Buying for charity: Who should pay the donation?

Buying for charity: Who should pay the donation?

A thesis in the field of cause-related marketing

Renée Jule Baeten

MSc Marketing

University of Groningen

Faculty of Economics and Business Economics

Supervisor: M.C. Leliveld

June 2019

Abstract

The purpose of this thesis was to study whether the payments structure of a CM campaign -who is paying for the donation? - affects people’s willingness to pay for the product and the brand liking. It also aimed to reveal if being transparent about the payment structure (explicit vs. implicit communication) moderates this effect. This was studied with a focus on the potential mediating influence of anticipated warm glow and causal inferences about the company’s motives. The study revealed that consumers assume that the company pays for the donation amount and when the company passes on the costs of the donation to the consumers, the consumers infer the company’s motives as manipulative intent and thereby their anticipated warm glow feeling decreases, which leads to a less positive attitude towards the brand and their purchase intentions drops. Moreover, the study revealed that when a company communicates explicitly that they pay for the CM donation themselves, people were overall more positive compared to when a company was not running a CM campaign. So, transparency could have an influence on the consumer response variable, however, according to the results transparency does not seem to moderate the effect of who is paying for the donation. Taken together, the study suggests that when a company wants to run a CM campaign the best option is to go with a payment structure in which the company pays and communicate explicitly that the company is paying for the donation.

Introduction

Almost every day you find yourself walking in the aisles of a supermarket doing your daily groceries. Looking around in the store you will find numerous products with promotional campaigns: from discounts, and giveaways, to contests (e.g., Share what you love about Kellogg's Frosted Flakes and win a VIP sports practice session with former New York Yankee Tino Martinez). Beside these campaigns, an increasing number of for-profit organizations engage in Cause-related Marketing (CM) initiatives (i.e., the purchase of a product is linked to support for causes) and thereby give the consumer the chance to do something good for something or somebody else by purchasing the organization’s product. For instance, if you are craving for a smoothie you can choose for the smoothies produced by Innocent, a brand that is fighting against bee mortality under the slogan “Buy me, save a bee. For every smoothie you buy, we save one bee”. Or when you are in need of diapers for your child you have the opportunity to choose Pampers from the brand P&G, a brand that donates one vaccine against tetanus to UNICEF for every pack of Pampers sold.

Such cause-related marketing has been formally defined 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 exchanged that satisfy organizational and individual objectives” (Varadarajan and Menon, 1988, p.58). Companies implement these campaigns for several reasons. Some companies want to contribute to society and provide additional benefits to their consumers (Koschate-Fischer, Huber and Hoyer, 2015). Others are motivated by creating a positive image for their firm by using the goodwill from a non-profit organization’s name which will eventually generate more sales (Nowak and Clarke, 2003). From the consumers’ side, companies investing in CM might be seen as moral competent, because CM is a form of Corporate Social Responsibility (CSR) (Marín, Cuestas and Román, 2016). It generates a positive image of the company; the company does something good for the world (Marín, Cuestas and Román, 2016; Müller, Mazar and Fries, 2016).

they are the ones that have to pay extra for the donation to the charitable cause? Isn’t it the social responsibility of the company to bear the cost of the donation? This can make consumers also suspect that the company only implemented this campaign to generate more sales and not because they care so much about charitable causes.

On the other hand, when the company is not paying for the donation, consumers themselves are the ones who actually do good by donating to charity. The transactional element of CM thus provides the consumer with the feeling of control, which might be interpreted as costly prosocial behavior (i.e., any behavior that benefits others which involves costs to the self, like money or time) and as such contributes to the warm-glow feeling (Gneezy et al, 2012). It has been suggested that this feeling could lead to favorable brand attitudes and has the potential to influence outcomes as purchase intentions and consumer loyalty (Müller, Mazar and Fries, 2016; Barone, Miyazaki and Taylor, 2000; Koschate-Fischer, Huber and Hoyer, 2015; Habel et al, 2016).

These two competing aspects might be the reason that CM campaigns could give consumers mixed feelings and these feelings might influence the outcome of the campaign. It might influence the purchase intentions and the attitude of the consumer towards the brand, and therefore it could possibly have an effect on the good name of the company and the number of sales.

Yet, there is not much known about the potential effects of being transparent towards the consumer about the CM payment structure. Prior research did not take into account the fact that companies often do not communicate price changes and payment structures explicitly in their campaigns. In fact, oftentimes research on CM simply assumes that the consumer pays the increase and the consumer is aware of it (cf., Habel et al, 2016; Koschate-Fischer, Huber and Hoyer, 2015; Krishna and Rajan, 2009).

