Positive Corporate Social Responsibility news influence on stock return:

(1)

Positive Corporate Social Responsibility

news influence on stock return:

Empirical evidence from the New York

Stock Exchange

University of Groningen

Faculty of International Business & Economics

Landleven 5

9747 AD, Groningen

The Netherlands

Perttu Juntura

Student number: 1659618

Email: junturap@gmail.com

(2)

ABSTRACT

While there is ongoing discussion about corporate social responsibility (CSR) and its importance to the companies and its stakeholders the issue has not being studied thoroughly. This study focuses on elaborating on the research area and testing if CSR actually has an influence on company value. Key issues and relevant literature is discussed in order to create a functional entity of information.

By looking into 310 companies in the New York stock exchange this thesis found indications that press release of CSR issues can have a adverse influence on the stock price. No statistically significant evidence of a positive CSR news publication’s influence on stock return was found, but consistent trend of negative CAR could be seen from the study results.

(3)

1 INTRODUCTION

1.1 Intro

Increased concern of environment has been a hot topic for years and the hype is not slowing down. Due to growing and expanding economies as well as population growth in the world, in developed and developing countries, corporations have the possibility to produce and sell larger quantities. As more production output there is, more harmful and un-recyclable waste from production, packaging, transportation and obsolete products are cumulating also. Nevertheless, the overall production amount is increasing, but the limits for pollution and use of un-recyclable materials is being cut down by municipal, national and international organizations policies. Also other social issues that concern large population of stakeholder are matters that companies have to deal with. Largest companies importance in overall economy is growing which leads to increased social demand for acceptable behavior of companies. Corporate social responsibility (CSR) promoting organizations develop, cooperate and monitor with other organizations, governments and authorities that the corporations and countries follow the international common standards of environmental policies. This is the troublesome market place where corporations are fighting to stay profitable.

(6)

Global warming is a current and ongoing disputed issue that is something concrete that will influence everybody around the globe. Pollution that the corporations and individuals are producing is increasing even though it causes problems already to the environment. Corporations producing goods for the market needs has being selected as the prime scapegoat and the responsibility for diminishing direct and indirect environmentally hazardous effects from the whole production and end of the product life cycle is on the corporations. It seems that a shift towards CSR has occurred. The causal relationship discussed above between changes in the legislations, changes in consumers and investors values, environmental pollution and its monitoring increases, and CSR, have led to a continuous changes in corporate operations.

MNCs operating in many countries with different regulations and different national level of enforcement of international environment agreements create a complex scenario. Due to this complexity, regulations are unavoidably weaker or completely missing compared to domestic economic scenario. Becchetti and Ciciretti (2006) claim that this is contributing to the increasing demand from individuals to expect corporations to act in socially responsible manner, because rules and regulations doesn’t seem to work. This can be seen from Corporate Social Responsibility Monitor’s studies, which records a substantial increase in European consumers demand for CSR (Corporate Social Responsibility Monitor, 2008). Also American investors take it seriously since investing decisions are influenced by ethical issues for over 25 percent of share-owing citizens (International Institute for Sustainable Development, 2003). What’s more, the demand has been real since Social Investment Forum reports that between 1995-2007 Socially Responsible Investing (SRI) has increased substantially (Social Investing Forum, 2007).

(7)

According to the survey corporate responsibility (CR) reporting has been increasing since 1993 and in 2005 80 percent of the largest global electronics and computer producers issued a CR report. (KPMG, 2005)

This thesis’ main objective is to find out if reporting of companies positive CSR

policies and achievements cause disturbances in stock prices?

By selecting New York Stock Exchange (NYSE), which is the largest stock exchange in the world, as research target it is possible to gain suitable conditions for doing a successful event study. Companies listed in NYSE, which engages in CSR actions, and additionally publishing their CSR activities in Corporate Social Responsibility Newswire are selected in this thesis as a data sample. With this combination a large number of companies was selected and therefore the reliability and visibility of the news improved. Because semi-strong form market efficiency is assumed in this thesis it is good to have news information that has publicity.

The following research question used in this thesis is:

1. Does publication of a company’s positive CSR activities influence its stock price?

(8)

1.2 Conceptual model

The conceptual model for this thesis is shown below in figure 1. In the conceptual model there are three parts that clearly show the process. This model is based on the efficient market hypothesis (EMH) theory, which means that every time when new information is published to the market it is available to everyone in the market immediately. Depending on the information and how stakeholders perceive it influences directly the stock prices.

New information in the market Influence of the new information to the stakeholders Stock return change

Figure 1: Conceptual model.

(9)

2 CAPITAL MARKET EFFICIENCY

Product markets and capital markets are connected to each other. When in product markets corporation makes an investment, it is the capital markets that finance these investments. Both markets enjoy fierce competition, which breeds efficiency in the market. The competition makes it hard to gain excessive returns in the market, but in product markets it is possible to gain excessive returns through excellent strategies. If this weren’t so, companies in product markets would not be able to seek return for their risk (e.g. product development that cannot determine if a good product will be developed would not get invested in), because only in inefficient product markets corporate strategies can be used to gain excessive returns. On the other hand, due to efficient capital markets, corporations should not be able to increase much value through financial decisions. (Hamberg, 2004)

According to Fama (1970) the optimal markets offers corporations production-investment alternatives and securities alternatives for investors that represent ownership of corporation’s activities, while the security prices correlate all available information out there.

(10)

Some researchers such as Grossman and Stiglitz (1980) argue that efficient financial market doesn’t exist, because information and trading is not free. Copeland et al. (2005) have a counterargument that the securities prices would still reflect all the available information even with trading costs. The trading prices would be the transparent signal of market price. Therefore the securities prices would efficiently drift according to the information available.

2.1 Financial markets role

Malkamäki (1989) listed the main purpose of financial markets:

1. Asset allocation from surplus spending unit (SSU) to deficit surplus unit (DSU)

2. Short-term dept transformation into long-term loan 3. Mediate information to all parties in the market

4. Risk management and diversification in financial issues

SSUs’ has more income than expenditure and real investments and therefore the surplus assets can be invested through financial institutions for the use of DSUs’. This way an efficient use of capital can occur. SSU gains return for its invested capital while DSU can borrow this excess capital for its spending and/or investment from the financial institution, which acts as a mediator. Because the timeframe for SSU and SDU alters, the financial markets can modify the SSUs’ supply for investments into SDUs’ demand for a loan. In order for both SSU and SDU to be able to know about financial markets opportunities and threats they need information. Financial markets mediate information to all parties, which helps them (SSU and SDU) to seek optimal solution and to diminish the overall risk. (Malkamäki, 1989)

(11)

the basic assumption for other efficient market models to start from (Fama, 1970). Fama (1970) defines a list of three ground rules for perfect capital market, which is further fulfilled by Copeland and Weston (1988).

