financial performance within ethics is weakening ESG investing’s positive social impact (Revelli, 2017).

The argument of business-driven versus ethics-driven approaches to ESG legal standards is an important consideration as more countries move towards implementation.

Ideally, future ESG investment reporting will provide transparent, uniform data that encourages positive social externalities from public companies and financially satisfies investors. The need for regulation sparks debates on the success of various control measures used to encourage ESG reporting necessary for informed investing. As countries led by France move towards implementing reporting laws for investment institutions, countries such as the United States are deviating from governed reporting

(Giamporcaro, Gond, & O’Sullivan, 2019).

6 Research evaluating the financial effects of ESG-focused policies is necessary to identify the most productive method for encouraging valid ESG investment. My research combines analysis of government regulation and ESG-focused portfolio returns to identify tangible financial effects of regulation for investors. Specifically, my data focuses on how the introduction of France’s Article 173-VI affected financial returns from public ESG-focused companies. Interpreting the results of my research requires an understanding of financial and regulatory studies on ESG investing and the implications for future government intervention. This literature review analyzes previous research debating ESG scoring mechanisms, measuring the effects of government regulation, and quantifying ESG’s financial impact for investors.

Quantifying and standardizing ESG scores

Standardizing reported scores and producing useful numerical outputs emerged as a heavily debated topic due to varying definitions of ESG factors among professionals and governing bodies as well as a lack of uniformity within ESG data organization

(Eccles & Stroehle, 2018). Although consumers and investment managers using ESG portfolios strategize according to reported scores, the variations in scores based on the provider results in mixed outcomes. Further complicating the ESG scoring process, ESG data and reporting became a market itself, subject to the nuances of consumer demand, profit-focused management and political intrusion. While many companies attempted to enter this market, a few prominent firms dominate market share (Huber, Comstock, and

Polk, 2017). This section provides an overview of key players in the market for reporting data and the effectiveness of current scoring methods.

7 ESG reporting companies use widely varying data points and rating systems to construct their overall ESG scores as evidenced in a report by Harvard's Huber,

Comstock, and Polk (2017). Several key players shape the market for reporting data, namely Bloomberg ESG Data Services, DowJones Sustainability Index, MSCI ESG

Research and Sustainalytics Company ESG Reports.

The Harvard report analyzed these data providers on a range of factors but the most relevant are methodology and rating scale. Methodology outlines how each provider evaluates companies and rating scale refers to the scoring system used (Huber, Comstock, and Polk, 2017). Excepting MSCI's letter rating system, all of the major providers used ratings scales out of 100. This finding provides a certain level of uniformity that indicates future adoption of this rating system across providers. Methodology showed diversity across the board. Although all providers pull ESG data from similar government and company documents, Huber, Comstock, and Polk found that each provider uses a distinct classification and weighting system. A 2015 study by Dorfleitner, Halbritter, and Nguyen had similar conclusions. These unique methodologies concur with skepticism surrounding ESG scoring principles and implies the need for increased regulation to validate the data.

In addition to the validity of ESG scores, critics often question how accurately these scores measure actual social impact, prompting research on the correlation between the two. Studies done on major players in the market challenge their scoring methods, often highlighting inefficiencies. Chatterji, Levine, and Toffel (2009) investigated MSCI scores and their correlation with each company's actual social impact. For example, the authors looked at correlations between KLD environmental scores and pollution

8 emissions. The authors found that KLD scores did not correlate with a more positive social impact. Chatterji, Levine, and Toffel credited KLD’s failure to weigh historical social impact as a major fault. Historical social impact was found to be a higher predictor of a company’s current social impact than company-reported practices and spending

(Chatterji, Levine, & Toffel, 2009). This study, while older than others I examined, is one of the first to question the validity of ESG scoring and its accuracy. The factors MSCI used to measure ESG in 2009 were consistent with the reported methodology from

Huber, Comstock, and Polk (2017), so these findings remain relevant.

Eccles and Stroehle (2018) conducted a similar and wider analysis of major ESG data providers, concluding that ESG scores more directly correlate with the values of the data provider than the companies being scored. The author’s conclusions tie back to the origin of discrepancies found in methodology studies. The lack of enforced regulation subjects ESG data to the agenda of providers instead of universally progressive ESG indicators. Within a subjective field such as ESG scoring, these inefficiencies negatively impact ESG investing’s overall social impact (Eccles & Stroehle, 2018).

Government intervention in sustainability practices

As researchers continually identify ESG reporting inefficiencies, the need for government intervention in these practices became clear. The United Nation's Global

Compact and France's Article 173-VI approach the concept of intervention, but investors across the globe are hesitant to comply with stricter regulations (Giamporcaro, Gond &

O'Sullivan, 2019). Research outlining the effectiveness of past government intervention in sustainability practices offers a clearer path for ESG regulation as it becomes a

9 prevalent topic of discussion. This section documents proven methods of government intervention in data reporting and historical government intervention trends in the French market, including initial findings from the implementation of Article 173-VI.

Several schools of thoughts exist relating to the logistics of government intervention within financial reporting. According to a historical analysis by

Giamporcaro, Gond and O'Sullivan (2019), the two main considerations for intervention in developed countries involve either ‘layering’ increasingly aggressive standardization measures over time or implementing strict standardized reporting laws. The authors found that, historically, both approaches produce positive results. This research provides clarity for proven techniques when implementing initial reporting laws.

Both of these standardization measures require agreement on the definition of collective ESG factors. Although some researchers argue that subjective ESG measures are impossible to define under the law, Sitkoff (2018) challenges this with his study relating ESG investing to the fiduciary duty law. Sitkoff (2018) looks at ESG investing as a legal construct, showing how standardization implementations could impact the way

ESG investments are managed. Under the fiduciary duty law, investors are required to make decisions in the best interest of their clients. There are debates as to whether ethical considerations should be taken into account or if these are subject to individual discernment (Sitkoff, 2018). Sitkoff found that, while ESG factors cannot currently be taken as a standard for 'best interest,’ the risk-adjusted returns and direct benefit of using

ESG criteria are often beneficial for the recipient. Therefore, this study indicates that legally defining measures can be based on or incorporated into the fiduciary law.

10 Critics of government regulation voice concerns that reporting standards encourage companies to find loopholes for compliance with reporting standards, thus counteracting the positive impact of ESG factors. Revelli (2017) theorized that government standards could exacerbate the negative effects of 'mainstreaming' ESG.

