How do investors value sustainability?
An investigation on the importance of corporate
sustainability before and after the financial crisis.
June 9, 2015
Name: Bas Baas
Student number: 10002406
Specialization: MSc Accountancy & Control, variant Control Thesis supervisor: G. Georgakopoulos
Date & Place: June 9, 2015, Amsterdam Thesis: Master Thesis
Word count: 16,420
2
Statement of Originality
This document is written by student Rudolf B. Baas who declares to take full responsibility for the contents of this document.
I declare that the text and the work presented in this document is 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.
3 Table of Content Abstract p. 4 Introduction p. 4 Background p. 4 Research question p. 6 Motivation p. 7
Literature review and hypotheses p. 9
Literature review p. 9
What is sustainability? p. 9
The Dow Jones Sustainability World Index p. 10 Social Responsible Investments p. 12 Sustainability strategies creates long-term firm value p. 13
The financial crisis p. 14
Resource-based view, legitimacy- and stakeholder theory p. 16
Hypotheses p. 17
Data p. 21
Research methodology p. 29
Regressions p. 29
Explanations and expectations of the variables p. 31
Results p. 33
Conclusion p. 39
Discussion p. 40
Interpretation of the results in light of existing literature p. 40
Suggestions for further research p. 43
4
Abstract
During the last decennia there is a shift in corporate sustainability strategies amongst large companies. Companies used to focus solely on the maximization of shareholders value but currently they tend to put more effort in strategies which include a focus on environmental
and social aspects. Companies have experienced the punishment by society when their operations are regarded as being unsustainable. On the contrary companies have experienced the customer loyalty when they are being socially accepted as being a corporate sustainable company. Sustainable companies focus their strategies on making profit but also on the environment and social aspects. But is this change in strategies accepted by investors? Are investors willing to invest in corporate sustainable companies or is this seen as a waste of money? Do the costs that companies make in order to be sustainable benefit the company in
such a way that these costs are faded out by residual income?
This paper shows that after the financial crisis in 2008 investors are more focused on investing in sustainable companies than before. After the crisis there is a significant increase in abnormal returns when companies are added on the Dow Jones Sustainability Index. When
investors would only invest in companies which are added on the Dow Jones Sustainability Index during the years 2009 and 2010 they would have made an abnormal return of 6.42% (compared to the expected returns based on the CAPM-model). These investors would have
made an actual return of 16.93% compared to the expected cumulative return of 10.51%. When investors would have invested in the companies on the S&P500 during the years 2009 and 2010 they would have made an actual cumulative return of 13.12% which is 3.81% lower
than the actual return resulted from investing in the companies included on the Dow Jones Sustainability Index.
1. Introduction
1.1 Background
Corporate sustainability reports are defined as “public reports by companies to provide
internal and external stakeholders with a picture of the corporate position and activities on economic, environmental and social dimensions” (WBCSD, 2002). This definition
emphasizes the necessity for companies to report on economic, environmental, and social aspects of sustainability. The motivations for corporate sustainability reporting include “enhanced ability to track progress against specific targets”, “greater awareness of broad environmental issues”, “improved all-round credibility from greater transparency”, “reputational benefits”, among others (Kolk, 2010). There is an interest in using the
5
information in sustainability reports to guide investment decisions. However investors require information that goes beyond the information within the corporate sustainability reports. Therefore indices linked to financial markets have emerged, including the Dow Jones Sustainability Index (DJSI) (Searcy et. al., 2012).
Hoon et. al. (2013) examined how firms from the financial sector are evaluated and rated via the Dow Jones Sustainability World Index (DJSWI). They also introduce the term Social Responsible Investment (SRI), which refers to: “the practice of directing investment
funds in ways that combine investors financial objectives with their commitment to social concerns.” Most major fund managers have introduced SRI funds, in 2010 professionally
managed assets based on SRI strategies had a value of $3.07 trillion. SRI funds have two different aspects: they may influence companies to change their behavior regarding
sustainability and SRI funds’ performance is not different from conventional investments in the short run but it is likely to be superior in the long run. Companies that put emphasis on Corporate Social Responsibility (CSR) will gain a competitive advantage and are likely to outperform their peers (Hoon et. al. 2013). For funds to invest in these companies investors can make profits from the high growth potential of sustainability-driven products.
Cheung (2011) investigated the effect of index inclusions and exclusions on corporate sustainable firms which are included or excluded from the DJSWI. He didn’t find strong evidence that the announcement (of companies being included or excluded by RobecoSAM) has a significant impact on stock returns, but he did measure a temporary shock on the day of change. As Cheung (2011) describes, if financial markets do not value the companies’ efforts on corporate sustainability, they do not have enough incentive to become or continue to be a corporate sustainable company. What Cheung (2011) wants to provide is an evidence on two issues: do investors value corporate sustainability? And in what way they value corporate sustainability? Cheung’s (2011) results show that index inclusion stocks result in an increase in stock returns, whereas index exclusion stocks have a decrease in stock returns. Cheung (2011) investigated the period 2002 – 2008, with this paper we would like to invest if there is a difference between the period before and after the financial crisis. Do investors regard sustainable companies of being more valuable after the financial crisis? And was therefore the financial crisis an event to shift the mindset of investors towards more sustainable companies?
The adoption of sustainability strategies should grant companies a competitive
advantage over companies who do not adopt these strategies (Adams and Zutshi, 2004). CSR practices (e.g. management quality, environmental management, brand reputation, customer loyalty, corporate ethics and talent retention) are beginning to become more important for
6
successful businesses to integrate in their strategies (Lopez et. al., 2007). Measures included in the term CSR (adopting ethical codes, better environmental practices, or human capital development) are usually considered a good strategy which should lead to better performance. Thus, focusing on CSR practices and developing a strategy based on the term CSR would enable long-term sustainable competitive advantages (compared to companies who doesn’t adopt these CSR practices (Lopez et. al., 2007).
Lopatta et. al. (2013) showed in their paper that the financial crisis in 2008 caused a shift in social responsible investments. When Lehman Brothers filed for bankruptcy in September, 2008 this positively affected the perception of corporate sustainability. As their analysis shows, the financial crisis led to a positive perception of corporate sustainability in industries that are exposed to higher environmental and social risks. Lopatta et. al. (2013) showed that companies in industries with high environmental and social risks should increase their focus towards corporate sustainability.
1.2 Research question
Based on prior literature it is interesting to investigate what the effect is on share prices of companies who are changing their sustainability practices. If companies choose to show the world they put more effort in being sustainable does the market value this, and if so what is the value of these changes? Thereby is there a change in importance of sustainability after the shock of the financial crisis in 2008? Therefore the research question of this paper is focused on answering the question what the value is of corporate sustainability before and after the financial crisis. By focusing on a widely used sustainability index and investigating the effect of an inclusion or exclusion on this index we try to calculate the value of corporate
sustainability. By investigating the difference between the value of corporate sustainability before and after the financial crisis we would like to establish the change in the way investors regard the importance of corporate sustainability between these two periods.