Prerequisite conditions for a perfect capital markets:

1. There are no transaction costs in trading securities (Fama) 2. All information is free to all participants (Fama)

3. All agree on the implications of current information for the current price and distributions of future prices for each security (Fama)

4. All investors are rational value maximizers (Copeland and Weston)

Capital markets can be divided into three processes from efficient markets perspective. First, allocation process is where primary markets guide assets from SSU to SDU. Second, securities in aftermarket are usually exchanged through stock exchange, which is called the exchange process. Finally, information process is also in a key element in this list. When information is mediated between participants in the market the information process occurs. (Malkamäki, 1989)

2.3 Efficient capital markets

(12)

1. Weak form test includes only information based on historical prices. It is impossible for investor to gain excessive abnormal stock returns with this method.

2. Semi-strong form test states that all public information will reflect into securities prices immediately. Therefore it is impossible for investor to gain excessive abnormal returns with this information set, due to instant reflection of new information to the security prices.

3. Strong form test has more comprehensive demands since it includes all the

relevant information, public and non-public, which reflect immediately to the security prices. This means it is impossible for any investor to gain excessive abnormal returns.

This listing was redone and named by Fama in 1991 in order to gain more functional tests. The weak-form test covered a specific and limited area before, but the new test covers a more general area. Previous weak-form test changed into tests for return

predictability, which includes the burgeoning work on forecasting returns with

variables (e.g. dividend yields and interest rates), not just forecasting on the basis on historical returns. The semi-strong form and the strong form test’s biggest change was the name change into event studies and tests for private information, respectively. By widening the category of tests by including more forecasting tests it is possible to gain more comprehensive view. (Khotari & Warner, 2006, Fama, 1991)

2.4 Stock market anomalies

(13)

Schwert (2002) states that there are two main reasons for anomalies to occur. First, the market is inefficient and therefore there are opportunities to make profits with a strategy. Second, there are inadequacies in the underlying asset-pricing model. Additional to these Schwert believes that behavioral finance (e.g. irrational behavior of investor) concept could explain some anomalies.

The most common anomalies are size effect, January effect, turn-of-the-month effect, day-of-the-week effect, and P/E effect (Malkamäki, 1989). Some of these are going to be discussed next in few words. Notice that when discussing about anomalies the word “effect” is used as describing the same thing.

2.4.1 Size anomaly

Rolf Banz was the first researcher who found out the size effect (Malkamäki, 1989). A small capitalization companies’ risk adjusted return was higher compared to companies with large capitalization. Even though the size difference of the return was nearly 20 percent in Banz’s research it was also noted that there were possible periods when the large capitalization companies return was higher than small capitalization companies (Banz, 1981). Therefore, small capitalization earned higher average returns than predicted by capital asset-pricing model (CAPM) (Schwert, 2002). Chen (1983) reports that Arbitrage Pricing theory (APT) can explain most of the excessive abnormal returns caused by size effect. Due to the fact that APT includes several systematic risk components it can explain stock returns better than CAPM.

(14)

2.4.2 January anomaly

Officer (1975) found out that stock returns were higher during January compared to other months among Australian stocks. Since then this anomaly has been found in other stock exchanges in the world (Malkamäki, 1989). Keim (1983) showed in his research that January effect, also called turn-of-the-year effect, is noticeable for small companies and companies who’s stock price has decreased during previous December and/or the whole year. January effect is most significant during the first days of the year. Roll (1983) conjunctures that due to higher volatility of small companies stocks they are more likely to experience paramount short-term capital losses, which the investors might want to realize for income tax purposes for the present year. Roll further elaborates that this extensive selling reduces the prices of small companies stocks, which leads for these stocks to be repurchased right after turn-of-the-year. Ritter (1988) revealed another explanation for January effect. According to Ritter the institutional investors sell the small capitalization stocks at the end of the year in order to decrease risk of the portfolio, which would look bad in the end of the year accounting.

2.4.3 P/E anomaly

(15)

3 STOCK VALUATION

Before the Wall Street stock market crash in 1929 it was generally thought that the stock exchange quotation of the day was the intrinsic value of a stock. Since then stock valuation research begun and the goal was to gain knowledge about the intrinsic value of a stock. (Suvas, 1989)

Several different valuation models exist, but in general as long as same assumptions are used in every model the end-result is the same. Therefore the main difficulty is to estimate the future cash flows, not the valuation model. If the cash flows increase it would increase the stock price also and vice-versa with decreasing cash flow (Hamberg, 2004). The beginning of the stock valuation theory is believed to have started from John Burr Williams in 1938 when he proposed a net present value (NPV) of a stock formulates by discounting all future dividends to present time (Suvas, 1989).

Value-oriented investors use their common sense combined with easily understandable methods based on theoretically correct investment principles while trying to maximize their returns. EMH states that this is impossible unless the stock is under-valued. (Fielitz & Muller, 1985)

Notice that in this thesis stock valuation and company valuation expressions are both used, but the meaning of both the expressions is the same.

3.1 Stock valuation models

(16)

3.1.1 Dividend based model

In 1938 J. B. Williams developed one of the first company valuation models, which got common acceptance among valuation literature. Williams proposed that the stock value is the present value of all net-dividends that is gained from the stock in the future. Therefore the future dividends are discounted to the present value with investors expected rate of return working as discount rate. (Suvas, 1989, Gordon & Shapiro, 1956).

Formula 1: (Gordon & Shapiro, 1956).

! P₀ = Dt (1+ k)t t =1 "

#

where: P0 = Stock price at t = 0

Dt = Expected dividend at time t

k = Rate of profit (investors expected rate of profit)

(17)

Formula 2: (Gordon & Shapiro, 1956). ! P₀= D0 k " g where: D0 = Dividend at t = 0

k = Expected rate of dividend growth g = Dividend growth

In order for this to work (k) must be larger than (g) (Gordon & Shapiro, 1956). Suvas (1989) also states that there is a problem that if growth (g) would be equal to rate of profit (k), then the value of the stock experiences a paramount increase. In addition to this, even though Gordon model was more straightforward there is a problem with the model’s expectations. The Gordon model assumes that the future dividends paid are stable, in order for the model to function better, but the future dividends might not stable. (Suvas, 1989, Kallunki et al., 1999)

3.1.2 Profit based model

Miller and Modigliani (MM) (1961) argued that stock value is independent from the dividend policy. MM created a principle for the valuation: “the price of each share

must be such that the rate of return (dividends plus capital gains per dollar invested) on every share will be the same throughout the market over any given interval of time”. Therefore, the model holds that the value comes from both dividends and price

change of a stock discounted to present value. According to MM model investor receives each year (t) a profit that is based on risk-free interest rate (rt) (Suvas, 1989).