Revelli's paper implies danger in continuing to spread ESG investment strategies among investment professionals. This opposing viewpoint examines the precautions that should be taken into account when implementing any reporting standards. It also forecasts potential causes of negative impacts, an important factor when examining the effectiveness of reporting standards.

Within the progressive French SRI market, Crifo and Mottis (2013) found that increased legislation involving SRI reporting standards led to an increase in ‘the diffusion of SRI into French mainstream asset management.’ Their study, analyzing how the

French SRI market functions in relation to other large players and shows positive correlations between reporting standards and reporting standardization improvements.

The valuable analysis of not only financial performance but the rate of adoption of these strategies provides an additional dimension encouraging legal involvement within ESG investment markets (Crifo & Mottis, 2013). Article 173-VI falls under SRI implementation, indicating positive adoption prospects as the law matures.

Relating specifically to Article 173-VI, Evain, Cardona and Nicol (2018) looked at climate reporting under France’s Article 173 for French insurance companies in the second year of the law's implementation. The study found an improvement in the standards of reporting but lacked clarity on how positively investors viewed ESG strategies (Evain, Cardona, & Nicol, 2018). This study provides one of the first

11 examinations of the impact and implementation of Article 173-VI by examining qualitative date on improvements in reporting standards and the ways companies include them in business practices. Although this article focuses on insurance companies specifically, it includes a cross section between understanding and implementation that provides context into the interpretation and practice of the law (Evain, Cardona, & Nicol,

2018).

ESG’s Financial Impact

The correlation between a company's ESG score and its long-term corporate financial performance (CFP) is the most contentious issue among ESG researchers.

Research on the financial effects of ESG investing is of chief importance among present and future ESG investors in light of mixed results. While some investors are willing to sacrifice financial performance for greater social impact, positive financial returns remain a decisive factor to most consumers and investment managers (Bannier, Bofinger, &

Rock, 2019). Supporters defend long-term ESG investment approaches by claiming ESG- focused companies provide long-term stability for both financial and non-financial business functions while skeptics argue that increased investment in sustainable practices leads to lower returns. Bannier, Bofinger and Rock (2019) found data from these studies regularly display conflicting results showing positive, negative and neutral correlations between ESG scores and financial returns. As previously mentioned, the lack of standardization among scoring systems is a major cause of the variation in correlation results. However, the relevance of ESG-CFP correlation means current research on this subtopic is comparatively robust and formed patterns of methodology and discussion.

12 Friede, Busch and Bassan (2015) recognized emerging patterns in ESG-CFP correlation in an analysis of over two thousand academic studies on the financial impact of ESG performance. Attempting to consolidate results from widespread research in this field, the authors analyze relevant studies dating from 1970 in both vote-count studies and meta-analysis3. The paper recognizes a majority non-negative finding across combined studies, adding credence to generalized ESG support. More importantly, the authors broke down portfolio and non-portfolio-based methodologies to find a statistically significant difference in the results, with portfolio-based research providing a lower incidence of positive-trending ESG-CFP correlation. The majority of portfolio- based analyses produced neutral or mixed results. This study is the first to analyze this specific methodology divide and provides important insight. Portfolio-based studies include effects of market changes and investment structures that may confound exclusive

ESG causation but more accurately mimic realistic financial performance (Friede, Busch

& Bassan, 2015). The effect of these findings hints at ESG as an investment factor operating in cohesion with the external economic and management shifts present in traditional, non-ESG focused mutual funds. As the largest aggregation of studies in this subtopic, these results provide a focused lens for the wider field of studies.

I identified several notable studies offering a relevant discussion of ESG's financial impact on the market and representing the most prominent mixture of results in the field. Niche research areas in specific industries and markets reveal patterns that are an important part of these discussions however, for the purpose of my research, I chose to

3 Friede, Busch and Bassan use vote-count methods, a simple count of correlational results with less sophisticated quantitative effects, to provide a high-level view of their findings. However, their meta- analysis is far more robust and the authors account for the shortcomings of vote-count results.

13 analyze portfolio-based studies applicable to developed financial markets4 5. In this section, I will discuss the importance of results from studies producing non-negative and negative correlation between ESG factors and CFP.

Non-negative correlation

Socially conscious investors eagerly encourage adoption of ESG factors in investment decisions by demonstrating a positive correlation between high ESG scores and attractive CFP. Neutral or mixed results are often discussed as positive results supporting ESG investment practices because of the additional positive externalities provided by ESG factors. As evidenced in Friede's report, thousands of studies support result in a positive ESG-CFP relationship. Therefore, I have identified only studies with particularly strong methodology and perspectives.

Sherwood and Pollard (2018) establish robust positive ESG-CFP correlations using portfolio methodologies and MSCI scores in their seminal research paper. Their statistically significant results validate the use of ESG factors as a core consideration in investment strategy. The combination of Sherwood and Pollard's well-established portfolio approach and ESG metrics makes this publication noteworthy. Toledo and

Bocatto (2014) added to this argument by identifying ESG levels as a unique risk premium. Their study indicated that ‘high-ESG’ firms were valued at 13% higher in terms of premia than ‘low-ESG’ firms (Toledo & Bocatto, 2014). By identifying ESG

4 Portfolio-based methodologies are widely respected within financial research. They primarily use Fama- French factors to obtain financial performance metrics specifically focused on investment returns subject to normal market constraints (Fama & French 1992). 5 Developed markets in the European Union and the United States are most relevant to the focus of my research and therefore are most prominent in this discussion.

14 level as a risk premium, Toledo and Bocatto weave ESG scores into results with relevant implications for investors.

Sheridan Titman, a pioneer ESG researcher, provides similar positive results from an inverse angle. Cao, Titman, Zhan, and Zhang (2019) identified companies that fail to address ESG risks and their long-term stock returns. Titman found that lack of attention to or incorporation of ESG priorities into corporate strategy leads to 3.5% lower returns when adjusted for risk in the long-run. Unlike other studies, this paper takes the approach of examining companies with little focus on ESG investing, adding credence to the argument for ESG strategy by proving results from a new lens. By factoring in multiple angles and finding clear positive benefits, Titman's study adds validity to the need for enforcement of ESG-investment strategies for both socially and financially focused initiatives. The findings provide a more holistic view of the argument about the validity of ESG investment strategy that is useful when conversing on opposing viewpoints. The combination of mixed methodologies and consistently positive results from well- established authors provides strong data for ESG supporters.

