Sustainability Index Inclusion Effects in the Emerging
Markets between 2014 and 2019: an event study
Jippe Wieringa 11009187
University of Amsterdam Economics and Business Supervisor: Richard Evers
Statement of Originality
This document is written by Student Jippe Wieringa (11009187) who declares to take full responsibility for the contents of this document.
I declare that the text and the work presented in this document are original and that no sources other than those mentioned in the text and its references have been used in creating it.
The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.
Abstract
In the past years investors have been increasingly aware of the ethical side of doing business leading to a rise in socially responsible investments. Past research (Consolandi et al., 2009; Orlitzky et al., 2003) into the area gives evidence for a positive relationship between companies that perform well in corporate social responsibility and financial performance, which is substantiated by stakeholder theory. As most of the research in this field is done into developed countries, this thesis researches the effect of inclusion in the Dow Jones
Sustainability Index for the Emerging Markets on stock returns in between 2014 and 2019 using an event study methodology. The research finds no significant relationship between the two in the days before the announcement or the announcement day itself, but a negative relationship is found in the days after the announcement as well as in the overall period. Therefore the hypothesis that there is a positive relation between the two is rejected. Based on this evidence, it can be concluded that shareholders in emerging market companies do not value investments into CSR, however it has to be noted that little research has been done in the workings of CSR in the emerging markets. Therefore it is recommended that in further research this, and the link between CSR and financial performance in the emerging markets is further investigated.
Keywords: corporate social responsibility, CSR, stakeholder theory, sustainability index,
Table of Contents
1. INTRODUCTION ... 5
2. LITERATURE REVIEW ... 6
2.1.CORPORATE SUSTAINABILITY ... 6
2.2.RELATIONSHIP BETWEEN CSR AND CORPORATE PERFORMANCE ... 7
2.3.INDEX INCLUSION EFFECTS ... 9
2.4.EMPIRICAL RESEARCH INTO SUSTAINABILITY INDEX INCLUSION ... 11
2.5.CSR AND THE EMERGING MARKETS ... 12
2.6.DOW JONES SUSTAINABILITY INDEX EMERGING MARKET AND HYPOTHESIS ... 14
3. METHODOLOGY ... 15
3.1.DATA ... 15
3.2.EVENT STUDY METHODOLOGY ... 17
4. RESULTS ... 22
5. CONCLUSION AND DISCUSSION ... 24
5.1.SUMMARY ... 24
5.2.CONCLUSION ... 24
5.3.LIMITATIONS ... 25
1. Introduction
Since the 1990s the popularity of socially responsible investment has increased significantly, and moved from a niche to a mainstream line of investment in which pension funds and institutional investors also partake. As shareholders started considering the ethical side of their investment strategy, corporations were forced to act as well, which led to corporate social responsibility gaining a more prominent place on the corporate agenda (Spakes & Cowton, 2004). In more recent years institutional investors are reevaluating their traditional risk-return beliefs and are also incorporating corporate social responsibility (CSR) factors into this equation. Investors do not only want to be shielded from reputational risk due to poor CSR performance, but also want to see environmental, societal and governance returns (KPMG, 2020).
After the introduction of the Dow Jones Sustainability Index for the World in 1999, and subsequently other regional indices for Europe and North America, RobecoSAM introduced the Dow Jones Sustainability Index Emerging Markets in 2013 (S&P Global, 2020). In their press release S&P Dow Jones Vice President Alka Banerjee stated: ”An important
strategic reference point for sustainability investors around the globe, the DJSI are continuously advanced to respond to market trends and requirements. The DJSI Emerging Markets, the first index of its kind in the market, is launched in response to the evolving needs of the global investment community.” (RobecoSAM, 2013). Which indicates that shareholders
would make decisions based on the inclusion in an sustainability index.
There has been past research in this field before. Consolandi et al. (2009) found that inclusion into the DJSI Europe had positive results on the share price, on the other hand Oberndorfer et al. (2013) replicated this research limited to German companies and found a negative relation between the two. Given these ambiguous results, and that most previous research has been done into developed countries, makes it interesting to research the question:
To what extent do shareholders in emerging markets value the inclusion in the Dow Jones Sustainability Index Emerging Markets (DJSIEM) in between 2013-2018?
The structure of this thesis will be as follows: the second section of this thesis focusses on the existing literature on CSR, stakeholder theory, CSR in relation with corporate performance, index inclusion effects, the effect of inclusion in sustainability, CSR in the emerging markets and the methodology of the Dow Jones Sustainability Index for the emerging markets. The third part will the event study methodology used of this research. In the fourth part the results will be presented, after which the conclusion and limitations of this research will be presented in the fifth part together with recommendations for further research.
2. Literature review
Before generating a hypothesis, it is important to understand previous research that has been done into the subject. Firstly it is important to know what the definition of corporate social responsibility is and how the term evolved during the years. Subsequently, the stakeholder model in which CSR is often placed is elaborated. Then the relationship between CSR and corporate performance needs to be clarified and effects of index inclusions in general after which the effects of sustainability index inclusions will be reviewed. Conclusively, there will be an overview of research on CSR in the emerging markets.
2.1. Corporate Sustainability
Van Marrewijk (2003) acknowledges the difficulties associated with developing a CSR definition. Everyone comes to a definition which best fits their situation, but this leads to a broad range of biased definitions which are hard to apply to research and in business. Contrarily, a too broad definition is also not feasible for the same reason. After elucidating the historical, philosophical and the practical background he comes to three definitions.