(18)

Formula 3: (Suvas, 1989).

!

r_t= dt+Pt+1" Pt

P_t

where:

dt = Dividend per stock, which corporation pays at the last day

of period t

Pt = Stock price on the beginning of period t (after distribution

of dividends on previous period t-1)

And therefore the formula to derive the stock price can be expressed as: Formula 4: (Suvas, 1989). ! P_t= 1 1+rt d_t+P_t+1

(

)

(19)

where:

nt = Amount of stocks in the beginning of period t

mt+1 = Amount of stocks sold in an new issue during period t (sold

after distribution of dividends of year t, at price Pt+1)

nt+1 = nt+mt+1

Vt = ntPt = Value of the company in the beginning of year t

Dt = ntdt = All the dividends paid at the end of year t to the

existing stock in the beginning of year t

In addition investments influence the valuation. To finance the investment, company can decrease dividends or issue more stocks. In formula 6 company’s dividend policy affects the value directly (if the amount of dividends change) or indirectly (amount of the external financing used) (Suvas, 1989). The indirect consequences can be expressed as:

Formula 6: (Suvas, 1989).

!

m_t+1P_t+1= lt" X

(

t" Dt

)

where:

lt = Amount of investment during year t

Xt = Net profit during year t

(20)

Formula 7: (Suvas, 1989). ! V_t = ntPt = 1 1+ rt X_t" lt+ Vt+1

(

)

MM model states that the company value is depending on the net profit and investments and company value in the future (Vt+1), not on dividends like Gordon

model suggests. Because future value of the company (Vt+1) is not depending on the

present and future dividends (Dt), then company’s dividend policy doesn’t influence

the company value. (Miller & Modigliani, 1961, Suvas, 1989)

3.1.3 Cash flow based model

Valuation models displayed above are problematic which gave the incentive to create another model. Dividend based model might not be accurate since the management of a company can keep the dividends equal regardless of the changes in yearly profits. Problem with the profit based model is that it is subjected to management’s discretionary decisions on financial statement issues. Cash flow based models focus only on the movements of the cash to and from the company. From the management perspective this means moving the focus from issues influencing short-term profit development to long-term maximization of company value. (Kallunki & Niemelä, 2007)

(21)

Formula 8: (Levin, 1998).

where:

NOPLAT = Net operating profit less adjusted taxes Dep. = Depreciation

CWC = Changes in working capital CE = Capital expenditures

After this weighted average cost of capital (WACC) is needed because it is used for discounting the expected future FCF in order to calculate the company value. WACC is based on weighted average of the costs of the capital that was for financing an investment (Hamberg, 2004). WACC can be calculated with formula 9.

Formula 9: (Hamberg, 2004).

where:

rd = Expected return on debt

req = Expected return on equity

D = Market value of the company’s debt EQ = Market value of the company’s equity

(22)

Formula 10: (Copeland, 1988).

where:

(23)

4 CORPORATE SOCIAL RESPONSIBILITY

4.1 Definition of CSR

World Business Council for Sustainable Development’s (WBCSD) has a long and extensive international cooperation history with MNCs’ around the world in sustainable development and CSR issues. WBCSD definition of CSR:

“CSR is continuing commitment by business to behave ethically and contribute to economic development while improving the quality of life of the workforce and their families as well as of the local community and society at large.” (WBCSD, 2000)

As seen from WBCSD’s definition of CSR it affects a large stakeholder group. Jamali and Mirshak (2007) also conclude that businesses need to pay regard to a wider scope of stakeholders. Companies need to understand values of stakeholders, who are affected by their operations, not just publish their own values (WBCSD, 2000).

Van Dijken (2007) reports variation of definitions of CSR from previous literature. Variation of these include: 1) companies decide voluntarily to contribute to a better society and cleaner environment, 2) companies are accountable for other things than just financial performance, 3) CSR is included in all corporate actions, behavior and its impact.

(24)

wide scope, and other participants who are affected. There is also a difference between CSR and acting on it, in other words the corporation is responding to social demand.

4.2 CSR concept developments

CSR has its roots since 1920’s and 1930’s from which on its gained more focus in business and academic research (Carroll, 1979, Frederick, 1994). By 1950’s CSR had gained visibility among businesses even though concept of CSR was still vague (Carroll, 1979). In 1962 and 1970 Milton Friedman’s controversial arguments challenged the CSR thinking. Friedman argued that social development is government’s task, not businesses (Warhurst, 2005, Carroll, 1979). Friedman’s opinion was that the businesses role was stockholder wealth maximization, not to engage into SRs (Carroll, 1979).

(25)

Three concentric circles are divided into inner circle, intermediate circle, and outer

circle. The inner circle (lightest gray circle in the figure 2) includes only basic

responsibilities to fulfill economic functions (e.g. production, jobs, and economic growth). The intermediate circle (mediocre dark shade circle in the figure 2) includes inner circle’s economic functions and additionally recognizes corporation’s influence on altering social values and priorities (e.g. environmentally friendly procedures, correct human resource actions, etc.). The outer circle is the most comprehensive of these. It consists all the functions in inner and intermediate circle in addition to futures significant unknown non-economic functions (e.g. poverty). (Carroll, 1979, Key, 1999)

In the 1970’s many academic researcher thought word “responsibility” moves the emphasis from performance to motive. They argue that acting on social demand is performing something, not just deciding to take action. (Carroll, 1979, Jamali & Mirshak, 2007)

(26)

Key (1999) summaries that research done about corporate correlation to external environment has produced literature that dealt with CSR, corporate responsiveness and corporate social performance (CSP). Main focus in this has been in developing a logical understanding of corporate behavior. Frederick’s model of explaining corporate behavior involves three parts called CSR1, CSR2, and CSR3. CSR1 stated corporate responsibility, stewardship, charity and philanthropy as an obligation to the members of the community. CSR2 is corporate responsiveness to societal demand in order to insure corporation legitimacy. CSR3 is corporate rectitude and ethical part, which should guide corporate behavior. (Key, 1999)

4.3 Regulative issues

Increased scrutiny of MNCs’ by variance of audience, non-governmental organizations (NGO) and the media has been developing during 21st century as well as global activism (Juholin, 2004). Van Dijken (2007) stated that as CSR has increased its popularity among businesses has the regulation and directives towards CSR also multiplied. Governments and NGOs have created legislations and regulations concerning CSR to push businesses towards SR behavior, which have tighten the corporations’ operating possibilities, but these are difficult to monitor in international markets.