Negative correlation

Research producing negative correlation between ESG factors and CFP highlight a less optimistic view of feasibly integrating social impacts into investment decisions.

While few recognized studies produce egregiously negative ESG-CFP relationships, all negative correlations caution against robust early adoption of ESG-focused investment strategies. The lack of standardized data and uncertainty surrounding future regulation present risks that could exacerbate negative financial performance. ESG skeptics discuss

15 the pitfalls of basing portfolio strategy primarily on today's ESG scores, particularly as reporting standards change in areas exploring government intervention.

ESG supporters often acknowledge these uncertainties by encouraging incremental adoption of ESG strategies based on specific non-financial factors, however the advantages of this method are not supported by data. Auer and Schuhmacher (2016) analyzed subcategories of ESG-CFP relationships by breaking ESG scores into three categories (environmental, social and governance) and using updated Sustainalytics data to create subsequent portfolios. Their methodology mirrors that of positively correlated

ESG-CFP research. The authors found a negative ESG-CFP correlation across all three categories in European markets and neutral correlation in the United States and Asia.

Their results challenge both the general discussion of positive effects of ESG factors and arguments for strategizing based on specific factors. Similar studies by Fatemi, Glaum, and Keiser (2017) corroborate Auer and Schuhmacher's findings.

The debate on ESG's financial impact provides insight into the volatility of findings on ESG-CFP correlation. For profit-seeking investors, these studies have a large impact on their desire to include ESG initiatives in investment decisions. Therefore, future research results on ESG portfolio's financial performance provide opportunities to encourage or dissuade widespread ESG adoption.

Conclusion

Unclear scoring methodologies, the need for government regulation and mixed financial performance results within an emerging industry provide increased research opportunities exploring interactions between government regulation and CFP within ESG

16 investing. Thorough research on existing laws contributes to progressive policy implementation as more countries look to regulate ESG investment reporting. The results of relevant research serve as motivation for implementing or amending ESG-focused investor reporting policies in developed markets across the world. By evaluating the financial performance of companies in a country encouraging ESG investment through reporting standards, I can make conclusions that shape policy in other countries seeking to implement similar controls. Reports released investigate the participation by investment institutions following Article 173-VI’s implementation, but data tying financial performance to any reporting standards does not yet exist.

RESEARCH METHODOLOGY

Introduction

The following section addresses the data collection and analysis techniques I used to assess correlation between government reporting standards, specifically those outlined in Article 173-VI of the Energy Transition for Green Growth Act, and ESG investment strategies in France. To gauge the financial performance of companies based on ESG metrics, I use a portfolio-based method to implement popular ESG investment strategies.

Specific data sets and time periods were chosen to best reflect the effects of the legislation. This section provides an overview of the data selected, research procedure and limitations.

Datasets

To properly conduct a portfolio-based approach, I aggregated annual ESG scores and historical stock prices for a specific set of French companies. I collected ESG scores and historical stock prices for selected companies using Bloomberg’s ESG dashboard and monthly historical adjusted closing prices from Yahoo Finance6. Yahoo Finance provides a reputable source for historical stock prices and monthly collection provided a robust sample of stock prices that could be annualized.

6 Adjusted stock prices account for various corporate actions such as dividends.

18 I chose Bloomberg’s ESG dashboard as my ESG data source because it has a consistent categorization of historical ESG scores that can be used to compare changes overtime. Bloomberg’s ESG dashboard provides a range of historical ESG scores from the past 12 years, both sourced by Bloomberg and reported by third-party agencies such as Sustainalytics. Bloomberg sources their data from public company filings, including

CSR reports, annual reports, websites and Bloomberg surveys, including requests for data

(Huber, Comstock 2017). The Bloomberg dashboard provides scores out of 100, making this information comparable to other companies and easily digested. Their focus on mid to large capitalization companies fits well within my subject range. By staying within a single source of ESG scoring, the ESG metrics will be most consistent. I specifically focused on the data Bloomberg sourced from Sustainalytics because of Sustainalytics’ reputability and consistency as a first-mover as an ESG data provider (Huber, Comstock

2017).

Bloomberg and Sustainalytics base ESG scores off a number of metrics, such as greenhouse gases per revenue and employee turnover, that are available through public company reports and surveys directly affected by the Energy for Green Growth Act7.

This database also produces individual scores for each category within the broader ESG industry. Therefore, each company has an individual environmental, social and governance score. Sustainalytics scores specifically take into account disclosure, preparedness and controversy surrounding ESG risk and opportunities from each individual company. Each company is scored on a scale from 0-100, with 100 being the highest score. A company receiving a higher score reported more favorable ESG metrics.

7 See Table 7 in the Appendix for a list of example calculations for each component of the ESG score.

19 These scores are grouped into the following risk categories: negligible (90-100), low (80-

89), medium (70-79), high (60-69) and severe (0-59). Investors have full access to the information comprising Bloomberg’s scores and investor attitude towards changes made to the information in these reports, specifically regarding ESG scoring metrics, would be reflected in the stock market.

Qualifying Subjects

To create a representative, robust sample with available data, I chose the largest public French companies based on performance within stock indexes. The companies included in my sample are public French companies that belong to the Société des

Bourses Françaises 120 Index (SBF 120) stock index and the Cotation Assistée en

Continu (CAC 40) as of December 31, 2019, screened within the Bloomberg terminal.

These two indices represent France’s highest performing companies. I chose to focus on public companies in these indices because the information on ESG scores are more widely available due to enforced standards of reporting, company maturity and consumer scrutiny. Additionally, companies within these indices are large enough to balance any risk associated with instability prominent in small companies.

Within the SBF 120 index, I screened each company within Bloomberg and

Yahoo Finance to assure that enough data was available for both ESG scores and historical stock prices. Some companies within the SBF 120 were eliminated due to lack of ESG scoring or stock price data. This left 77 companies included in the data pool. All

40 companies in the current CAC 40 index produced sufficient data and are included in each data set.

