The importance of social and environmental disclosures for
stakeholders: The potential added value of being sustainable
for shareholders
Student: Maikel Lust Student Number: 10001560 Date: 12 August 2014
Education: MSC Accountancy and Control, Accountancy track and Control track Institution: University of Amsterdam, Faculty of Economics and Business
2
ABSTRACT: This study examines the potential added value of being sustainble for shareholders. This is done at two ways. The first method is by looking at the financial benefits. These benefits are measured by economic value added (EVA). The other way is by looking if the shareholder demands for a sustainable firm. If they want it, it should add value. This is measured by agency conflicts, for which I’ve used four proxies: size; risk; leverage; and free cash flows. Both methods are tested using a sample of firms listed on the S&P 500 in the years 2010-2013. This research found no existing relation between sustainability and firm value. This study also found no relation between sustainability and agency conflicts.
3
Table of contents
1. Introduction ... 5
2. Literature review ... 9
2.1 sustainability ... 9
2.2 Theories concerning sustainability ... 12
2.2.1 Legitimacy theory ... 12
2.2.2 Stakeholder theory ... 14
2.3 Agency conflicts ... 15
2.4 Shareholder value ... 16
2.5 EVA ... 18
2.6 Sustainability Yearbooks of RobecoSam ... 18
2.7 Hypotheses... 20
3. Research design ... 22
3.1 Quantitative research method ... 23
3.2 Sample ... 23 3.3 Research method ... 24 3.4 Measures ... 24 3.4.1 Sustainability ... 25 3.4.2 EVA ... 25 3.4.3 Agency conflicts ... 27 4. Results related to H1 ... 29 4.1 Summary statistics ... 30 4.2 Univariate test ... 33 4.3 Linear regression ... 34 4.4 Sensitivity analysis ... 36 5. Results related to H2 ... 39 5.1 Summary statistics ... 39
4 5.2 Univariate test ... 41 5.3 Robustness test ... 43 6. Analysis ... 44 6.1 H1 ... 45 6.2 H2 ... 46 7. Conclusions ... 49 8. References ... 51
5
The importance of social and environmental disclosures for the stakeholders: The potential added value of being sustainable for shareholders.
1. Introduction
Since the Brundtland report was published in 1987, society cares much more about the environment, compared to the period before the Brundtland report (Robinson, 2004). This increased public interest in environmental and societal issues is for example reflected in the eight millennium development goals (UN, 2014). These goals, developed by the United Nations aim at eradicating global poverty, human well-being and supporting the nature and include the following goals: eradicate extreme poverty and hunger; achieve universal primary education; promote gender equality and empower women; reduce child morality; improve maternal health; combat HIV/AIDS, malaria and other diseases; ensure environmental sustainability and global partnership for development (UN, 2014). The intension is that these goals are achieved in 2015. Governments have the primary responsibility for achieving these goals, but there is also increasing interests of companies to be part of the solutions (Nelson and Prescott, 2008).
Since the mid-1980s (Adams, 2004), organizations started reporting on the social and environmental impacts of its operations in a SER (social and environmental report). The most important reasons for organizations to report are community concerns and shareholder rights (Dobbs and van Staden, 2012). The study of Dobbs and van Staden also shows that
organizations have social and environmental disclosures, because they want to show the society that they act sustainable, rather than that they want to act legitimate in the view of the society.
The legitimacy theory argues that the values of the organization have to be congruent with the values of the society (Sethi, 1975; Dowling and Pfeffer, 1975). This is one of the most used theoretical frameworks in sustainability reporting research. The organization has to act legitimate in the view of the society, otherwise the society will no longer support the organization. The society is the aggregate of people that live together in a community. The society consists of different groups that can affect the firm, the stakeholders. Stakeholders are any group or individual who can affect or is affected by the achievement of the organization’s objectives (Freeman, 1984). Or a more narrow definition of stakeholders:
6
the individuals and groups on which the organization is dependent for its continued survival (Freeman and Reed, 1983). Stakeholders theory is a theory that is closely linked to the legitimacy theory. Stakeholder theory argues that the organization has to act legitimate in the view of their stakeholders rather than society at large (Freeman, 1984). There are a lot of different stakeholders, including the owners, the customers, the employees, the community, the competitors, the suppliers, social activists groups, public at large and others (Carroll, 1991). In my thesis I will focus on the owners of the organization, the shareholders. Shareholders are one of the major stakeholder groups.
Prior literature suggests that organizations should engage more with their stakeholders when reporting on it’s social and environmental impacts (Edgley et al., 2010; Gray, 2000). One of the major stakeholders the organization has to engage with, are the shareholders. However, the question is whether they want to be involved. People wants to maximize their own utility. Self-interest is a determinant of human behavior (Miller, 1999). From this point it can be concluded that shareholders only want to engage with the management and in the SER process if being sustainable adds value for themselves (Beard and Beil, 1994). Therefore I want to examine the following research question: In which degree adds being sustainable, value for the organization in the eyes of the shareholder? To measure this relation, I will use economic value added (EVA). When a company is sustainable, does it leads to a higher EVA. EVA is one of the best measures to measure the value creation for a shareholder and the performance of an organization (Ittner, 1997).
Relatively little research has focused on the relationship between a sustainable firm and it’s economic performance, there are some but are using other measures of economic performance. For example Bampton et al. (2011) has examined the relation between a sustainable organization and the net profit of an organization. The research of Bampton et al. was a case study at the Royal Dutch Shell Plc. They linked key performance indicators in Royal Dutch Shell Plc’s sustainability report with profitability. They found a positive relation between sustainability and financial performance. However a weak point in their study are the profitability measures they have used. The measures Bampton et al. used in their study are revenue, net income and earnings per share. EVA is a better measure to evaluate the performance of an organization, because it’s much more future oriented (Ittner, 1997). Another limitation in the study of Bampton et al. is that they have only used one company.
7
In previous research there is an absence of common conclusion between firm performance and sustainability (Cottrill, 1990). Some researchers find a positive relation (Peters and Waterman, 1988; Freeman and Gilbert, 1987) and some find that there is no relation (Aupperle et al. 1985; Alexander and Bucholtz, 1987). They all used, as variable for profitability, traditional performance measures like net income and return on assets which are much more past oriented than EVA is (Ittner, 1997). EVA is more able to measure the value creation for the shareholders. Therefore I want to contribute prior research by examining the relation between EVA and sustainability.
My thesis responds to a call in prior literature. Gray et al. (2006) call for more research concerning the interests of the financial markets for more sustainable firms. At the moment there is an absence in research on how financial markets can be persuaded to act more sustainable. If my thesis find evidence that there is a positive relation between shareholder value and sustainability, this can help to persuade them to act more sustainable.
My thesis also has a societal contribution. In 2011 there was a survey from Laughland and Bansal (2011). They asked to organizations worldwide why they are still not sustainable. The most common reason of these organizations was that it costs them to much money. They only want to apply being sustainable when it adds value for them. My study wants to
counterattack their statements by showing that being sustainable adds value to a firm. My thesis also contributes to prior literature by examining a relation which is never made before. The relation between sustainability and agency conflicts, as a proxy for
shareholder value. This problem may arise when there is a gap between what the shareholder wants and what the organization does.