(27)

Media, NGOs and other national and international agencies are making an effort to bring corporate misbehaviors public in order to stop these from happening (Van Dijken, 2007), and this transparency has put pressure on MNCs to report and act in a way that fulfills the CSR demands (Juholin, 2004).

In order to fulfill societies demands of CSR there would have to be standard criteria for this. Problem is that between countries, cultures and even generations the definition of CSR can alter. Therefore MNCs have a troublesome task to do if it tries to act in CSR manner in its global presence (Juholin, 2004). In 1997 United Nation’s Environmental Program (UNEP) and Coalition for Environmentally Responsible Economies (CERES) organization developed a Global Reporting Initiative (GRI) that was one of the first attempts to create an approved reporting regulations for environmental and social responsibility reporting (Niskala & Tarna, 2003).

According to Niskala and Tarna (2003) GRI directives development is a stakeholder relationship process, which has included the central stakeholders of reporting. The most important goal of GRI directives is to improve corporations’ CSR reports comparability while the GRI organizations goal has been to become an independent worldwide institution, which guides CSR reporting.

(28)

Wide category of NGOs has made an extensive variation of standards and directives, which are based on voluntary participation. These include organizations such as United Nations, European Union, International Labour Organization, International Chamber of Commerce, International Standards Organization, etc. (Niskala & Tarna, 2003)

4.4 CSR reporting

A “triple bottom line” concept emerged during the scandals of 1980s and it is the core of corporate responsibility and corporate citizenship. Key point of “triple bottom line” is to measure and report company’s performance with a respect to economic prosperity, social justice and environmental quality. This means increased transparency in terms of reporting performance and risk. Through time this development changed the attitude from 1980s way of thinking that environmental protection issues of a corporation are just something extra giving positive image into today’s more comprehensive consideration of corporations positioning in the society. International companies utilize the concept corporate citizenship and develop socially acceptable policies, strategies and reporting principles. (Warhurst, 2004)

(29)

Van Dijken (2007) states that CSR is most beneficial when it shapes the strategy to create value. What this means is that by implementing CSR the company will perform well while also doing good for wider stakeholder group. Niskala and Tarna (2003) continue that the corporate structures and supply chains are complex in strategic and operational sense and therefore reporting is a vital tool to inform about corporate actions to internal and external stakeholders. Due to this, CSR reporting is a useful tool to inform about risks and opportunities in financial, social, and environmental issues.

Financial reporting is a standard protocol that corporations issue to give information about the economic condition of the corporation. As there are many financial performance indicators displaying an image of the economic state of the corporation, CSR reporting includes other performance indicators. CSR reporting indicators paint a more comprehensive picture of the corporation’s state, not just financial performance. These include environmental, social, political and economic performance indicators, which provide information concerning non-financial issues. Social demand for non-financial indicators has created a situation when corporate reports display a more transparent idea of the overall operations of the corporation. In order for corporations to produce these reports it needs to involve wide variation of business departments in the reporting process. Investor relations, research and development, human resources and environmental departments give their input, so thorough view of operations can be produced. (Jamali & Mirshak, 2007, O’Connor & Spangenberg, 2007, Reynolds & Yuthas, 2008)

(30)

Niskala and Tarna (2003) explain thoroughly CSR reporting and relevant issues. Next the figure 3 is explained to give comprehensive idea of it. Business environments

external pressures (e.g. investor and stakeholder demands, authorities instructions)

and internal pressures (e.g. internal reporting, SR management) determine implementation of SR reporting. The meaning of SR for the corporation should be clear in order to define strategic goals, significant SR issues and vital stakeholders, which are related to the operations of the corporation. This is the base for defining which SR processes and procedures are implemented. Commitment to the CSR issues comes after the corporation has published it commitments externally and internally. Previous steps build up the corporate approach to the issue. SR management and reporting are a seamless entity. The management process starts from setting

objectives, visions, strategies, and CSR policies. In order to get to the objectives set,

corporate needs to plan the actions and resources needed to get to the objectives. The outcome of planning is initiated by taking concrete actions. These actions are then

followed-up on the basis of objectives set in the beginning. Final phase of

management process is to evaluate the overall process. (Niskala & Tarna, 2003) SCR reporting process starts with setting objectives for the report (e.g. what are the main issues reported, is it done according to specific instructions, what is the target group of the report). When it is known what to aim for the reporting needs to be

planned. Corporate needs to plan if the report done internally or externally and who is

responsible, timeframe, what information is submitted and from which areas, who analyses and verifies the report. In implementation the structure, presentation, fulfillment of set requirements, analysis issues are considered and selected.

Distribution of the report is done depending who it is for. Due to huge amount of time

(31)

Figure 3: Issues influencing the implementation SR reporting (Revised from Niskala & Tarna, 2003).

4.4.1 Theories behind voluntary reporting

In the next chapters few relevant theories concerning voluntary reporting will be discussed.

4.4.1.1 Organizational legitimacy theory

(32)

assumed to be under the influence and also have influence on the society where it operates. Therefore corporate reporting policies are considered to be a good vehicle to influence society’s perception of the corporate. This is vital since if society perceives that the corporation has broken the “social contract” the corporation’s survival is threatened. “Social contract” means the same as idea of legitimacy, because if society believes corporate is not legitimate then it is not operating according to society’s values. To summarize, corporate is perceived as legitimate organization by the society if their values are parallel. Targeted and strategically correct reporting can thereby be used as a tool by a corporation to paint a deliberate picture of the organization to the society. Even go far as alter the structure and operations towards socially accepted values. This is not a static position either, since society’s values can change during time which makes it much more difficult for corporations. (Deegan, 2002)

As the society provides legal standing, possibility to own and use natural and human resources it does not mean that the corporation would have any inherent rights to these. In order for the society to allow corporation access to these resources and produce goods and services as well as waste, there must be more pros than cons (Deegan, 2002). Therefore corporate survival depends not only on profits, but also society’s acceptance of the final product and operations needed to produce that (Milne & Patten, 2002). Deegan et al. (2002) argue that media has the capability to influence society’s interest on specific issues and that managers believe mass media can alter society’s expectations. They further argue that due to this reasoning management focuses in the annual reports on issues they believe society is concerned about.