Firstly he introduces the term Corporate Societal Accountability as an alternative for CSR. This definition constructed by Göbbles (2002) is purely linguistic. According to him both the words social and responsibility are too narrow to reflect what they actually should mean, therefore he replaces them with broader terms societal and accountability. Societal better emphasizes the environmental factor of CSR.
Secondly, he poses a definition where corporate sustainability (CS) is the highest goal, and corporate social responsibility is where companies balance the societal, environmental and business aspects. Lastly, he poses that CSR and CS are two sides of the same coin. They used to be different, but in the past decades the definition of both terms grew to each other. He comes with the following definition: “ In general, corporate sustainability and, CSR refer to company
activities – voluntary by definition – demonstrating the inclusion of social and environmental concerns in business operations and in interactions with stakeholders.” (Van Marrewijk, 2003,
p.102). After that he splits up the broad definition into five ambition levels companies can adopt depending on their “awareness and ambitions (Van Marrewijk, 2003, p.103)”. He concludes that there is not one set of features that describes CSR, but a certain ambition level that comes with certain kinds of activities.
Dahlsrud (2006) states that the biases van Marrewijk (2003) sees in the definitions are not supported by any empirical evidence. He poses that CSR is a social construct, and therefore it’s not possible to be defined unbiased. However, we can research how it is defined in the existing definitions. CSR was split into five dimensions, the environmental, the social, the economic, the stakeholder and the voluntariness dimension. After this 37 definitions of CSR were analyzed and reviewed based on these dimensions. He finds that CSR was always embedded in the company’s policy automatically, but because of the rapidly globalizing world, new stakeholders with new expectations are introduced, which asks for a different balance between the dimensions.
Most CSR definitions include all five dimensions, while sometimes phrased differently, they have the same meaning, which makes the problem of defining CSR less significant. The issue is not how CSR is defined, it’s how it’s constructed in a specific context and how to implement it into a business strategy.
2.2. Relationship between CSR and corporate performance
The relationship between corporate performance and CSR is a heavily debated subject in finance, a discussion which starts in the 60s and goes on until present day. At the heart of the debate lies the debate between shareholder and stakeholder theory. Friedman first posed the shareholder value theory which is best summarized by the following quote: “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 ... engages in open and free competition, without deception or fraud“ (Friedman as cited in Smith, 2003, p. 85), which can be seen as an attack on companies performing CSR, as the direct effect on profitability of companies is not always clear.
Contrary to this, Freeman (1984) introduced stakeholder theory, which is used to link CSR to corporate performance in most of the research in this field. Donaldson and Preston (1995) explained that in stakeholder theory “all persons or groups with legitimate interests
participating in an enterprise do so to obtain benefits and that there is no prima facie priority of one set of interests and benefits over another.” (Donaldson & Preston, 1995, p.68.) They
argue that stakeholder theory can be split up in three aspects, the descriptive, the instrumental and the normative. They put the normative component at the heart of theory, which means that serving only shareholders is amoral, and other stakeholders should be considered as well. The instrumental component of the theory describes how stakeholder theory can be used to perform
better financially. This is the field were most of the research described in this review can be situated in. Lastly, the descriptive aspect of the theory describes how companies currently operate. Jones (1995) deepens the instrumental aspect of stakeholder theory. He describes how stakeholder theory, and by extension CSR, can give companies a competitive advantage. He builds this argument by stating that the firm is a “nexus of contracts” managed by the managers. As markets tend toward equilibrium, efficient contracting becomes a more important factor. Normally contracting brings commitment problems, firms that solve these in an ethical way will mitigate these problems, from which a competitive advantage over other firms can follow.
The earliest empirical research into the link between CSR and corporate performance is very ambiguous. Arlow and Gannon (1982) summarize earlier research into the topic and come to the conclusion that while businesses became increasingly aware of the importance of “social responsiveness” there was no evidence to support a strong positive economic relationship between the social responsiveness and corporate performance in the short run. Moreover, as all seven researches used different performance measures, and some companies used questionable and old-fashioned measures of social performance, the results were not completely trustworthy.
Cochran and Wood (1984) recognize that earlier research made mistakes with incorrect financial performance proxies, small sample sizes and questionable methodology. They use a reputation index as CSR performance measure and use accounting data to measure the financial performance of the firms. Overall they conclude that there is a weak positive relation between CSR and financial performance. They recommend that the causality of the relationship is an important field to study.
Following this Waddock and Graves (1997) pose two theories regarding the causal relation between CSR and financial performance. Firstly, they pose the slack resources theory, suggesting that better financial performance results in slack resources which can be used to improve the CSR of the firm. Conversely, they pose the good management theory, which argues that attention to CSR results in better stakeholder relationships which results in a better overall performance of the firm. This research recognizes the multidimensional features of CSR and use a better measurement of CSR recognizing community relations, employee relations, environmental performance, product characteristics, and the treatment of women and minorities as well as involvement in South Africa during the apartheid regime military contracting and participation in nuclear power. Financial performance was measured using return on assets, return on equity and return on sales. They find proof for both the slack
resources as the good management theory, which means that CSR influences financial performance and vice versa.