This study found a higher EVA for the panel that consists of the sustainable firms, than for the panel of the non-sustainable firms. However this difference is not significant. I also found that there is no significant relation between agency conflicts and sustainability. A reason for this could be that because sustainability doesn’t lead to a higher firm value, shareholders also don’t demand for this. People are still to much self-interested.
Overall I came to the conclusion that there is no relation between sustainability and shareholder value. At this moment being sustainable doesn’t add value for a firm.
The remainder of this paper is structured as follows. First a literature review is provided, which shows the different theories regarding sustainability and aims to better
8
understand the hypotheses and findings of this study. Second, the research methodology with the different models and measures that will be applied in this thesis. Third, the results of the tests are presented and analyzed. Finally, the conclusion is provided.
9
2. Literature review
In this chapter, I will describe the theoretical concepts applied in my thesis. First of all, I will describe the concept of sustainability. Secondly, I will describe two theories related with sustainability: Legitimacy theory and Stakeholder theory. These two theories can be seen as a reason why there might be a relation between sustainability and financial performance of a firm. In the third paragraph of this chapter, I will describe the concept of shareholders’ value. What the drivers of shareholders’ value are and how being sustainable can influence the shareholders’ value. In the fourth paragraph, I will describe the measure of financial performance I will apply in my thesis: EVA. I will show the advantages of EVA in
comparison with the more traditional financial performance measures which are used in prior literature. I will end this chapter with a description of the agency theory. How being
sustainable can narrow the gap between the interest of the managers of the firm and the owners of the firm with is related with my third hypothesis.
2.1 sustainability
In this section, I will describe the concept of sustainability.
There is not one definition of sustainability. There are several definitions of sustainability available in prior literature which are all referring to a more humane, more ethical and a more transparent way of doing business. (Marrewijk, 2003; Carroll, 1999). The definition of sustainability that I will apply in my thesis is the definition used by the company RobecoSam. RobecoSam is the company that constructed the database I will use in my thesis. “Corporate sustainability is a business approach that creates long-term shareholder value by embracing opportunities and managing risks deriving from economic, environmental and social developments” (sustainability-indices, 2014).
Sustainability, in the sense of corporate social responsibility (SER) has been first introduced in the 1950s (Carroll, 1999) by Bowen. He said that social responsibility refers to the obligations of businessmen to pursue those policies, to make those decisions, or to follow those lines of actions which are desirable in terms of the objectives and values of our society (Bowen, 1953). The concept of sustainability became popular in 1987 in the Brundtland Report (WCED, 1987). In the Brundtland Report the concept of sustainability is described as
10
“development which meets the needs of present generations without compromising the ability of future generations to meet their own needs”. For a sustainable development, there always has to be a balance between planet (the environment), profit (the economy) and people (the society) (Elkington, 1998). In practice this means that an increase in profit isn’t good when it has a negative impact on the nature. Organizations have responsibilities to the society they exist in (McDonald and Puxty, 1979).
Whether a firm choose to be sustainable or not depends on different aspects.
According to Aras and Crowther (2008) there are four dimensions of sustainability that are important when a company has to make a decision, these dimensions are: societal influence; environmental impact; organizational culture; and finance. All these dimensions are equally important. Societal influence deals with the impact that society has on a firm. This is also described by the legitimacy theory in the next paragraph. When all the customers of a firm demands for a sustainable firm, the firm will be a sustainable firm. Otherwise it might loose all of its customers and go bankrupt. Environmental impact deals with the effects of the actions of the firm on its environment. Organizational culture is the interaction of the organization with it’s internal stakeholders. The dimension finance deals with the issue that firms demand for an adequate return for the additional risk they have taken.
11 Source: Aras and Crowther, 2008, p. 438
As mentioned before, an increase in profit isn’t good when it has a negative impact on nature. An accounting practice that tries to deal with this issue is the full cost accounting approach of Herbohn (2005). In this approach, every impact on the environment is quantified. When these impacts are negative, they are recorded as a loss and when these impacts are positive, they are recorded as a profit. Due to this approach, the impacts of an activity on the environment is taken into account in the decision making process. Therefore, there is the possibility of a sustainable development. Things like the dumping of millions of tones of dredging sludge in the waters of the Great Barrier Reef (The Sydney Morning Herald, 2014), are not possible, because it’s not a profitable project anymore.
However there is no agreement on whether an organization has to act sustainable. Not everyone agrees with the responsibility of firms to be socially and environmentally
sustainable. According to Friedman (1962), the only social responsibility of business is to increase their profit. Organizations don’t have a social responsibility. This is also, in some way, mentioned by Ackerman (1975). He says that organizations have to adapt to the new social climate but that the core orientation of business still has to be on the financial results. When there is a positive relation between sustainability and the financial results, these propositions are not in conflict with the theories that demand for more sustainable
organizations. When being sustainable leads to a higher profit, Friedman and Ackerman also demands for more sustainable firms.
Another theory against sustainability comes from the psychological field. Miller (1999) argued that self-interest is a powerful determinant of human behavior. When people do something, they want material benefits in exchange. Other theories also mentioned the self- interest of business which is an argument against sustainability. Gray and Milne (2004) mentioned that in the most part, capitalism punishes non-economic behavior, like being sustainable if it run counter to the self-interest of business. The nature of capitalism is that there is a conflict between moral and financial criteria, therefore just comply with the law. However, again, these theories are not in conflict with sustainability if being sustainable leads to a higher profit.
In previous research there is an absence of common conclusion between firm performance and sustainability (Cottrill, 1990). Some researchers find a positive relation,
12
some find a negative relation and some find that there is no relation between the financial performance of a firm and sustainability. Ruf et al. (2001) found that there is a positive relation. They found that when management meets the demands of their stakeholders, they will receive financial benefits. The authors found evidence on the short-term and also on the long-term. The financial measures they used were return on equity, return on sales and growth in sales.
López et al. (2007) have also examined the relation between the corporate financial performance and the sustainable development. They have used the same sustainability index as the index that will be applied in this thesis. As financial performance measures they have used the growth of profit before tax and the growth in revenue. As result the authors found evidence that there is a negative relation between the use of sustainability practices and the financial performance of a firm. My study differs from this study, by using another measure for shareholder value (EVA instead of the growth of profit before tax). A measure which is seen as the best measure of shareholder value (Yeniyurt, 2003). More about the measure EVA in section 2.5.
In conclusion this paragraph has shown that there are theories for and theories against sustainable business. This conflict in theories can be overcome when being sustainable leads to better corporate financial performance. In prior literature there is an absence of common conclusion between firm performance and sustainability.
2.2 Theories concerning sustainability
In this paragraph, I will give a brief description of the legitimacy theory and the stakeholder theory. These two theories will be used to provide a better understanding of the findings of this thesis.
2.2.1 Legitimacy theory
In accordance to the legitimacy theory, the value of the organization has to be congruent with the values of the society (Sethi, 1975; Dowling and Pfeffer, 1975). The legitimacy theory is one of the most used theoretical frameworks in sustainability reporting research. The
13
organization has to act legitimate in the view of the society, otherwise the society will no longer support the organization. An organization is dependent from the society to survive. The legitimacy theory is derived from the concept of organizational legitimacy of Dowling and Pfeffer (1975). They defined organizational legitimacy as “a condition or status which exists when an entity’s value system is congruent with the value system of the larger social system of which the entity is a part. When a disparity, actual or potential, exists between the two value systems, there’s a threat to the entity’s legitimacy”. When there is a difference between how the organization act and how it should act according to the society, there is a legitimacy gap (Lindblom, 1994). When there is an legitimacy gap, the organization has to undertake some actions to solve this gap.