4.4.1.2 Political economy theory

(33)

powerful and have more assets to battle in and out of court about controversial issues with different individuals and organizations of the society.

Also arguments are made that when investigating political economy much wider social issues should be focused which impacts the corporate operations and disclosures. Disclosures are a useful tool since they have the possibility to inform about social, political, and economic matters. (Deegan, 2002)

4.4.1.3 Stakeholder theory

The stakeholder theory is trying to explain the relationship between corporation and its external environment and how corporations behave within the environment. Previous academic literature has tried to come up with logics of explaining organizations behavior with prescriptive approach (e.g. Frederick, 1994, Carroll, 1979, etc.). Stakeholder theory is descriptive as it focuses primarily on the actors in the environment (Key, 1999).

(34)

4.5 CSR & stock performance

CSR actions can be beneficial for the corporate, until the costs of doing so threaten the corporate survival in competitive markets. SR actions are not a “free lunch” as it commonly means compromising between the shareholder value maximization and the welfare of stakeholder, which usually implicates that there are additional costs from doing this (Becchetti & Ciciretti, 2006). Becchetti et al. (2007) listed positive and negative sides of CSR. Some of the positive sides are the higher employee moral and productivity, which is created due to better employee relations and HR policies. When the consumers see the positive CSR actions sales growth might occur, due to consumers increased purchase of the corporation’s products. In addition more positive relationship between the stakeholders and the corporate and corporate performance can be seen. Increased awareness of CSR issues among the stakeholders has made CSR issues count in corporation’s benefit. Main negative side is of course the resources imputed (expenses) into CSR and that deviates from the shareholder value maximization.

(35)

Two ways of how the new information can influence the investors’ perception. First, the disclosure effect means that new information is produced outside the corporation and it presents how well is the corporation fulfilling the legally demanded environmental laws. Relevance of this comes from the fact that it can influence the investors’ perception about future corporate expenditures to fulfill environmental legislation. Second, regulatory effect comprises the outside produced information’s influence to the probability that the environmental legislation will become tighter and increase the penalties given for not following the new legislation. Regulatory effect might alter investors’ perception, since the new information can mean that the corporation has more expenditure in the future, due to the stricter legislation. (Shane & Spicer, 1983)

Jaggi and Freedman (1992) found contradicting results about corporate pollution disclosures effect to stock prices. In order for management to pursue increased profitability they should cut down on expenditures related to lower pollution. Therefore management should be encouraged to cut down pollution in order to do good for the society. Since in previous research investors perceived that the most profitable large corporations had the lowest total risk. As stated above Gordon and Rogene (1978) and Niskala and Tarna (2003) found a link that CSR is associated with good management, which means lower risk. Warhurst (2005) states that it is better for the corporation to be proactive and manage the risks, not just do crisis management when the adverse incident has happened. Rationality behind this is that the corporation can notice any opportunities associated with these and these can be used to enhance the value. What’s more, by acting in responsible manner the corporation can give a positive outlook of themselves to consumers, employees, and business partners (Warhurst, 2005).

(36)

Numerous studies concerning CSR and stock returns have been made from variation of viewpoints. Pava and Krausz (1996) made interesting findings when looking into 21 previous studies. Most of these studies, even thought they used alternative measures, almost all of them found CSR and traditional financial performance positively linked. Over half of the studies that focused on stock returns showed a positive link between CSR and stock return. The studies were based on different kinds of data, but the positive linkage cannot be overlooked. Pava and Krausz (1996) conclude that the SR corporations are at least as good performers, and some cases even better, compared to corporations that are not SR.

Investors’ interest towards the SR corporations has increased during time. Social Investment Forum (SIF) is a national membership association in United States that is working to develop, practice, and increase the SRI. In SIF’s 2007 Report on SRI trends in the United States they state that SRI assets, using SRI strategies, has increased during 1995-2007 from 639 billion USD to 2.71 trillion USD. This is an excessive 324 percent increase compared to 260 percent increase in broader universe of professionally managed assets. These latter assets cumulated total of 25.1 trillion USD in 2007, which is over nine times more compared to SRI assets (SIF, 2007). Similar trends of SRI are published by other equivalent organization, for example: The Canadian Association for Socially Responsible Investing, European Social Investment Forum, and Association for Sustainable and Responsible Investment in Asia.

(37)

(38)

5 DATA & METHODOLOGY

5.1 Event Study

Event studies have been used extensively during last decades and it has gained its place in financial economics. The event studies focus on movements of stock prices around specified events related to a corporation. Event studies are based on the fact that EMH works. Therefore systematical nonzero abnormal returns following a specified corporate event should not occur under the expectation that EMH works. Because this is the case sometimes a short-horizon event study can be used to understand corporate policy decisions. (Khotari & Warner, 2006)

Numerous event study methods are used by researchers, which can differ from each other. Bowman (1983) presented a model that includes five parts for doing an event study.

1. Identifying the event of interest

2. Creating and explaining a model of the security price reaction 3. Estimating of the excess returns

4. Organizing and grouping the excess returns 5. Analyzing the returns

5.1.1 Research limitations and definitions

(39)

CSR issues discussed in earlier sections of the thesis. This is done by taking CSR news that a company has published themselves and additionally paid to get it published in NGO based CSR promoting company’s Internet page. CSRwire is the NGO of CSR promoting company in question. This creates the additional definition and reliability to this thesis, because CSRwire is a leading source and known promoter of CSR issues.

By having positive CSR news that is announced by a company and additionally in CSRwire’s Internet pages it is argued in this thesis that the news has gained enough publicity that it fulfills the EMH semi-strong conditions. Therefore the influence of these news should be reflected to the stock prices immediately under the semi-strong form of EMH.

The event window is seven days and this is displayed in figure 4. One day (-1) before the publication of the CSR news, the day (0) CSR news published and the next five days from the publication (+5). Only business days are used, because these are the days when the stock exchange is open and trading can occur. Therefore, if a publication has occurred on non-business day then the next business day will be the event day.