Aiming to address this research gap, our study examines whether the payments structure of a CM campaign -who is paying for the donation? - affects people’s willingness to pay for the product and the brand liking. And does being transparent or not about this structure moderate this effect. I will study this research question with a focus on the potential mediating influence of anticipated warm glow and causal inferences about the company’s motives. First, I assume that consumers make the right assumption about who is paying for the donation during a CM campaign. Hereafter, I will examine whether implicit or explicit communication about the payment structure has an influence on the effect.

THEORETICAL BACKGROUND

Payment structure of CM campaigns

When implementing a CM campaign, companies can choose between different payment structures. A key question in this process is how different payment structures will affect consumer’s perceptions of the campaign and the brand (Koschate-Fischer, Huber and Hoyer, 2015). To set the price, the marketers try to predict how consumers are likely to respond to price changes (Campbell, 1999).

Most common is to set a fixed charitable donation amount for each unit sold of a product (Pecorino, 2016). Companies can pay the promised donation amount per product sold themselves, by extracting it for instance from company profits, advertising budgets, or top management salaries (Habel et al, 2016). Second, the company can increase the price of the product during the period of the campaign and pass on the costs to the consumer. The consumer will finance all of the provision via the price increase (Pecorino, 2016). This option is beneficial to the company, in the sense that it will cost the company less. Finally, the donation can be paid by both the company and the consumer in various distributions. In this case, the donation will be larger than the price increase (Koschate-Fischer, Huber and Hoyer, 2015).

In this paper the focus lies on the first two payment structures, the company pays or the consumer pays. I chose the two extremes because I want to be able to disentangle the effects of when the company pays and when the consumer pays. In the third payment option, in which both parties pay a part of the donation, there are countless possible combinations in the percentage of who is paying which part. This would complicate the interpretation of the effect and the possible motives.

current study I will not specify the motives of the company upfront, so people will make their own causal inferences about the motives of the company.

The problem with the payment structure in which the consumer pays is that it may

produce a conflict: the company wants to help the charitable cause, but the consumer may react negatively if they infer that the price increased (Habel et al, 2016; Koschate-Fischer, Huber and Hoyer, 2015). When asked about the expectations of the consumers, people believed that some companies pass along the costs of CSR activities to their consumers in the form of an increased price (Habel et al, 2016). In a prior study it has been suggested that the likelihood of purchasing a brand that runs a CM campaign decreases as the price of the product increases compared to its alternatives (Subrahmanyan, 2004). Consumers may deteriorate in their perception of price fairness, because in the period of a CM campaign they pay a higher price for the same product. As a consequence, they may be less willing to respond positively to the campaign, which could result in lower sales (cf., Marín, Cuestas and Román, 2016; Campbell, 1999). However, in these cases the consumer has to be aware of the price difference in the first place.

When taking into account the consumers’ benefit perception and price markup inferences, it has been suggested that CSR engagement has mixed effects on the price fairness evaluated by the consumer and thereby on subsequent outcomes such as customer loyalty (Habel et al, 2016). However, in this study they do not specifically look at CM campaigns and the effect on brand likability and willingness to pay. Another study draws the conclusion that if consumers value the supported charitable cause and understand the link between their own purchase and the increase in provision of the cause, they might be willing to pay a price premium (Pecorino, 2016). However, they only researched companies that would donate a percentage of its profit and compared the CM product to similar products. In this study I will use a fixed donation amount and do not compare it to other products.

Based on the previous findings, I propose that in a CM context with a fixed price of the donation:

H1: Compared to not having a CM campaign, the willingness to pay and the brand

Causal inferences about company’s motives

In general consumers are suspicious towards for-profit organizations, assuming that these organizations try to persuade them to buy their products. To understand and cope with the marketers’ actions of these organizations, consumers build knowledge structures about marketing, also known as persuasion knowledge (Campbell and Kirmani, 2000). This has been defined as ‘a loose set of beliefs or intuitive theories about persuasion, and these beliefs may be accurate or inaccurate’ (Campbell and Kirmani, 2000, p.70). Persuasion knowledge includes ideas about persuasion motives (e.g. what the company is trying to achieve), as well as ideas about persuasion tactics (e.g. how the company is trying to achieve it). Consumers may use this knowledge to infer the extent to which the motives underlying certain behavior or actions, for instance a campaign of a company, involve the intent to persuade. The consumer may infer that a company has hidden motives and these suspicions could result in less favorable perceptions of the company. When people infer that a company has ulterior motives for their behavior or actions, the company can be perceived as manipulative and insincere (Campbell and Kirmani, 2000).

Consumers are likely to make causal inferencesabout why the company increased the price (i.e., payment structure = consumer pays) and there is a great risk that negative motives, such as profit-driven and cause-exploitive motives, will be attributed to the company (Koschate-Fischer, Huber and Hoyer, 2015). Consumers are critical to CM efforts; they could question whether the company’s engagement with a social cause is designed to benefit the cause (i.e. cause beneficial) or the company (i.e. cause exploitive) (Barone, Miyazaki and Taylor, 2000). Put differently, consumers could infer that the company is trying to exploit or take advantage of the consumer with the goal of generating sales (Gneezy et al, 2012).