There are four phases an organization can exist in regard to legitimacy: establishing legitimacy; maintaining legitimacy; extending legitimacy; and defending legitimacy. Establishing legitimacy is the first phase. In this phase the firm is transforming to a more sustainable way of doing business (Hearit, 1995).
The second phase is maintaining legitimacy. In this phase the firm is already perceived as sustainable and the firm wants to maintain this. This doesn’t mean that this is easy, because the expectations of society always changing. When a firm wants to maintain their legitimacy, it has to change it’s actions to this changing expectations. Legitimacy is a dynamic construct (Deegan et al., 2002).
Extending legitimacy is the third phase. In this phase the organization is entering new markets. In this new markets the organization will face the expectations of a ‘new society’. To meet the expectations of this new society it might change their actions (Ashford and Gibbs, 1990).
The fourth phase an organization can exist in is defending legitimacy. An organization can come in this phase when an internal or external incident happen that requires action. The task of the management is to counter the threat, otherwise it actions might not be seen as legitimate anymore (Ashford and Gibbs, 1990).
In conclusion the legitimacy theory will lead to sustainable firms if the society
14 gap and the firm might not survive.
2.2.2 Stakeholder theory
A theory that is related to the legitimacy theory is the stakeholder theory.
The society consists of different groups that can affect the firm, the stakeholders. Stakeholders are any group or individual who can affect or is affected by the achievement of the organization’s objectives (Freeman, 1984). Or a more narrow definition of stakeholders: the individuals and groups on which the organization is dependent for its continued survival (Freeman and Reed, 1983). The stakeholder theory says that the organization has to act legitimate in the view of their stakeholders (Freeman, 1984). There are a lot of different stakeholders. The owners, the customers, the employees, the community, the competitors, the suppliers, social activists groups, public at large and others (Carroll, 1991). In my thesis I will focus on the owners of the organization, the shareholders. Shareholders are one of the primary stakeholder groups (Freeman, 1984). Without this stakeholder group, the firm might even not survive. The interaction of an organization with the different stakeholders is shown in figure 1 at the end of this paragraph.
The stakeholder theory says that the organization has to act legitimate in the view of their stakeholders (Freeman, 1984). It implies that managers must pay attention to it’s most important constituencies that can effect the firm to have value maximization (Jensen, 2010). Some of these interests might be in conflict with each other. However the stakeholder theory has no answer on the question what to do when there is a conflict. A theory, which is not part of the stakeholder theory, is the theory of Sundaram and Inkpen (2004). Their theory suggests that if the goal of the organization is value maximization, you should only listen to the
shareholders and ignore all the other stakeholders.
In conclusion, the stakeholder theory can be seen as a more practical specification of the legitimacy theory. The main difference is that under the legitimacy theory, the values of the organization has to be congruent with the values of the whole society. Under the
stakeholder theory, the values of the organization has to be congruent with the values of it’s primary stakeholders. If the organization doesn’t meet the demands of their primary
stakeholders, the organization might even not survive.
15 Figure 2
Source: Donaldson and Preston, 1995
2.3 Agency conflicts
In this paragraph, I will describe the theory concerning agency conflicts. This theory is important, because it’s the basis of the third hypothesis.
The agency problem occurs when cooperating parties have different goals and division of labor (Ross, 1973). An agency conflict may occurs in settings in which one party (the principle) delegates work to another (the agent), who perform the work. In these settings the goals of the principle and the agent can be in conflict, because the principle and the agent have different risk preferences which lead to a difference in preferred actions. Another problem is that it’s difficult or expensive for the principle to verify what the agent is actually doing (Eisenhardt, 1989). The question here is, is the agent really doing what he’s supposed to do and what he’s paid for (information asymmetry).
There’s a link with the self-interest theory (Miller, 1999). Both parties want to maximize it’s own utility which can lead to a difference in actions. For example the problem that may exits in an organization. In an organization the shareholders are the principle and the manager is the agent. The preferred goal of the shareholders is shareholder value
16
this setting, the manager might undertake some actions to earn his bonus but at the same time these actions may hurt the shareholders’ value.
There is also a link with the stakeholder theory. The agency theory can be seen as an outcome of the stakeholder theory. As described in the previous section, stakeholder theory says that the organization has to act legitimate in the eyes of the stakeholders. When the actions of the organization are not in line with the expectations of the stakeholders, there is a gap. This gap results in the agency problem. The action the principles want, is not in line with the actions the agent performs.
The study of Dey (2008) found that there are less agency conflicts when there is a strong corporate governance which has mechanisms to minimize these conflicts. Corporate governance refers to the system of structures, right, duties and obligations by which
corporations are directed and controlled. It deals with the way in which shareholders can control the company to get a return on their investment (Shleifer and Vishny, 1997). In conclusion the agency theory describes the conflict that may arise by the self-interest between the principle and the agent. A way to overcome this problem is by a strong corporate governance.
2.4 Shareholder value
In this paragraph, I will describe the concept of shareholder value. This is an important
concept, because the aim of this thesis is to find the potential added value of being sustainable for shareholders.
A shareholder is an individual or institution with shares or stocks in a public or private corporation (businessdictionary, 2014). The shareholders are the owners of the firm. You can measure the value of the shareholders by looking at the shareholder value.
In history there were different ways of calculating the shareholder value. Traditional measures like net profit, earnings per share, dividend per share, etc. and newer measures of shareholder value, like EVA and MVA (market value added) (Ittner, 1997). Stern (1993) found that EVA is the best measure of shareholder value. He found a correlation of 50 percent with MVA. O’Byrne (1996) found that over a period of 10 years, changes in EVA values explained 74 percent of the changes in market values. Other measures had a lower score. The
17
second best was ROA (return on assets) with a score of 13 percent. A more recent study of Nagar (2007) found that EVA had a correlation with the shareholder value of 28,57 percent, while traditional measures like earnings per share and dividend per share had a correlation of respectively 0,04 and 0,03 percent.
There are different drivers of shareholder value. According to Srivastava et al. (1998) the key drivers of shareholder value are: An acceleration of cash flows (earlier cash flows are preferred because risk and time adjustments reduce the value of later cash flows); An increase in the level of cash flows (e.g., higher revenues and/or lower costs, working capital, and fixed investments); reduce volatility and vulnerability of cash flows; and enhance residual value of cash flows (long-term value can be enhanced, for example, by increasing the size of the customer base). A research conducted by Deloitte in comparison with the Boston Consulting Group found that the key driver of shareholder value is revenue growth (Deloitte, 2014a) which leads to an increase in cash flows.
The preferred goal of each company should be shareholder value maximization (Sundaram and Inkpen, 2004). Reasons why maximization of shareholder value should be the preferred goal of each company are, according to Sundaram and Inkpen: The goal of
maximizing shareholder value is pro-stakeholder; maximizing shareholder value creates the appropriate incentives for managers to assume entrepreneurial risks; having more than one objective function will make governing difficult, if not impossible; it is easier to make shareholders out of stakeholders than vice versa; and in the event of a breach of contract or trust, stakeholders, compared with shareholders, have protection (or can seek remedies) through contracts and the legal system. Maximizing the shareholder value leads to a maximization of the value of the whole firm. The firm should ignore all the other
stakeholders. This is a different approach as the stakeholder theory, who says that you have to listen to all the primary stakeholders.