With this short-horizon model the results of stock returns can be seen clearly. Khotari and Warner (2006) state that by selecting a short-horizon method the results represent more reliable and clean evidence of efficiency and therefore this method has lower possibility of having bias in the events (Khotari & Warner, 2006).

(40)

5.1.2 Hypotheses

It is argued by academic researchers that investors are concerned only gaining maximum profit for their investment and therefore CSR might have a negative influence or no influence on the stock price. Nevertheless previous research in this area has produced conflicting results for and against companies’ CSR policies influence on stock price. Good example of this is a research done by Pava and Krausz (1996) that is concerning CSR issues and its influence on stock prices. They focused on previous research done in this area and concluded that the results are complex and vague. Still they found out that in over half of the previous studies in this area companies that were perceived as fulfilling the SR criteria did actually perform equal or better compared to companies that weren’t seen as SR. Since the previous studies has not produced a clear and systematical results of the influence of the CSR policies to stock returns the following hypotheses are produced for this thesis:

H1: Positive CSR news has no influence on the stock return

H2: Positive CSR news has a positive influence on the stock return

5.2 Data

In the following sections data and limitations of the study will be discussed.

5.2.1 Stock Data

(41)

1. Company is listed in NYSE at the time of the publication of the news 2. Company had additionally published the positive CSR news in CSRwire’s

Internet page

3. Financial information could be found in Datastream

4. News announcement occurred during 1st of January 2000 to 1st of April 2008 The data was gathered by first taking a list of companies listed in NYSE and then crosschecking these companies to CSRwire’s database in order to get a group of companies, which were listed in NYSE and had also published their CSR news in CSRwire’s Internet page. Then the financial information is searched from Datastream. With these limitations 310 events was found and used in the research.

Two additional sub-samples were created and tested. In order to see if the type of news had influenced the stock return the news were divided by using Niskala’s and Tarna’s (2003) CSR dividing structure. Their structure divides CSR into three different categories. Economical, environmental, and social are the three types of CSR activities company can get involved. From the data there were only two types of news by Niskala and Tarna categorization, since economical news was not found. Therefore, news was divided into environmental and social and total number of news that could be used was 86 and 213, respectively. In the process of dividing the news they were already categorized in CSRwire’s Internet page. Since the number of companies in one of the group decreased under a hundred, further or more comprehensive dividing of news was not done in order to keep the rules of grouping clear. In addition it is not beneficial to make groups small because of the testing for statistical significance of the groups.

(42)

larger groups size and clear split of groups. Therefore mid cap companies will move into small cap or large cap according to their market capitalization at the event day. With these dividing rules for market cap groups there were 124 companies in small cap and 184 in large cap.

5.2.2 New York Stock Exchange Data

The NYSE is the world’s largest and most liquid exchange group with its wide variation of financial services and products. The NYSE Composite index is weighted using free-float market capitalization on both price and total return basis (NYSE, 2008). Free-float market capitalization is calculated by multiplying equity’s price by the number of shares readily available in the market. Free-float method is a better way of calculating market capitalization, due to its ability to reflect accurately market movements (Investopedia, 2008).

The NYSE Composite Index will be used as a market return in this thesis. The NYSE Composite index includes 1901 companies with total market capitalization of 18510 billion USD while average market capitalization is 9.7 billion USD (NYSE, 11.06.2008). The NYSE Composite Index includes over 60 percent of the total market capitalization of all publicly traded companies in the world (Bloomberg, 2008). It is designed to measure the performance of all common stocks listed in the NYSE (NYSE, 2008).

5.3 Abnormal Return and Cumulative Abnormal Return

(43)

Formula 11: (Khotari & Warner, 2006).

where:

t = Time

i = Sample security

Rit = Actual return of a security (i) during time (t)

Kit = Expected return of security (i) during time (t)

eit = Abnormal return of security (i) during time (t)

Secondly, the cross-sectional average abnormal return (AR) in calculated to see if there are abnormal returns in cross-sectional distribution. With formula 12 this can be done in order to see if the actual and expected returns are the same. Usually a null hypothesis is used to see if average abnormal return equals zero at time (t). (Khotari & Warner, 2006)

Therefore this thesis assumes that mean AR is zero, under the null hypothesis, which means that actual returns equal expected return at time (t).

where:

N = Number of sample securities

(44)

results of AR with formula 13, which is a typical calculation when making time-series aggregation in event studies (Khotari & Warner, 2006). If there are abnormal returns before the event day then possibility of information leakage might have occurred. Also if significant abnormal returns occur days after the event day then it can be interpreted as inefficient market, because the information does not effect the security prices immediately as assumed on the basis of semi-strong form of EMH. With CAR it is possible to get overview of the abnormal returns around the event.

where:

CAR (t1,t2) = Cumulative abnormal return starting at time t1

through time t2

Different test intervals are used to investigate the event by calculating CAR for each interval for the whole event sample and the two sub-samples.

Intervals are [-1, 0], [-1, +1], [-1, +2], [-1, +3], [-1, +4], [-1, +5], and [0, +1].

5.4 Statistical testing of abnormal returns

(45)

Formula 14: (Heikkilä, 2004). where: t = T-value ! x = Sample mean µ = Expected value

(46)

6 EMPIRICAL RESULTS

The research event data is gathered during an eight-year period from 2000 to 2008 from the NYSE. The NYSE Composite Index, which also acts as market return, has being significantly decreasing and increasing during this period (see figure 5). Therefore, the market has experienced both bull and bear market conditions. As the index is acting as market return in this thesis the stocks are expected to follow the index. Due to a substantial amount of events, 310 to be specific, external shocks that can cause the market to change direction are not expected to create a significant bias in the results.

Figure 5: NYSE Composite Index (Yahoo! Finance, 2008).

In the following sections of chapter six the empirical results of the study, which are calculated with previously explained methods, are discussed. First the results of the whole data sample are discussed and then the sub-samples. Finally, in the conclusion part of chapter six the assumed hypotheses are discussed in terms of results.

(47)

6.1 Stock return results of positive CSR news

In this part the entire set of observations are used for testing the influence of the positive CSR news. All the 310 observations are used for CAR calculations, which were also statistically tested. A histogram of the total data sample is displayed in figure 6. During the event day there was 151 negative AR and 159 positive AR, which suggest here already that there might not be a significant AR and/or CAR occurring as a result of the event.

Figure 6: Histogram of data sample.