20 Time period

The time period I used as the cutoff for pre and post-legislation analysis is limited by when Article 173-VI passed. The Energy Transition for Green Growth Act went into effect in mid-2015 but Article 173-VI went into effect on January 1, 2016 (Article 173-

VI, 2016). I chose Article 173-VI’s passing as the cutoff date because this specific article had the most relevance for investors and the finance industry. Data was collected after the law from 2016-2019. After screening ESG data within the Bloomberg terminal, I focused on February 1, 2014 as a start date due to poor ESG scoring history before this date.

Therefore, I used a complete time period of February 1, 2014- December 31, 2019 to include three years before the law passed and three year after. The time period from

February 1, 2014- December 31, 2015 is considered ‘pre-legislation’ while the time period from January 1, 2016-December 31, 2019 is considered ‘post-legislation’.

Procedure overview

My research procedure involves collecting ESG scores, company information and stock price reports from Bloomberg’s ESG Sustainalytics dashboard. For each of the 77

French companies selected from the SBF 120 index, I collected an annual ESG score and monthly stock prices from February 1, 2014 to December 31, 2019. After gathering relevant data points, I grouped the companies based on ESG score, forming three mock portfolios for high, mid and low-scoring ESG companies. The portfolio method popularized by Fama & French (1992) is a data analysis tool for determining financial performance based on specific factors. For each portfolio within my data set, I calculated the average rate of return (r) as seen in Equation 1 below. After completing my calculations, I annualized the returns for each portfolio in pre and post-legislation time

21 periods. I then compared these two sets of portfolios, culminating in a final chart tracking changes in stock valuation from 2014-2019.

Following the creation of these three portfolios, I implemented a popular ESG strategy of investing ‘long’ in high-scoring ESG companies and ‘short’ in low-scoring

ESG companies. This created a ‘high-minus-low’ portfolio that subtracted returns from the low-scoring portfolio from those of the high-scoring portfolio. I collected monthly returns for this portfolio. These returns were annualized based on pre and post-legislation time periods. After testing the differences in ESG scores for each company pre and post- legislation, the sorted companies remained in their respective portfolios and, therefore, did not require rebalancing.

Additionally, I calculated the Sharpe ratio (Equation 2) of the ‘high-minus-low’ portfolio to assess the portfolio’s excess returns taking into account the risk-free rate of return and standard deviation for the same time period. The risk-free rate of return was calculated using the average returns of 10-year French government bonds for each time period (France Government Bond 10Y, 2020). Sharpe ratio calculations assume a .99% pre-legislation risk free rate of return and a post-legislation risk free rate of return of .6%.

Finally, I tested the pre and post-legislation average returns for statistical significance using a two sample mean t-test. This test provided evidence required to reject or not reject the null hypothesis. In this scenario, the null hypothesis states there is no significant difference between average annual returns for the ‘high-minus-low’ portfolio for pre and post-legislation returns.

To add robustness and eliminate excessive idiosyncratic risk, this procedure was repeated step-by-step with two additional filters. The first procedure isolated CAC 40

22 companies from the SBF 120 data set. Following this procedure, the companies were broken into three ESG portfolios evaluated by returns year after year, as discussed above.

The reason for isolating this specific index is to further assess the impact of legislation on highly scrutinized firms. By isolating the highest performing companies from the data set,

I could assess the impact of the Energy Transition for Green Growth Act and Article 173-

VI on the largest, most visible firms. Government legislation may have a greater impact on the returns for these key companies within the French market. Additionally, should these companies’ returns be significantly impacted by this legislation, this finding could indicate early adoption success from larger companies within the market.

The second filter further isolated specific factors within the data set by using only the environmental score for CAC 40 companies. Because the Energy Transition for

Green Growth Act primarily targets environmental metrics for reporting, this filter attempts to eliminate and test if targeting legislation disproportionately impacts specific factors within the broader ESG industry. Three portfolios were created based on each

CAC 40 company’s environmental score and follow the same procedure discussed above.

Equations

Equation 1

(�� ) − $1 � = $1

Equation 2 (Sharpe Ratio)

�(�) − �(�) �(�) = ��(�) x = the investment r(x) = average rate of return of x R(f) = Risk free rate of return

23 StdDev(x)=The standard deviation of r(x)

The outlined calculations are focused on a portfolio-based financial analysis using the principles of Fama-French market and portfolio studies as well as William Sharpe’s measure of adjusted risk to produce quantitative measures of ESG changes. The Fama-

French design takes market risk into account when evaluating stock returns (Wharton, n.d.).

Research limitations

The main limitation in my research is the lack of available data. Ideally, I would include data for longer time periods and more companies. Due to the recent nature of

ESG scoring systems, only larger companies are efficiently scored. Therefore, it is not currently feasible to produce a larger, more robust data set to analyze. Additionally, consistent ESG scores are only available starting in 2014. The short time period and lack of data available for more companies limits my ability to analyze the Energy Transition for Green Growth Act’s impact on a larger scale. Similarly, the small number of companies included in each portfolio limits my ability to balance risk within each portfolio.

Another limitation is inconsistency in how ESG scores are calculated and the analysis of global and national factors affecting various industry stock prices. ESG scores are a relatively new concept, taking place mainly within the past 15 years. Due to changes in definitions and the bias of company self-reporting, factors used to calculate

ESG scores may have shifted. Using a consistent source for ESG scores, Bloomberg, alleviates some of this volatility. However, the short history of such scoring systems adds some level of uncertainty to the validity of the ESG scores used.

24 Finally, the conclusions drawn from my research are limited by extraneous factors. The created portfolios are not large enough to eradicate random, unrelated variables that may affect certain companies. Any outliers, positive or negative, could have a disproportionately large affect on the overall rate of return. I did not identify any affecting outliers within these data sets, but this limitation remains important to note.

RESEARCH RESULTS

Introduction

This section analyzes and discusses the conclusions from the research I did to understand how France’s Energy Transition for Green Growth Act affected portfolio performances for ESG investment strategies. The data is divided into three sections, each with a different set of companies and focus. Section 1 creates three ESG-based portfolios for 77 companies from the Société des Bourses Françaises 120 Index (SBF 120), implements a ‘high-minus-low’ investment strategy and discusses the results and implications. Section 2 focuses repeats this process for all 40 companies of the Cotation

Assistée en Continu (CAC 40) to assess the index’s role as a key driver for the French

ESG market with highly scrutinized companies. Section 3 creates new portfolios from the

CAC 40 index using only their environmental score from the ESG reports to highlight specific factors targeted by the legislation.