Research question:
What is the relation between the inclusion or exclusion of a company on the Dow Jones Sustainability World Index and the share price of North American listed companies before
7
1.3 Motivation
1.3.1 Scientific point of view
The Lopez et. al. (2007) paper describe the importance of companies listed on the DJSWI. It is a way to show investors that the company is long-term oriented and this could create a better long-term performance and therefore long-term value creation. But does the stock market also react towards an inclusion or exclusion of a company on the DJSWI, both
positively and negatively? And if it does, how much value do these investors appoint towards these changes? And finally, has the importance of sustainability changed after the financial crisis? With the research performed for this paper those questions will be answered. Robinson et. al. (2011) describe the effect on the share price of companies included on the DJSI or excluded from the DJSI. They found a significant increase in share prices when the company was included on the DJSI but they didn’t find any results to proof that the share price would decrease when the company was excluded from the DJSI. Robinson et. al. (2011) investigated the period 2003 – 2007 and therefore their time period ended before the financial crisis. Lopatta et. al. (2013) investigated the effect of the financial crisis on the valuation of corporate sustainability. They didn’t use the DJSI as the sustainability index but they used the international rating agency: Global Engagement Services (GES). Lopatta et. al. (2013) did find a change in abnormal returns when a company is proven to be corporate sustainable after the financial crisis compared to before the financial crisis. On the contrary Lopatta et. al. (2013) didn’t find any proof that companies who are no longer operating in a corporate sustainable manner are being “punished” by investors. Therefore this paper is investigating if an inclusion and exclusion of a company on the DJSI, both affect the stock market positively and negatively. And thereby if these effects have increased after the financial crisis. The analysis of Lopatta et. al. (2013) also shows that the financial crisis led to a positive
perception of corporate sustainability in industries that are exposed to higher environmental and social risks. Lopatta et. al. (2013) showed that companies in industries with high environmental and social risks should increase their focus towards corporate sustainability. Therefore this paper also includes an investigation to discover if there are any differences in abnormal returns amongst the different industries. Moreover, is there a higher difference in abnormal returns between the different industries after the financial crisis compared to before the financial crisis.
8
1.3.2 Societal point of view
Academics have started to identify the importance of corporate sustainability, but companies are mainly focused on whether shareholders are aware of their corporate sustainable choices Cheung (2011). If financial markets do not value the effort companies put in corporate sustainability, companies don’t have enough incentives to become or continue to be a corporate sustainable company (Cheung, 2011 and Consolandi et. al., 2010). Therefore it is very important to measure and understand the value of corporate sustainability. What is the value increase (decrease) when a company is included (excluded) on the DJSWI? If we can answer this question we can motivate companies to put more emphasis on being included to the DJSWI and maintain their position. Because the main focus of companies is long-term profitability, it would be desirable if this long-term profitability could be connected with companies’ corporate sustainability. This way all its stakeholders benefit and not only the shareholders.
This paper is organized as follows. First, it presents the literature review where there is a short introduction on what sustainability entails, thereafter it is explained what the Dow Jones Sustainability Index is and how it is composed. Furthermore the literature review consists of an explanation about Social Responsible Investments, long-term value creation caused by corporate sustainability, the financial crisis and lastly different theories which have a connection with corporate sustainability will be explained. Secondly, it is described what the different hypotheses of this paper are and how they are supported by existing literature.
Thirdly, it presents the data which is used and how this was obtained. Furthermore, this part will describe the abnormal returns of companies included or excluded from the DJSI.
Thereafter the Research Methodology will describe the different regressions and analysis used for this paper. Thereafter the different results will be presented, the abnormal returns of companies included on the DJSI and excluded from the DJSI; the difference in abnormal returns of companies included on the DJSI after the financial crisis compared to before the financial crisis; the different abnormal returns amongst the different industries. Finally, the paper presents the conclusion and discussion. In this part it is described what can be concluded from this investigation and how this is connected with existing literature.
9
2. Literature review and hypotheses
2.1 Literature review
2.1.1 What is sustainability
Originally the concept of sustainability was created in forestry, where it means never
harvesting more than what the forest yields in new growth (Kuhlman et. al., 2010). Currently this is called sustainable development which means in the Commission’s words (WCED, 1987): “development that meets the needs of the present without compromising the ability of
future generations to meet their own needs”. There have been major developments in the
concept of sustainability, it’s interpretation in terms of three dimensions: social, economic and environmental.
Elkington (1994) created the idea for sustainability having three ‘pillars’ from the Triple Bottom Line concept. The bottom line (profit) should be added care for the
environment (planet) as well as taking care of the social aspect (people). Companies need to attain a certain minimum performance in those three areas in order to be regarded as
‘sustainable’. Sustainable business requires not only change within the company, but also a new type of interaction. Being socially accepted by the community in which the company operates is crucial to sustainable business. This is sometimes referred to as a firm’s ‘license to operate’ i.e. is the firm accepted by society (Cramer, 2002). This ‘license to operate’ is also in line with legitimacy theory. Legitimacy theory can be defined as an assumption that the actions of a company are desirable within some socially constructed system of norms and values (Suchman, 1995). The society where the company operates has a system of norms, values and beliefs. When a company operates within these norms, values and beliefs it is possible the company will be regarded as a legitimate company. However when the firm lacks legitimacy this can result in a decline in profit or even bankruptcy of the company because customers or suppliers will refuse to do business with that company (Lopatta et. al., 2013).
As a result of several negative experiences in the past, firms are reluctant to promote a socially responsible image. Firms that are communicating their social and environmental virtues might be attacked when this is proven to be wrong (Simintiras, 1994). Cramer (2002) gives as an example the ‘trade not aid’ policy adopted by The Body Shop. This policy involved an arrangement where they (The Body Shop) only bought nuts from Brazilian Indians and not from any other source. This ‘good publicity’ was turned into ‘bad publicity’ by sceptics. One of the accusations was that the amount of money involved in this particular arrangement was so small that it formed only a negligible part of the firm’s overall
10
trading transformed into a company who had difficulties in being socially accepted. Another example of the economic consequences of violating customers’ values and beliefs is the Shell boycott (Lopatta et. al., 2013). This boycott started after Shell announced plans to dump its oil storage in the Atlantic Ocean. None of its customers were directly affected by these oil
dumps, although many refused to buy gasoline at Shell gas stations. Customers did not wanted to be associated with such a company and therefore refused to do business with Shell. This caused revenues and share prices to decline significantly (Sandhu, 2010).
So when is a company ‘sustainable’? How do investors know which companies are focused on all the three aspects of sustainability: profit, people and planet? This was one of the reasons the Dow Jones Sustainability World Index was introduced in the late 90s (Hoon et. al., 2013). This index calculates the sustainability score of the largest companies and therefore gives investors a good indication on how sustainable these companies really are. The DJSWI is focused on three dimensions: economic, environment and social. The DJSWI specifies different industries and appoints different criteria to these industries. Every industry is different and therefore the DJSWI set different weights to the different criteria amongst the industries, therefore the DJSWI obtains the perfect sustainability score for that specific industry. In the next paragraph the DJSWI will be explained, how it started and how the industry criteria is composed.