On the basis of previous literature, one can hypothesize that if CAR is present it should be positive. This assumption is tested with one-tailed t-tests to see if the observed CAR values are statistically significantly positive. Table 1 displays the results of the tests for given intervals.

(48)

occurred after the event day. Nevertheless, this negative trend is not statistically significant. Therefore it can be stated that there exists a consistent trend of negative stock return, but it cannot be statistically proven that this trend has a connection to the positive CSR news publication.

CAR results of a positive CSR news influence on stock return

Interval CAR t-value significance day AR

-1, 0 0.28% -0.024703 0.490150 -1 0.29% -1, +1 -0.08% -0.886294 0.187958 0 -0.01% -1, +2 -0.15% -1.028821 0.152053 +1 -0.36% -1, +3 -0.11% -0.891720 0.186513 +2 -0.07% -1, +4 -0.20% -0.923844 0.178061 +3 0.04% -1, +5 -0.30% -1.063843 0.144027 +4 -0.09% 0, +1 -0.37% -0.827243 0.204238 +5 -0.10%

Table 1: CAR results Table 2: Daily AR

The CAR has a more negative return on interval (0, +1) compared to interval (-1, 0) that supports the fact that there was a noticeable turn towards negative return on event day. In addition, when looking at table 2 the movement of the AR towards negative occurred on event day and increased substantially during day (+1). This suggests that the market is efficient, as assumed by semi-strong EMH, since the influence of the new information was reflected instantly to the stock price. These are implications that can be derived from the results from tables 1 and 2, even though no statistically significant movement occurred.

6.2 Social news vs. environmental news

(49)

insignificant (tables 3 and 5). Again, the interval (0, +1) shows higher negative CAR compared than interval (-1, 0) in both types of news. Social news (table 3) has steeper negative turn during interval (0, +1) compared to environmental news (table 5), which can indicate that investors perceived the social CSR news as more negative or environmental news as positive. Also environmental news (table 5) group shows positive CAR on longer intervals (-1, +4, and -1, +5) and positive AR (table 6) on later days (+4 and +5).

Results for social news

-1, 0 0.32% -0.172615 0.863040 -1 0.38% -1, +1 -0.15% -0.952736 0.341461 0 -0.07% -1, +2 -0.23% -1.069493 0.285685 +1 -0.47% -1, +3 -0.24% -1.039027 0.299629 +2 -0.08% -1, +4 -0.43% -1.145832 0.252875 +3 -0.01% -1, +5 -0.57% -1.285055 0.199891 +4 -0.20% 0, +1 -0.54% -0.796001 0.426573 +5 -0.14%

Results for environmental news

-1, 0 0.17% 0.213318 0.831348 -1 0.07% -1, +1 -0.03% -0.194588 0.845976 0 0.10% -1, +2 -0.10% -0.328652 0.742887 +1 -0.20% -1, +3 -0.13% -0.351346 0.725868 +2 -0.07% -1, +4 0.02% -0.075979 0.939556 +3 -0.03% -1, +5 0.04% -0.047833 0.961929 +4 0.15% 0, +1 -0.10% -0.358813 0.720188 +5 0.02%

(50)

6.3 Small cap companies vs. large cap companies

Companies in this thesis were divided into two groups according to their market value at the event day. These results deviate from results discussed above. In the small cap companies (table 7) group, that consists 124 companies, all the CAR values are positive, but it has negative t-values. This shows that there was some negative movement in the stock return, but the results are more or less ambiguous. Daily AR table (table 8) displays the negative and positive movements, but these are also scattered since the returns are positive and negative in turns.

Results for small cap companies

-1, 0 0.60% 0.328002 0.743195 -1 0.40% -1, +1 0.55% 0.235235 0.814236 0 0.20% -1, +2 0.32% -0.122032 0.902981 +1 -0.05% -1, +3 0.51% 0.161650 0.871730 +2 -0.24% -1, +4 0.08% -0.326417 0.744506 +3 0.20% -1, +5 0.11% -0.293680 0.769362 +4 -0.43% 0, +1 0.16% -0.064070 0.948968 +5 0.02%

In the large cap company group there is 184 companies and the results of the test are shown in table 9 and 10. These results are more consistent with overall positive CSR news (see table 1 and 2) than small cap group’s results. All the t-values are negative which suggests that the stock return trend was negative. Also the AR (table 10) turns negative on event day and the CAR turns negative forward from event day. Even though there are a positive CAR values on interval (-1, +4 and -1, 0) and positive AR on days (-1, +2 and +4) the results are mainly negative.

Results for large cap companies

-1, 0 0.05% -0.534231 0.593530 -1 0.21%

-1, +1 -0.10% -0.941294 0.347290 0 -0.16%

-1, +2 -0.05% -0.733259 0.463995 +1 -0.15%

(51)

Differences between small cap and large cap groups are quite distinct. Even though there are no statistically significant CAR values some observations can be made from the results. Small cap group has more positive results than large cap group. Also the event seems to have more instant and bigger impact on the stock return on large cap companies than small cap. Therefore it can be concluded that large cap companies experienced more obvious negative stock return, which is consistent the results with positive CSR in general. From other point of view small cap companies were not influenced by the CSR news publication as large cap companies or the total data sample.

In chapter 2, size anomaly was shown to influence the performance in similar manner as our findings, but not in CSR context. Small cap group has the most distinct difference in results to any other sub-sample group (see tables 3-6 and 9-10) or the main research (see tables 1-2). The small cap group has the most positive CAR results. However only speculations can be made that the size anomaly would be the case here since there is no statistical significance in small cap results (table 7). Therefore the noticeable difference in the trend of CAR movement towards positive in small cap group is useful for making assumptions, but not to generalize the results.

6.4 Testing hypotheses

Based on the relatively ambiguous previous research concerning the influence of CSR news publication on stock price, two hypotheses were produced. According to previous research, SR companies’ stock return is on par or better when comparing to companies that are not SR. Since all the events in this thesis are categorized as positive CSR news the hypotheses expect the SR companies to have equal or better return compared to market return.

(52)

returns in table 1. Most importantly, it was observed that the CAR and AR turned negative on event day. Interval (0, +1) has the highest negative CAR while the result for interval (-1, 0) is positive. Table 2 displays the largest decreasing AR values on event day and the day after. CAR values did not show any statistically significant values and therefore these remarks are made on the basis on general trend in the results. The results and general remarks deviate from previous literature based initial assumptions.