In conclusion, the goal of each company should be shareholder value maximization which measures the value of the organization for the shareholder. An increase in cash flows is the key driver of shareholder value.
18 2.5 EVA
In this paragraph, I will describe the concept of economic value added (EVA). This concept is important, because it’s the dependent variable in my thesis that has a potential relation with sustainability. More about the application of this variable in the next chapter.
EVA as measure of shareholder value is first introduced in 1991 by the company Stern Steward & co (Stewart, 1991). This relatively new measure replaced the traditional measures like return on investment (ROI) and earnings per share (EPS) (Mäkeläinen, 1998). The big advantage of EVA related to the traditional measures of performance is that EVA provides a better understanding of the value creation capability of the organization and that it’s better in predicting the future than the traditional measures (Yeniyurt, 2003; O’Byrne, 1996). The idea behind EVA, is to make managers act like shareholders and thereby reducing the agency conflicts (Tully, 1993). “Good is no longer positive operating earnings, it’s only good when you beat the cost of capital”.
EVA can be calculated by the operating profit after tax minus the total annual cost of capital. There are three ways by which EVA can increase: Earn more profit without using more capital; use less capital; and/or invest capital in high-return projects. When the earnings are more than the cost of capital, the company creates true value for the shareholders (Dierks and Patel, 1997).
In conclusion, EVA has as advantage, in relation to tradition performance measures, that it’s better able to measure the value creation capability and can better predict the future. EVA can be calculated by the net operating profit after tax minus the cost of capital.
2.6 Sustainability Yearbooks of RobecoSam
In this paragraph, I will describe the methods RobecoSam has applied to made their sustainability yearbook each year. These yearbooks of RobecoSam will be applied in my thesis to determine if a firm is sustainable or not. I will use the sustainability index of RobecoSam, because this index is seen as one of the leading sustainability-indices in the world (Lo and Sheu, 2007). Firms want to make it to this index, because then they can benefit from the growing demand for sustainable firms. They will gain the reputation of being an industry leader.
19
Since 2004, every year, RobecoSam has published a sustainability yearbook. In this sustainability yearbook, RobecoSam showed the firms which are seen as sustainable. The 2500 largest companies within the S&P Global Broad Market IndexSM are invited to participate in the test (sustainability-indices, 2014).
Assessment, whether a firm made it to the sustainability yearbook or not, is based on the RobecoSam Questionaire, which consists of 80-120 questions (sustainability-indices, 2014). These questions consists of economic, environmental and social questions which all focus on long-term value creation. The different dimensions are all equally balanced. These can be seen in the figure below (figure 2).
Only the firms who scores best in the assessment made it to the sustainability yearbook (sustainability-indices, 2014). These firms are seen as the best in their industry in the case of sustainability. This are somehow relative scores, because it’s based on the firms’ own industry it operates in.
Figure 2
20 2.7 Hypotheses
In this paragraph, I will describe the hypotheses that will be applied in my thesis.
My first hypothesis is related to the possible relation between sustainability and EVA. EVA is the measure of shareholder value that I will apply in my thesis. The goal of the shareholder and the goal of the company is shareholder value maximization (Sundaram and Inkpen, 2004). The key driver of shareholder value is an increase in revenue which will lead to an increase in cash flows (Deloitte, 2014). Revenues can be increased by increasing the selling price or by increasing the amount of products the company sells.
In accordance to the legitimacy theory, the values of the organization should be aligned with the values of society (Sethi, 1975; Dowling and Pfeffer, 1975). If these values are not congruent with each other, society won’t support the company any longer. In the case of revenues, this means that the society will not longer buy the products of this company. Conclusion of this is that the revenues of the company are lower, when the values of the company are not congruent with the values of the society.
The same can be predicted based on the stakeholder theory. The message of the stakeholder theory is that the organization has to act legitimate in the eyes of the stakeholders (Freeman, 1984). The stakeholder theory suggest that when the stakeholders are satisfied, the future profit of the organization will increase (Berman et al, 1999). One of the primary stakeholders of a firm are the customers. When the firm doesn’t act legitimate in the view of the customers, the customers won’t support the organization any longer. This will lead to a decrease in revenues.
The customer value based theory of the firm also says that an organization has to meet the demands of the customers to be successful (Slater, 1997).
Customers demand for a more sustainable firm. Customers are becoming more aware of their personal environmental impacts. They want to become more environmentally friendly and also demand this from the companies. (Rao, 2002; Melville, 2010).
In conclusion the values of an organization has to be congruent with the values of it’s customers for a growth of EVA. Customers demand for more sustainable firm. Therefore I expect a positive relation between sustainability and EVA.
21
H1: There is a positive relation between sustainability and EVA
Aras and Crowther (2008b) mentioned that both corporate governance and
sustainability are essential for an organization to survive. For a firm its important to have a strong corporate governance, otherwise it will face a lot of risks it can’t mitigate. Corporate governance procedures determine the role of management and give control mechanisms to increase shareholder value and to make the other stakeholders satisfied. It tries to find a balance between the social and economic goals of a company. Good governance levels can lead to higher trust from the public and to confidence in the political environment. As a result the expectations of the society are more aligned with the performance of the organization. The same is said by the legitimacy theory. The legitimacy theory said that the values of the firm has to be the same as the values of the society. The society wants a more sustainable world. Therefore Aras and Crowther developed the hypothesis that a good corporate governance will foster sustainability. They found evidence that a relation exists between corporate governance and sustainability.
The other article that is important, is the article of Dey (2008). As mentioned before, corporate governance can increase the shareholder value. It also leads to higher trust and therefore lower the risk of a ‘wrong’ financial report. Shareholders are satisfied if their value increased. Therefore, he founds that there is a negative relation between corporate governance and agency conflicts.
There is prior literature that finds a relation between sustainable reporting and corporate governance (Aras and Crowther, 2008b). There is also prior literature that found a relation between corporate governance and agency conflicts (Dey, 2008). So there is an indirect link between sustainability and agency conflicts. I want to examine this relation. Do the owners of the company want a sustainable firm. When managers then are sustainable, they make the owners satisfied.
This relation can also be seen in a more direct way as described in section 2.3. When the actions of the organization are not in line with the expectations of the stakeholders, there is a gap. This gap results in the agency problem. The action the principles want, is not in line with the actions the agent performs. From this point it can be said that when shareholders see sustainability as a value increasing way of doing business, they will demand for it. When
22
firms don’t respond to this call, an agency problem occurs. So I come up with the following hypothesis:
23
3. Research design
In this chapter, I will describe the methodology that I will apply in my thesis.
3.1 Quantitative research method
In general there are two primary kind of researches in the economic field. Quantitative and qualitative research. Both have their specific advantages and disadvantages. The debate of which method is best is to big to outline in this thesis. Therefore I will only provide a brief introduction. In my thesis, I will apply the quantitative research method.
Quantitative research is mostly numerical and designed to ensure objectivity,
generalizability and reliability (Zawawi, 2008). Whereas qualitative research can be described as ”an array of interpretive techniques which seek to describe, decode, translate and otherwise come to terms with the meaning, not the frequency, of certain more or less naturally occurring phenomena in the social world” (Maanen, 1983).