Thus, I propose that consumers infer different motives when the consumer pays for the donation than when the company pays for the donation:

H2: The consumer will infer motives of manipulative intent when the consumer

pays for the donation, compared to when the company is paying for the donation.

a company’s activities consumers may find the brand likeable (Nguyen et al, 2015). In a similar vein, consumers support companies of which their CSR perception or beliefs are positive. The perceptions will impact corporate outcomes, such as brand evaluations and purchase intentions (Marín, Cuestas and Román, 2016).

The findings above implicate that the inferred motives of why the company provides support could play an important part in the reaction of consumers on CM activities (cf., Barone, Miyazaki and Taylor, 2000). Therefore, it is crucial for marketing managers to understand how social responsibility activities can affect the consumer’s overall perception of the company’s value by attributing a positive motive to the CM campaign. If they are not careful, the consumers may infer that the company’s true motives to support a charitable cause is to sell more products rather than acting on behalf of its customers. It could in fact backfire on the company (Marín, Cuestas and Román, 2016). For these reasons, I propose that the causal inferences about company’s motives will mediate the effect of payment structure on corporate outcomes, such as willingness to pay and brand likeability:

H3: When the consumer pays for the donation (vs. when the company pays) the

consumer will infer motives of manipulative intent, which will have a negative effect on the willingness to pay for the product, and on the brand likeability.

Anticipated warm glow

Besides having certain ideas about why a company is running a CM campaign, consumers could also experience certain feelings when deciding to buy a CM product. Prior literature indicates that people could experience joy from donating (Andreoni, 1989; Erlandsson, Jungstrand and Västfjäll, 2016; Giebelhausen et al, 2017). This feeling might have an influence on how people react to CM campaigns overall, therefore it is also an important factor to look at.

Cause-related marketing campaigns in essence provide consumers a chance to do a small good deed while shopping. These small good deeds could be seen as prosocial consumer behavior and have the ability to provide the consumer with a “warm glow” feeling (Giebelhausen et al, 2017). This feeling can be described as a feeling of joy. The amount of joy (warm glow) the helper receives from helping is an important motivator of acting prosocial (Andreoni, 1989; Andreoni, 1990; Verteramo Chui et al, 2017; Erlandsson, Jungstrand and Västfjäll, 2016).

Jungstrand and Västfjäll, 2016). Consumers could experience such anticipated warm glow as when buying a CM product: they expect they will feel this warm glow upon purchasing a product of which they know that the money they will spend supports a charitable cause (cf., Habel et al, 2016). As such, anticipated warm glow is known to positively affect customer satisfaction, willingness to pay and customer loyalty (Giebelhausen et al, 2017; Habel et al, 2016). According to Koschate-Fischer, Stefan and Hoyer (2012) the willingness to pay increases the higher the donation amount, particularly when a consumer experiences these strong warm glow motives. However, they only examine the warm glow motives by comparing different donation levels and use the payment structure in which the company pays a part and the consumer pays a part of the donation to a charitable cause.

In a similar vein, I expect that who pays for the donation can also affect anticipated warm glow. Note that previous research suggests that it matters whether the prosocial behavior consumers engage in is costless for the consumer (cf. company pays) or costly (cf. consumer pays). Gneezy et al. (2012) showed a theory about costless and costly prosocial behavior. When prosocial behavior is costless (i.e., any action aimed at benefiting others which does not involve any costs to the self) it will influence the warm glow feeling of the consumer less than when the prosocial behavior is costly (i.e., any behavior that similarly benefits others which involves costs to the self, like money or time) (Liu and Aaker, 2008; Gneezy et al, 2012). In a similar vein, Erlandsson, Jungstrand and Västfjäll (2016) tested in their research if anticipated warm glow if helping and anticipated guilt if not helping were similarly or differently affected by different types of responsibility-manipulations. They found, amongst others that anticipated warm glow is higher in situations where the responsibility to help is low. This situation entails that the helping behavior is effortful and we are doing more than what is expected of us. Based on these prior research studies, I expect that consumers who intent to buy a CM product and infer that they pay themselves might feel better about themselves than people who decide to buy a CM product when they infer that it is costless:

H4: When the consumer pays for the donation the consumer will feel a

stronger anticipated warm glow, compared to when the company is paying for the donation.

stronger when the person’s donation is greater (Andreoni, 1989; Gneezy et al, 2012; Erlandsson, Jungstrand and Västfjäll, 2016), so I assume that when the payment structure implies that consumer pays for the donation the anticipated warm glow is higher:

H5: When the consumer pays for the donation (vs. when the company pays) the

consumer will feel a stronger anticipated warm glow, which will result in a positive effect on the willingness to pay for the product, and on the brand likeability.