2.1.2. The Dow Jones Sustainability World Index
The Dow Jones Sustainability World Index (DJSWI) was introduced in 1999, at the time it was the first global sustainability benchmark (Hoon et. al., 2013). The Dow Jones
Sustainability Indexes (The Dow Jones Sustainability World Index, The Dow Jones STOXX Sustainability Index, The Dow Jones Sustainability Index North America and the Dow Jones Sustainability Asia Pacific Index) were established to track the performance of company leaders in the field of corporate sustainability (Searcy et. al., 2012). The DJSWI tracks the performance of the top 10% of the 2,500 largest companies. The DJSWI uses information from annual surveys, annual reports, sustainability reports, media coverage, and other company documents to rate the CSR performance of those companies (Hoon et. al., 2013).
The DJSWI is divided into 59 different industries (RobecoSAM, 2014). Each industry has different weights appointed to the different measures. There are three different
dimensions: Economic, Environmental and Social. The emphasis on each dimension is
dependent on the industry. Each dimension has different criteria where the company can score points, with each criteria assigned a different weight. Examples of criteria are: antitrust
11
policy, codes of conduct, corporate governance, customer relationship management, risk & crisis management and tax strategy (economic dimension); climate strategy, environmental policy/management system, environmental reporting, operational eco-efficiency and product stewardship (environmental dimension); enabling local development, human capital
development, labor practice indicators & human rights, occupational health & safety, social impacts on communities, stakeholder engagement and talent attraction & retention (social dimension) (RobecoSAM, 2014).
The importance of each dimension is different amongst the industries. By assigning weights to every criteria the importance of the dimension is taken into account when assessing the total score. For instance the banks industry weighs the economic dimension a total of 41%, the environmental dimension a total of 23% and the social dimension a total of 36%; the chemicals industry weighs the economic dimension a total of 36%, the environmental
dimension a total of 32% and the social dimension a total of 34%; whereas the pharmaceutical industry weighs the economic dimension a total of 45%, the environmental dimension a total of 10% and the social dimension a total of 45% (RobecoSAM, 2014). These differences in weights affect the total score differently between the different industries. If the company is determined to be included on the DJSI they are forced to put more focus on the dimension with the highest weights. Therefore banks should put more focus on the economic dimension than on the environmental dimension, which has a difference of 18% (in determining the total score). However when a company is included in the pharmaceutical industry the difference between the economic or the environmental dimension is 35%.
Another important aspect of the DJSI assessment is the ongoing monitoring of media and stakeholder commentaries and other publicly available information. On a daily basis monitoring is conducted. This includes host of issues such as corruption, fraud, human rights violation, workplace safety, labor disputes, catastrophic accidents and environmental
disasters. If the company is accused of something that could harm its reputation, resulting in financial consequences. It would require a reaction from the company in order to minimize the negative impact on the company’s image (Hoon et. al., 2013). As the examples of The Body Shop and Shell show in the previous chapter, firms can be punished by their customers and suppliers if they are operating outside the norms, values and beliefs of those stakeholder groups (Lopatta et. al, 2013). These effects are included in the DJSI assessment.
The DJSI is a widely used benchmark of social responsible investments (SRI). ‘SRI’ refers to the practice of managing investment funds in ways that combine investors’ financial objectives and their commitment to social concerns (Hoon et. al., 2013). In the next section
12
SRI will be reviewed, and it will be explained what the contribution of SRI is to the change in share prices caused by the inclusion of companies on (or exclusion of companies of) the DJSI.
2.1.3 Social Responsible Investments
Since the 90s social responsible investments (SRI) have grown rapidly around the world. An important influence on this growth was caused by ethical consumerism, which means that consumers pay a premium for products that meet their personal values (Renneboog et. al., 2008). Things like environmental protection and human rights have become important issues within the SRI investments. Another important factor which plays a role in the growth of SRI are the changes in regulation regarding the disclosure of social responsible and sustainable practices for listed companies. Renneboog et. al. (2008) describes what SRI investors aim to do. They aim at promoting strong social and environmental corporate behavior. Where they avoid companies who operate and exploit employees either in developed or developing countries, which is called ‘negative screening’. Therefore they only select companies with strong social and environmental records and with a good corporate governance, which is called ‘positive screening’. To summarize, SRI investors expect companies to not only focus on shareholder’s interest but to focus on social welfare in addition to value maximization.
As Bakshi (2007) describes the SRI is growing rapidly. In 1984 the total assets involved in social investing in the United States of America was $40 billion. By 1995, this was increased towards $639 billion. After this year SRI funds kept on growing, in fact they have grown 40% faster than the total investment market. In 2003 there was a total of $2.3 trillion SRI assets in the US, from which $2 trillion was based on negative screening
Renneboog et. al. (2008). Positive screening is also a widely used method to select companies with a good record concerning energy usage or community involvement. This is often
combined with a ‘best in class’ approach, where firms are ranked based on CSR criteria. Negative and positive screens are referred to as the first and second generation of SRI screens (Renneboog et. al., 2008).
As Renneboog et. al. (2008) describes the most important question regarding the SRI movement: is a firm’s aim to maximize shareholder value or social value? The maximization of shareholder value does conflict (most of the time) with the social welfare criterion. By maximizing shareholder value, firms are not taking care of the interest of their other stakeholders. If a firm maximizes shareholder value it needs to be low on costs, therefore stakeholders as employees and suppliers will suffer from these cuts. But can a company survive when it focuses on all its stakeholders? A firm lowering its profits to pursue social
13
and environmental goals may not survive the competition. Another company could acquire this firm and replace the management with a value-maximizing one (Tirole, 2001).
In 2010, professionally managed assets following SRI based strategies had a total value of $3.07 trillion, which was an increase of more than 380% since 1995 ($639 billion). During the financial crisis, from 2007 till 2010, SRI assets showed healthy growth while the overall size of professionally managed assets remained somewhat flat (Hoon et. al., 2013). In the short term the financial performance of SRI funds is no different from conventional investments. But in the long run SRI funds are likely to be superior. As Hoon et. al. (2013) describes: “companies taking a lead in CSR will gain a competitive advantage and likely
outperform their peers”.
2.1.4 Focusing on Corporate Social Responsible and Corporate Sustainable strategies creates long-term competitive advantage and value creation
Lopez et. al. (2007) describes the fact that society has begun to demand a change in
companies behavior toward sustainable development. Sustainability philosophy suggests that we should forget the narrow version of classical economic theory and focus on a wider group of stakeholders. Friedman (1970) describes classic economic theory as the only social
responsibility a company has is maximization of company value. Companies should only focus on shareholder interest. From this point of view, any social act is associated with costs that would reduce profit, and therefore it would prejudice shareholders. From the
sustainability philosophy point of view we should develop corporate strategies which are not only focused on shareholder’s interest, but on the demands of a wider group of stakeholders (Lopez et. al., 2007). Stakeholder theory is focused on this point of view (where the business focuses on all its stakeholders), this theory will be explained in paragraph 2.1.6.