The main reason why I will apply the quantitative research method is the
generalizability. Qualitative research is often limited to the setting it’s conducted in. Whereas the outcomes of quantitative research can be applied in many settings (Tewksbury, 2009). Therefore the outcomes of study is applicable for many firms, which can persuade them to transform to a more sustainable organization.
3.2 Sample
As sample, I will use data from large U.S. firms from the S&P 500 (based on the year 2013) from the years 2010 to 2013. The main reason why I will choose this sample is the availability of data. The most of the firms that are in the Dow Jones Sustainability index of RobecoSam are American firms and all the firms listed on the S&P 500 are invited for the assessment. The total sample of firms will be divided in three different panels. Panel A consists of all the firms of the S&P 500 minus the firms with missing data. Panel B consists of the S&P 500 firms that are listed in the Sustainability Yearbook. Panel C consists of the S&P 500
24
firms that are not listed in the Sustainability Yearbook. These firms will be seen as not sustainable.
Table 1: Panel size
2010 2011 2012 2013 Total
Panel A: All firms 348 349 332 282 1311
Panel B: Sustainable firms 46 57 52 50 205
Panel C: Non-sustainable firms 302 292 280 232 1106
3.3 Research method
The model that I will apply in my thesis for my first hypothesis can be estimated by: EVA = α + β1’S + β’Cit + ε. In this formula is EVA the dependent variable, which measures the firm
value. S is the sustainability dummy. This dummy equals 1 when the firm is sustainable and equals 0 when the firm is not sustainable, according to the Sustainability Yearbook of RobecoSam. The variable Cit represents the different control variables. β1 and β are the
different regression coefficients. ε is the error variable. This all lead to the following model:
For my second hypothesis, I will use a student t-test. There are different proxies for agency conflicts: SIZE, RISK, LEVERAGE and FCF. In my student t-test, I will examine if the mean of these values are lower in sustainable firms, which means that there are less agency conflicts in sustainable firms. More about these proxies in paragraph 3.3.3.
3.4 Measures
In this paragraph, I will give a description of the different measures that I will apply in my thesis.
25
3.4.1 Sustainability
Sustainability is the most important variable of my thesis. “Corporate sustainability is a business approach that creates long-term shareholder value by embracing opportunities and managing risks deriving from economic, environmental and social developments”
(sustainability-indices, 2014).
To measure sustainability, I will apply the same method as used by Lo and Sheu (2007). Therefore I will make a dummy variable for sustainability. This variable equals 1 when a firm is listed in the Dow Jones Sustainability Index (DJSI) and 0 otherwise. To know if a firm is sustainable, I will manually collect information from the sustainability yearbooks from RobecoSam. If a firm has made it to the yearbook, the firm is a sustainable firm. If an invited firm hasn’t made it to the yearbook, I will assume that the firm is not sustainable.
3.4.2 EVA
The first hypothesis goes about financial performance (EVA) related to sustainability. I expect that there is a positive relation between EVA and sustainability.
EVA can be calculated by the net operating profit after taxes (NOPAT) minus the cost of capital. A more practical approach defines EVA as NOPAT – (WACC (weighted average cost of capital) * capital employed) (Chen and Dodd, 1997). The database I will use for these variables is Compustat. Capital employed is the sum of shareholders’ equity and non-current liabilities. NOPAT will be calculated as EBIT (earnings before interest and taxes) times ( 1 – (income taxes / earnings before income taxes). The WACC can be calculated as the return on equity times (1 – debt / total capital) + return on debt x (1 – income taxes) x total capital). Whereas return on equity is the risk free rate + beta x (return on market – the risk free rate) (Deloitte, 2014b) and return on debt will be calculated as the ratio of interest expenses on debt. Beta measures how volatile a stock is. It shows the riskiness of a firm (Yahoo, 2014). I will get the beta of the S&P 500 firms from the finance part of yahoo. For simplicity, I assume that the beta of a firm won’t change in a period of four years. The market return comes from 1stock1 (2014).
26 ( ( ) ( ) ( )
To compare the variable EVA among the different firms, I will divide EVA by the enterprise value of those firms. This results in the variable CEVA.
However sustainability is not the only factor that can increase the value of a firm. Therefore I will include six control variables in my model that can also have an effect on the value of a firm.
The first control variable that I will apply in my thesis is size (SIZE). This will be calculated as the logarithm of total assets. Firm size is negatively related with firm value (Lo and Sheu, 2007).
The second control variable is access to financial market (AFM). This variable will be measured by using a dummy variable. This dummy variable equals 1 when the firm has paid a dividend and 0 otherwise. This dummy variable is expected to be negatively related with firm value (Lo and Sheu, 2007).
The third control variable is leverage (LEVERAGE). Leverage will be measured by the debt to equity ratio. This can be calculated as total liabilities divided by the total equity of the firm.
The fourth control variable is profitability (PROFITABILITY). Profitability will be measured by the return to assets ratio (ROA). ROA can be calculated as net income divided by the total assets of the firm. The higher the profitability of a firm, the higher the value of a firm (Lo and Sheu, 2007).
The fifth control variable that I will apply in my thesis is sales growth (SALESG). Sales growth can be measured by the percentage difference in the sales of the present year
27
compared to the sales of the previous year. Growth in sales is positively correlated with firm value (Lo and Sheu, 2007).
The last control variable that I will use is investment growth (INVESTG). Investment growth can be calculated as the total of capital expenditures divided by sales.
The following table represents all the variables that will be applied in my thesis for the first hypothesis. H1: There is a positive relation between sustainability and EVA.
Table 2: Variables H1
Variable Description
SUSTAINABILITY Sustainability equals 1 when a firm is in the Sustainability Yearbook and equals 0 otherwise.
EVA Economic value added. The measure of firm value. CEVA EVA divided by the book value of the firm.
SIZE The firm size.
AFM Access to financial markets. LEVERAGE The capital structure of a firm. PROFITABILITY The profitability of a firm. SALESG The growth of sales.
INVESTG Future investment opportunities. Investment growth.
3.4.3 Agency conflicts
The second hypothesis focuses on agency conflicts. I expect that there is a negative relation between sustainability and agency conflicts. To examine this relation, I will use the measures of agency conflicts as applied by Dey (2008).
There are four measures of agency conflicts that I will apply in my thesis. According to Dey it’s difficult to measure agency conflicts in a firm, because there are a lot of different firm-specific situations. Therefore there’s not just one measure of agency conflicts.
The first measure of agency conflicts that I will apply in my thesis is firm size (SIZE). Firm size can be measured by the natural logarithm of sales. Bigger firms have more complex ownership structures with more different owners. Therefore the interests of these owners are also more defers. Another reason why there are more agency conflicts when the firm is
28
bigger, is that larger firm have more different operations which provides greater incentives and opportunities for managers to shirk (Demsetz and Lehn, 1985).
The second measure of agency conflicts that I will apply in my thesis is volatility in operating environment (RISK). This will be measures by looking to the SIC codes of the different companies. When a firm has a SIC code between 3400 and 4000 the company will be seen as a high risk firm (Titman and Wessels, 1988; de Jong and van Dijk, 2002). Demsetz and Lehn (1985) argued that managers of more volatile firms are more difficult to monitor. The reason behind this is information asymmetry and an incentive for managers to act more risky, because they want to achieve their goals and they want to outperform. This is not in line with the interests of the owners. They don’t like these risky projects (Skinner and Sloan, 2002).
The third measure of agency conflicts that I will apply in my thesis is leverage (LEVERAGE). This measure can be calculated as the ratio of long-term debt to total assets. Managers of firms with a higher long-term debt to total assets ratio are more likely to engage in earnings management, because they want to meet debt covenants (Watts and Zimmerman, 1990). This will lead to incorrect numbers which are not showing a true and fair view of what has happened. This can reduce the value of the firm. From this it can be concluded that the more leverage in a firm, the more agency conflicts (Dey, 2008).
The last measure of agency conflicts that I will apply in my thesis is free cash flow (FCF). This can be calculated as the cash flow from operations minus the capital expenditures of the year. This has to be divided by current assets. Dey (2008) expects that the greater levels of free cash flows from operations, the more agency conflicts.
The following table represents all the variables that will be applied in my thesis for the second hypothesis. H2: There is a negative relation between sustainability and agency
29 Table 3: Variables H2
Variable Description
SUSTAINABILITY Sustainability equals 1 when a firm is in the Sustainability Yearbook and equals 0 otherwise.
SIZE The firm size.
RISK The firm will seen as risky when it has a SIC code between 3400-4000.
SIZE The firm size.
LEVERAGE The capital structure of a firm.
FCF Free cash flow (cash flow from operations – capital expenditures) / current assets.
30
4. Results related to H1
In this chapter I will present the results that are related to my first hypothesis. H1: There is a positive relation between sustainability and EVA.
4.1 Summary statistics
In the following table, I present the summary statistics related to my first hypothesis. The summary statistics are divided in three different panels. Panel A consists of all the firms listed on the S&P 500. Panel B consists only of the S&P 500 firms that are listed in the
Sustainability Yearbook of RobecoSam. Panel C consists of the S&P 500 firms that are not listed in the Sustainability Yearbook.
31 Table 4: Summary statistics H1
Variables Mean Std. dev. Median Skewness Kurtosis
Panel A: all firms CEVA -0,2949834 10.86521 0.0079855 -36.11661 1306.608 SIZE 9,636212 1.207468 9.478136 0.6558351 3.572602 AFM 0,8093059 0.3929983 1 -1.574683 3.479627 LEVERAGE 2,722951 29.82613 1.409082 27.12594 924.2409 PROFITABILITY 0,0652851 0.0598738 0.0631927 -0.7467498 13.33832 SALEG 0,0959089 0.331051 0.061922 17.23298 416.3734 INVESTG 0,0811852 0.1541874 0.0346391 4.466127 27.36501 Panel B: sustainable firms CEVA 0.0519062 0.4425888 0.0218958 12.03174 160.0696 SIZE 10.25357 1.18515 10.07226 0.9203538 4.492797 AFM 0.9121951 0.2837037 1 -2.912927 9.485146 LEVERAGE 2.094411 1.919256 1.540636 2.833472 12.30469 PROFITABILITY 0.0760022 0.0495011 0.0747611 0.1442048 3.095615 SALEG 0.0680193 0.1256098 0.0591675 0.9977985 11.66149 INVESTG 0.0644592 0.0649203 0.0409197 2.583726 11.85437 Panel C: non-sustainable firms CEVA -0,3592803 11,82757 0,0059406 -33,17899 1102,56 SIZE 9,521782 1,177006 9,358363 0,6376533 3,347139 AFM 0,7902351 0,4073249 1 -1,425724 3,032688 LEVERAGE 2,839453 32,46336 1,378526 24,93137 780,4952 PROFITABILITY 0,0632987 0,0614191 0,0604954 -0,8007845 13,9006 SALEG 0,1010783 0,3561502 -0,0631097 16,36298 367,5702 INVESTG 0,0842854 0,1653622 0,0327431 4,216233 24,16509
The summary statistics in the table above support my expectations that the EVA of sustainable firms is higher than the EVA of non-sustainable firms. Panel B consists of the sustainable firms and panel C consists of the non-sustainable firms. The mean of the variable CEVA of the sustainable firms is 0,0519062. This is 0,4111865 higher than the mean of CEVA of the non-sustainable firms, which is -0,3592803.
32
However it’s not possible to take conclusions only based on these summary statistics, because other factors also have a potential influence on the firm value which is measured by EVA. The summary statistics supports this theorem, because also the other variables show differences. For example the profitability of sustainable firms is higher than the profitability of non-sustainable firms.
The mean value of the variable CEVA is higher than the median value in the different panels, which suggests that the distribution of CEVA is skewed. To control for the skewness, I will use the natural logarithm of CEVA in my regression.
The following table represents a correlation matrix. Gujarati (1988) mentions that there’s a problem with the variables when there is a correlation coefficient between two independent variables of at least 0,8. All of the coefficients shown in table 5 are lower than 0,8. This indicates that no issues are expected regarding correlation. A test of
33 Table 5: Correlation matrix H1
Variables CEVA SIZE AFM LEVER PROFIT SALEG INVEST
Panel A: All firms
CEVA 1,0000 0,0347 -0,0131 0,0016 -0,0139 -0,0045 0,0079 SIZE 0,0347 1,0000 0,2492 0,0508 -0,2672 -0,0600 0,0263 AFM -0,0131 0,2492 1,0000 0,0416 0,0252 -0,0917 0,0019 LEVERAGE 0,0016 0,0508 0,0416 1,0000 -0,0134 -0,0100 -0,0178 PROFITABILITY -0,0139 -0,2672 0,0252 -0,0134 1,0000 0,0415 -0,1290 SALEG -0,0045 -0,0600 -0,0917 -0,0100 0,0415 1,0000 -0,0033 INVESTG 0,0079 0,0263 0,0019 -0,0178 -0,1290 -0,0033 1,0000
Panel B: Sustainable firms
CEVA 1,0000 -0,0956 0,0291 -0,0644 -0,0134 0,0707 -0,0384 SIZE -0,0956 1,0000 0,1667 0,4049 -0,2869 0,0505 0,1603 AFM 0,0291 0,1667 1,0000 -0,0912 0,0630 0,0038 0,1281 LEVERAGE -0,0644 0,4049 -0,0912 1,0000 -0,3810 -0,1130 -0,0867 PROFITABILITY -0,0134 -0,2869 0,0630 -0,3810 1,0000 0,0902 -0,0607 SALEG 0,0707 0,0505 0,0038 -0,1130 0,0902 1,0000 -0,0237 INVESTG -0,0384 0,1603 0,1281 -0,0867 -0,0607 -0,0237 1,0000 Panel C: non-sustainable firms CEVA 1,0000 0,0360 -0,0155 0,0018 -0,0158 -0,0042 0,0087 SIZE 0,0360 1,0000 0,2417 0,0545 -0,2940 -0,0621 0,0298 AFM -0,0155 0,2417 1,0000 0,0455 0,0119 -0,0933 0,0011 LEVERAGE 0,0018 0,0545 0,0455 1,0000 -0,0101 -0,0100 -0,0182 PROFITABILITY -0,0158 -0,2940 0,0119 -0,0101 1,0000 0,0429 -0,1316 SALEG -0,0042 -0,0621 -0,0933 -0,0100 0,0429 1,0000 -0,0045 INVESTG 0,0087 0,0298 0,0011 -0,0182 -0,1316 -0,0045 1,0000 4.2 Univariate test
I will test my hypothesis by using a t-test with a 95% confidence interval on the mean and the median of CEVA. Both are needed, because, as explained in the previous section, my results are skewed.