In sum, I expect that causal inferences about company’s motives and the anticipated warm glow will both mediate the effect of the payment structure of CM campaigns on the willingness to pay and the brand likeability, but in opposing directions. On the one hand I expect higher inferences on manipulative intent when the consumer pays, having a negative effect on willingness to pay and brand likeability. On the other hand, I expect higher anticipated warm glow when the consumer pays, having a positive effect on willingness to pay and brand likeability. So, the question now is which mediator will be stronger. I expect that the effect of inferences about the manipulative intent will be stronger than the effect of the anticipated warm glow, because the consumer perception of why the company provides support to the cause might be a key determinant of the responsiveness to a CM campaign (Barone, Miyazaki and Taylor, 2000). Therefore, I assume that this underlying process will be more negative than positive when the consumer pays. The negative responses towards the campaign might dampen the positive feelings. This effect between the mediators will be included in the model, but will be exploratory and so I do not dedicate a specific hypothesis to this effect.

Transparency

structure change how people feel about the campaign and the for-profit organization, or prevent the pre assumed consequences?

It is a challenging task to provide the consumer with the right communication strategy, regarding that the consumer perceives a company’s motive in a positive light. When a company pays for the donation themselves, it has been suggests that the best option might be to make explicit statements that a company’s CM donation is extracted from top management salaries, advertising budgets, or company profits (Habel et al, 2016). This might prevent backlash effects, so consumers will not infer that the company passes on the costs to the consumers (Habel et al, 2016). However this assumption has not been empirical tested. When the payment structure implies that the consumer pays for the donation, it has been suggested that if a justification is provided the outcome might be perceived as fairer than when there is no justification being offered (Campbell, 1999). However, companies should be aware that they could not simply frame their motives as intrinsic, because this will actually arouse customers’ suspicion that a company is trying to persuade them into buying (Habel et al, 2016).

Suppose a company communicates explicitly that the consumer pays for the donation. Based on the previous, I propose that this will heighten the positive effect of anticipated warm glow, because the consumers are made aware of the fact that they are the one’s doing something good instead of the company. Moreover, I propose it will affect inferences of manipulative intent. If companies are explicit about not paying the donation themselves, the motives of the company will most likely be more clear to the consumer and therefore consumers might infer more manipulative intent:

H6: When the company is transparent, it will heighten the “who pays” effect on

anticipated warm glow and manipulative intent compared to the effect when a company is not transparent.

lead to a greater likelihood of purchasing a CM brand. Subrahmanyan (2004) found that when, in an Asian culture, a brand clearly states that the price difference would go to a charity, the company still gets strong support from the consumer. However, the difference in ethics could have an influence on the outcomes, since this study will be conducted in a western culture.

Thus, I expect that transparency will moderate the effect of the payment structure of CM campaigns on the willingness to pay and the brand likeability. As mentioned before, the hypotheses H1, H2, H3, H4 and H5 are based on the assumption that consumers make the right assumptions about who is paying for the donation. Comparing the effect of explicit and implicit communication, I propose the following hypothesis. When the company explicitly communicates that the consumer pays for the donation, the company will be perceived as using more manipulative intent than when the company is not explicit about it. At the same time the anticipated warm glow will increase, because the consumers are made aware of the fact that they are the ones doing something good when buying the product. Therefore, the willingness to pay and brand likeability will decrease more than when the company is not transparent. When the company communicates explicitly that they are paying the donation themselves, the causal inferences about the company’s motives will be more positive, but the anticipated warm glow will probably decrease, compared to not being explicit about it. As a consequence the willingness to pay and brand likeability will increase more than when the company is not transparent:

H7: When the company is transparent, it will heighten the “who pays” effect on the willingness to pay and brand likeability compared to the effect when the company is not transparent.

In sum, in this study I will focus on whether the payments structure of a CM campaign -who is paying for the donation? - affects people’s willingness to pay for the product and the brand liking. And does being transparent or not about this structure moderate this effect. This research question will be studied with a focus on the potential mediating influence of anticipated warm glow and causal inferences about the company’s motives.

Fig. 1. Conceptual Model

METHODS

Design & participants

The study used a 2(payment structure: company pays vs. consumer pays) x 2(transparency: transparent vs. not transparent) + control group (no CM campaign) between subjects design, resulting in a total of five conditions. Participants were randomly assigned to one of the conditions. I started with 390 MTurkers, of which 63 failed the manipulation check or/and attention check (see for more details the result section & Appendix 4). Therefore, I based the final analysis on remaining 327 United States residents (44% female, 56% male) with a mean age of 37 years (Mage=37.42, SD=11.68) who completed an online questionnaire about CM campaigns in exchange for a small payment. Most of which were in possession of a high school degree (35.2%) or Bachelor’s degree (52.9%) and had a full-time job (61.5%).

Stimuli

In selecting a brand for the experiment, it is important that the product itself will not influence the consumers’ perception of the CM campaign beforehand. In order to avoid second-guessing the product, I chose to use a utilitarian product to which everyone can relate to buy and has little emotional attachment. Therefore, the product hand soap of the brand Softsoap was used for the advertisements in this study. Moreover, it is important that the collaboration between the profit organization and the charity is a good fit (Koschate-Fischer, Stefan and

DVs Potential mediators Transparency Payment structure of CM campaigns Causal inferences about company’s motives Anticipated warm

glow Willingness to pay

Hoyer, 2012; Marín, Cuestas and Román, 2016); I wanted to avoid that a misfit will influence the research results. Because Softsoap provides products concerning personal hygiene, I chose school water points as they make it possible for children in third world countries to have access to clean water and sanitary facilities. School water points thus also take care of their personal hygiene; it should be a good fit. All in all, the CM campaign in this research implies that the for-profit organization (Softsoap) collaborates with the non-profit organization “Save the Children”. The advertisement stated that Softsoap makes a donation to the charity organization to help them build school water points in third world countries.

Second, I determined the price of the donation based on research by Kang and Leliveld (2019) and Fischer, Huber and Hoyer (2015). In the study of Koschate-Fischer, Huber and Hoyer (2015) no difference was found between a donation amount of 25 cents and 40 cents on a product price around €1.00, which suggest that the consumer perceives the price as fair. In a more related study, Kang and Leliveld (2019) used toilet paper (2.29 euro) and either had an increase of 30 cents or no increase after implementing the campaign. So the price increase was 13% in case the consumer paid. I used price points that are quite similar to the price points Kang and Leliveld (2019) used in their study, because the descripted situation is so similar and the donation amount lays between 25 cents and 40 cents. For the donation structure, there are a multiple options. Donations can for instance be set to a fixed value or can be set at a percentage of the product price or at a percentage of the net profits (Guerreiro, Rita and Trigueiros, 2015; Koschate, Huber and Hoyer, 2015; Haruvy and Popkowski Leszczyc, 2009). I choose to set the donations at a fixed value ($0.30) and to make use of the donation structure ‘Purchase 1 product = $ (amount) donated of the price to charity program’, to control for donation magnitude and framing effects (cf., Guerreiro, Rita and Trigueiros, 2015).

Procedure

Fig. 2. Sample ads: (on the left) company pays, transparency, (on the right) company pays,

no transparency.

Participants then responded to questions measuring the constructs of interest: anticipated warm glow, causal inferences about the company’s motives, willingness to pay, and the brand likeability. Anticipated warm glow was measured with 4 items (e.g., “I would experience a warm pleasurable feeling if I bought the product.”; cf., Erlandsson, Jungstrand and Västfjäll, 2016) and was rated on 7-point Likert-scales where 1 = not at all and 7 = very much (Cronbach’s α = .94). Causal inferences about company’s motives were measured with 8 items, of which 3 items were recoded (see Appendix 4) and for which the last 3 items were based on the study of Kang, Leliveld and Ferraro (2019) (e.g., “Softsoap tried to manipulate the audience in ways that I do not like.”; cf., Kang, Leliveld and Ferraro, 2019) (Cronbach’s α = .87). The items were rated on 7-point Likert-scales where 1= strongly disagree and 7=strongly agree. Note that the higher the score of causal inferences about company’s motives the more the motive was inferred as manipulative and the lower the score the more the motive was inferred as not manipulative. The willingness to pay was measured with 4 questions (e.g., “I am willing to pay the price mentioned in the advertisement.”; Cronbach’s α = .97) and was rated again on 7-point Likert-scales where 1=strongly disagree and 7=strongly agree.

another brand. Moreover, the consumer’s likeability of the brand was measured with 9 items (e.g., “This brand is very attractive.”; cf., Nguyen et al, 2015), and assesses the appeal a customer has for a brand (Nguyen, Melewar and Chen, 2013) and was also rated on 7-point Likert-scales where 1= strongly disagree and 7=strongly agree (Cronbach’s α = .95).

Finally, participants stated their gender, their current age and other demographics. Moreover, I used an attention check and a manipulation check to see if participants were paying attention during the questionnaire. The participants were asked to indicate the original price of the product and if there was a price difference between the price in the advertisement and the price in the introduction text. For a full overview of all measures, see Appendix 4.

RESULTS

Manipulation check

Out of the 390 participants, 63 participants were dropped from the analyses because they failed one or both control questions about their given condition or the attention check (see Appendix 4 for control questions and attention check). A total of 18 participants failed the attention check, although all of them answered the control questions correctly. A total of 45 participants failed one or both control questions; 34 participants failed only one out of the two control questions and 11 participants failed both control questions. The numbers of participants who failed one or two control questions were spread evenly across the conditions. A chi-square test of independence was performed with all five conditions and the outcome of the control questions (0= failed one or both control questions, 1= passed control questions) to analyze if there was a significant difference between the number of participants who failed per condition. The relation between these variables was significant (X2(4, N = 390) = 16.03,

p = .003, Cramer’s V = 0.20). Compared to the control condition (7.60%), the number of

(‘Was there a different price indicated as original price than in the advertisement?’). Therefore, the analysis is based on the remaining 327 participants.

Choice of brand

At first, I analyzed whether or not the choice for brand Softsoap was influenced by our manipulations, and whether having a CM campaign in general changed people’s choice compared to the control condition, in which the brand ran no CM campaign. A chi-square test of independence was performed with all five conditions and brand choice (1= Softsoap, 2= other brand).

Table 1. Brand choice per condition in percentages.

The relation between these variables was significant (X2(4, N = 327) = 18.59, p = .000, Cramer’s V = 0.24). Compared to the control condition (in which 90.00% choose Softsoap), participants in the condition in which the company pays and is transparent about it (97.20%) were more likely to choose the brand Softsoap over other brands than participants in the conditions in which the consumer pays and the company is transparent about it (79.70%) and in which the consumer pays and the company is not transparent about it (77.40%). The brand choice of the participants in the control condition (90.00%) compared to the condition in which the company pays and is not transparent about it were almost the same (93.80%). The effect size is small. See Table 1 for an overview of the results.

A binary logistic regression was performed to explore the relationship between the payment structure and the brand choice, as well as the relationship between transparency and brand choice in more detail. There was a significant association between payment structure and brand choice (X2_{(1) = 0.17, p = .000). Participants in the conditions in which the company} pays were less likely to choose another brand hand soap than participants in the conditions in which the consumer pays were. However, there was no significant association between transparency and brand choice (X2(1) = 1.32, p = .480), neither was there a significant relation

Transparency

Being transparent Softsoap Other brand

Company pays Yes 97.20% 2.80% No 93.80% 6.20% Consumer pays Yes 79.70% 20.30% No 77.40% 22.60% 90.00% 10.00% Payment

structure

Brand choice

between the interaction effect of payment structure and transparency, and the brand choice (X2(1) = .38, p = .085).

Willingness to pay and brand likeability

In order to analyze the influence of the payment structure and transparency on the willingness to pay and brand likeability, 2(payment structure: company pays vs. consumer pays) x 2(transparency: transparent vs. not transparent) ANOVA’s on willingness to pay and on brand likeability were performed. Consistent with H1, there was a significant main effect of payment structure on the willingness to pay (F(1,257) = 47.13, p < .001) and on brand likeability (F(1,257) = 18.76, p < .001). The participants in the company pays conditions had a significantly higher willingness to pay (M = 6.15, SD = 1.18) and significantly liked the brand Softsoap more (M = 5.28, SD = 1.08) than did participants in the consumer pays conditions (Mwillingness to pay = 4.85, SD = 1.83; Mbrand likeability = 4.59, SD = 1.46). There was no

significant main effect of transparency on willingness to pay (F(1,257) = 0.68, p = .410), or significant interaction effect of payment structure and transparency on willingness to pay (F(1,257) = 3.05, p = .082). Neither was there a significant main effect of transparency on brand likeability (F(1,257) = 0.05, p = .821), or significant interaction effect of payment structure and transparency on brand likeability (F(1,257) = 2.04, p = .155).

In order to test whether having any kind of CM campaign would affect willingness to pay and brand likeability differently compared to having no campaign, a new variable was created with 5 categories: 1) company pays, transparent, 2) company pays, not transparent, 3) consumer pays, transparent, 4) consumer pays, not transparent, 5) no CM campaign. One-way ANOVA’s with this variable were used to test the effects on willingness to pay and brand likeability. See Table 2 for cell means and standard deviation of all the conditions. There was a statistically significantly difference between the conditions when comparing the willingness to pay (F(4,322) = 14.44, p < 0.001). A simple custom contrasts test comparing the control condition with each level of CM campaigns revealed that the willingness to pay was statistically significantly lower when the consumer pays for the donation in both transparent (M = 4.60, p = .000) and not transparent (M = 5.09, p = .005) conditions compared to the control condition (M = 5.81). There was no statistically significantly difference between the control condition and the conditions in which the company was transparent about that they pay themselves for the donation (p = .081) or when they were not transparent about it (p = .314).

the brand likeability was statistically significantly higher when the company communicated that they were the ones who pay for the donation (M = 5.41, p = .011) compared to the control condition (M = 4.87). There was no statistically significantly difference between the control condition and the condition in which the company did not communicated that they pay for the donation (p = .205). Neither was there a statistically significantly difference between the control condition and the condition in which the company did communicate that the consumer pays (p = .080) or did not communicate that the consumer pays (p = .374).

Table 2. Cell means (standard deviations in parentheses).

*Different uppercase letters show significant difference between conditions within that column (p < .05).

Causal inferences about company’s motives

To analyze the influence of payment structure and transparency on causal inferences about company’s motives, a 2(payment structure) x 2(transparency) ANOVA on manipulative intent was performed. The effect of payment structure on manipulative intent was significant (F(1,257) = 23.16, p < .001). Participants inferred the company’s motives as less manipulative when the company pays for the donation amount (M = 3.27, SD = 1.06) than the participants did when the consumer pays for the donation (M=4.02, SD=1.43). This was in line with H2. There was no significant effect of transparency on causal inferences about company’s motives (F(1,257) = 0.01, p = .906), or significant interaction effect of payment structure and transparency on causal inferences about company’s motives (F(1,257) = 1.81,

p = .179).

A one-way ANOVA with the new created variable with 5 categories was used to test the effect on manipulative intent. See Table 2 for cell means and standard deviation of all the

Transparency

Being transparent Company pays Yes

No Consumer pays Yes No Payment structure Willingness to pay Brand likeability Causal inferences about company's motives Anticipated warm glow

Control condition (no CM campaign)

conditions. There was a statistically significant difference between conditions (F(4,322) = 7.15, p < 0.001). A simple custom contrasts test comparing the control condition with each level of CM campaigns revealed that the manipulative intent was statistically significantly lower when the company was transparent about that they were the ones who pay for the donation (M = 3.17, p = .040) compared to the control condition (M = 3.57). In the condition in which the consumer pays for the donation and the company was transparent about it, the manipulative intent was statistically significantly higher (M = 4.13, p = .008) compared to the control condition (M = 3.57). There was no statistically significantly difference between the control condition and the conditions in which the company was not transparent about that they pay for the donation (p = .377) or when they did not communicated that the consumer pays for the donation (p = .0.78).

Anticipated warm glow

To analyze the influence of payment structure and transparency on the anticipated warm glow, a 2(payment structure) x 2(transparency) ANOVA on anticipated warm glow was performed. The effect of payment structure on anticipated warm glow was significant (F(1,257) = 18.56, p < .001). Participants experienced a higher anticipated warm glow when the company pays for the donation amount (M = 5.36, SD = 1.29) than the participants did when the consumer pays for the donation (M = 4.53, SD = 1.81). This was not in line with H4. There was no significant effect of transparency on anticipated warm glow (F(1,257) = 0.46, p = .496), or significant interaction effect of payment structure and transparency on the anticipated warm glow (F(1,257) = 2.74, p = .099).

A one-way ANOVA with the same variable with 5 categories was used to test the effect on anticipated warm glow. See Table 2 for cell means and standard deviation of all the conditions.There was a statistically significant difference between the conditions (F(4,322) = 7.00, p < 0.001). A simple custom contrasts test comparing the control condition with each level of CM campaigns showed that the anticipated warm glow was statistically significantly higher when the company pays for the donation in both transparent (M = 5.45, p = .000) and not transparent (M = 5.27, p = .004) conditions compared to the control condition (M = 4.50). There was no statistically significantly difference between the control condition and the conditions in which the consumer pays for the donation and the company is transparent about it (p = .440) or is not transparent about it (p = .351).

Mediation effect

glow. Given the results of the ANOVA’s, I left transparency and the interaction effect between transparency and payment structure out of the mediation analysis. First, I ran a simple mediation analysis with model 4 of PROCESS based on 5000 bootstrapped samples using bias-corrected and accelerated 95% confidence interval (Hayes, 2013) with payment structure (0 = company pays, 1 = consumer pays) as the independent variable, causal inferences about company’s motives and anticipated warm glow as parallel mediators, and the willingness to pay as dependent variable. The payment structure had a positive significant effect on the causal inferences about company’s motives (β = .75, 95% CI = [.45, 1.06]) and the effect of causal inferences on willingness to pay was significant and negative (β = -.50, 95% CI = [-.62, -.38]). So, when the consumer pays for the donation (vs. the company pays), the consumer inferred the motives more towards manipulative intent than genuine intent. The more the motives are inferred as manipulative, the less willing consumers were to pay for the product. Overall, there was a significant indirect effect of payment structure on the willingness to pay through causal inferences about company’s motives (β = -.38, 95% CI = [-.58, -.21]). This is in line with H2 and H3.