Ruf et al. (2001) investigate the causal relationship between CSR and financial performance by using stakeholder theory as a framework. In this study, stakeholder theory is seen in the instrumental view, as a tool with which companies can reach higher financial performance. Stakeholders also have demands from the company (like shareholders) and can enforce them with boycotts, lawsuits and protest, which could influence shareholder value as well. Stakeholders considered in this research are employees, consumers, environment, community and society. In contrast to earlier research, here the change in CSR is used, and not the level. The study found that there is a short-term positive relationship between CSR and growth in sales, which indicates consumers are aware of the company taking CSR measures and support this. In the long-term a positive relationship with return of sales and return on equity are observed. From this they conclude that by meeting demands by stakeholders, shareholders can financially benefit as well. Lastly, they note that some firms only perform CSR for opportunistic reasons, but others also do it for philanthropic reasons.
Finally, Orlitzky et al. (2003) conducted a meta-analysis on 52 studies to get a more general view on the relationship between CSR and financial performance. They found a significant positive relationship between the social component of CSR and financial performance and a smaller but still significant relationship between the environmental component of CSR and financial performance. Furthermore both the slack resources and the good management theory (called instrumental stakeholder theory here) are found to be true. Financially sound performing companies will spend more because they can afford it, but they’re also more successful because they invest in CSR.
2.3. Index inclusion effects
To research the effect of inclusion in a sustainability index, it is important to know the general effects of index inclusion on stock prices. Multiple theories of index inclusion exist of which the most important ones will be laid out below. Most of the research in this field has been done into the inclusion effects to the S&P500.
The Efficient Market Hypothesis says that all publicly available information is included in stock prices. An underlying assumption is that all stocks are (nearly) perfect substitutes for each other and have nearly horizontal demand curves, which means that a shift of the demand curve, shouldn’t have an effect on the stock pricing (Shleifer, 1986). However, several different
researches have shown that this doesn’t hold and share prices (temporarily) rise after the inclusion in the S&sP500.
Firstly, Shleifer (1986) posed the downward sloping demand curve theory. He assumes that inclusion in the S&P500 doesn’t signal anything to the market about the future performance of a firm. Index funds try to replicate the S&P500, which causes the demand to increase when the inclusion is announced, and causes the demand curve to move to the right. If the demand curve had been horizontal as is assumed in the efficient market hypothesis, there would not have been any price change. However, he observed that inclusion into the index had a significant upward effect on the share price of firms. If it is still assumed that inclusion in the S&P500 doesn’t take into account any future performance measures or unknown information, the demand curve has to be downward sloping. Important to note is that according to Shleifer (1986) this effect is permanent.
Alternatively, Harris and Gurrel (1986) have done research over a similar period, but found only a temporary effect positive effect of the inclusion, which they call the price pressure hypothesis. They find that in the long term the efficient market hypothesis indeed holds, but in the short term the stock price rises. They suggest that this happens to persuade passive investors to offer their shares, which they later, when the price drops back, can buy back, recording a net profit.
Both theories explained above assume that the announcement of the inclusion doesn’t contain any information not known to the market. However, when we relax this assumption, several other hypotheses are possible. Merton (1987) introduced the information cost theory hypothesis, arguing that for example inclusion into the S&P500 could bring information to the broader public, that was publicly available but costly to obtain. This is then reflected in the share price which explains the increase after the announcement. Jain (1987) first introduced the theory that the publicity (e.g. inclusion in an index) might signal information to the market that wasn’t known before. Later research showed that analysts increased their earnings per share forecast for selected stocks, but the eventually realized earnings were also higher after the inclusion which may mean that management is motivated by the inclusion, but it can also mean that information is used that is not available for other market participants (Denis et al. 2003). Lastly, some evidence shows that the rise of share price might be due to the rise in liquidity which is observed after the introduction of stocks into the index. The increase in liquidity lowers the costs of trading which is incorporated in the share price (Hedge & McDermott 2003).
2.4. Empirical research into sustainability index inclusion
Most of the aforementioned theories are based on the S&P 500. However past research has also been done into the inclusion into several major sustainability indices, which will be summarized below.
The first to apply event study methodology to sustainability indices were Curran and Moran (2006). They researched the effect of the inclusion and deletion of companies in the FTSE4GOODUK index in 2002 and found an insignificant positive effect on the inclusion and negative effect on the deletion of the index. However, they note that at the time the index wasn’t well known which might be why the effect was insignificant. Furthermore, they note that the inclusion also gives information on the economic status of the company, which might explain the effects on the stock price.
Later research from Consolandi et al. (2009) found a significant relationship between inclusion in a sustainability index and stock returns. They researched the effects of inclusion (deletion) from the Dow Jones Sustainability Stoxx Index, the sustainability index for Europe, between 2006-2008 and found a significant positive effect when a firm was added to the index, and an even stronger results when it was deleted. They suggest this might be due to the Information Hypothesis, the inclusion in the index introduces new information into the market which is factored into the share price. Another suggestion is the fact that socially responsible investors will notice the announcement and will act on it but other investors might not which results in a net positive effect.
Cheung (2011) researched the effects for US companies included in the Dow Jones Sustainability World Index between 2002 and 2008. He concludes that the in- or exclusion into the index had significant impact on the stock on the day of change, included stocks rose, and excluded stocks dropped. He finds that the results are consistent with price pressure hypothesis as he also observes a rise in liquidity on the effective date. Furthermore, he finds evidence for a price reversal 1 week after the in- or exclusion.
Oberndorfer et al. (2013) researched the effect of inclusion into the DJSI STOXX and DJSI World on stock performance for German companies in the short run in between 1999 and 2002. They find that inclusion in the DJSI STOXX doesn’t have a significant impact on share price, but inclusion into the world index has significant negative impact on the share price, the average decrease is more than 2%. However, in the time period they considered the selection process wasn’t optimized yet. The DSJI STOXX and World indices are selected from the best performing companies in the Dow Jones STOXX and World indices. At the time, lots of
companies in these indices weren’t assessed, so couldn’t be included. They conclude that the more visible a sustainability index is, the larger the negative impact on the stock price.
Following Consolandi et al. (2009), Van Stekelenburg et al. (2015) did research into the DJSI Europe in the subsequent period from 2009-2015 and found results similar to Cheung (2011). After the effective day of change, the newly included stock returns increased, and vice versa. These effects where temporary, with the price going back to the initial level after 60 days. Like Cheung (2011) suggests as well, this is consistent with the price pressure hypothesis. This research also looked at the returns of companies that were announced industry group leaders by the DJSI which finds that these company have a long-term high increase in abnormal stock turns. This is explained by the earlier mentioned slack resources theory, which explains that if a company invests in CSR, it’s underlying financial performance should be robust to be able to do so. CSR performance could be an indicator of otherwise good financial performance.
2.5. CSR and the Emerging Markets
CSR in the emerging markets is a subject on which to the best of the authors knowledge seems to be little existing literature and most of the existing literature focusses on one country specifically and not on the group of countries as a whole. Moreover, most of the literature describes the CSR reporting and doesn’t provide any evidence of a link with financial performance. However, as this thesis focusses on the emerging markets, it is still meaningful to get a perspective on how CSR is interpreted and applied in the emerging markets.
Doberts and Halme (2009) researched CSR practices in a broad range of developing nations, ranging from Africa to the BRIC countries. They note that as CSR is not only the relationships between firms and other stakeholders, but also focusses on the responsibility of the firm in social and economical context. As these contexts are different in different societies, CSR is different in every country and region around the world. As CSR originates from Western countries, most of the theory and concepts focus on countries with strong institutions and efficient regulation and law enforcement. In developed countries, good CSR is usually regarded as doing more than the law requires a company to do. However, in the context of the emerging markets CSR can be totally different. In countries with weaker institutions, following the law might be already be good regarded CSR.
This corroborates earlier research by Belal (2001). He investigates CSR in Bangladesh. He also states that Western CSR practices might not be applicable for developing and emerging markets, as the social, political regulatory, and economic context in which decisions are made
are different. He describes the challenges the country faces due to industrialization. In poorly unionized sectors, labor rights are often denied, industrial pollution has been identified as big environmental hazard, corruption and moral degradation are rising. He concludes that while almost all companies disclose CSR information, the quality of the provide information is questionable, which is caused by a lack of regulatory oversight, political instability and widespread corruptions
Baskin (2006) compares the reported CSR of companies in the emerging market to companies in developed countries using inclusion in de Dow Jones Sustainability Index, registration with the Global Reporting Initiative and the uptake in ISO1400, the global standard for environmental management systems as indicators. The type of company Baskin researches largely corresponds to the type of firm investigated in this thesis, large stock listed companies in the emerging markets. He found that the reported CSR from companies in the emerging markets is more extensive than commonly believed and there is no significant difference in the approach to CSR. Nevertheless, CSR is less embedded in the corporate strategy, less pervasive and less rooted in politics. Differences between the best and worst- in-class companies are bigger in the emerging markets. Baskin (2006) notes that leading emerging market companies usually have more in common with each other than with other companies in their own region. He concludes his research with some general inferences on drives of CSR in the emerging markets. According to Baskin (2016) the focus on the social part of CSR is mostly driven by the lack of government involvement in this part of society. This part of CSR is most prevalent in countries with high inequality, low state provision and active civil society. The increasing global focus on sustainable development with the millennium goals also have a big influence on CSR in the emerging markets. To be able to compete globally, companies have to take these standards in to account. The uptake in governance is driven by the capital markets, as countries with higher governance standards are more attractive for outside investment.
Ortas et al. (2012) researched the financial performance of the Brazil Corporate Sustainability Index (BSCI) compared to the market index. They found that financial
markets in the emerging markets have a positive view on companies that improve their CSR. Moreover, they found evidence stating the BCSI is less risky than the market index while giving similar returns. According to stakeholder theory, companies not taking sustainable development into account could limit their investment opportunities and following from this their financial performance. Following from this it seems logical that the BSCI firms will perform better than other firms as they do take sustainable development into account.
Furthermore, the fact that companies included in the BSCI have to be transparent about their corporate governance issues makes the stocks included in the index more
valuable. In the emerging markets the effect of good governance is higher than in developed countries because the differences across companies are much bigger, and corporate
governance is not necessarily regulated by the state. Lastly, the strong CSR policies that all the companies in the index adopt makes them less vulnerable for ethics and environmental scandals which have a big impact on the market value of companies.
2.6. Dow Jones Sustainability Index Emerging Market and hypothesis
Dow Jones Indexes and SAM, a sustainable investment specialist, claim to have created the first global sustainability index in 1999. Interesting is that at the time of the launch, SAM viewed sustainability as ‘business approach that creates long term shareholder value ’ which seems more in line with the shareholder value approach of Friedman (1962), than with the stakeholder approach sustainability is usually associated with. Over the years, regional and national indices were added, including the Dow Jones Sustainability Index for the Emerging Markets in 2013 (Fowler and Hope , 2007).
Fowler and Hope (2007) point out that the Domini 400 Social Index launched in 1990. However, this is not a global index and is only focused on the United States. In their research they estimate that about 10% of all investment funds have some sort of social screening. This can be a negative screening, where just industries deemed unethical are disregarded, but also a positive approach which only includes the companies that have proven to do good.
RobecoSAM uses a positive best-in-class approach to construct the indices under their management. For every index the starting point is the so called Invited Universe, these are the companies invited to fill in the Corporate Sustainability Assessment (CSA). This contains general questions as well as industry specific questions on economic, environmental and social criteria. The weights of the three dimensions changes every year, as well as the questions asked in the questionnaire. Important to note that included in this CSA the ongoing screening of media and stakeholder commentaries on the companies. By looking at this data, RobecoSAM checks data supplied in the assessments translate into actual performance. All the data together is used to calculate a Total Sustainability Score (TSS). For the DJSIEM, the 800 largest emerging markets companies from the S&P Global Broad Market Index are used combined
with any existing companies in the DJSIEM with a market cap above $500m (RobecoSAM, 2020).
From the invited universe, the assessed universe are all companies that have a TSS. All companies with a score lower than 40% than the highest score within the universe are deleted. After this companies are ranked in descending order by TSS score per industry. Per industry the 10% highest ranked companies are added, combined with the existing constituents that are in the top 15%. The index is weighted by float-adjusted market cap with 10% maximum (RobecoSAM, 2020)
Following the reasoning from Jones (1995) on instrumental CSR, CSR should bring a competitive advantage over other companies. Generally speaking, research into the link between CSR and financial performance has found a positive link between the two. Index inclusion theory also indicates a positive relationship between the inclusion and share price. Inclusion into the DJSI EM gives new information to the market, or at least bring it under the attention of the general public. But even if this wouldn’t be the case, according the price pressure hypothesis a (temporary) rise in share price is to be expected. Taking all of this together, results in the hypothesis:
Hypothesis: Inclusion of a stock in the Dow Jones Sustainability Index Emerging Markets has a positive influence on stock returns in between 2013 and 2018
3. Methodology 3.1. Data
The sample used are the inclusions into the Dow Jones Sustainability Index Emerging Markets (DJSIEM) between 2014 and 2019. The index was first published in 2013, but data on the first inclusions isn’t available, the last published results are for the 2019 index. The inclusions are derived from the historical components sheets downloaded from the SPGlobal website, the stock returns are downloaded from Factset. The amount of inclusions is slightly higher in the earlier years as observed in Table 1:
Table 1: Number of inclusions over time Year Amount 2014 20 2015 14 2016 16 2017 10 2018 11 2019 12 Total 83
Important to note is that some emerging economies have restrictions on foreign investment, and therefore have separate investment derivatives for foreign investment. In this research, the common stock was always chosen, but this may have influence on the results. Some companies shares are traded on multiple exchanges, in this research the home exchange of the company is used. Some companies trade both on the Hongkong and a China mainland exchange. The shares on the mainland are only available to domestic investors and therefore less liquid, therefor the Hongkong shares were chosen. A breakdown of the inclusions per countries can be found in Table 2.
Table 2:Inclusions per Country
Country Amount Taiwan 15 Greece 2 Colombia 5 Thailand 18 South Africa 4
Brazil 4 Malaysia 2 Mexico 6 India 1 Chile 8 Turkey 2 Arabic Emarites 1 Russia 1 Hongkong 3 Singapore 1 Source: RobecoSAM
3.2. Event study methodology
An event study measures the effect of a specific event on the value of a firm, or multiple firms. This event can be announcements, merger or any other event which brings new information to the market. Through the efficiency of the marketplace (efficient market hypothesis) the information is immediately incorporated in the price, which makes it possible to measure the economic impact by measuring the share prices over a relatively short period of time (MacKinlay, 1997).
In an event study, the expected returns around the event date are compared with the realized returns around the event date. The difference between the two, is called the abnormal return. When this is done on one stock, around one event date, no general conclusion can be derived from the data, but when the abnormal returns on multiple stock, around multiple event dates of the same kinds are aggregated, a general hypothesis can be tested (De Jong, 2007). For an event study, three important steps have to be followed. First the event of interest and it’s timing should be identified, as well as its timing, secondly a benchmark model should be specified to calculate normal stock behavior. Lastly, the abnormal returns around the event date should be calculated and analyzed (De Jong 2007).
Following the model used by Consolandi et al. (2009), two event dates are defined, the announcement date (AD) and the effective date (ED), when the index is actually changed. A 10-day window before the announcement is included to account for anticipation of leakage of information, after which the first trading date following the announcement date itself will be tested, following this the period in-between the announcement and the effective date will be tested, then the date of the actual effective revision will be tested, and lastly the 10-day period after the effective date for any delayed effects on the share price, as depicted in figure 1
Figure 1: Estimation and event windows
The announcement and effective dates were obtained from the press release RobecoSAM does on the announcement day, after market closure in Amsterdam. As the companies in the DSJIEM come from around the world, markets are still open in some countries when then announcement is made. Therefore a broader window is taken for the announcement date namely the date, and the date thereafter (AD+1). On the effective date this is not necessary as this is always on a Monday. In Table 3 are the announcement and effective date obtained from the press releases from which follow the sub periods.
Table 3: Applicable Windows (AD and ED)
Year Pre-AD AD AD-ED ED Post-ED
2014 27/08-10/09 11/09-12/09 15/09-19-09 22/09 23/09-7/10 2015 26/08-09/09 10/09-11/09 14/09-18/09 21/09 22/09-6/10 2016 25/08-07/09 08/09-09/09 12/09-16/09 19/09 20/09-5/10 2017 24/08-06/09 07/09-08/09 11/09-15/09 18/09 19/09-4/10 2018 30/08-12/09 13/09-14/09 17/09-21/09 24/09 25/09-9/10 2019 30/08-12/09 13/09-14/09 17/09-21/09 23/09 25/09-9/10
Secondly, De Jong (2007) states that a model should be specified with which the benchmark returns will be calculated. Abnormal returns are defined as the returns, minus the benchmark (normal returns)
𝐴𝑅!" = 𝑅!"− 𝑁𝑅!" (1)
Again following Consolandi et al. (2009) the market model will be used to calculate the normal returns.
𝑁𝑅!" = 𝑎'!+ 𝛽*𝑟#$" (2)
𝑎' and 𝛽* are OLS estimates of the regression coefficients, while 𝑟#$" represents the return of the market benchmark. The market benchmark will be the S&P Emerging Broad Market Index. Other model possibilities include the mean adjusted returns, the market adjusted returns and the CAPM-model as a benchmark. The first two are very limited and don’t consider the beta of the stock. The CAPM model ran out of fashion for use in event studies in the seventies as the validity it imposes on the market model are questionable. When using CAPM not only are you testing the hypothesis, but also the model itself (McKinlay, 1997). Furthermore, in the CAPM risk-free rates are used, and considering that this research is over multiple years in a broad range of countries, no sufficient data is available to use this model. Another possibility is the use of a multifactor model but according to McKinlay this isn’t useful in a situation where the companies are spread out over different countries and industries (McKinlay, 1997).
While most of the event studies use the market model as a benchmark model, the estimation windows used are very diverse. Consolandi et al. (2009) use an estimation window of one year,
where Cheung uses 235 trading days and Oberndorfer et al. (2013) use a 100 day estimation window. In this thesis the benchmark returns are calculated over a period of 100 trading days before the pre-AD time window, from AD-110 until AD-100.
Lastly the results the abnormal returns should be calculated and analyzed. To do this the abnormal returns calculated using (1) and (2) need to be aggregated as most of the periods are over multiple days, these are called the cumulative abnormal returns (CAR):
𝐶𝐴𝑅! = 𝐴𝑅!,"&+ . . +𝐴𝑅!,"' = / 𝐴𝑅!" "'
"("&
(3)
Now we have the CAR, but these are just the results from one company, as we average these over the cross-section of events, we obtain the cumulative average abnormal returns (CAAR)
: 𝐶𝐴𝐴𝑅 = 1 𝑁/ 𝐶𝐴𝑅 ) !(& (4)
When we have obtained the CAAR, we can perform a t-test given that the Central Limit Theorem applies. According to the Central Limit Theorem the distribution of the mean of a random sample is approximately normal, when the sample size is large enough. For event study, it is assumed that any sample size over 30 is large enough (de Jong, 2007). As heteroskedasticity and serial correlations are common problems in finance research, robust standard errors are used in the t-test (Petersen, 2009).
However as in almost all stock return series leptokurtosis is observed (de Jong, 2007), which could violate the normality of the distribution, a non-parametric test is added as well to add robustness to this thesis, the Corrado rank test, with correction for multiday events a described by Cowan (1992). Firstly all abnormal returns, including the estimation period are ranked from lowest value until highest value. After this the following formula is applied to obtain a z-statistic:
𝑧 =
𝐾%
!− (𝑀𝑅)
+∑
""#$(
𝐾%
"%− (𝑀𝑅)/𝑡)
(5)
Where 𝐾2* is the average rank across all stocks during the whole event period, MR is the mean rank, and 𝐾2" is the average rank of all stocks on day t of the combined event and estimation window.
4. Results
In this research an event study was performed to research the impact of inclusion into the Dow Jones Sustainability Index for the Emerging Markets on stock price. Abnormal returns are calculated using the market model with an estimation period of 100 trading days before the inclusion. In this study multiple event periods are observed to consider the various effects of the inclusions on the returns. The research hypothesis is that given the previous research into this and related areas a positive reaction is expected on the inclusion in the DSJIEM.
As displayed in Table 3, the overall influence of the inclusion into the DJSIEM on the returns of companies during the event window is negative of 1.82% (AD-10-ED+10), with a significance of p<0.05 on the Corrado Rank test. The first event window of 10 days before the announcement day was included to observe any effects from leaking information, but given the small and insignificant negative results, there is no reason to assume any leakage of information before the announcement day or any anticipation from investors. During the announcement days a small, insignificant positive effect is seen, but there is not enough evidence to assume any effect of the announcement on the abnormal return of the stock.
The most interesting finding of this research is the significant negative result found on the event window in between the effective and announcement date (AD+1 – ED-1) , which may indicate a delayed reaction from shareholders on the news. On the effective date itself an insignificant negative result is found, but in the period thereafter a significant result is found, which could also indicate a delayed response of shareholders on the inclusion. This research finds that if there is a significant impact from inclusion into the DJSIEM on share price, it is negative.
Table 4: Cumulative Abnormal Returns and Tests
Event Window CAAR(%) robust standard
error
t-test rank test(z)
AD-10 – AD -1 -0.225 .7009169 -0.32 -1,4216815
AD – AD+1 0.338 .3201747 1.05 0,94115962
ED -0.160 .165103 -0.97 -0,8437777
ED – ED+10 -0.797 .6146827 -1.30 -2,4567786*
AD-10-ED+10 -1.823 1.293472 -1.41 -4,1109769*
*denotes significance at 5%
Figure 2 shows the rolling average of the CAAR over the event period to give a better
overall view of the results. In the pre-announcement window, the abnormal returns are stable around 0%, but in the days after the announcement of the inclusion, a sharp decline is seen in the abnormal returns. According to the theory, you could expect an reversal to normal returns after the announcement of the inclusion in the index, but this is also not observed, the
abnormal returns stay negative until the end of the event period.
Figure 2: Cumulative Average Returns over the event period
-2,5 -2 -1,5 -1 -0,5 0 0,5 1 -1 0 -8 -6 -4 -2 AD AD+1 ED 2 4 6 8 10 % Ab no rm al R et ur ns Time
5. Conclusion and discussion
This thesis researched the relationship between corporate social responsibility and stock performance in the emerging markets. This was done by doing an event study on the effect of inclusion into the Dow Jones Sustainability Index for the Emerging Markets on stock returns in between 2014 and 2019.
5.1. Summary
The hypothesis is that based on the instrumental view on stakeholder theory, a positive relationship between inclusion in the DJSIEM and stock returns was expected. However, in this research a mostly negative relationship between the two was found, with negative results over the whole event period except for the announcement day. In the event window between the announcement day and effective day a significant negative effect of -0.978% was found and in the period following the effective date a significant negative effect of -0.797% found. In the whole event period from 10 days before the announcement until 10 days after the effective date of the inclusion a significant negative effect of 1.8% was found. The results do not indicate any effect on the announcement date itself or the window leading up the announcement date.
5.2. Conclusion
Previous research into the link between CSR and financial performance suggest a positive relationship between the two where CSR works as a tool to obtain higher financial performance. Two of the possible explanations for the results of this research on this context are that shareholders don’t think the benefit of having good CSR policies in place don’t weigh up against the investments it requires, or the relationship between CSR and financial performance works different in the context of the emerging markets than it does in developed countries.
Most previous theory into index inclusion also suggests a positive relation between inclusion effects in general and stock returns. Previous research in index inclusion found several positive inclusion effects which are based on whether the inclusion signals any new information to the market. If we assume that new information is given to the market, and investors use this information to come to a new share price, it can be concluded that inclusion in the DSJIEM is regarded as negative by investors. More broadly, if we take inclusion into
an sustainability index as a proxy for having good CSR, this means that investors do not value companies performing well on CSR, and investing in CSR in the emerging markets. Where most of the research into sustainability index inclusion has positive results,
Oberndorfer et al. (2013) found similar results in Germany. They suggest that the actions performed to be in a sustainability index are purely symbolic, and cost more money than they deliver and therefore the inclusion is seen as bad news by the market. Another explanation for the negative result of this research may be related to the research done by Doberts and Halme (2009). They pose that CSR in emerging markets happens in a different context, and therefore takes another form. As the DJSIEM is built on a view on CSR from developed economies, it might be that investors don’t value the application of Western CSR standards outside of the western world. In developed countries going beyond the law is what is expected from companies performing good CSR, where in developing nations simply following the law can be seen as good CSR.
While the hypothesis is based on the instrumental view of stakeholder theory, the results of this research are more in line with Friedman’s shareholder value theory in which shareholder value is the only responsibility of the company and good CSR is mostly seen as a distraction from this.
5.3. Limitations
This research however is limited in several ways. One of the limitations of this research is the way in which the DSJIEM is composed. The index is composed using a best-in-class approach, which contains the best performers in terms of CSR of each industry. A direct result from this approach is that some companies are included from industries that are regarded by many as having naturally bad CSR, like the oil or weapons industry.
Furthermore, if the mean CSR level in a certain industry is low, then becoming the best in class company does not indicates much about the actual level of CSR of a company. Lastly, as the DJSI is composed by a European company it inherently looks at CSR from a western viewing point, and uses western metrics to rate CSR policies.
Another limitation is the fact that the emerging markets are a broad group of countries with huge variations in government, level of development, and many more factors. It could be that this group of countries is simply too diverse to asses in one event study, and should be investigated separately. However, research into the emerging markets found that the sort of
companies researched in this thesis often have more in common with each other than with other companies in their own country (Baskin, 2006).
Lastly, the post-event window researched in this thesis might not be long enough to see longer term results of the inclusion. The research of Cheung (2011) found price reversal 7 days after the announcement which is why the post-event period in this thesis was chosen to be 10 days. However, research from Stekelenburg (2015) found the price reversing 60 days after the announcement, and therefore found evidence for supporting the price pressure hypothesis.
In the future, more research into the workings of CSR in the emerging markets is recommended. Past research already stated that CSR takes different forms depending on the economic and societal context in countries, but doesn’t tell much about these different forms. Furthermore, research in the linkage between CSR and corporate performance in the
emerging markets is also needed. As currently most of the research is based on western economies, not much is known about the benefits and drawbacks of investing in CSR in the emerging markets.
Bibliography
Arlow, P., & Gannon, M. (1982). Social Responsiveness, Corporate Structure, and Economic Performance. Academy of Management Review, 7.
Baskin, J. (2006). Corporate responsibility in emerging markets. Journal of corporate citizenship, (24), 29-47.
Belal, A. R. (2001). A study of corporate social disclosures in Bangladesh. Managerial Auditing Journal.