34 Table 6: Sample t-test H1
Sustainable firms Non-sustainable firms Difference T-statistics
Panel A: difference in means
Mean 0,0519062 -0,3592803 -0,4111865 -0,4975
Std. dev. 0,0309117 0,4425888
Panel B: difference in medians
Median 0,0218958 0,0059406 0,0159552 0,181052
Number of observations N
N 205 1106
The mean EVA of sustainable firms is 0,0519062 and the mean EVA of the non-sustainable firms is -0,3592803. This result is a premium of 0,4111865 for the non-sustainable firms. However this premium is not significant, because the t-value of -0,4975 is lower than 1,96. Based on this test, my hypothesis is not supported.
In panel B I’ve tested my hypothesis by using the median of EVA. The sustainable firms have a higher median than the non-sustainable firms, 0,0218958 compared to
0,0059406. This premium of 0,0159552 has a t-value of 0,181052 which is not significant. Therefore my hypothesis is not supported. Both results are inconsistent with my hypothesis which predicts that the EVA of sustainable firms is significantly higher than the EVA of the non-sustainable firms.
4.3 Linear regression
In addition, I’ve made a linear regression to examine the relation between EVA and sustainability. In this regression EVA is the dependent variable and sustainability is the independent variable. Table 7 presents the result of this regression which is based on the following model:
35 Table 7: Linear regression H1
logceva Coef. Std. Err. t P>t [95% Conf. Interval]
sustainability 0,051375 0,197132 0,260000 0,794000 -0,335353 0,438104 _cons -6,298567 0,077953 -0,800000 0,000000 -6,451493 -6,145641
The coefficient of sustainability is 0,051375, which means that when a firm transforms to a sustainable firm, the firm value will increase with 0,051375.
The results of this regressions don’t support my first hypothesis. The effect of
sustainability on EVA is not significant (the t-value of 0,26 is lower than 1,96). The power of this model is very low. It has an adjusted R-square of -0,0007, which is very low. The p-value of F is 0,7944 which is much more than the F value of 0,07. From this it can be concluded that, overall, the model can not significantly predict EVA.
To improve my model, I’ve included my control variables in my regression model. EVA is the dependent variable and sustainability is the most important independent variable. The other independent variables are the firm size, the access to the financial markets (afm), the capital structure of the firm (leverage), profitability, sales growth and investment growth. The results of this test are shown in table 8.
36
Table 8: Linear regression H1 including control variables
Including my model with the control variables has increased the adjusted R-square of the model to 0,0487. This implies that this model is a better predictor of firm value than the previous model. The model explains 4,87% of the variability of EVA. Also the p-value of F (0,0000) is much lower than the F-value of 10,57. From this it can be concluded that overall the applied model can significantly predict EVA.
In this model the coefficient of sustainability is 0,284527 which implies that when a firm became sustainable, the value of the firm will increase with 0,284527. The t-value of the variable sustainability is 1,43 which is smaller than 1,96. This means that the t-value is not significant. My hypothesis is not supported.
The variable SIZE has a significant t-value of -5,19. SIZE is negatively related with the value of a firm. This implies that bigger firms have a relatively lower firm value than the smaller firms. Also the variable PROFITABILITY has a significant effect on firm value (a t-value of 4,18). This means that the more profitable a firm is, the higher the firm t-value of a firm. Which is consistent with almost all of the economic literature. All the other control variables: AFM; LEVERAGE; SALEG; and INVESTG have not a significantly effect on the firm value of a firm.
4.4 Sensitivity analysis
Since 2009 there is an increase in companies who are getting a reward for being sustainable.
Ceva Coef. Std. Err. t P>t [95% Conf. Interval] Beta
sustainability 0,284527 0,199610 1,430000 0,154000 -0,107065 0,676119 0,039892 size -0,332170 0,064059 -5,190000 0,000000 -0,457840 -0,206500 -0,154768 afm -0,243781 0,185304 -1,320000 0,189000 -0,607308 0,119747 -0,036969 leverage 0,000114 0,002347 0,050000 0,961000 -0,004489 0,004718 0,001316 profitability 5,173680 1,238248 4,180000 0,000000 2,744502 7,602858 0,119531 saleg 0,206942 0,212196 0,980000 0,330000 -0,209342 0,623225 0,026436 investg 0,965338 0,457184 2,110000 0,035000 0,068441 1,862236 0,057435 _cons -3,373166 0,625932 -5,390000 0,000000 -4,601110 -2,141452 .
37
Even more and more firms are transforming in a sustainable firm. A reason for this increase in sustainable firms might be that more customers will give a higher value to the products of the sustainable firms compared to the products of the non-sustainable firms. Therefore I will apply a regression analysis with only the data from the most recent year: 2013. The results of this test are shown in table 9.
Table 9: sensitivity analysis H1
logceva Coef. Std. Err. t P>t [95% Conf. Interval]
sustainabi~y -0,041748 0,393786 -0,110000 0,916000 -0,816979 0,733483 size -0,215320 0,138692 -1,550000 0,122000 -0,488358 0,057718 afm 0,319159 0,426872 0,750000 0,455000 -0,521207 1,159526 leverage -0,095337 0,031011 -3,070000 0,002000 -0,156387 -0,034287 profitabil~y -7,182239 3,069198 -2,340000 0,020000 -1,322445 -1,140032 saleg -1,616361 1,197676 -1,350000 0,178000 -3,974179 0,741456 investg 2,058035 1,117102 1,840000 0,067000 -0,141159 4,257228 _cons -3,019235 1,367143 -2,210000 0,028000 -5,710675 -0,327795
The t-value of sustainability is -0,11. This is lower than 1,96. From this it can be concluded that my hypothesis is not supported. There is no relation between sustainability and EVA. This is the same result as the result of table 8 in which all the years are included. The main difference of the results of this sensitivity test compared to the test with all the years included is that the variable SIZE has not a significant effect on the firm value anymore. A counter intuitive result of this sensitivity test is that the variable
PROFITABILITY has a negative impact on the firm value. This means that the less profitable a firm is, the higher the firm value of that firm.
All the applied tests lead to the result that my hypothesis is not supported. There is no relation between sustainability and EVA. A reason for this is that there is now a process in the direction to sustainability (Sustainability Yearbook, 2014). Which leads to high costs in the
38
first years of the transformation process to a sustainability process. When this study will be repeated over 10-20 years, it might come to a different conclusion.
39
5. Results related to H2
In this chapter I will present the results that are related to my first hypothesis. H2: There is a negative relation between sustainability and agency conflicts.
5.1 Summary statistics
In the following table, I present the summary statistics related to my first hypothesis. The summary statistics are divided in three different panels. Panel A consists of all the firms listed on the S&P 500. Panel B consists only of the S&P 500 firms that are listed in the
Sustainability Yearbook of RobecoSam. Panel C consists of the S&P 500 firms that are not listed in the Sustainability Yearbook.
Table 10: Summary statistics H2
Variables Mean Std. dev. Median Skewness Kurtosis
Panel A: all firms
SIZE 9.468251 1.059122 9.376786 0.3073554 2.712649 RISK 0.2517182 0.4341867 0 1.144155 2.309091 LEVERAGE 2.471269 30.70004 1.313238 28.51954 925.6169 FCF 0.1908772 0.3378548 0.1983725 -0.8598361 42.45953 Panel B: sustainable firms SIZE 10.10275 0.9424787 9.976156 -.031873 2.234496 RISK 0.2135417 0.4108782 0 1.398016 2.95445 LEVERAGE 1.822793 1.322389 1.522322 2.319695 10.58378 FCF 0.2414972 0.1885987 0.2193737 1.184104 6.272984 Panel C: non-sustainable firms SIZE 9.342788 1.036061 9.199482 0.4183231 2.980355 RISK 0.2592593 0.4384537 0 1.098701 2.207143 LEVERAGE 2.599495 33.5947 1.280136 26.06318 773.0165 FCF 0.1808679 0.3593443 0.1952 -0.8296376 39.53619 The results of the summary statistics related to H2 show a higher mean of the variable
40
SIZE for the sustainable firms (10,10275) compared to the non-sustainable firms (9,342788). Prior literature has shown that the bigger the firm, the more agency conflicts occur in the firm (Demsetz and Lehn, 1985). In conclusion this result shows that there are more agency
conflicts in the sustainable firms than in the non-sustainable firms. This is not in relation with the predictions of my second hypothesis. My second hypothesis predicts that there exists a negative relation between sustainability and agency conflicts.
The second variable, RISK, shows a lower risk in sustainable firms (0.2135417) than in the non-sustainable firms (0.2592593). This result is supporting my expectations. Prior literature showed that the more risky a company is, the more agency conflicts occur in that company (Demsetz and Lehn, 1985). Shareholders don’t prefer those managers who are more difficult to monitor, due to information asymmetry.
The third variable, LEVERAGE, is lower for the sustainable firms (1.822793) than it is for the non-sustainable firms (2.599495). Prior literature showed that the more leveraged a firm is, the more agency conflicts may occur in that firm (Dey, 2008). In conclusion this result shows that there are more agency conflicts in the non-sustainable firms than in the sustainable firms. This supports my hypothesis.
The fourth variable, FCF, is higher in the sustainable firms (0.2414972) than in the non-sustainable firms (0.1808679). Dey (2008) showed that there is a positive relation
between free cash flows from operations and agency conflicts. My results shows that there are more free cash flows in the sustainable firms than in the non-sustainable firms. In conclusion, based on the variable FCF, there are more agency conflicts in the sustainable firms, This result doesn’t support my hypothesis.
Overall, I can’t make a common conclusion based on these summary statistics. The results of the variables RISK and LEVERAGE are supporting my hypothesis and the results of the variables SIZE and FCF are not supporting my hypothesis that there are more agency conflicts in non-sustainable firms.
In the following table the coefficients of the correlation test are provided. All these coefficients are lower than 0,8. This indicates that no issues regarding correlation are expected. A test of multicollinearity is therefore not needed (Gujarati, 1988).
41 Table 11: Correlation matrix H2
Variables SIZE RISK LEVERAGE FCF
Panel A: All firms
SIZE 1,0000 -0,1263 0,0369 -0,0214 RISK -0,1263 1,0000 0,0411 0,0124 LEVERAGE 0,0369 0,0411 1,0000 -0,0226 FCF -0,0214 0,0124 -0,0226 1,0000
Panel B: Sustainable firms
SIZE 1,0000 0,0329 -0,0535 -0,1333 RISK 0,0329 1,0000 0,1356 -0,1842 LEVERAGE -0,0535 0,1356 1,0000 -0,1165 FCF -0,1333 -0,1842 -0,1165 1,0000 Panel C: Non-sustainable firms SIZE 1,0000 -0,1459 0,0445 -0,0325 RISK -0,1459 1,0000 0,0431 0,0346 LEVERAGE 0,0445 0,0431 1,0000 -0,0222 FCF -0,0325 0,0346 -0,0222 1,0000 5.2 Univariate test
I will test my hypothesis by using a t-test with a 95% confidence interval on the difference in the mean.
42 Table 12: Sample t-test H2
Sustainable firms Non-sustainable firms Difference T-statistics 95% confidence interval SIZE: Mean 10.10275 9.342788 -0.7599673 -9.4217 -0.9182251 -0.6017096 Std. dev. 0.9424787 1.036061 RISK: Mean 0.2135417 0.2592593 0.0457176 1.3337 -0.0215374 0.1129726 Std. dev. 0.4108782 0.4384537 LEVERAGE: Mean 1.822793 2.599495 0.7767027 0.3202 -3.982536 5.535941 Std. dev. 1.322389 33.5947 FCF: Mean 0.2414972 0.1808679 -0.0606292 -2.2762 -0.1128906 -0.0083679 Std. dev. 0.1885987 0.3593443 Number of observations: N 192 971
This test shows mixed results regarding my second hypothesis. The difference in mean of the different variables are already shown and discussed in the summary statistics section. In this sample t-test I’ve investigated whether the observed differences are also significant differences.
The premium of the variable SIZE for sustainable firms is significant (-9,4217 is more than -1,96). This is an indication that my hypothesis can be rejected. It indicates that there are more agency conflicts in the sustainable firms than in the non-sustainable firms.
The premium of the variable FCF for sustainable firms is also significant (-2,2762 is more than -1,96). This is an indication that my hypothesis is supported. It indicates that there are less agency conflicts in the sustainable firms compared to the non-sustainable firms.
The difference in mean of the variables RISK and LEVERAGE is not significant. There is not a significantly difference in the mean of risk in sustainable and non-sustainable firms.
43
Overall, I still can’t make a common conclusion. Based on the results of the sample t-test provided above, there are only significant differences for the variables SIZE and FCF. The variable FCF supports my hypothesis. It shows that there are less agency conflicts in the sustainable firms, however the variable SIZE shows that there are more agency conflicts in the sustainable firms. The other variables, RISK and LEVERAGE, don’t show a significant difference. The fact that it’s not possible to make a common conclusion, may indicate that there is no relation between sustainability and agency conflicts.
5.3 Robustness test
I will perform an additional test to test the robustness of the significant relation between free cash flows and sustainability. Therefore I will use a linear regression analysis.
Applying this model leads to the following results:
Table 13: Robustness test H2
FCF Coef. Std. Err. t P>t [95% Conf. Interval] sustainabi~y .0602932 .0267059 2.26 0.024 .0078959 .1126905 _cons .1808679 .0108273 16.70 0.000 .1596246 .2021113 The t-value of sustainability is higher as 1,96 which means that my hypothesis still hold stand in the case of FCF. However FCF is only one of many proxies for agency conflicts. Therefore I can’t conclude only based on this variable that there is a positive relation between sustainability and agency conflicts.
Overall the different tests which are applied, leads to a rejection of my hypothesis. There is no relation between sustainability and agency conflicts. This is related to my conclusion based on my first hypothesis, which is also rejected. When being sustainable doesn’t add financial benefits for a firm, shareholders don’t demand their firm to be
sustainable. This result supports the theory of Friedman (1962), Ackerman (1975) and Miller (1999). All these researchers mentioned that the only responsible of a firm is to increase their
44
firm value. People and firms want to maximize their own utility. When sustainability is not congruent with this behavior, they don’t want to be sustainable.