H2: Positive CSR news has a positive influence on the stock return

H2 is more defined hypothesis compared to H1, but as stated above the results deviated from expected results. H2 expects companies that released a positive CSR news report to have higher stock return compared to market return. Therefore H2 is rejected on the basis of the results. By having all the CAR values negative (see table 1) after the event day the publication of positive CSR news appear to have an adverse influence on the perception of the investors of CSR news. Nevertheless, the results (table 1) were not statistically significantly negative, but noticeable amount so that the H2 is rejected.

Due to lack of previous literature on the topic of CSR’s influence on stock return and more specifically on stock return based on type of CSR news or market capitalization no hypotheses were produced for samples. In general the results of both sub-samples did not provide proof of statistically significant CARs.

(53)

7 CONCLUSION

As presented in chapter 1, the goal of this thesis was to find answer if publication of a company’s positive CSR activities influence the stock price. In view of the findings from the empirical results in chapter 6 and the conceptual approach adopted, there were indications that positive CSR news publication decreased company’s stock price.

Complexities of the CSR, as academic researchers have stated it also, are that CSR include a wide variation of different issues and these issues all have variation of degrees of importance. In addition many of the CSR issues are on voluntary basis and there is no general international legislation that would cover and control all the huge amount of different SR matters. In international market place there are differences between local, national and regional SR policies and legislation, due to various reasons. Therefore a clearly determined and general definition of CSR issues are not produced which makes this issue more intangible and complex to understand and manage.

(54)

Among the scarce amount of comparable previous studies there are ambiguous results of CSR’s influence on stock price. In this study there is a trend of negative stock return after the event. Interestingly the sharpest negative turn occurs on event day or day after it. Therefore in semi-strong form efficient market, as in this study market is assumed to possess, all the public information is reflected directly to the stock price. Also it can be argued that the selected event might have an influence on the stock price due to the general negative trend in the main test and most of the sub-samples. Nevertheless there were no statistically significant results found, which is why the generalization of the results is not possible, so the results can be taken as indicative. Another interesting deviation in the results was that small cap group had distinctly more positive CAR values (see table 7) on tested intervals when comparing to other sub-sample groups (see tables 3, 5, and 9) or the main test (see table 1).

This study was conducted using functional amount of data sample in regards to event study method. When dividing the sample to sub-samples the amount of data per group decreased, but was at tolerable level. Since there is a lack of definition of CSR issues studied in regards of stock return to date, dividing of the data into sub-samples according to their characteristics should be done in transparent and consistent manner. In this study, two hypotheses were made on the basis of previous literature. The first hypothesis stated that positive CSR news should not affect the stock performance. This is hypothesis was not rejected, despite a consistent negative trend in the findings. According to the second hypothesis, the positive CSR news should influence the stock performance positively. This hypothesis was not supported by the observations and it was therefore rejected. As a matter of fact, the studied firms underperformed in comparison with the market index.

(55)

be that Friedman’s (1970) argument holds true and investors think that companies’ task is to maximize shareholder value, not CSR. This would defend the argument that CSR activities are not part of maximizing shareholder value, but rather unbeneficial expenses or worthless activities. Friedman’s (1970) argument would also support the ambiguous previous results because some shareholders might be either for or against CSR. With these uncertainties the future research should be extensive in order to be able to find out conclusive results that could be generalized.

Future research should focus on creating a consistent and clear way of defining the studied CSR event and companies. By creating quality research in this field would assist companies in CSR issues influence on stock return, but also the investors in case there is a systematical stock return trend found. The case of solving the influence of positive CSR news to stock market should be studied further in order to gain statistically significant results. Also the negative CSR news could be investigated to get an overall knowledge in this field of science.

The small capitalization sub-sample had quite different test results comparing to other tests. This is something that could be investigated in the future with larger data sample in order to get large groups in all levels of market capitalization.

(56)

REFERENCES:

Association for Sustainable and Responsible Investment in Asia (2008). http://www.asria.org

Banz, R. (1981). The Relationship Between Return and Market Value of Common Stocks. Journal of Financial Economics 9:1, 3–18.

Basu S. (1983). The Relationship between Earnings' Yield, Market Value and Return for NYSE Common Stocks: Further Evidence. Journal of Financial Economics. 12:1, 129–56.

Becchetti, L. & R. Ciciretti (2006). Corporate Social Responsibility and Stock Market Performance. Centre for International Studies on Economic Growth, Research Paper Series. 27:79, 1–30.

Becchetti, L., R. Ciciretti & I. Hasan (2007). Corporate Social Responsibility and Shareholder’s Value: An Event Study Analysis. Federal Reserve Bank of Atlanta. 1– 33.

Blacconiere, W. G. & D. Patten (1994). Environmental Disclosures, Regulatory Costs, and Changes In Firm Value. Journal of Accounting and Economics. 18, 357– 377.

Bloomberg (2008). http://www.bloomberg.com/apps/quote?ticker=NYA:IND

Brigham, E. & J. Houston (2004). Fundamentals of financial management. Concise 4th edition. South-Western Publishing. ISBN: 0-324-25872-0.

(57)

Carroll, A. B. (1979). A Three-Dimensional Conceptual Model of Corporate Performance. The Academy of Management review. 4, 497–505.

Canadian Association for Socially Responsible Investing (2008). http://www.socialinvestment.ca

Chen, N.F. (1983). Some Empirical Tests of the Theory of Arbitrage Pricing. Journal of Finance. 38:5, 1393–1414.

Committee for Economic Development (2003). Reducing Global Poverty: Engaging

the Global Enterprise. 1–34.

http://www.ced.org/docs/report/report_globalization_enterprise.pdf

Copeland, T. & F. Weston (1988). Financial Theory and Corporate Policy. 3rd edition. Addison-Wesley Publishing. ISBN: 0-201-10648-5.

Copeland, T., F. Weston and K. Shastri (2005). Financial Theory and Corporate Policy. 4th edition. Pearson Education Inc. ISBN: 0-321-22353-5.

Corporate Social Responsibility Monitor (2008). http://www.bsdglobal.com

Deegan, C. (2002). The Legitimising Effect of Social and Environmental Disclosures – A Theoretical Foundation. Accounting, Auditing & Accountability Journal. 15:3, 282–311.

Deegan, C., M. Rankin & J. Tobin (2002). An Examination of the Corporate Social and Environmental Disclosure of BHP from 1983-1997: A Test of Legitimacy Theory. Accounting, Auditing & Accountability Journal. 15:3, 312–343.

Positive Corporate Social Responsibility news influence on stock return: