Doing good or buying cheap? : innovative auction-based research into consumer social (ir)responsibility

Doing good or buying cheap?

Innovative auction-based research into consumer social (ir)responsibility

Student: Fanar Al-Obaidy Student Number: 10003544

Institution: University of Amsterdam Faculty: Amsterdam Business School Supervisor:

Second reviewer:

Lars Moratis Mark van der Veen Due date: The 1st of July, 2014

Abstract

This paper examines to what extent consumers will actually pay more for a product if it was produced in a socially responsible manner. This issue has been addressed by countless researchers before. However, these researchers mostly used surveys and small scale controlled experiments to answer this question. Especially in the field of ethical decision making those techniques can be prone to social desirability answering bias by respondents. To address this issue we conducted an online real life experiment on a large scale that measured consumer willingness to pay for CSR. The real life setting guaranteed the measurement of actual behavior and ruled out the possibility of social desirability bias. This innovative method led us to major new insights about the consumer’s willingness to pay for CSR. Furthermore it gave us the ability to prove the presence of the attitude-behaviour gap in the area of ethics.

F

OREWORD

Ethics, one of the most difficult area’s that can be studied in science. The attitude-behavior gap in ethics research is just as infamous as difficult to overcome. Studies in the past (and history itself) have proven that the human race is not as well natured as many people want to believe it is. The Stanford prison experiment is a good example of this (Haney et al, 1973). This experiment, where regular people took on the roles of prisoners and guards in a prison, showed how cruel human nature can be in certain situations. What makes this study so special, valid and reliable is the fact that it was a controlled experiment. By conducting this type of experiment actual behavior was measured in causal relationships and not just attitudes or opinions (like with surveys or interviews). Had a survey been conducted, the results would probably not be the same (and less realistic). After all, who would admit in a survey to being capable of torturing another human being?

In my opinion this type of controlled experiment is the only reliable and valid way of doing research in business ethics. This is why I decided to conduct a real life controlled experiment to measure consumer willingness to pay for CSR. Of course this would be impossible without the help of professors Bram Kuiken and Anouar El Haji from the University of Amsterdam. I am very thankful for the chance they gave me to utilize the Veylinx tool to conduct my experiment. Furthermore I also want to thank Anouar El Haji for helping me with the data collection and analysis. I would also like to thank the UvA for making the great and innovative Veylinx tool available for students.

Last but not least, I would like to thank my supervisor Lars Moratis for supporting me and providing information and advice to me throughout the process of writing my thesis. His patience, knowledge and understanding were essential in the past couple of months.

C

ONTENTS

Foreword ... 2

1. Introduction ... 4

2. Theoretical background ... 6

2.1 About CSR ... 6

2.2 Socially Responsible Consumer Behavior (SRCB) ... 8

2.3 The attitude-behaviour gap ... 14

2.4 Consumer skepticism towards CSR ... 16

2.5 Increasing CSR credibility towards consumers ... 18

2.5.1 Credibility by transparency ... 18

2.5.2 Credibility by association or partnering ... 20

2.5.3 Cause-related CSR ... 21 2.6 Overview of hypotheses... 23 3 Methodology ... 24 3.1 Overall approach ... 24 3.2 Experiment design ... 25 3.3 Auction treatments ... 25 3.4 Data analysis ... 27

3.5 Validity and reliability ... 27

4 Results ... 29

4.1 The data set ... 29

4.2 Main research question ... 30

4.3 Findings concerning the reliability moderators ... 30

4.4 Effects of the demographic and personality moderators ... 33

5 Discussion ... 34

5.1 Overall consumer willingness to pay for CSR ... 35

5.2 Credibility moderators of willingness to pay for CSR ... 36

5.3 Demographic and personality moderators of willingness to pay for CSR ... 37

6. Conclusions and implications ... 38

6.1 Academic implications ... 38

6.2 practical implications ... 39

6.3 limitations and suggestions for future research... 40

References ... 43

1. I

NTRODUCTION

“Actions that appear to further some social good, beyond the interests of the firm and that which is required by law”.

This is how McWilliams and Siegel (2001) defined Corporate Social Responsibility. CSR is nowadays considered a mainstream part of organizational decision making. A recent study showed that more than sixty percent of firms have CSR strategies (Oekom, 2013). Many studies have investigated this widespread phenomenon of businesses claiming to care about their societal responsibilities. In the past decade multiple researchers tried to create a standard framework for CSR design and

implementation (Werre, 2003; Maignan et al., 2005; Maon et al., 2009; Asif et al., 2011).

Many firms however keep struggling with the strong level of skepticism towards CSR initiatives by the different stakeholders (Bernstein, 2009). A lot of research has therefore been done on skepticism towards CSR and ways to increase the credibility of CSR initiatives. This has led to a number of considerations that should be made when implementing CSR strategies. The International

Organization of Standardization, which tries to promote global standards for industry, commerce and propriety, is one of the bodies that integrated the management of CSR credibility in their ISO 26000 guidelines (ISO, 2010).

Ha-Brookshire and Hodges (2009) studied CSR from a consumer perspective. They looked at

“Socially Responsible Consumer Behavior” (SRCB). Mohr et al. (2001) had defined this as the extent to which a consumer bases his or her purchase, consumption and disposal decisions on the desire to minimize harmful impacts and maximize beneficial effects to society and the environment. Previous

research has suggested that on average consumers are willing to pay more for socially responsible products (e.g. Ha-Brookshire and Norum, 2011). However, this research measured attitudes, not actual behavior. Ha-Brookshire and Norum (2011) for example conducted a survey by phone, asking participants if they were willing to pay a premium for socially responsible products. More than half the

respondents indicated that they were willing to pay more for these products. The reliability of this type of research can be questioned as there is a major discrepancy between consumer attitude and behavior. Especially when it comes to ethical or socially responsible actions there tends to be an

attitude-behavior gap. This gap is the result of people’s tendency to give socially desirable answers (De

Pelsmacker et al, 2005). The consequence of this attitude-behavior gap is that on average people tend to portray their attitude in a more positive manner than their actual behavior suggests.

This study will try to find a causal relationship between the presence of CSR initiatives and the consumer willingness to pay for the product. We shall take moderating demographic and personal factors of the consumers into consideration in our study. Furthermore we will research if credibility enhancement tactics by firms can moderate this causal relationship. Our research question will be: Does a causal relationship exist between CSR communications in the product ad and consumer

willingness to pay? And if so, what are the moderators of this relationship? We will try to answer this

question by conducting an experiment that measures consumer willingness to pay.

This article will start in chapter 2 by providing an overview of the existing literature concerning Corporate Social Responsibility (CSR) and Socially Responsible Consumer Behavior (SRCB). An

extensive methodology section can be found in chapter 3. The results of this research will be shown in chapter 4 and discussed in chapter 5. Finally, in chapter 6 a summary can be found of the conclusions and implications of this research.

2. T

HEORETICAL BACKGROUND

2.1 A

CSR

Corporate social responsibility has been discussed in the academic literature since the 1950’s (De Bakker et al, 2005). During this long period however there was never one prevailing definition. Existing definitions vary from the loosely defined “actions that appear to further some social good, beyond the

interests of the firm and that which is required by law” (McWilliams and Siegel, 2001) to “The social responsibility of business encompasses the economic, legal, ethical, and discretionary expectations that society has of organizations at a given point in time” (Caroll, 1979) and “societal expectations of corporate behavior; a behavior that is alleged by a stakeholder to be expected by society or morally required and is therefore justifiably demanded of a business” (Whetten et al, 2002).

Looking closely at these three definitions of the same phenomenon you can conclude that there are more differences than similarities. McWilliams and Siegel (2001) appear to claim that CSR practices are “beyond the interests of the firm” while the definitions of Caroll (1979) and Whetten et al (2002) explicitly state that those practices are expected by society and thus not beyond the interests of the firm. Whetten et al (2002) even go so far as to state that those actions can not only be expected but also justifiably demanded. One cannot consider all three definitions as equally right, as there is a clear difference in the motivation of CSR that they present (Own initiative vs. expected by society). Whetten et al (2002) also state that the stakeholders are the link between society and the business. This introduction of a mediator has significant consequences for CSR literature and the business. In essence it means that businesses should focus on the demands of stakeholders and not the entire society, giving the stakeholder a spokesman role. Sen and Bhattacharya (2001) state that consumers have a direct influence on firm decisions in their role of primary stakeholders (through purchasing decisions).

The definition used for the remainder of this paper shall include the role of the stakeholders and therefore commit to stakeholder theory. Stakeholder theory states that businesses have

obligations to shareholders as well as to other interest groups. Those interest groups can include customers, employees, suppliers or the community as a whole (Carroll, 1991; Freeman, 1984; Jamali, 2008; Mitchell et al., 1997). Our definition for CSR was originally derived from Gray et al (1996) and reframed to its current form by Maon et al (2009): “CSR is a stakeholder-oriented concept that extends

beyond the organization's boundaries and is driven by an ethical understanding of the organization's responsibility for the impact of its business activities, thus, seeking in return society's acceptance of the legitimacy of the organization”. To make this abstract concept more concrete we can use the

explanation given by Bhattacharya et al (2009) who explain CSR initiatives as corporate supported actions intended to improve the welfare of society. This corporate support can be in the form of money, labor or other corporate resources.

Many authors find it difficult to define ethical behaviour, ethical retailing and ethical consumption (Singhapakdi et al, 1999; Cherrier, 2005; Clavin and Lewis, 2005). However, despite this ongoing debate about ethics, a number of common issues can be found in ethics literature. The most common ones are:

- Usage of organically grown and processed materials (Shaw et al, 2006; Tsakiridou et al., 2008). - Labor and production practices in developing nations (Anniss, 2003; Joergens, 2006)

- Depleting earth’s natural resources (Sanfilippo, 2007)

In the existing literature a clear distinction can be made between the authors that focus on solving incidental CSR issues (e.g. Macagno, 2013) and the authors that view CSR as a continuous and integrated part of overall corporate strategy (e.g. Asif et al, 2011). Roughly speaking the distinction can be made between the reactive and the proactive streams of CSR, the former focusing on fixing existing issues and the latter trying to prevent the aforementioned issues.

Despite all the different definitions and approaches of CSR one can say that the academic literature nowadays agrees upon the necessity of CSR initiatives by organizations. The main reason for

this unanimity lies in stakeholder theory. In chapter 2.2 the phenomenon of CSR shall be described from the consumer point of view (Socially Responsible Consumer Behaviour).

2.2 S

R

C

B

(SRCB)

A lot of research has been done in the past decades on CSR from a strategic business perspective. However, the attention to CSR from a consumer behavior perspective is a relatively new phenomenon. Mohr et al. (2001) defined SRCB (socially responsible consumer behavior) as the extent to which a consumer bases his or her purchase, consumption and disposal decisions on the desire to minimize harmful impacts and maximize beneficial effects to society and the environment. The major difference between CSR and SRCB is therefore that SRCB studies CSR from a consumer perspective.

A survey study in the United States in 1999 (Cone Communications Press Release) showed that about eighty percent of those surveyed have a more positive attitude towards a company that supports CSR activities. Smith and Alcorn (1991) had already concluded that forty-nine percent of their survey sample took CSR into consideration into their purchase decision making. Ross, Patterson and Stutts (1992) concluded that customers are willing to pay more for products with a cause related CSR initiative. In a more recent study Maignan (2001) found that German and French consumers partly base their purchase decisions on the social responsibility of businesses. Maignan and Ferrell (2003) came to the same conclusion as Maignan had in 2001. They also found that European consumers are significantly more willing to actively support businesses that take socially responsible actions. Recently Ha-Brookshire and Norum (2011) conducted a phone survey to measure consumer willingness to pay for socially responsible products. They found that more than half of their respondents were willing to pay more for a t-shirt if it was produced organically and in a sustainable manner (on average five dollars more for a shirt with thirty dollar retail value).

Not all researchers in the area of SRCB used surveys and interviews. Experiments were also used to determine customer attitude towards CSR and the extent to which customers base their

purchase decisions on the presence of CSR initiatives. However, the results with this method were mixed. An experiment conducted by Holmes and Kilbane (1993) suggested that customers have a more positive attitude towards companies which promise a donation to charity, but this did not affect their purchase intention. Lafferty and Goldsmith (1999) experimented by showing test subjects positive or negative newspaper articles about a firm’s social contribution.

They found that the test subjects had a more positive attitude towards the ad and brand if they were shown the positive newspaper article. Furthermore this positive attitude also had a positive effect on their purchase decision. Folkes and Kamins (1999) did a similar experiment and had similar findings as Lafferty and Goldsmith. Their research findings even suggested that consumers found ethical information more important than some important product attributes. They tested this by showing the participants a hypothetical opinion of a friend about a phone brand. The participants valued the ethical information in the opinion even more highly than the sound quality, which is an important product attribute for a phone.

Creyer And Ross (1997) conducted a study that showed that consumers are prepared to pay higher prices if ethical behavior is evident and lower prices if unethical behavior was seen.

Our first hypothesis will be:

H1: Consumers are willing to pay more for a product if CSR is communicated in

the product ad.

FIGURE 1. HYPOTHESIS 1

The existing literature also names demographic factors as potential moderators of sensitivity to CSR campaigns. The factors that have been examined in previous studies include gender, age and level of education. Since the 1970’s researcher have tried to identify the socially conscious consumer based on

demographic factors. Anderson and Cunningham (1972) and Berkowitz and Lutterman (1968), studied the profile of the socially responsible consumers. Their research portrays a highly socially conscious person as an educated pre-middle aged female. Since the 1970’s many researchers tried to find evidence for this profile. However, the results of these studies were inconclusive or contradictory to each other. Recent studies in the area of environment and ecology showed that women tend to be more conscious of sustainability and campaigns addressing this issue (McIntyre et al, 1993; Banerjee & McKeage, 1994; Parker, 2002). However, this is contradicted by other studies that show that men are more aware of environmental issues and that men are more sensitive towards campaigns in this topic (Reizenstein et al, 1974; Balderjahn, 1988). Therefore, we shall put this issue to the test in our

experiment. We hypothesize the following:

H2: Women are more sensitive to CSR communications in product ads than men.

FIGURE 2. HYPOTHESIS 2

The same inconclusiveness that exists for the gender issue also presents itself for other demographic factors such as age. Research in the 1970’s showed the socially conscious consumer as younger than average (Berkowitz and Lutterman, 1968). Otmann (1993) stated that the socially conscious consumer is likely to be between the ages of 30 to 49. Other research contradicted this by profiling the socially

conscious consumer as being older than average (Samdahl and Robertson, 1989; Vining & Ebreo, 1990; Hines & Ames, 2000). About the age of the socially conscious consumer we hypothesize:

H3: Consumers between the ages of 30 to 49 are more sensitive to CSR

communications in product ads than consumers in other age categories.

FIGURE 3. HYPOTHESIS 3

The third demographic trait that has been hypothesized to be of influence on socially responsible consumer behavior is whether the consumer has children or not (Otmann, 1993; Laroche et al, 2001). Therefore, we shall take this demographic factor into consideration as well. We hypothesize:

H4: Consumers with one or more children are more sensitive to CSR

FIGURE 4. HYPOTHESIS 4

Level of education is also one of those demographic factors that researchers have been trying to link to the socially conscious consumer profile. However, just like with gender and age, the link was never shown clearly in scientific research. Many researchers (e.g. Van Liere and Dunlap, 1981) found a positive correlation between social consciousness and the level of education. Others like Samdahl and Robertson (1989) found a negative correlation. Sixteen years later Dickson (2005) also found that consumers with lower education are more sensitive to social issues. This contradictory evidence brought most researchers to conclude that level of education is not a good predictor of socially conscious purchase behavior. Furthermore, it can be said that earlier research was prone to social desirability bias due to its survey methods. For our own research we shall hypothesize as follows:

H5: Consumers with high education (college/university) are more sensitive to

FIGURE 5. HYPOTHESIS 5

Aside from the abovementioned demographic factors, there is one more important factor that according to some studies can moderate the relationship between the presence of CSR

communications in the product ad and the consumer willingness to pay for the product. Suchard and Polonski (1991) state that consumers that already show socially conscious behaviour in some of their actions may also show this tendency for other actions (e.g. recycling and purchasing green products). However, other studies show that socially conscious behavior in other situations may not be a good predictor for socially conscious behavior in the situation at hand (e.g. Pickett et al, 1993). We hypothesize:

H6: Consumers that donate to charity are more sensitive to CSR communications

FIGURE 6. HYPOTHESIS 6

2.3 T

-

The main issue with most previous research into socially responsible consumer behaviour is the reliability and validity of the research methods. Survey data typically suffers from tendency of participants to give socially desired answers. Respondents may want to portray their attitudes and behaviors in a more positive manner than reality. This creates a significant bias in the research results (De Pelsmacker et al, 2005). The cost of answering questions is lower than the cost of actual behavior and therefore surveys tend to overestimate the effect of CSR on product decisions.

A study conducted by Cowe and Williams in 2000 showed that nearly thirty percent of British consumers considered themselves “ethical purchasers”. However, ethically accredited products like Fair Trade only achieved a three percent market share in the UK. Cowe and Williams (2000) named this the “30:3 phenomenon” considering about thirty percent of the consumers claimed to care about ethical purchasing and only three percent actually behaved in such a manner. Other authors also noted this phenomenon under various names such as “the attitude-behaviour gap” (Kim et al, 1997) and the “ethical purchasing gap” (Nicholls and Lee, 2006).

Bray et al (2011) made an attempt to investigate more reasons for this attitude behaviour gap, apart from the well-known social desirability answering bias. They found a number of causes that

could explain why consumers in the end did not buy the ethically responsible products. Price sensitivity topped the list with reasons for not purchasing socially responsible products. Participants admitted that in the end they cared more for financial values than ethical values. Other reasons that surfaced in this study were: (1) “I can’t make a difference on my own”, (2) The lack of information about the topic, (3) inertia and (4) cynicism towards CSR.

Another author known for his work in this field is Icek Ajzen. In 1985 he was the first to introduce the Theory of Planned Behaviour in an article that has become the standard in this research area. The Theory of Planned Behaviour (Ajzen, 1991) offers a theoretical framework that can be utilized to study the influence of attitudes, personal determinants and volitional control on a consumer’s intent to buy socially responsible products.In the 29 years since its introduction, the TPB model has been used in a wide variety of studies ranging from physical activities to quitting smoking and blood donation. A review by East (1997) indicated that the model had provided solid estimates in those studies.

The TPB model is presented in figure 7. In this figure the three main determinants can be seen that lead to intention and eventually to behaviour. The first determinant is the attitude to behaviour (AB) which implies the extent to which the planned behaviour is expected to lead to the desired outcomes. The second determinant is that of the Subjective Norms (SN), which takes into account the perceived opinion of external parties on the planned behaviour and the consumer’s tendency to conform to this external pressure in his or her behaviour (acting like others think you should act). The third determinant in the TPB model is the Perceived Behavioral Control (PBC) which considers the extent to which a consumer believes that he or she can conduct the planned behaviour effectively. In short, this model states that actual behaviour depends on the expectancy that the behaviour will lead to the desired results, peer pressure/sensitivity to this pressure and the extent to which a consumer beliefs he is capable of performing the planned behaviour.

FIGURE 7. THE THEORY OF PLANNED BEHAVIOUR (AJZEN, 1985, 1991)

2.4 C

CSR

Organizations all over the world are increasingly interested in adopting CSR initiatives in their strategy. MIT Sloan and the Boston Consulting Group recently published a report showing that in the period 2002-2012 more than fifty percent of the surveyed organizations changed their business practices to include CSR initiatives (MIT Sloan & BCG, 2013). This report underlines the significance and importance of CSR in modern day business practices. “Doing well by doing good” is no longer an idea applied by a few companies. It has become a mainstream, visible and common practice

(Skarmeas & Leonidou, 2012).

Despite the increasing attention by companies for social responsibility, consumer skepticism toward corporate responsible efforts remains high (Bernstein, 2009). This is mainly caused by the continuing occurrence of incidents of socially irresponsible behavior by businesses (Carson, 2003; Wagner et al., 2009). These incidents (e.g. fraud, environmental disasters, scams, corporate misconduct) attract a high level of media interest and are therefore highly visible. Consumers often find it difficult to make the distinction between the socially responsible and irresponsible businesses (Bernstein, 2009; Parguel et al, 2011). It should also be taken into consideration that negative

information is more diagnostic and weighs more to consumers than positive information (Baumeister et al., 2001). Based on this previous literature it can be hypothesized that consumers that watch the news on a daily basis will be more skeptic towards CSR initiatives by businesses. Therefore they will

have a more negative view of CSR initiatives in product ads than their counterparts that do not watch the news on a daily basis. We will test the following hypothesis:

H7: Consumers that watch the news on a daily basis are less sensitive to CSR

communications in product ads than people that do not watch the news on a daily basis.

FIGURE 8. HYPOTHESIS 7

Skepticism is defined as “a person’s tendency to doubt, disbelieve and question” (Boush et al., 1994; Forehand & Grier, 2003). The word comes from the Greek “skeptomai” which means to think, examine and consider. The phenomenon skepticism is studied by multiple disciplines like psychology (e.g., Forgas & East, 2008); philosophy (e.g., McGrath, 2011) and politics (e.g., Taber & Lodge, 2006). Skepticism leads to a number of negative customer responses towards businesses and their actions such as distrust (Darke & Ritchie, 2007); suspicion (Ferguson et al., 2011); cynicism (Chyllinsky & Chu, 2010); outrage (Lindenmeier et al., 2012) and boycott (Klein et al., 2004). Some researchers view skepticism as a personality trait, a continuous state of distrust and suspicion that cannot be influenced (Boush et al., 1994). However, most studies focus on situational skepticism, a consumer state, partly created by personality traits but also influenced by context and situation (Forehand & Grier, 2003; Mohr et al., 1998; Vanhamme & Grobben, 2009). Skeptical individuals consider and analyze the information given to them thoroughly, therefore they can also change their minds if enough evidence

is given to proof them wrong (Mohr et al., 1998). In chapter 2.5 a summary will be given on the strategies which can be used to increase CSR credibility, according to ISO 26000 (the most recent global standard for CSR).

2.5 I

CSR

The International Organization for Standardization (ISO) -an international body aimed at promoting global proprietary, commercial and industrial standards- has created its most recent standard for the implementation of CSR. A significant difference between ISO 26000 and other CSR standards is that it also discusses CSR credibility and provides suggestions for firms as how to increase this credibility (Moratis, 2013). The ISO 26000 and a number of other sources shall be used to develop a set of CSR credibility enhancement tactics which could influence the consumer willingness to pay for CSR. The credibility enhancement tactics mentioned in ISO 26000 and other literature cover the entire CSR design, implementation and communication process. However, for this research we are only interested in credibility tactics towards consumers at the point of purchase and therefore we shall only focus on marketing related suggestions. This also means that we shall not discuss CSR credibility towards other internal and external stakeholders.

2.5.1 C

A suggestion mentioned in ISO 26000 to increase CSR credibility is credibility by transparency. Specifically mentioned are the following types of transparency: Raising internal and external

awareness, showing how the organization is meeting its commitments, providing information about the organization’s impact, facilitating comparison to peer organizations and enhancing an

organization’s reputation (ISO, 2010: 76; Moratis, 2013). The implementation model of Maon et al. (2009) mentions “creating internal awareness” in the first stages of the process. They argue that this “sensitizing” process is necessary to convince management of the importance of sustainability issues. Furthermore it will give management the chance to find and overcome any resistance towards the

change. For transparency towards the external stakeholders it is very important to consider signaling theory that states that a communication chain consists of three parts: The signaler, the signal and the receiver (Connelly et al., 2011; Lester et al., 2006; Spence, 2002). The essence of signaling theory is an asymmetry in the information level that two or more parties possess about a topic. To reduce this asymmetry the party that has more knowledge (signaler) should communicate (signal) with the other party (receiver). The success of this process is defined by a number of factors such as signal strength, visibility, intensity, clarity, fit and consistency (Connelly et al., 2011). The ability to effectively

communicate with stakeholders is essential to the success of a company’s CSR efforts (Du et al., 2010).

H8: Consumers are willing to pay more for a product when it is explained to

them in the product ad how the CSR initiative is carried out, compared to just a basic mentioning of the CSR initiative in the ad.

FIGURE 9. HYPOTHESIS 8

The credibility effect of transparency can be increased by letting an external body verify the CSR claims made by the organization. This gives a perception of credibility and trustworthiness (Maon et al., 2009). Independent bodies like NGO’s and accountants are more trustworthy to the public eye when it comes to the verification and communication of CSR performance by businesses. They can take away part of the natural skepticism that consumers have towards businesses in general (Boush et

al., 1994; Forehand & Grier, 2003). The implementation of CSR should be audited on a continuous basis. Maon et al. (2009) state: “The value of such audits increases if the process appears rigorous,

which may be achieved by involving external auditors”. Firms should involve secondary stakeholders

that have no economic interest in the organization and its performance. Because they lack economic interest in the organization, these secondary stakeholders are seen as reliable reporters (Crane & Ruebottom, 2011) or reliable signalers (Connelly et al., 2011).

2.5.2 C

Also a suggestion that can be found in ISO 26000 to increase CSR credibility is credibility by

association or partnering (ISO, 2010: 79; Moratis, 2013). This involves firms voluntarily joining

organizations, initiatives, networks or other coalitions aiming for corporate social responsibility implementation. This can range from a simple agreement to an integrative partnership (Yuan, Bao & Verbeke, 2011). This way firms can promote their behavior and seek recognition of their efforts. This can significantly increase the credibility of the CSR initiatives and the firm as a whole (Deegan & Caroll, 1993; Mutch & Aitkin, 2009; Seitanidi & Crane, 2009). This is also known as the spillover effect of brand alliances on consumer brand attitudes. In the literature a brand alliance is defined as a short or long term association or combination of two or more products, brands or other distinctive proprietary assets (Rao and Ruekert, 1994). In the CSR context this would entail an alliance between a corporate body and an external organization or coalition (e.g. an NGO or a certification body). A corporation can create more positive brand associations by forming an alliance with an external party with a good reputation on one or more areas of expertise (Lafferty & Goldsmith, 2005; Rao & Ruekert, 1994; Simonin & Ruth, 1998; Washburn, Till, & Priluck, 2004). Taking this into account we hypothesize:

H9: CSR communications in product ads have a greater effect on willingness to

pay when they mention an external independent partner that is involved in the CSR initiative.

FIGURE 10. HYPOTHESIS 9

A notable finding considering the effect of certification is that it might actually reduce the practical focus on CSR because management will focus too much on meeting the minimum requirements for the certification and not so much on the primary goal of “doing good” (Tenbrunsel et al, 2000).

2.5.3 C

-

CSR

Cause-related CSR is not merely a credibility increasing add-on to CSR. It can be seen as a complete subtype of CSR. This type of CSR involves a promise made by a business that for every sold product a donation will be made to a worthy cause (Van de Brink et al., 2006). It was defined by Varadarajan & Menon (1988) as: “The process of formulating and implementing marketing activities

that are characterized by an offer from the firm to contribute a specified amount to a designated cause when customers engage in revenue-providing exchanges that satisfy organizational and individual objectives”. Nowadays CRM is one of the most-used types of CSR in the world (Van den Brink et al.,

2006). Even though some literary sources are available about this topic (e.g. Varadarajan & Menon, 1988), the academic coverage is very limited.

A Cause-related marketing (CRM) campaign has two objectives. The first one is to support a social cause and the second one is to improve marketing performance, e.g. to sell more products

1990’s CRM was the fastest growing type of marketing (Smith, 1994). While some studies have shown that CRM has a positive effect on customer attitude towards businesses (Sen and Bhattacharya, 2001), other studies did not always find a positive relationship between the presence of CRM and attitude towards businesses. Mohr et al. (2001) found that CRM had a significantly smaller impact on consumer behavioral intentions than regular CSR. Mohr et al. (2001) propose that a reason for this lack of effect can be the perception that businesses that use CRM are trying to exploit a cause for promotional activities. Barone et al. (2000) found that consumers were only prepared to accept minor price and product trade-offs for cause-related CSR. Strahilevitz and Myers (1998) showed that CRM campaigns are more effective in the promotion of hedonic products than practical products.

H10: Cause related CSR communications have a greater effect on willingness to

pay when shown in product ads, compared to regular CSR communications.

FIGURE 11. HYPOTHESIS 10

In chapter 2.6 an overview can be found of the hypotheses that were developed for this research. These hypotheses will be tested using a real-life auction experiment. More details about this experiment can be found in chapter 3 of this article.

2.6

O

3 M

ETHODOLOGY

3.1 O

For this study we decided to test the effect of CSR and the different credibility enhancement tactics on consumer willingness to pay. The aim is to see if there is a causal relationship between the presence of CSR in its different forms and the price that consumers are willing to pay for the product. A lot of research has been done before in this area. Previous studies mostly used surveys and in-depth interviews as a research method. However, these research methods only measure opinions, attitudes and behavioral intentions. As we also explained in chapter 2.3, surveys are prone to social desirability bias (De Pelsmacker et al, 2005). This implies that respondents may give (partly) false descriptions of their behavior to meet the social standard.

This difference between attitude and behavior is called the attitude behavior gap (Kim et al, 1997). Bray et al (2011) found that price sensitivity is one of the main causes that refrain consumers from buying ethically responsible products. An explanation as to why consumers do make the effort to claim social responsibility, but do not make the effort to act socially responsible, can be found in the fact that the cost of answering questions is lower than the cost of actual behavior (De Pelsmacker et al, 2005).

We do not wish to measure attitude or opinions but will solely focus on actual behavior. Most experiments that aim to measure willingness to pay end up measuring attitudes, opinions or intent. The reason for this is that most experiments take place in simulated environments and use

hypothetical situations and choices. For this reason we have decided to use a new revolutionary method of experimentation developed especially to measure the willingness to pay factor. The revolutionary method mentioned before is an auction site called Veylinx. Through a sophisticated scientific setup and a controlled online environment Veylinx can measure willingness to pay in real life situations, yet in a controlled environment.

3.2 E

For this experiment a scientific auction was conducted. This was done using the Veylinx auction tool. Veylinx has a large panel of members that receive a weekly invitation-mail to bid on a certain product. They can then make one bid on the item. The risk of using an auction for scientific reasons is the possibility of causing the “anchoring-effect” (Chapman & Johnson, 2002). The anchoring-effect can be defined as a psychological process which causes unrelated numbers to affect the consumer’s product evaluation and willingness to pay. To prevent the anchoring effect from playing a role in the auction some precautionary measures were taken. The bidders cannot see other people’s bids and they only have five minutes to place their bid, so they have no time to do research on the product. No price or any other indication of the product worth is mentioned in the product ad.

Furthermore, a relatively new and unknown product was chosen to be auctioned off (Princess Smoothie Maker) to minimize the existing knowledge available to respondents. In essence it comes down to the fact that the panel members have to bid the amount that they are willing to pay for the product based solely on the product information provided in the ad that is shown to them. If they are not willing to pay anything for the product they can place a zero as bid. The auction is not a simulation so the highest bidder will actually purchase the product.

The key factor that makes this an interesting experiment for researchers is that they can create multiple treatments for the bidders. Not all the panel members get the same offer. By manipulating some independent variables a lot of data can be gathered about the dependent variable “willingness

to pay” in different situations. To make the concept more attractive for panel members they only have

to pay the second-highest bid if they win the auction.

3.3 A

For this experiment we shall create five different treatments using the same product. We shall use the following five treatments (in a combination of text and visual cues). The actual ads can be found in the appendix of this paper:

1. Baseline treatment. The product offer without any mention of CSR (see appendix)

2. CSR basic treatment. The product offer that mentions a CSR initiative on a very basic level (see appendix)

3. CSR through partnership. The product offer that mentions a CSR initiative through a partnership with an external party like an NGO (see appendix)

4. Cause-related CSR. The product offer that mentions that a certain amount of each sale shall be invested in a CSR initiative (see appendix)

5. CSR transparent. The product offer that mentions a CSR initiative and also gives some more information about the mentioned CSR initiative (see appendix)

As explained earlier in this chapter each bidder gets to see one of these five treatments. They have five minutes to place their bid and to confirm their bid they also have to answer a set of five questions. The answers to these questions can form an extra source of information in the data analysis. The questions do not directly mention CSR, so that respondents will not know what the topic of the experiment is. This way the effect of socially desirable answering bias can be minimalized. These are the questions that each auction bidder will be asked after he or she places a bid.

- Do you regularly eat fruit? - Do you watch the news daily?

- Do you sometimes donate to charity? - Have you ever been outside of Europe? - Do you have children?

The abovementioned questions can be answered by selecting a simple yes or no. This way the effort for respondents is minimalized and unrealistic answers are prevented.

3.4 D

The data gathered during the course of the experiment shall be analyzed and the findings shall be shared in the findings section of this paper. Most data will be numerical and therefore quantitative. Because this is an experiment and not a survey, no sophisticated calculations and formulas are needed. To prove a causal relationship between an independent and a dependent variable only a mean needs to be calculated. This will be done using SPSS software and the specialists of Veylinx will provide the necessary assistance for the data analysis. The means of the different treatments shall be compared to see what effects the manipulations have on the dependent factor willingness to pay. Furthermore we will also take a closer look at other factors which could influence the willingness to

pay like gender and age.

3.5 V

The basis shall be the same for each of the five treatments. The product, product-specifications, online environment and timing shall be exactly the same for all five treatments. The only differences between the five treatments are the manipulations (independent variables) mentioned in the treatments. This way any differences that are measured can be attributed almost completely to the manipulations without the risk of contamination or bias. Participants have limited prior knowledge about the product they will bid on. This, combined with the fact that participants only have five minutes of decision time from the first moment that they see the ad, reduces the risk of contamination bias.

Compared to the traditional experiment methods Veylinx can perform better in the area of validity. Because of the real-life component and the fact that it’s not a simulation invalidity is reduced to a minimum. We do not wish to measure attitudes, opinions or intentions. This study is merely interested in actual behavior and this innovative method can measure just that. Taking into account the combination of a real-life environment and a controlled online environment, the phenomenon of socially desirable answers can be ruled out.

Reliability in scientific research is also achieved through numbers. A common problem that experiment designers face is the limited number of test subjects. Due to their expensive nature experiments are often limited to small numbers of participants. Because of the online environment, the vast panel and the easy accessibility it is expected that we can find a significant number of participants for this experiment. Any findings derived from the data will therefore be deemed as reliable and scientifically credible.

4 R

ESULTS

4.1 T

The experiment was completed successfully by 854 participants (N=854). For the data analysis we solely used the data of the participants that had completed the experiment. The average age of the sample was 42 years (mode 24; standard deviation 16,142; minimum 16; maximum 76). The sample consisted out of 468 males (54.8%) and 386 females (45.2%). Each of the 854 participants was shown only one of the five available treatments. So each treatment was seen by about 20 percent of the sample.

Before any statistical tests could be conducted, the normality of the dataset had to be

determined. This is a necessary step because parametric statistical tests assume a normal distribution in the data they analyze. Using parametric tests (on a normal distribution) increases the chances of significant findings. The data was found to be not normally distributed. An algorithm was used to transform the dataset into a relatively normal distribution that SPSS would be able to analyze better.

A One-Way ANOVA test was utilized to compare the bid amount means of the multiple

treatments (Hypotheses 1, 8, 9 and 10). Further analysis was performed using Fisher’s Least Significant Difference test (LSD). The original table with the mean differences between all treatments can be found in the appendix. To test hypotheses 2, 3, 4, 5, 6 and 7 the readily available demographic data about the respondents was used. By using the “select cases” option in SPSS sub-groups of respondents were created to test if they provided different results in willingness to pay compared to the general sample. (See appendix for corresponding advertisements and relevant tables).

4.2 M

H

: CSR communications -> consumer willingness to pay

At first sight there seemed to be a small positive mean difference of .061 between treatment 2 (a basic CSR communication in the product ad) and treatment 1 (baseline).

FIGURE 14. COMPARISON TREATMENT 2 TO 1

However, this mean difference was found to be not significant (.310). Thus, it seems that CSR communications in product ads do not influence consumer willingness to pay for the product. No support is found for hypothesis 1 and therefore it can be rejected.

4.3 F

H

: Transparent CSR communications -> Consumer willingness to pay

No causal relationship was found between transparent CSR communications in the product ad and consumer willingness to pay for the product. This was tested by comparing the mean of treatment 5 (transparent CSR communications) to the mean of treatment 1 (no CSR communications).

FIGURE 15. COMPARISON TREATMENT 5 TO 1

As the figure shows, no significant relationship was found between the presence of a transparent CSR communication in the product ad and consumer willingness to pay for the product. As a result, hypothesis 8 can be rejected as well.

H

: Certified CSR communications -> consumer willingness to pay

Treatment 3, which consisted out of the product ad that mentioned a CSR initiative through a partnership or certificate with an external party, provided some surprising results.

It showed a negative mean difference in bid amount of -.152 compared to treatment 1 (no CSR communications). Significance of this mean difference was .008 and therefore it is a significant result. Effectively this entails that consumers pay 15.2% less for a product when a certified CSR initiative is mentioned in the product ad. Moreover, treatment 3 showed negative mean differences compared to all other treatments. The average bid for this treatment was 21.3% lower than the basic CSR

treatment, 14.3% lower than the cause related CSR treatment and 16.1% lower than the transparent CSR treatment.

TABLE 1. COMPARISON TREATMENT 3 TO ALL OTHER TREATMENTS

This suggests that CSR communications involving an external partner or certificate have a

counterproductive effect on consumer willingness to pay when placed in a product ad. Certification through partnership does not have a positive moderating influence on the relationship between CSR communications and consumer willingness to pay. Therefore, hypothesis 9 is rejected.

H

: Cause related CSR communications -> consumer willingness to pay

Treatment 4 (A cause related CSR communication) showed no significant mean differences in bind amount compared to treatment 1 (no CSR communications).

FIGURE 17. COMPARISON TREATMENT 4 TO 1

The data collected in this experiment does not support hypothesis 10, which consequently will be rejected.

4.4 E

An attempt was made to discover relationships between demography/personality moderators and

willingness to pay for the different treatments (H2, H3, H4, H5, H6 and H7). A mean comparison was conducted with the selected cases to discover any significant differences. However, no significant mean differences in the bids were found for any of these moderators. Therefore, hypothesis 2, 3, 4, 5, 6 and 7 were also rejected.

5 D

ISCUSSION

In this chapter our findings will be discussed in an attempt to answer the research question. Furthermore, we shall compare the results of this study to existing literature to explain their

significance and contribution to the current academic debate. In chapter 6 the scientific conclusions and practical implications of this research will be presented. And finally, in chapter 6.3, the limitations of this research will be acknowledged and suggestions will be given for potential future research.

The research question that led to this paper was: Does a causal relationship exist between CSR

communications and consumer willingness to pay? And if so, what are the moderators of this relationship?

As seen in the literature review, this is not the first attempt to answer this question. Not so strange, considering it is the most straightforward way of investigating to what extent consumers value ethics in the purchase decision making process. It’s a very simple trade-off between costs and socially responsible consumer behavior that consumers have to decide upon in almost every purchase decision.

To answer the research question an online real-life controlled experiment was conducted on a large scale. Five treatments were used to investigate consumer willingness to pay in different situations. This study carried the positive aspects of a real-life experiment whilst at the same time increasing validity due to the controlled environment. By using an actual auction with real money, and combining this with an anonymous online setting, the social-desirability answering bias was

significantly reduced. By using a controlled online environment with a preset number of cues and limiting the amount of thinking time available to participants, validity was significantly boosted. The large number of participants (N=854) significantly increases the reliability of our findings. A relatively new and unknown product was chosen (princess smoothie maker) to reduce the existing knowledge that participants had about the product attributes and its price. By using a five minute time limit for participants to place their bids, it was very difficult for them to conduct any research about the

product prior to placing their bid. These measures were taken to rule out the influence of pre-existing information or consumer research on the internet.

5.1 O

CSR

To investigate if a causal relationship existed between the presence of CSR communications in the product ad and consumer willingness to pay, five different treatments were offered in our experiment. The baseline treatment consisted out of a product ad without CSR communications. The four other treatments all had different types of CSR communications embedded (see appendix for ads). Taking existing literature into consideration, we expected that consumers would place higher bids for the treatments that included CSR communications. However, the data analysis showed that there was no significant positive difference in the average bids of the different treatments. Our results therefore contradict most existing literature concerning consumer willingness to pay for CSR (Cowe & Williams, 2000; Creyer & Ross, 1997; Folkes & Kamins, 1999; Ha-Brookshire & Norum, 2011; Lafferty &

Goldsmith, 1999; Maignan, 2001; Maignan & Ferrel, 2003; Ross et al, 1992)

These unexpected results lead us to believe that the infamous attitude-behaviour gap in ethics research played a significant role in previous studies. As previous studies mainly asked participants if they were willing to pay more for socially responsible products, they merely measured attitude and not actual behaviour. Our experiment tested actual behaviour in a real life situation and found no evidence for increased willingness to pay. Therefore a difference in attitude and behaviour can be claimed, the so called attitude-behaviour gap. Possible explanations for this difference in attitude and behaviour can be found in the Theory of Planned Behaviour (Ajzen, 1985). This model states that actual behaviour depends on three determinants (Attitude to Behaviour, Subjective Norms and Perceived Behaviour Control). If one or more of these determinants are not met attitudes will not lead to behaviour. For an in-depth explanation of this model, see chapter 2.3 of this article.

5.2 C

CSR

As part of our experiment, four different types of CSR communications were put to the test. They were:

- CSR basic treatment. The product offer that mentions a CSR initiative on a very basic level (see appendix)

- CSR through partnership. The product offer that mentions a CSR initiative through a partnership with an external party like an NGO (see appendix)

- Cause-related CSR. The product offer that mentions that a certain amount of each sale shall be invested in a CSR initiative (see appendix)

- CSR transparent. The product offer that mentions a CSR initiative and also gives some more information about the mentioned CSR initiative (see appendix)

As can be seen in the findings section of this paper, none of these treatments produced significantly higher bids than the other treatments. However, one treatment did produce some very surprising results. The “CSR through partnership” treatment showed a very surprising negative tendency compared to other treatments.

The bids for this treatment were 15.2% lower than for the baseline treatment, 21.3% lower than the basic CSR treatment, 14.3% lower than the cause-related CSR treatment and 16.1% lower than the transparent CSR treatment. What makes these numbers even more astonishing is the fact that they were all found significant in SPSS. As a matter of fact, it was the only treatment in the whole

experiment that produced significantly different bids than the other treatments. For some reason the mentioning of a CSR certificate in the product ad attracts very low bids. As to what the exact reasons for this result may be, we can only speculate. Certification and partnership are mentioned by various researchers as a way to increase CSR credibility and reduce consumer skepticism towards CSR including the ISO 26000 (ISO, 2010: 76).

5.3 D

CSR

This study was not able to find any relationships between the different demographic factors, personality factors and willingness to pay for CSR. This is not a very surprising result, considering the fact that previous studies are very conflicting when it comes to the influence of demographic factors on socially responsible consumer behavior. Some researchers have claimed that demographic factors such as age, gender and education level have a moderating influence (Hines & Ames, 2000; Parker, 2002; Dickson, 2005) while others state that demographic factors are poor predictors of ethical behavior (De Pelsmacker et al., 2005; O’Fallon & Butterfield, 2005). The results of this experiment show that demographic factors and personality factors do not significantly influence willingness to pay for CSR.

6. C

ONCLUSIONS AND IMPLICATIONS

In this chapter the conclusions of this study will be elaborated. Furthermore in chapter 6.1 the academic implications will be clarified and in chapter 6.2 the practical relevance of this research will be explained. The causal nature of this study, combined with the large and varied sample that participated in the experiment, make the results of this study worthwhile from an academic perspective as well as a business perspective. This research produced some interesting conclusions that can have significant implications for current beliefs about CSR and related business practices. Furthermore it is necessary to also mention the practical implications of these results. The three main conclusions that can be drawn from this study are (1) that consumers are not willing to pay more for a product when CSR is communicated in the product ad, (2) that advertising a certificate from an external body for a CSR initiative leads to a negative evaluation of the related product and (3) that the

attitude-behaviour gap presents itself in a very clear manner when it comes to ethical purchase

decisions.

6.1

A

From an ethical point of view it is very surprising that consumers are not willing to make the trade-off between socially responsible consumer behaviour and extra costs. It is not surprising that

consumers do not wish to incur vast extra costs when acting socially responsible. Previous studies showed that consumers are only willing to make small trade-offs between socially responsible consumer behavior and costs. However, we did not expect a complete lack of willingness to make even the smallest trade-off in order to “do the right thing”.

Furthermore, from an academic perspective it is very interesting to consider the

attitude-behaviour gap and the reasons for its existence. Comparing this research to other studies that used

surveys or interviews, it has to be concluded that consumers in reality do not behave as they claim to do. Where those other studies found evidence for increased willingness to pay for CSR, this study shows that actual behaviour contradicts these claims. This leaves us with the following question: To

what extent are surveys and interviews useful indicators for ethical issues? These methods are generally accepted in scientific research concerning the social sciences, even though their pitfalls are known and widely reported. Real-life experiments are not widely used as a research tool in the social sciences.

This type of experiment is sometimes associated with high costs, a long duration, low reliability and low validity. As shown in this article, these drawbacks can be eliminated for a large part in the digital age we currently live in. The online setting keeps the costs low and decreases the time that needs to be spent in monitoring the participants. Furthermore the large sample significantly increases reliability and the “clean” online environment increases validity. By keeping all cues the same (ceteris paribus) except for one independent variable, any differences in outcome between treatments can be attributed to the manipulated independent variable. The real-life setting guarantees the measurement of actual behavior and not attitudes. And finally, although it should be obvious: only experiments can show causal relationships.

6.2

From a business perspective perhaps the negative influence of advertising a CSR certificate is the most important conclusion of this study. As many products and corresponding product ads contain some kind of certificate for CSR, health, organic production etc. it is of the utmost importance to investigate the effect of these certificates further. Currently firms expect those certificates to have some positive effect on the corresponding product. This is underlined by the fact that ISO 26000 (2010) clearly mentions partnerships and association as a way to increase CSR trustworthiness. However, this research clearly shows that this is not (always) the case. On the contrary, it actually reduces consumer willingness to pay.

It is also important for firms to reconsider the value of advertising corporate social responsibility strategies in product ads. However, first it is crucial to interpret this study and its outcomes in a correct manner. This study does not show that CSR is undesired or unwanted. Nor does this study

show that consumers do not care for socially responsible firms. What this study does show is that consumers are not willing to pay more for a product if CSR is communicated in the product

description. This outcome is relevant for every firm that communicates CSR in their ads and products. The following question rises for these firms: To what extent will consumers accept price premiums for products that are produced, consumed and distributed in a socially responsible manner? The general thought seems to be that consumers are willing to accept some cost trade-offs in order to behave socially responsible. This study shows that this is not the case. Of course there are other reasons for promoting CSR, such as creating positive brand associations and satisfying primary and secondary stakeholders. However, perhaps it is more effective to do this on a brand level instead of on a product level.

6.3

As mentioned earlier in this paper, many measures have been taken to produce a solid research. However, considering the countless factors that can influence scientific research we understand that there are many points for improvement. Furthermore, we also have suggestions for future research as this study creates even more questions than it answers.

- For our experiment we auctioned off a Princess Smoothie Maker, which is a product in the consumer electronics market. Some researchers state that not all product categories are equally sensitive to CSR claims and therefore this experiment should be repeated for different product categories.

- This experiment was conducted in The Netherlands, and studies in the past have shown differences in consumer sensitivity to CSR in different regions in the world (Ottman, 1992; Peattie, 1992). For generalizability purposes this same study should be repeated in different regions of the world or perhaps even on an international scale (which is possible because of the online setting).

- To test the effect of certification and partnership in CSR we showed a certificate for safe labor in factories. It is possible that a different type of certificate or partnership could produce a different result.

- This study measured willingness to pay but it did not investigate the underlying reasons for this willingness to pay. Considering the Theory of Planned Behaviour (Ajzen, 1985) it can be stated that a difference between attitude and behaviour can be caused by three determinants (Attitude to Behaviour, Subjective Norms and Perceived Control). We suggest a study into which of these determinants played a significant role in the outcomes of this research. - In our experiment we tested only one aspect of CSR, namely the labor conditions in factories.

Other aspects of CSR such as environmental claims, the usage of natural resources and

organic production were not investigated. Therefore these factors should be studied as well to determine whether they have a different impact on consumer decision making.

- This research used an online anonymous environment for the experiment. It is possible that other purchase decisions concerning ethics are made when anonymity is not an option or when others are present (e.g. in a physical environment such as a store). This study should be repeated in a more public environment to test the influence of anonymity on the findings.

Other interesting points for future research include the question as to why exactly the CSR with partnership treatment produced such negative results. It contradicts the current belief that external support from an NGO or certifying body leads to a more credible CSR claim. Furthermore, it would be interesting to dive deeper into the attitude-behaviour gap that shows itself in ethics research. What are the major obstacles that prevent consumers from being socially responsible consumers even though they claim to be just that? We encourage other researchers to continue using and improving this innovative auction method. It is an opportunity to prove causal relationships on a larger scale than previously possible. Furthermore, we feel that it is difficult to match the validity of this real life auction setting with any controlled and simulated experiment. However, it is important to note that an

pay that amount. Perhaps in the future these auctions could be combined with in-depth interviews to also study the “why” question.

R

EFERENCES

Ajzen, I. (1985). From intentions to actions: A theory of planned behavior. In J. Kuhl & J. Beckman (Eds.), Action-control: From cognition to behavior (pp. 11-39). Heidelberg: Springer. Ajzen, I. (1991). The theory of planned behavior. Organizational Behavior and Human Decision

Processes, 50, 179-211.

Anniss, E.: 2003, ‘Sweatshops vs. Goldmines of Expertise’, Financial Times, Main section, 1st Edition, December 13.

Asif, M., Searcy, C., Zutshi, A., Fisscher, O.A.M., 2013, An integrated management systems approach to corporate social responsibility, Journal of Cleaner Production 56, 7e17

Balderjahn, I. (1988), “Personality variables and environmental attitudes as predictors of ecologically Responsible consumption patterns”, Journal of Business Research, Vol. 17 No 1, pp. 51-6 Banerjee, B. and McKeage, K. (1994), “How green is my value: exploring the relationship between

environmentalism and materialism”, in Allen, C.T. and John, D.R., Advances in consumer

research, Association for consumer research, Provo, UT, Vol. 21, pp. 147-52

Barone, M.J., Miyazaki, A.D. and Taylor, K.A. (2000), “The influence of cause-related marketing on consumer choice: does one good turn deserve another?”, Journal of the Academy of

Marketing Science, Vol. 28 No. 2, pp. 248-62.

Baumeister, R. F., Bratslavsky, E., Finkenauer, C., & Vohs, K. D. (2001). Bad is stronger than good. Review of General Psychology, 5(4), 323–370.

Berkowitz, L. and Lutterman, K.G. (1968), “The traditional socially responsible personality”, Public

Opinion Quarterly, Vol. 32, pp. 169-85

Bernstein, D. (2009). Rhetoric and reputation: some thoughts on corporate dissonance. Management

Decision, 47(4), 603–615.

Through Mutually Beneficial Corporate Social Responsibility Initiatives’, Journal of Business

Ethics 85, 257–272.

Boush, D. M., Friestad, M., & Rose, G. M. (1994). Adolescent skepticism toward TV advertising and knowledge of advertiser tactics. Journal of Consumer Research, 21(1), 165–175. Carroll, A. B. (1979). A three-dimensional conceptual model of corporate social performance.

Academy of Management Review, 4, 497-505.

Carroll, A.B., 1991. The pyramid of corporate social responsibility: toward the moral management of organizational stakeholders. Business Horizons 34 (4), 39e48.

Carson, T. L. (2003). Self-interest and business ethics: Some lessons of the recent corporate scandals.

Journal of Business Ethics, 43(4), 389–394.

Chapman, G. B., & Johnson, E. J. (2002). Incorporating the irrelevant: Anchors in judgements of belief and value. In: Gilovich, T., Griffin, D., Kahneman, D. (Eds.), Heuristics and Biases. The

Psychology of Intuitive Judgement (pp. 120-138). Cambridge: Cambridge University Press.

Choi, J., & Wang, H. (2009). Stakeholder relations and the persistence of corporate financial performance. Strategic Management Journal, 30(8), 895–907.

Chyllinski, M., & Chu, A. (2010). Consumer cynicism: Antecedents and consequences.

European Journal of Marketing, 44(6), 796–837.

Cone Communications Press Release. 1999. (March 8), Boston, MA.

Connelly, B. L., Certo, S. T., Ireland, R. D., & Reutzel, C. R. (2011). Signaling theory: A review and assessment. Journal of Management, 37(1), 39‐67.

Cowe, R. and S. Williams; 2000, ‘Who are the ethical consumers?’ Ethical consumerism Report, co- operative bank

Crane, A. & Ruebottom, T. (2011). Stakeholder Theory and Social Identity: Rethinking Stakeholder Identification. Journal of Business Ethics, 102:77-87

Creyer, Elizabeth and William T. Ross Jr. 1996. The Impact of Corporate Behavior on Perceived Product Value. Marketing Letters, 7, 2: 173- 185.