Cheung, A. W. K. (2011). Do stock investors value corporate sustainability? Evidence from an event study. Journal of Business Ethics, 99(2), 145-165.
Cochran, P. L., & Wood, R. A. (1984). Corporate social responsibility and financial performance. Academy of management Journal, 27(1), 42-56.
Consolandi, C., Jaiswal-Dale, A., Poggiani, E., & Vercelli, A. (2009). Global standards and ethical stock indexes: The case of the Dow Jones Sustainability Stoxx Index. Journal of
Business Ethics, 87(1), 185-197.
Curran, M. M., & Moran, D. (2007). Impact of the FTSE4Good Index on firm price: An event study. Journal of environmental management, 82(4), 529-537.
Dahlsrud, A. (2008). How corporate social responsibility is defined: an analysis of 37 definitions. Corporate Social Responsibility and Environmental Management, 15(1), 1–13. https://doi.org/10.1002/csr.132
De Jong, F. (2007). Event studies methodology. Lecture Notes.
Denis, D., Mcconnell, J., Ovtchinnikov, A., & Yu, Y. (2003). S&P 500 Index Additions and Earnings Expectations. Journal of Finance, 58(5), 1821–1840.
Donaldson, T., & Preston, L. E. (1995). The stakeholder theory of the corporation: Concepts, evidence, and implications. Academy of management Review, 20(1), 65-91.
Friedman, M. (1962). Capitalism and freedom. University of Chicago press.
Göbbels, M.: 2002, Reframing Corporate Social Responsibility: The Contemporary Conception of a Fuzzy Notion.
Jain, P. (1987). The Effect on Stock Price of Inclusion in or Exclusion from the S&P 500. Financial Analysts Journal, 43(1), 58–65.
Harris, L., & Gurel, E. (1986). Price and volume effects associated with changes in the S&P 500 list: New evidence for the existence of price pressures. the Journal of Finance, 41(4), 815-829.
Hegde, S., & Mcdermott, J. (2003). The liquidity effects of revisions to the S&P 500 index: an empirical analysis. Journal of Financial Markets, 6(3), 413–459.
Jones, T. M. (1995). Instrumental stakeholder theory: A synthesis of ethics and economics. Academy of management review, 20(2), 404-437.
KPMG. (2020). Sustainable Investing: Fast-forwarding its Evolution. Retrieved from:
https://assets.kpmg/content/dam/kpmg/xx/pdf/2020/02/sustainable-investing.pdf. 2020. MacKinlay, A. C. (1997). Event studies in economics and finance. Journal of economic literature, 35(1), 13-39.
Merton, R. (1987). A Simple Model of Capital Market Equilibrium with Incomplete Information. The Journal of Finance, 42(3), 483-510. doi:10.2307/2328367
Oberndorfer, U., Schmidt, P., Wagner, M., & Ziegler, A. (2013). Does the stock market value the inclusion in a sustainability stock index? An event study analysis for German firms. Journal of Environmental Economics and Management, 66(3), 497-509.
Orlitzky, M., Schmidt, F. L., & Rynes, S. L. (2003). Corporate social and financial performance: A meta-analysis. Organization studies, 24(3), 403-441.
Ortas, E., Moneva, J. M., & Salvador, M. (2012). Does socially responsible investment equity indexes in emerging markets pay off? Evidence from Brazil. Emerging Markets Review, 13(4), 581-597.
Petersen, M. (2009). Estimating Standard Errors in Finance Panel Data Sets: Comparing Approaches. The Review of Financial Studies, 22(1), 435–480.
RobecoSAM (2013). RobecoSAM and S&P Dow Jones Indices introduce DJSI Emerging Markets. Retrieved from https://www.prnewswire.com/news-releases/robecosam-and-sp-dow-jones-indices-introduce-djsi-emerging-markets-192270281.html
Ruf, B. M., Muralidhar, K., Brown, R. M., Janney, J. J., & Paul, K. (2001). An empirical investigation of the relationship between change in corporate social performance and financial performance: A stakeholder theory perspective. Journal of business ethics, 32(2), 143-156.
Shleifer, A. (1986). Do Demand Curves for Stocks Slope Down? Journal of Finance, 41(3), 579–590.
Smith, H. J. (2003). The shareholders vs. stakeholders debate. MIT Sloan Management Review, 44(4), 85-90.
Sparkes, R., & Cowton, C. J. (2004). The maturing of socially responsible investment: A review of the developing link with corporate social responsibility. Journal of Business Ethics, 52(1), 45-57.
S&P Global (2020). Dow Jones Sustainability Indices Methodology guidebook. Retrieved from https://www.spglobal.com/esg/csa/csa-resources/csa-methodology
S&P Global (2020). Celebrating 20 Years of the Dow Jones Sustainability Indices. Retrieved from https://www.spglobal.com/esg/csa/about-us/history
Van Marrewijk, M. (2003). Concepts and definitions of CSR and corporate sustainability: Between agency and communion. Journal of business ethics, 44(2-3), 95-105.
Stekelenburg, A., Georgakopoulos, G., Sotiropoulou, V., Vasileiou, K., & Vlachos, I. (2015). The relation between sustainability performance and stock market returns: An empirical analysis of the Dow Jones Sustainability Index Europe. International journal of Economics and Finance, 7(7).
Waddock, S. A., & Graves, S. B. (1997). The corporate social performance–financial performance link. Strategic management journal, 18(4), 303-319.