In the short term firms are only able to use existing resources to apply sustainable practices, since the time frame is not enough for obtaining supplementary financing. In the long run, resources needed to carry out CSR strategies can be financed. If the changes remain over time, they may create a differentiation which results in long-term competitive advantages (Gladwin et. al., 1995). An important factor for companies to choose a social responsible strategy is the economic conditions of that company (Campbell, 2007). Weak corporate financial performance reduces the probability of a company to obtain a corporate sustainable strategy, whereas the level of competition increases this responsible behavior (Campbell, 2007).
14
Indexes linked to financial markets have emerged because of the importance of
sustainability practices for investors’ investment portfolios. The basic idea of these indexes (such as the DJSWI) is that sustainability practices contribute to long-term value creation. These practices support the development of opportunities and manage economic,
environmental and social risks (Lopez et. al., 2007). To be included to sustainability indexes, firms are required to disclose information that reflects their sustainability practices (which usually appear in their sustainability reports).
Consolandi et. al. (2010) describe two different views regarding financial performance of companies putting effort in being sustainable, the skeptical view and the positive view. The skeptical view states that the focus on corporate social responsibility would increase costs and therefore would reduce financial performance. On the contrary, the positive view states that the standards of corporate social responsibility reached by a firm can be seen as a sign of good management, a management who is able to mediate the interest of different stakeholder groups in a long-term perspective. According to the positive view the evaluation of corporate social responsible performance could be considered a useful criterion for asset allocation. Excellent corporate social responsible standards would therefore increase demand of a stock because of the company’s focus on long term firm value. Because of this increasing demand of stock caused by higher corporate social responsible standards managers are incentivized to further strengthen its social responsible standards. Consolandi (2010) states that this virtuous circle has a positive effect on the sustainability of firms and of the entire economy. Therefore companies should focus more and more on non-financial aspects of corporate performance.
A good example of a period when there was a strong short-term focus were the years before the financial crisis. Bank managers were focused on selling mortgages to every US citizen to increase their (short term) bonuses. As stated in the previous paragraph it was during the financial crisis that SRI assets showed healthy growth while the overall size of professionally managed assets remained somewhat flat (Hoon et. al., 2013). Therefore it is possible that after the financial crisis the returns of sustainable companies (which are included on the DJSI) are higher than their unsustainable peers because there is a focus shift towards sustainable companies.
2.1.5 The financial crisis
From August 2007 till August 2008 the U.S. mortgage crisis became a deep financial crisis. In the third quarter of 2008, the United States entered a recession (Grigor’ev et. al., 2009). On September 15, 2008 (at the time the fourth largest investment bank) Lehman Brothers went
15
bankrupt. The bankruptcy of the Lehman Brothers triggered this financial crisis in the United States (Arner, 2009). One of the main causes of the 2008 financial crisis was the high
amounts of leverage among institutional investors and the public. High consumer leverage was stimulated by low interest rates. On top of that were the unregulated mortgage products which made it easy for every US citizen to get a mortgage. These aspects led to the rise of prices within the housing market (Brummer, 2010).
The financial crisis was preceded by a significant boom: during the period 2003 – 2008 international gross domestic product (GDP) grew by one-third. By 2008 the
characteristics of a cyclical crisis had been created: overaccumulation during an intensive boom, growing sector imbalances, and increased inflation (Grigor’ev et. al., 2009). This boom lost its rise in August 2007. The problem of worthless American assets increased during the coming 12 months. By September 2008 the write-downs totaled to some hundreds of billions of dollars (Grigor’ev et. al., 2009).
The financial crisis should lead to a conformation towards financial professionals that they should adapt their view on the road they should be taking. As van de Ven et. al. (2011) describes the road they should be taken include: the raise of transparency; not only be focused on serving their own interests and the interest of short-term focused shareholders, but put more emphasis on the interests of other stakeholders; reduce the degree of risk taking by lowering leverage; moderate the bonuses of the bank managers so that they will be more long-term focused; show a willingness to learn from past mistakes. Those aspects are intertwined with the aspects of the three pillars of sustainability. Companies who are proven to be sustainable are more transparent, focus on all types of stakeholders, don’t have a high risk profile, and are focused on the long run. This would indicate that after the crisis investors should have learned from their mistakes and would be investing more in sustainable companies.
Lopatta et. al. (2013) investigated the effects of sustainability on the market value of MSCI (Morgan Stanley Composite Index) World firms by using the international rating agency: Global Engagement Services (GES). Their results showed (at the beginning of the sample period: 2003 – 2007) a negative relation. When companies were more sustainable this would have a negative effect on the market value of MSCI World firms. But they found a change after the financial crisis. After the bankruptcy of Lehman Brothers (September, 2008) they found that the previously negative perception of corporate sustainability was positively affected. The financial crisis may have been a trigger towards recognizing the benefits of sustainability. Results of Lopatta et. al. (2013) show that sustainability is relevant to firm
16
valuation. And therefore managers who have not recognized the change in perception are not following a value-maximizing strategy when they are not focused on corporate sustainability. There are several theories which establishes a link between corporate sustainability and the market value of a firm, namely, the resource-based view, legitimacy theory and stakeholder theory. In the next paragraph it is explained how these theories are able to be used to link sustainability with the market value of a firm.
2.1.6 Resource-based view; legitimacy theory and stakeholder theory
The resource-based view sees expenditures which are focused on the environment or socially desirable business practices as investments that increase the reputation of the company and therefore lead to higher long-run profits. This theory states that corporate sustainability contributes to the reputation of the company and can therefore be regarded as a profitable management practice (Lopatta et. al., 2013). Maurer et. al. (2010) describes resource-based view as something that may create or destroy value. When firms set a strategy that becomes associated with a social issue it creates a risk for the firm. It can have a positive or a negative effect because the social issue expands the reach of the firm, when this is done properly it will be rewarded by society, but if it’s done wrong society will punish the company (as was shown with the example of Shell in paragraph 2.1.1). But firms that recognize the interplay between their resources and their environmental/social context are able to engage in cultural work and thereby preserve their strategy’s economic value.
From a legitimacy theory point of view, corporate sustainability can be regarded as a managements’ intention to ensure a going concern (Lopatta et. al., 2013). It increases the life expectancy of the firm and therefore increases the future cash flows, which causes to increase firm value. Legitimacy theory can be described as a generalized perception that the actions of an entity are desirable within the socially constructed norms and beliefs (Suchman, 1995). If the behavior of the firm is not according these socially constructed norms and beliefs it is possible that customers or suppliers refuse to do business with this company. Then, the company lacks legitimacy, which can result in a decline in profit or even bankruptcy (Hybels, 1995). In other words a company’s performance is legitimate when it is judged to be fair and worthy of support, e.g. when it is socially accepted (Eugenio et. al., 2013). But when society is not convinced that the company is operating in a legitimate manner, a legitimacy gap can arise. A legitimacy gap exists when there is a difference in a corporations’ actions and the society’s beliefs of what these actions should be (O’Donovan, 2002). In other words, a legitimacy gap is the difference between the societal expectations and the perceptions of
17
business behavior. To improve its legitimacy within a society, a company wants to minimize this legitimacy gap (Panwar et. al., 2014). In order to narrow this gap, a company must investigate this gap in light of its own definitions of “what ought to be” and “what is”
appropriate behavior. Throughout this investigation the legitimization gap can be reduced by knowing what is expected from the company by society (Panwar et. al., 2014).
Stakeholder theory has been developed over the last thirty years as a response to the dominant mindset where corporations are seen as the property of their owners (e.g.
shareholders in public corporations). If stakeholder theory is stripped down to its essentials it emerges out of four ideas: the separation fallacy, the open question argument, the integration thesis, and the responsibility principle (Freeman et. al., 2010). The separation fallacy states that business and ethical decisions can be regarded as separate decisions. However there are implications of rejecting the separation fallacy, almost any business decision has some ethical content. The open question argument is based on asking the questions “if a business decision is being made what value is created and destroyed?” and “who is harmed and who does benefit from this decision?” But an answer could be “only value to shareholders count”. Therefore the integration thesis is needed as a theory, where it is stated that most business and ethical decisions are intertwined. And finally the responsibility principle states that most people want to accept responsibility for the effects of their actions on others (Freeman et. al., 2010). It is clear to see that stakeholder theory is a mix of the integration thesis and the responsibility principle. Stakeholder theory implies that the interest of these groups are joint and that to create value, companies must focus on how value gets created for every
stakeholder. Godfrey (2005) used stakeholder theory to describe how corporate sustainability can create shareholder value by creating and maintaining positive stakeholders surrounding the company. These positive stakeholders can serve as a protection for the company’s
relationship intangible assets. Because the company is aware of this positive relationship and therefore is able to reduce its exposure to stakeholder risk (loss of customers, loss of
suppliers) it is able to set a lower cost of capital and therefore (ceteris paribus) higher firm value.
2.2 Hypotheses
Prior literature (Lopez et. al., 2007) shows corporate sustainability as a way to proof towards the investors that the company is focused on the long term. The DJSI is a widely used index for investors to know which companies can be regarded as being sustainable and which companies can’t. Prior literature also shows that the financial crisis could have triggered
18
investors to focus more on corporate sustainable companies. As Lopatta et. al. (2013) show, the negative perception of corporate sustainability has made a shift towards a positive perception after the financial crisis in 2008.
Therefore I expect a difference in abnormal returns before and after the financial crisis in 2008. As Hoon et. al. (2013) describes, there was a growth in investments in sustainable companies after the financial crisis compared to the overall size of professionally managed assets. Because the financial crisis shocked the world and the investors I would think that they are more aware of good and bad investments and that they tend to focus more on long-term healthy and sustainable companies. A good measurement of long-term healthy and sustainable companies is the DJSI, I therefore expect a shift in abnormal returns when a company is included on the DJSI after the financial crisis compared to before.
I will investigate two type of shocks, the inclusion of a company on the DJSI and the exclusion of a company on the DJSI. Do both shocks have an influence on share price and if so, which type of shock has a larger effect on share prices? Thereafter I will examine if there is a difference in abnormal returns between the years before the financial crisis and the years after. Finally I will investigate if there is a difference amongst the different industries listed on the DJSI.
Hypothesis 1 and Hypothesis 2 are constructed on the findings of Cheung (2011), Robinson
et. al. (2011) and Lopez et. al. (2007). Cheung (2011) did not find strong evidence that the announcement of inclusion has a significant impact on stock return. But he did find a
significant but temporary increase (decrease) when a company was included (excluded) from the DJSI. Cheung investigated a sample of US stocks which are included or excluded from the DJSI over the period 2002 – 2008. This paper focuses on the period 2006 – 2010, where the financial crisis plays an important role and which is incorporated in Hypothesis 3 and
Hypothesis 4.
Lopez et. al. (2007) state that sustainability practices contribute to the value creation of the company on the long-term. Thereby they believe that society has begun to demand a change in companies behavior, which is a more corporate sustainable behavior. These two factors, indicated by Lopez et. al. (2007), were important by constructing Hypothesis 1 and
Hypothesis 2. Because long-term value creation by a company (by implementing
sustainability practices) should be reflected in the share price of that particular company. Therefore it is hypothesized that the inclusion of a company on the DJSI should indicate an increase in sustainability practices and therefore should result in a higher share price
19
decrease in sustainability practices and therefore should result in a decrease of the share prices.
Robinson et. al. (2011) investigated the change in share prices caused by the inclusion or exclusion from the DJSI during the years 2003 – 2007. Robinson et. al. (2011) results show that the mean cumulative abnormal returns for companies included on the DJSI are positive and significant at 2.1%. Thereby they found that companies excluded from the DJSI have a positive but insignificant mean cumulative abnormal return. Therefore the results of Robinson et. al. (2011) show that being included on the DJSI is a positive event for a company, but being excluded from the DJSI is neutral for a company. If the results of this paper are similar to the results of Robinson et. al. (2011) this would result in a support for Hypothesis 1 but no support for Hypothesis 2.
Waddock et. al. (1997) describe the financial effect on corporate social performance and the corporate social performance effect on future financial performance. Therefore they state that in order to fulfill corporate social operations funds are needed, but once these
corporate social operations are implemented this will benefit the future financial performance. This statement is an additional underpinning for Hypothesis 1.
Hypothesis 1:
The inclusion of a company on the Dow Jones Sustainability Index has a positive effect on the share price of a company. The Cumulative Abnormal Returns (CAR) are higher than zero
when the company is added on the DJSI.
H0: E (CAR) = 0 H1: E (CAR) > 0
Hypothesis 2:
The exclusion of a company on the Dow Jones Sustainability Index has a negative effect on the share price of a company. The Cumulative Abnormal Returns (CAR) are lower than zero
when the company is deleted from the DJSI.
H0: E (CAR) = 0 H1: E (CAR) < 0
Hypothesis 3 and Hypothesis 4 are based on the findings by Lopatta et. al. (2013) and on the
paper by Hoon et. al. (2013). Lopatta et. al. (2013) showed that when companies were more sustainable this would have a negative effect on the share prices, during the period 2003 –
20
2007. But they found a change after the financial crisis. The financial crisis could have been a trigger towards accepting the benefits of sustainability by investors. Lopatta et. al. (2013) show that sustainability is relevant to firm valuation and therefore should result in a change in share price when companies’ sustainability practices are changed. Hoon et. al. (2013) describe the findings by the Social Investment Forum Foundation (2011) that during the financial crisis the overall size of professionally managed assets remained somewhat the same, but social responsible investments assets showed a steady growth. Therefore stating that investors tend to move towards companies with corporate sustainability practices. Based on the findings by Lopatta et. al. (2013) and the theory described by Hoon et. al. (2013) Hypothesis 3 and
Hypothesis 4 state that there are higher abnormal returns when a company is included
(Hypothesis 3) and there are lower abnormal returns when a company is excluded (Hypothesis
4) during the two years after the financial crisis compared to the two years before the financial
crisis.
Hypothesis 3:
There are higher abnormal returns (when a company is added) during the two years after the shock of the financial crisis (in 2008) compared to the two years before the financial crisis.
H0: E [CAR (2006 – 2007) = E [CAR (2009 – 2010)] H1: E [CAR (2006 – 2007) < E [CAR (2009 – 2010)]
Hypothesis 4:
There are lower abnormal returns (when a company is deleted) during the two years after the shock of the financial crisis (in 2008) compared to the two years before the financial crisis.