The portfolio method used for each set of analyses follows an investment strategy of

‘long’ high-scoring ESG portfolios and ‘short’ low-scoring ESG portfolios. To implement this investment strategy, I created a ‘high-minus-low’ portfolio for each section by subtracting stock returns from the ‘low-scoring’ ESG portfolio from stock returns from the ‘high-scoring’ ESG portfolio. These portfolios are further divided into pre-legislation (2014-2015) and post-legislation (2016-2019) time periods as noted in the

26 methodology. The overall ‘high-minus-low’ portfolio is analyzed and discussed in these two time periods based on the following factors:

• Average ESG score

• Annualized returns

• Sharpe ratio

• Statistical significance

Section 1: SBF 120 Companies

Overview

In this section, I examine the annualized returns of three ESG-based portfolios composed of the 77 relevant SBF 120 companies in pre and post-legislation time periods to assess the impact of the Energy Transition for Green Growth Act on the effectiveness of using ESG metrics as a portfolio factor. I aggregated and annualized the company’s monthly stock prices from February 2014 through December 2019 as well corresponding annual ESG scores. The statistics are summarized both in graphs showing historical returns for each portfolio over time and tables summarizing the effectiveness of a ‘high- minus-low’ investment strategy over time8. The conclusion discusses the results and their implications for my hypothesis.

8 As discussed in the Research methodology section, the composition of each portfolio remains the same for pre and post-legislation analysis. For example, portfolios 1.1 and 1.4 are composed of the same companies using different time periods.

27 Figure 1: SBF 120 ESG Portfolio Performance (High-Scoring Minus Low-Scoring)

Stock Returns of High-Scoring Minus Low-Scoring Portfolio

10.0%

5.0%

0.0%

-5.0% Monthly Portfolio Returns

-10.0% 3/1/14 6/1/14 9/1/14 3/1/15 6/1/15 9/1/15 3/1/16 6/1/16 9/1/16 3/1/17 6/1/17 9/1/17 3/1/18 6/1/18 9/1/18 3/1/19 6/1/19 9/1/19 12/1/14 12/1/15 12/1/16 12/1/17 12/1/18 12/1/19 Date of Stock Price

Figure 2: SBF 120 High-Scoring ESG Portfolio Performance

Stock Returns of High-Scoring Portfolio

10.0%

5.0%

0.0%

-5.0% Monthly Portfolio Returns

28 Figure 3: SBF 120 Low-Scoring ESG Portfolio Performance

Stock Returns of Low-Scoring Portfolio

10.0%

5.0%

0.0%

-5.0% Monthly Portfolio Returns

Table 1: SBF 120 Portfolios Pre-Legislation Annualized Summary

Average Annual Returns Portfolio Average Annual ESG Score Standard Deviation

1.1: High-Scoring ESG 9.77% 93.58 companies 4.86 %

8.56% 1.2: Mid-Scoring ESG 83.61 1.95%

13.52% 1.3 Low-Scoring ESG 57.52 4.14%

Overall Portfolio Return of -3.74% 79.70 1.1 minus 1.3 5.86%

Table 2: SBF 120 Post-Legislation Data Annualized Summary

Average Annual Returns Portfolio Average Annual ESG Score Standard Deviation

14.64% 1.4: High-Scoring 96.62 4.47%

8.89% 1.5: Mid-Scoring 86.26 2.07%

9.44% 1.6: Low-Scoring 58.05 3.12%

Overall Return of 1.4 minus 5.19% 80.51 1.6 6.74%

Conclusion

Post-legislation annual returns, 5.19%, were higher than pre-legislation annual returns, 3.75%. This result was statistically significant at an 80% confidence level with a p-value of .11 and a t-stat of 1.62 on a two sample mean test comparing pre and post- legislation returns. This rejects the null hypothesis that pre and post-legislation returns for the ‘high-minus-low’ portfolio is equal with 80% confidence. For this data set, the high- scoring ESG portfolio outperformed the low-scoring ESG portfolio in the post-legislation time period while underperforming the pre-legislation time period.

The average annual ESG scores of the high, mid and low-scoring portfolios placed the average company for each portfolio in the negligible, low and severe risk categories respectively for both time periods. The average annual ESG score increased during post-legislation for each portfolio. Assuming an average risk-free rate of return of

.99%, the pre-legislation Sharpe ratio was -.81. Assuming a risk-free return rate of .6%, the post-legislation Sharpe ratio was .68.

The changes in annual ESG scores and annual returns between the two time periods offered the most interesting results from this section. While higher confidence levels (at least 90%) are generally preferred to conclude statistical significance within research and this level of significance is not enough to claim a distinct outcome, the lower statistical significance of these annual returns at 80% offers indications for future research and a contrast to the other data sets. This outcome suggests that a difference in returns for the ‘high-minus-low’ portfolio strategy occurred before and after the legislation was passed. While there are many factors to consider for the cause of the difference in returns, this result indicates potential for future research opportunities to

31 strengthen these findings by adding more companies or increasing the time period of analysis. Additionally, the increase in ESG scores for each portfolio suggests that, on average, ESG scores for these 77 companies are improving and can be used to support the positive non-financial impact of the law.

Section 2: CAC 40 Companies

Overview

In this section, I isolate the 40 most significant stocks of the stocks analyzed in

Section 1 to further study the impact of France’s legislation on specific company types.

Due to their size, performance and visibility as part of the CAC 40 index, the companies included in these portfolios are subject to increased scrutiny and this section aims to assess if the increased pressure on the largest firms, specifically those that drive the

French stock market, were subject to a greater difference in returns following the passage of legislation. This section follows a similar methodology and data collection strategy as

Section 1.

Figure 4: CAC 40 ESG Portfolio Performance (High-Scoring Minus Low-Scoring)

Stock Returns of High-Scoring Minus Low-Scoring Companies

10.00%

5.00%

0.00%

-5.00% Monthly Portfolio Returns

-10.00% 3/1/14 6/1/14 9/1/14 3/1/15 6/1/15 9/1/15 3/1/16 6/1/16 9/1/16 3/1/17 6/1/17 9/1/17 3/1/18 6/1/18 9/1/18 3/1/19 6/1/19 9/1/19 12/1/14 12/1/15 12/1/16 12/1/17 12/1/18 12/1/19 Date of Stock Price

Figure 5: CAC 40 High-Scoring ESG Portfolio Performance

Stock Returns of High-Scoring Portfolio

10.0%

5.0%

0.0%

-5.0% Monthly Portfolio Returns

Figure 6: CAC 40 Low-Scoring ESG Portfolio Performance

Stock Returns of Low-Scoring Portfolio

10.0%

5.0%

0.0%

Monthly Portfolio Returns -5.0%

Table 3: CAC 40 Pre-Legislation Annualized Summary

Average Annual Returns Portfolio Average Annual ESG Score Standard Deviation

2.1: High-Scoring ESG 8.34% 94.38 companies 5.49%

12.35% 2.2: Mid-Scoring ESG 88.23 4.06%

12.47% 2.3: Low-Scoring ESG 66.44 3.32%

Overall Return of 2.1 minus -4.13% 82.72 2.3 7.48%

Table 4: CAC 40 Post-Legislation Annualized Summary

Average Annual Returns Portfolio Average Annual ESG Score Standard Deviation

2.4: High-Scoring ESG 15.08% 97.00 companies 4.07%

8.44% 2.5: Mid-Scoring ESG 88.08 2.41%

11.70% 2.6: Low-Scoring ESG 62.67 3.08%

Overall Return of 2.4 minus 3.38% 82.72 2.6 8.44%

Conclusion

Post-legislation annual returns, 3.38%, were higher than pre-legislation annual returns, -4.13%. This result was not statistically significant with a p-value of .3 and a t- stat of 1.01 on a two sample mean test comparing pre and post-legislation returns. This result does not reject the null hypothesis that pre and post-legislation returns for the

‘high-minus-low’ portfolio is equal. Similar to Section 1 companies, the high-scoring

ESG portfolio underperformed in the pre-legislation time period while outperforming in the post-legislation time period.

The average annual ESG scores of the high, mid and low-scoring portfolios placed the average company for each portfolio in the negligible, low and high-risk categories respectively for both time periods. The average annual ESG score did not increase or decrease during post-legislation to a large extent for each portfolio. Assuming an average risk-free rate of return of .99%, the pre-legislation Sharpe ratio was -.55.

Assuming a risk-free return rate of .6%, the post-legislation Sharpe ratio was .40.

The lack of statistical significance in this particular data set indicates several key takeaways. The largest and most visible companies did not see significant changes in annualized returns after the passing of the legislation. There are several possible explanations for this. The lack of statistical significance could indicate that the legislation had no impact on the financial outcomes of ESG-focused investment strategies and therefore this particular government legislation did not produce a more efficient system of ESG investing. The lack of statistical significance could also indicate that ESG scoring systems were efficient within the market and additional legislation to control such systems do not make a significant impact.

36 Section 3: CAC 40 Companies – Environmental

Overview

The second filter for CAC 40 companies further isolated specific factors within the data set by using only the environmental score for CAC 40 companies. Because the

Energy Transition for Green Growth Act primarily targets environmental metrics for reporting, this filter attempts to eliminate extraneous factors and test if targeting legislation disproportionately impacts specific factors within the broader ESG industry.

Additionally, environmental factors within the ESG score receive the most attention from consumers and investors (see Table 7). This section implements an alternative investment strategy of investing based on specific factors within the broader ESG industry. Three portfolios were created based on each CAC 40 company’s environmental score and follow the same procedure as the sections discussed above.

37 Figure 7: CAC 40 Environmental Portfolio Performance (High-Scoring Minus Low-Scoring)

Stock Returns of High Minus Low-Scoring Companies 6.0%

4.0%

2.0%

0.0%

-2.0%

-4.0%

-6.0% Average Monthly Returns -8.0% 3/1/14 6/1/14 9/1/14 3/1/15 6/1/15 9/1/15 3/1/16 6/1/16 9/1/16 3/1/17 6/1/17 9/1/17 3/1/18 6/1/18 9/1/18 3/1/19 6/1/19 9/1/19 12/1/14 12/1/15 12/1/16 12/1/17 12/1/18 12/1/19 Date of Stock Price

Figure 8: CAC 40 High-Scoring Environmental Portfolio Performance

Stock Returns of High-Scoring Companies 12.0% 10.0% 8.0% 6.0% 4.0% 2.0% 0.0% -2.0% -4.0%

Average Monthly Returns -6.0% -8.0% -10.0% 3/1/14 6/1/14 9/1/14 3/1/15 6/1/15 9/1/15 3/1/16 6/1/16 9/1/16 3/1/17 6/1/17 9/1/17 3/1/18 6/1/18 9/1/18 3/1/19 6/1/19 9/1/19 12/1/14 12/1/15 12/1/16 12/1/17 12/1/18 12/1/19 Date of Stock Price

38 Figure 9: CAC 40 Low-Scoring Environmental Portfolio Performance

Stock Returns of Low-Scoring Companies 10.0% 8.0% 6.0% 4.0% 2.0% 0.0% -2.0% -4.0%

39 Table 5: CAC 40 Environmental Pre-Legislation Annualized Summary

Average Annual Returns Average Annual Portfolio Environmental Score Standard Deviation

3.1: High-Scoring ESG 12.26% 95.64 companies 3.90 %

11.33% 3.2: Mid-Scoring ESG 81.88 2.82%

10.13% 3.3 Low-Scoring ESG 61.66 5.90%

Overall Portfolio Return of 2.13% 79.62 3.1 minus 3.3 5.93%

Table 6: CAC 40 Environmental Post-Legislation Annualized Summary

Average Annual Returns Average Annual Portfolio Environmental Score Standard Deviation

3.4: High-Scoring ESG 10.50% 95.59 companies 4.60%

8.58% 3.5: Mid-Scoring ESG 81.88 4.13%

12.69% 3.6 Low-Scoring ESG 61.71 2.24%

Overall Portfolio Return of -2.19% 79.62 3.4 minus 3.6 6.20%

40 Conclusion

Post-legislation annual returns, -2.19%, were lower than pre-legislation annual returns, 2.13%. This result was not statistically significant with a p-value of .5 and a t-stat of -.61 on a two sample mean test comparing pre and post-legislation returns. This result does not reject the null hypothesis that pre and post-legislation returns for the ‘high- minus-low’ portfolio is equal. In contrast to the other data sets, the high-scoring portfolio outperformed the low-scoring portfolio pre-legislation and underperformed post- legislation.

The average annual environmental scores of the high, mid and low-scoring portfolios placed the average company for each portfolio in the negligible, low and high- risk categories respectively for both time periods. The average annual environmental score stayed essentially equal during post-legislation for each portfolio. Assuming an average risk-free rate of return of .99%, the pre-legislation Sharpe ratio was .19.

Assuming a risk-free return rate of .6%, the post-legislation Sharpe ratio was -.44.

The lack of statistical significance and differences from the first two data sets indicates that targeted legislation does disproportionately impact certain ESG metrics.

For countries seeking to efficiently raise returns for ESG investors while regulating the scoring system, this particular study does not find significant results. These portfolios also produced lower Sharpe ratios than sections 1 and 2, indicating more risk with this investment strategy. Therefore, targeting specific metrics or categories may not be a suitable investment strategy.

In contrast to the first two sections, the high-scoring portfolio underperformed in post-legislation time period. This could result from several scenarios. One possibility is

41 that increased scrutiny caused by the legislation specifically targeted towards environmental metrics prompted increased investment in improving these metrics, therefore lowering short-term stock returns. Another possibility is that the alternative method of sorting companies for these portfolios displays new groupings compared to the first two sections that follow normal return shifts similar to rebalancing a portfolio.

FUTURE RESEARCH OPPORTUNITIES

Based on the research I conducted on France’s Energy Transition for Green

Growth Act and its effect on ESG investment strategies, I identified several opportunities for future related research, including studies expanding the time period of collected data, incorporating a larger sample size of companies and using data from multiple countries.

Average ESG scores increased following the implementation of legislation. While this study cannot identify causation for these positive changes, further studies could attempt to replicate these results in other countries or with a wider set of companies and provide stronger support. If legislation correlated with stronger ESG performance, this could encourage other countries to increase ESG-related reporting standards and legislation.

As ESG data and scoring systems become increasingly robust and widespread, larger samples and expanded time periods for returns will offer opportunities for increasingly sound analysis of the Energy Transition for Green Growth Act and Article

173-VI’s impact. The increased level of statistical significance in a larger sample size indicates that adding additional companies and portfolios could produce strong results.

ESG scores are becoming mainstream and the continuation of scoring systems like

Sustainalytics and Bloomberg means ESG data will be available for more companies and for a longer time period. Future research could expand upon this paper’s findings by

43 adding smaller companies, increasing the data set for SBF 120 and CAC 40 companies, and tracking financial changes as ESG systems and popularity evolves.

Analyzing ESG portfolio effectiveness in other developed countries with various social systems, industries and styles of legislation would add to the overall available studies of government intervention on ESG investing. As more laws regulating ESG data are implemented and this investment strategy increases in popularity, there will be multiple laws in different countries and longer time periods to analyze returns from. This allows researchers to compare the effectiveness of various legislation types on ESG investment techniques. Additionally, researchers could examine legislation that targets specific metrics or factors within the broader ESG category as the Energy Transition for

Green Growth Act targeting environmental factors. By combining legislation with the highest level of financial performance, governments can analyze the most efficient way to encourage responsible, financially stable methods of investing in ESG-focused companies.

APPENDIX

Table 7: SBF 120 portfolio compositions

Portfolio 1.1: High-Scoring ESG 1.2: Mid-Scoring ESG 1.3: Low-Scoring ESG Composition Rexel SA BNP Paribas SA Ingenico Group SA Gecina SA Orange SA Unibail-Rodamco- STMicroelectronics NV Electricite de France SA Westfield Accor SA Capgemini SE LVMH Moet Hennessy AXA SA CNP Assurances Louis Vuitton SE Peugeot SA Credit Argicole SA Air Liquide SA Sodexo SA Klepierre SA Cie Generale des Carrefour SA Renault SA Etablissements Michelin EssilorLuxottica SA Bouygues SA SCA SEB SA TOTAL SA Veolia Environnement SA Kering SA TechnipFMC PLC Publicis Groupe SA Legrand SA L’Oreal SA Cie de Saint-Gobain Lagardere SCA Pernod Ricard SA Ipsen SA BioMerieux Bureau Veritas SA Alstom SA Atos SE Societe Generale SA Arkema SA ArcelorMittal SA Airbus SE Edenred Sanofi Solvay SA Faurecia SE Suez Natixis SA Vinci SA Valeo SA Getlink SE Engie SA ICADE Casino Guichard SES SA JCDecaux SA Perrachon SA Eutelsat Communications Schneider Electric SE Eurazeo SE SA Thales SA Dassault Systemes SE SPIE SA Covivio Vivendi SA Wendel SA Danone SA SCOR SE Safran SA Air France-KLM Iliad SA Remy Cointreau SA Eiffage SA Dassault Aviation SA Bollore SA Hermes International

i Table 8: CAC 40 portfolio composition

Portfolio 2.1: High-Scoring ESG 2.2: Mid-Scoring ESG 2.3: Low-Scoring ESG Composition STMicroelectronics NV Danone SA Vivendi SA Accor SA BNP Paribas SA Unibail-Rodamco- AXA SA Orange SA Westfield Peugeot SA Capgemini SE LVMH Moet Hennessy Sodexo SA Credit Agricole SA Louis Vuitton SE Carrefour SA Renault SA Air Liquide SA EssilorLuxottica SA Bouygues SA Cie Generale des Kering SA TOTAL SA Etablissements Michelin Legrand SA TechnipFMC PLC SCA Atos SE L'Oreal SA Veolia Environnement SA ArcelorMittal SA Pernod Ricard SA Publicis Groupe SA Sanofi Societe Generale SA Cie de Saint-Gobain Schneider Electric SE Airbus SE Vinci SA Thales SA Engie SA Safran SA Dassault Aviation SA Hermes International

Table 9: CAC 40- Environmental portfolio compositions

Portfolio 3.1: High-Scoring ESG 3.2: Mid-Scoring ESG 3.3: Low-Scoring ESG Composition Carrefour SA BNP Paribas SA Veolia Environnement SA Atos SE Airbus SE EssilorLuxottica SA Accor SA Sodexo SA Vinci SA AXA SA Renault SA Cie Generale des Orange SA Unibail-Rodamco- Etablissements Michelin Pernod Ricard SA Westfield SCA TOTAL SA Peugeot SA Bouygues SA TechnipFMC PLC Air Liquide SA Safran SA STMicroelectronics NV Kering SA Societe Generale SA Schneider Electric SE L'Oreal SA Vivendi SA Thales SA Credit Agricole SA Sanofi Danone SA Engie SA Legrand SA Capgemini SE Publicis Groupe SA ArcelorMittal SA LVMH Moet Hennessy BNP Paribas SA Cie de Saint-Gobain Louis Vuitton SE

ii Table 10: Examples of ESG metrics per category

Category Environmental Social Governance Examples of calculated Greenhouse gases/Revenue Women employees to % Independent directors metrics management ratio Energy/Revenue Director average age % Women employees Water/Revenue % Director meeting % Employee turnover attendance Waste/Revenue % Employees unionized Board size Water recycled/Revenue Total incident rate Carbon reserves Fatalities/1000 employees Fossil fuel generation capacity

iii

REFERENCES

Auer, B. R., & Schuhmacher, F. (2016). Do socially (ir)responsible investments pay? New evidence from international ESG data. The Quarterly Review of Economics and Finance, 59, 51–62. doi: 10.1016/j.qref.2015.07.002

Balatbat, M. C. A., Siew, R. Y. J., & Carmichael, D. G. (2012). ESG scores and its influence on firm performance: Australian evidence. The University of New South Wales. Retrieved from: https://www.researchgate.net/publication/260907931

Balibar, S. (2017). Energy transitions after COP21 and 22. Comptes Rendus Physique, 18(7-8), 479–487. doi: 10.1016/j.crhy.2017.10.003

Bannier, C. E., Bofinger, Y., & Rock, B. (2019). Doing safe by doing good: ESG investing and corporate social responsibility in the U.S. and Europe. University of Giessen. Retrieved from https://www.uni- giessen.de/fbz/fb02/fb/professuren/bwl/bannier/team/190424ESGPaper.pdf

Boerner, H. (2012). The corporate ESG beauty contest continues: Recent developments in research. Corporate Finance Review.

Cao, J., Titman, S., Zhan, X., & Zhang, W. E. (2019). ESG preference and market efficiency: Evidence from mispricing and institutional trading. SSRN Electronic Journal. doi: 10.2139/ssrn.3353623

Chatterji, A., Levine, D. I. I., & Toffel, M. W. (2009). How well do social ratings actually measure corporate social responsibility? The Journal of Economics and Management Strategy . doi: 10.2139/ssrn.993094

Crifo, P., & Mottis, N. (2013). Socially responsible investment in France. Business & Society, 55(4), 576–593. doi: 10.1177/0007650313500216

Dorfleitner, G., Halbritter, G., & Nguyen, M. (2014). Measuring the Level and Risk of Corporate Responsibility - An Empirical Comparison of Different ESG Rating Approaches. SSRN Electronic Journal. doi: 10.2139/ssrn.2536265

Eccles, R. G., & Stroehle, J. C. (2018). Exploring social origins in the construction of ESG measures. Saïd Business School, University of Oxford. Retrieved from http://dx.doi.org/10.2139/ssrn.3212685

Evain, Julie, Cardona, Michel, & Nicol, Morgane (2018). Article 173: analysis of

iv insurers' climate reporting (INIS-FR--19-0206). France

Fama, E. F., & French, K. R. (1992). The Cross-Section of Expected Stock Returns. The Journal of Finance, 47(2), 427–465. doi: 10.1111/j.1540-6261.1992.tb04398.x

Fatemi, A., Glaum, M., & Kaiser, S. (2018). ESG performance and firm value: The moderating role of disclosure. Global Finance Journal, 38, 45–64. doi: 10.1016/j.gfj.2017.03.001

France Government Bond 10Y. (2020). Retrieved from https://tradingeconomics.com/france/government-bond-yield

Friede, G., Busch, T., & Bassen, A. (2015). ESG and financial performance: Aggregated evidence from more than 2000 empirical studies. Journal of Sustainable Finance & Investment, 5(4), 210–233. doi: 10.1080/20430795.2015.1118917

Giamporcaro, S., Gond, J.-P., & O’Sullivan, N. (2019). Orchestrating governmental corporate social responsibility interventions through financial markets: The case of French socially responsible investment. Retrieved from https://journals.aom.org/doi/pdf/10.5465/AMBPP.2019.211.

Glooner, S. (2017). ESG risks and the cross-section of stock returns. SSRN Electronic Journal. doi: 10.2139/ssrn.2980917

Huber, B. M., Comstock, M., & Davis Polk & Wardwell LLP. (2017, July 27). ESG reports and ratings: What they are, why they matter. Retrieved from https://corpgov.law.harvard.edu/2017/07/27/esg-reports-and-ratings-what-they- are-why-they-matter/.

Jemel-Fornetty, H., Louche, C., & Bourghelle, D. (2011). Changing the dominant convention: The role of emerging initiatives in mainstreaming ESG. Critical Studies on Corporate Responsibility, Governance and Sustainability Finance and Sustainability: Towards a New Paradigm? A Post-Crisis Agenda, 85–117. doi: 10.1108/s2043-9059(2011)0000002011

Klober, D. (2015). Millennials and wealth management - Inside article. Retrieved from https://www2.deloitte.com/content/dam/Deloitte/lu/Documents/financial- services/lu-Millennials-wealth-management-trends-challenges-new-clientele- 0106205.pdf.

M., M., Sitkoff, & H., R. (2018, September 6). Reconciling fiduciary duty and social conscience: The law and economics of ESG investing by a trustee. Retrieved from https://papers.ssrn.com/sol3/Papers.cfm?abstract_id=3244665.

Ortas, E., Garayar, A., & Álvarez, I. (2015, February 11). The environmental, social,

v governance, and financial performance effects on companies that adopt the United Nations Global Compact. Retrieved from https://www.mdpi.com/2071- 1050/7/2/1932.

Revelli, C. (2017). Socially responsible investing (SRI): From mainstream to margin? Research in International Business and Finance, 39, 711–717. Doi: 10.1016/j.ribaf.2015.11.003

Sherwood, M. W., & Pollard, J. (2018). The development of ESG risk and the ESG rating system. Responsible Investing, 29–56. doi: 10.4324/9780203712078-3

Toledo, E. P. D., & Bocatto, E. (2014). The Impact of environmental, social and governance (ESG) standards on the value of cash holdings: Evidence from Canadian firms. SSRN Electronic Journal. doi: 10.2139/ssrn.2379895