H0: E [CAR (2006 – 2007) = E [CAR (2009 – 2010)] H1: E [CAR (2006 – 2007) > E [CAR (2009 – 2010)]
The DJSI measures every criteria differently amongst the industries. Every dimension (within the DJSI) is constructed by using specific criteria with different weights depending on the type of industry. Hypothesis 5 is focused on finding a difference between the industries. Waddock et. al. (1997) investigated the difference in behavior amongst the wide range of industries. They found a clear difference in performance and level of research and
development among different industries. This could indicate that there is a difference between the industries on the DJSI. Lopatta et. al. (2013) showed with a moderated regression analysis that the financial crisis increased the positive perception of corporate sustainability, especially
21
in industries that are exposed to higher environmental and social risks. Hypothesis 5 is stated to investigate whether there is a difference between the different industries, during the periods 2006 – 2010, 2006 – 2007 and 2009 – 2010.
Hypothesis 5:
There is a significant difference in abnormal returns between the different industries.
H0: CAR (Industry X) = CAR (Industry Y) H1: CAR (Industry X) ≠ CAR (Industry Y)
3. Data
The sample period is from 2005 – 2010, where 2005 is used to detect which companies were deleted in 2006. The information about which companies are included on the DJSI and excluded from the DJSI were obtained by requesting this information from RobecoSAM in Zurich. The data consists of 103 companies which were included on the DJSI and 74 companies which were excluded from the DJSI. After analyzing which data was unusable, there were 86 included companies and 43 excluded companies remaining. For every company there were 2 different ‘windows’ the estimation window and the event window. The
estimation window consists of 205 trading days and the event window consists of 75 trading days. This paper investigates the years 2006 – 2010, therefore the total amount of data points analyzed for the included companies is 120,400 (280 days * 5 years * 86 included
companies). The total data points analyzed for the excluded companies is 60,200 (280 days * 5 years * 43 excluded companies). Throughout this investigation extreme values weren’t deleted. Every change in share prices during the investigation period is adopted in this paper. Because the DJSI only tracks the 2,500 largest companies in the Dow Jones Global Total Stock Market Index that lead the field in terms of sustainability (Hoon et. al., 2013) it was not necessary to select companies based on their market value, number of shares outstanding, number of employees or any other measurement. Every event when a company was included on the DJSI or excluded from the DJSI was therefore taken into account for this investigation.
After it was determined which companies were included and excluded (in the years 2006, 2007, 2008, 2009 and 2010) the data regarding the different share prices of those companies were obtained using DataStream. Thereafter the share prices needed to be
transformed into daily returns in order to be able to calculate the cumulative abnormal returns. To calculate the abnormal returns the market returns and risk free rates were needed. For this
22
investigation the S&P 500 is used for the market returns and the risk free rates were obtained from the website of the U.S. department of the treasury. For the risk free rates the daily treasury yield curves were used with a fixed maturity of 3 months.
DataStream was used to get the daily stock returns of those companies from 250 days before the announcement date till 60 days after the announcement date. These 250 days partly consists of the estimation window which contains observations from t = - 250 to t = - 45. The event window starts from t = - 15 till t = 60. The 15 days before the announcement date detects the existence of an anticipation effect before the announcement. The days are trading days, so 15 days are 3 weeks and 60 days are approximately 3 months. Throughout this paper when “days” are mentioned these are referring to the ‘trading days’.
In DataStream the Total Return Index of the different companies were requested, thereafter they are transformed in daily returns using Formula 1.
Formula 1: 𝑅𝑖,𝑡 = 𝑅𝑎,𝑡+1 - 𝑅𝑎,𝑡 / 𝑅𝑎,𝑡
Where 𝑅𝑖,𝑡 are the daily returns, 𝑅𝑎,𝑡+1 the stock price on day t+1 and 𝑅𝑎,𝑡 the stock price on day t.
Thereafter the market returns (for which the S&P 500 index has been used) and the risk free rates were required to calculate the expected returns with the CAPM-model (Formula 2).
Formula 2: (E (𝑅𝑖)) - 𝑅𝑓 = α + β * (𝑅𝑚 - 𝑅𝑓)
Where (E (𝑅𝑖)) is the expected return, 𝑅𝑓 is the risk free rate and 𝑅𝑚 are the market returns. By using Formula 2 the alpha’s and the beta’s of the different companies were calculated, thereafter these were used in the same CAPM-formula but in a different form. By using the market returns, the risk free rate and the alpha’s and beta’s calculated in Formula 2 the
expected returns were calculated by using Formula 3. The alpha’s and beta’s are calculated by using SPSS and Excel. In Excel the market risk premium per trading day is calculated by subtracting the (daily) risk free rate from the (daily) market return (𝑅𝑚 - 𝑅𝑓) during the estimation window (t = - 250 to t = - 45). Thereafter the alpha’s and beta’s are calculated by using SPPSs linear regression function. With the company which is included on the DJSI (or the company which is excluded from the DJSI) as the dependent variable and the market risk
23
premium as the independent variable. After this analysis every included company on the DJSI (and excluded company from the DJSI) has a determined alpha and beta which is then used in
Formula 3.
Formula 3: (E (𝑅𝑖)) = α + β * (𝑅𝑚 - 𝑅𝑓) + 𝑅𝑓
Thereafter, the actual return is compared with the expected return, resulting in the abnormal return (Formula 4). The abnormal return is the difference between the expected return of the company and the actual return of the company. The abnormal return shows the effect of the inclusion and exclusion of a company on the DJSI.
Formula 4: 𝑅𝑒 – (E (𝑅𝑖)) = 𝐴𝑅𝑖
Lastly the cumulative abnormal returns are calculated (Formula 5):
Formula 5: 𝐶𝐴𝑅𝑖 = Ʃ𝑡−15𝑡60 𝐴𝑅𝑖
Where 𝐶𝐴𝑅𝑖 is the cumulative abnormal returns, Ʃ𝑡−15𝑡60 which is the summation of the days during the event window and 𝐴𝑅𝑖 the abnormal returns.
In order to determine the difference in returns before the credit-crisis (2006 – 2007) and after the credit-crisis (2009 – 2010) there is made a distinction. The different time periods are compared with each other to check if there is a difference in valuing sustainability before and after the credit-crisis.
The different returns regarding the added companies [actual (2006 – 2010), CAR (2006 – 2010), actual (2006 – 2007), CAR (2006 – 2007), actual (2009 – 2010) and CAR (2009 – 2010)] are shown in Table 1. The actual return of the added companies during the period 2006 – 2010 has a mean of 2.1% and an abnormal return of 0.7%. This means that during the event window (15 days before the announcement date till 60 days after the announcement date) the added companies did have 0.7% higher returns than expected. The actual return of the added companies two years before the credit-crisis (2006 – 2007) has a mean of 5.5% and an abnormal return of - 0.4%. Which means that the added companies did have a lower actual return during the event window (2006 – 2007) compared to the expected
24
returns. On the contrary the abnormal return of the added companies two years after the credit-crisis is 8.1% (with an actual return of 17.5%).
Table 1 (Included Companies)
Minimum Maximum Mean Standard deviation Number of firms Actual 2006 – 2010 - 87.487 47.4567 2.1143 26.0478 86 CAR 2006 – 2010 - 65.1353 29.0325 0.6758 16.482 86 Actual 2006 – 2007 - 17.8514 30.0809 5.4920 11.1666 35 CAR 2006 – 2007 - 21.7142 25.5757 - 0.3928 11.411 35 Actual 2009 – 2010 - 6.7373 47.4567 17.5339 12.0614 36 CAR 2009 – 2010 - 26.9177 29.0325 8.1063 11.3929 36
The average abnormal returns of the included companies from 15 days before the
announcement day till 60 days after the announcement day will give an estimation of the average cumulative abnormal return (CAR). Which means that the investor who invested in the included companies 15 days before the announcement and sold his investment 60 days after the announcement day (and he would repeat this strategy in every year during that period) should make this cumulative return (on average).
Figure 1 shows the average cumulative return on an investment in the years 2006,
2007, 2009 and 2010. The actual average cumulative return in those years is 11.15%, the expected average cumulative return is 8.41% which results in an abnormal cumulative return of 2.74%. This means that when an investor would invest in companies which are included on the DJSI 15 days before the announcement date in 2006, then sells his investment 60 days after the announcement date in 2006 and then continues this strategy in the years 2007, 2009 and 2010 he would make an abnormal cumulative return of 2.74%. The average cumulative return on an investment in the years 2006 – 2007 and 2009 – 2010 is shown in Figure 2 and
Figure 3 respectively. When performing the same strategy as described above, so buy 15 days
before the announcement date in 2006, sell 60 days after the announcement date and repeat this strategy in 2007 the investor would make an abnormal cumulative return of – 0.93%. The actual average cumulative return in the years 2006 – 2007 was 5.37% compared with an expected average cumulative return of 6.3%. Whereas the years 2009 – 2010, the two years
25
after the financial crisis, resulted in a positive abnormal cumulative return of 6.42%, with an actual cumulative return of 16.93% and an expected cumulative return of 10.51%. Therefore investing in companies which are added on the DJSI after the financial crisis results in a 6.42% higher return than investing in the S&P 500 companies (when using the same strategy as described above).
In Figure 1, Figure 2, and Figure 3 number 16 on the horizontal axe is the announcement date. On this day RobecoSAM announces the DJSI and publishes the companies on the list. In every period (2006, 2007, 2009, 2010; 2006 – 2007 and 2009 – 2010) there is not a ‘large shock’ on the announcement date. The cumulative abnormal returns in Figure 3 are gradually composed, there are no ‘large shocks’ in the 75 days measured.
Figure 1 (2006, 2007, 2009, 2010) 95 100 105 110 115 120 1 4 7 10 13 16 19 22 25 28 31 34 37 40 43 46 49 52 55 58 61 64 67 70 73 76 Expected Actual Abnormal
26
Figure 2 (2006, 2007)
Figure 3 (2009, 2010)
Table 2 shows the returns of the deleted companies [actual (2006 – 2010), CAR (2006 –
2010), actual (2006 – 2007), CAR (2006 – 2007), actual (2009 – 2010) and CAR (2009 – 2010)]. During the period 2006 – 2010 the average abnormal return of the companies which are excluded from the DJSI is -1.3%. Whereas the two years before the crisis show a positive average abnormal return of 0.8%, and the two years after the crisis also show a positive average abnormal return of 1.3%.
95 100 105 110 115 120 1 4 7 10 13 16 19 22 25 28 31 34 37 40 43 46 49 52 55 58 61 64 67 70 73 76 Expected Actual Abnormal 95 100 105 110 115 120 1 4 7 10 13 16 19 22 25 28 31 34 37 40 43 46 49 52 55 58 61 64 67 70 73 76 Expected Actual Abnormal
27
Table 2 (Excluded Companies)
Minimum Maximum Mean Standard deviation Number of firms Actual 2006 – 2010 - 91.0701 55.3783 .5197 25.1912 43 CAR 2006 – 2010 - 33.2352 46.2028 -1.3349 14.5945 43 Actual 2006 – 2007 - 29.171 18.2244 5.9676 12.8518 13 CAR 2006 – 2007 - 30.8629 14.986 0.8141 11.9317 13 Actual 2009 – 2010 - 9.8355 55.3783 10.8163 14.8948 23 CAR 2009 – 2010 - 23.0321 46.2028 1.2624 15.0009 23
The average cumulative abnormal returns when a company is excluded from the DJSI in the years 2006, 2007, 2009 and 2010 is 0.75% (Figure 4), with an actual cumulative return of 9.16% and an expected cumulative return of 8.41%. The cumulative abnormal return is the result if an investor would invest in all the companies which are excluded from the list. Investing 15 days before the exclusion of the company till 60 days after the exclusion of the company. When an investor would only invest in the deleted companies in the years 2006 and 2007 he would have an average cumulative abnormal return of 1.12% (Figure 5), with an actual cumulative return of 7.42% and an expected cumulative return of 6.30%. In 2009 and 2010 the investor would have an average cumulative abnormal return of 0.38% (Figure 6), with an actual cumulative return of 10.9% and an expected cumulative return of 10.51%.
In Figure 4, Figure 5 and Figure 6 number 16 on the horizontal axe is the
announcement date by RobecoSAM. On this day RobecoSAM announces which companies are on the DJSI, therefore Figure 4, Figure 5 and Figure 6 show on this day the shock on the share prices when the exclusion of those companies is made publicly available. Figure 4 and
Figure 6 show a large positive shock a few days before the announcement date. This is not in
line with Hypothesis 2 “The exclusion of a company on the Dow Jones Sustainability Index
has a negative effect on the share price of a company. The Cumulative Abnormal Returns (CAR) are lower than zero when the company is deleted from the DJSI”. In chapter 4 it will
be investigated if this positive abnormal return caused by the exclusion of companies from the DJSI is significant.
28 Figure 4 (2006, 2007, 2009, 2010) Figure 5 (2006, 2007) 95 100 105 110 115 120 1 4 7 10 13 16 19 22 25 28 31 34 37 40 43 46 49 52 55 58 61 64 67 70 73 76 Expected Actual Abnormal 95 100 105 110 115 120 1 4 7 10 13 16 19 22 25 28 31 34 37 40 43 46 49 52 55 58 61 64 67 70 73 76 Expected Actual Abnormal
29
Figure 6 (2009, 2010)
4. Research methodology
1.) Regressions
It will be investigated if the market responds to the inclusion and/or exclusion of a company from the Dow Jones Sustainability Index. It will be tested if the returns are different from the expected returns by using the CAPM-model. Analysis 1 will test (by using the One-Sample T-test) if the actual returns, from added companies, are significantly higher than the expected returns (actual returns – expected returns = abnormal returns). This analysis will be performed in three different periods: 2006 – 2010; 2006 – 2007; 2009 – 2010. Thereafter it will be investigated (by using Regression 1) if the abnormal returns of the included companies are significantly higher than the abnormal returns of the excluded companies (by using a dummy variable which equals 1 if the company is included on the DJSI and 0 if the company is excluded from the DJSI). This regression will also be performed on all three periods.
Regression 2 will test if there is a significant difference between the returns of the included
companies between the different industries. Every industry has a dummy variable which equals 1 if the company is listed in that particular industry, and 0 otherwise. The dummy variable of the industry Banks are omitted from this sample to avoid the dummy variable trap (perfect multicollinearity). Analysis 1, Regression 1 and Regression 2 are performed for the different periods (a: 2006 – 2010, b: 2006 – 2007, c: 2009 – 2010). Regression 3 tests if there is a significant difference between the abnormal returns two years before the credit-crisis (2006 – 2007) and two years after (2009 – 2010) regarding the included companies (2008 is omitted for this regression; only the years 2006, 2007, 2009 and 2010 will be used). Events in
95 100 105 110 115 120 1 4 7 10 13 16 19 22 25 28 31 34 37 40 43 46 49 52 55 58 61 64 67 70 73 76 Expected Actual Abnormal
30
the years 2009 and 2010 have a dummy variable which equals 1, events in the years 2006 and 2007 have a dummy variable which equals 0. Analysis 2 will test if the actual returns, from the excluded companies, are significantly lower than the expected returns. Regression 4 will test if there is a significant difference between the abnormal returns two years before the credit-crisis (2006 – 2007) and two years after (2009 – 2010) regarding the excluded companies.
Throughout this investigation there is no test for any control variables. This is because this paper is (partly) based on other similar work that does not test for any control variables (Cheung, 2011). Moreover, the assumption is made that the DJSI takes into account most of the information regarding corporate sustainability. If this investigation would only focus on the environmental part then we should have taken into account some control variables which would correct for the other ‘2 pillars’ (economic & social aspects). But because the DJSI uses all the information available regarding corporate sustainability this paper focuses on using the inclusion and exclusion of companies on the DJSI solely. Finally the CAPM-model, which is used for this investigation, takes into account the markets daily returns (during the estimation window) and the daily risk free rate (during the estimation window). Therefore the calculated expected returns corrects for the ‘normal’ shifts in the stock market. To conclude, this paper does not use any control variables because most of the information (regarding corporate sustainability) is included in the information from the DJSI and because the CAPM-model corrects for the stock index overall fluctuations.
Analysis 1:
𝑡 = 𝑥𝐼𝑛𝑐𝑙𝑢𝑑𝑒𝑑− 𝑢0 𝑠/√𝑛
Regression 1:
CAR = α + 𝛽1 Included Company + ɛ
Regression 2:
CAR = α + 𝛽1 Included Company + 𝛽2 Industry + ɛ
Regression 3:
31
Analysis 2:
𝑡 = 𝑥𝐸𝑥𝑐𝑙𝑢𝑑𝑒𝑑− 𝑢0 𝑠/√𝑛
Regression 4:
CAR (Excluded companies) = α + 𝛽1 After Crisis + ɛ
2.) Explanations and expectations of the variables
Analysis 1 and 2 have 4 different variables: 𝑥 , 𝑢0, 𝑠 and 𝑛. Where 𝑥 is the sample mean, 𝑢0 is the “population” mean, 𝑠 is the standard deviation (the square root of the variance) and 𝑛 is the number of observations. For Analysis 1 I expect positive abnormal returns. I think that companies which are added to the DJSI will increase their share price value and therefore this test should give positive abnormal returns. On the contrary I expect Analysis 2 to have
negative abnormal returns. Because I expect investors to withdraw money from their investments if the company is proven to be unsustainable.
Regression 1 will test the difference between the abnormal returns of excluded and
included companies. As shown in the formula of Regression 1 the “𝛽1 Included
Company”-factor shows how much higher (or lower) the abnormal returns are when the company is included (compared to when the company is excluded) to the DJSI by using a dummy variable which equals 1 if the company is included on the DJSI and 0 if the company is excluded from the DJSI. This regression is performed by using Excel and SPSS. With Excel all the relevant data for the specific time period is calculated, this contains all the data of the included and excluded companies (for period 2006 – 2010 the total number of companies is: 129, included: 86, excluded: 43). By using the alpha and beta (calculated with Formula 2) and the market risk premium, the expected returns are calculated by using Formula 3. Thereafter the actual returns are compared with the expected returns which results in the abnormal returns. Finally this data is used in SPSS to perform a linear regression. With the abnormal returns (of all the companies: included and excluded) as dependent variable and as independent variable a dummy variable which equals 1 if the company is included on the DJSI (86 companies) and 0 if the company is excluded from the DJSI (43 companies). Therefore with Regression 1 the abnormal returns of the companies which are included on the DJSI are compared with the abnormal returns of the companies which are excluded from the DJSI. This regression will be performed for every different time period (2006 – 2010, 2006 – 2007 and 2009 – 2010).
32
The alpha (α) measures the constant, or where the regression starts and the error term (ɛ) corrects for errors. I expect that the abnormal returns for included companies are higher during every period and the highest after the credit crisis. Because I do think that investors have invested more cautious after the crisis shock in 2007 and therefore put more money in sustainable companies instead of unsustainable companies based on the results by Lopatta et. al. (2013). Cheung (2011) and Robinson et. al. (2011) both investigated the inclusion of companies on the DJSI and the exclusion of companies from the DJSI. Cheung (2011)
investigated the time period 2002 – 2008, Robinson et. al. (2011) investigated the time period 2002 – 2007. The focus of this paper is whether or not the financial crisis has influenced the valuation of corporate sustainability. Lopatta et. al. (2013) did use a different type of
sustainability index but investigated the effect of the financial crisis on abnormal returns of included and excluded companies. Therefore this paper is grounded on the Lopatta et. al. (2013) paper.
Regression 2 is used to examine the difference between the different industries when a
company is added to the DJSI. Where 𝛽1 measures the difference between the added
company’s abnormal return (compared to the excluded company) and where 𝛽2 measures the difference between the abnormal returns amongst the different industries (compared with the Banks-industry). The same as with Regression 1 the alpha (α) measures the constant and the error term (ɛ) corrects for errors. Every industry has a dummy variable which equals 1 if the company is listed in that particular industry, and 0 otherwise (where the industry Banks is omitted from this sample to avoid the dummy variable trap).
Regression 3 will test if there is a significant difference between the abnormal returns
(of the companies which are included from the DJSI) before and after the credit crisis of 2008. Where 𝛽1 measures the difference between the abnormal returns before (2006 – 2007) and after (2009 – 2010) the credit crisis. For Regression 3 the sample consists of all the
included companies in the period 2006 – 2007 and the period 2009 – 2010. A dummy variable is used to test the difference between these two periods, 1 when the company was included in the period 2009 – 2010 and 0 if the company was included in the period 2006 – 2007. The total amount of companies which are included on the DJSI during the years 2006, 2007, 2009 and 2010 is 71 companies. From those 71 companies 36 were included after the financial crisis and 35 companies were included before the financial crisis. For this regression the data was analyzed in SPSS by using a linear regression with the abnormal returns as dependent variable and the dummy variable as independent variable.