The results also showed that anticipated warm glow mediated the payment structure effect. The payment structure (0 = company pays, 1 = consumer pays) had a negative significant effect on anticipated warm glow (β = -.84, 95% CI = [-1.22, -.45]) and the effect of anticipated warm glow on the willingness to pay was significant and positive (β = .42, 95% CI = [.33, .52]). So, when the consumer pays for the donation, the consumer experienced less anticipated warm glow. This is contrary to my expectations. The less experienced anticipated warm glow, the less willing consumers were to pay for the product. The indirect effect of payment structure on willingness to pay through anticipated warm glow was also significant (β = -.35, 95% CI = [-.59, -.17]). The mediators could account for more than half of the total effect (Pm = .56). This suggest that the payment structure in which the consumer pays can result in negative downstream consequences for the company such as lowering willingness to pay by the inference of manipulative intent and decreasing anticipated warm glow.

inferred the motives of the company as more manipulative. Therefore, the brand likeability went down. Overall, there was a significant indirect effect of payment structure on brand likeability through causal inferences about company’s motives (β = -.38, 95% CI = [-.57, -.23]) and through anticipated warm glow (β = -.27, 95% CI = [-.43, -.14]). The direct effect of payment structure on brand likeability is not significant (β = -.04, 95% CI = [-.24, .16]). The mediators could account for almost the whole total effect (Pm = .94). This suggests that when the consumer pays, the brand likeability decreased primarily by the inference of manipulative intent and the decreasing anticipated warm glow. These results support H3, but are in fact the opposite of H5.

Sequential mediation effect

To analyze if one mediator affects the other mediator, I ran a sequential mediation analysis (model 6; N = 5000 bootstrapped samples; 95% confidence interval) (Hayes, 2013) with payment structure (0 = company pays, 1 = consumer pays) as the independent variable, causal inferences about company’s motives and anticipated warm glow as sequential mediators, and the willingness to pay as dependent variable. Causal inferences about the company’s motives had a negative significant effect (β = -.75, 95% CI = [-.87, -.62]) on anticipated warm glow. However, in this case the payment structure (0 = company pays, 1 = consumer pays) did not have a significant effect on anticipated warm glow (β = -.27, 95% CI = [-.59, .05]), so there was no indirect effect through anticipated warm glow only (β = -.12, 95% CI = [-.28, .02]). Overall, the indirect effect of payment structure through manipulative intent and anticipated warm glow on willingness to pay was significant and negative (β = -.23, 95% CI = [-.42, -.12]). So, when the consumer pays for the donation (vs. the company pays), the consumer inferred the company’s motives more towards manipulative intent, thereby lowered in their feeling of anticipated warm glow, and decreased in their willingness to pay.

When taking anticipated warm glow as the first mediator, it had a significant negative effect on manipulative intent (β = -.48, 95% CI = [-.56, -.40]). Overall, the indirect effect of payment structure through anticipated warm glow and manipulative intent on the willingness to pay was significant and negative (β = -.20, 95% CI = [-.35, -.10]). So, when the consumer pays for the donation (vs. the company pays), the consumer experience less anticipated warm glow, thereby inferred the company’s motives more towards manipulative intent, and decreased in their willingness to pay.

Moderated mediation effect

Although the analyses of the ANOVA’s did not yield significant main or interaction effects with transparency, I did test if there was a moderated mediation effect. First, I ran a mediation analysis (model 8; N = 5000 bootstrapped samples; 95% confidence interval) (Hayes, 2013) with payment structure as independent variable, transparency as moderator, causal inferences about company’s motives as mediator and, willingness to pay and brand likeability as dependent variables. The moderated mediation effect on the willingness to pay was not significant for causal inferences about company’s motives (index = -.21, 95% CI [-.55, .09]). The moderated mediation effect on brand likeability was also not significant for causal inferences about company’s motives (index = -.21, 95% CI [-.56, .86]).

Secondly, I ran a similar mediation analysis (model 8; N = 5000 bootstrapped samples; 95% confidence interval) (Hayes, 2013) with the same independent variable, moderator and dependent variables but now anticipated warm glow as mediator. The moderated mediation effect on the willingness to pay was not significant for anticipated warm glow (index = -.27, 95% CI [-.64, .05]). Also the moderated mediation effect on brand likeability was not significant (index = -.21, 95% CI [-.48, .04]). This was in line with the results of the ANOVA’s, which also yielded no moderation effects. H6 and H7 were thereby also rejected based on these analyses.

GENERAL DISCUSSION

This thesis studies the influence of company’s cause-related marketing campaign approach on the consumers’ willingness to pay and the brand likeability. There are different approaches to run a CM campaign. Some companies pass on the costs of the donation for charity to the consumer. Other companies choose to pay the promised donation amount themselves. Companies also have to decide if they explicitly communicate the chosen payment structure towards the consumer or not. I examined if the way a company structures the campaign (who is paying for the donation?) and how they communicate it towards the consumers (transparent or not) plays an role in how people react to the campaign and thereby influence their attitude towards the brand and their purchase intent.

willingness to pay and the brand likeability slightly increased when a company runs a CM campaign and pays for the donation. However, it decreased when the consumer herself pays for the donation. The willingness to pay decreases even more than the brand likeability. In this case, people are not very convinced about how they feel about the brand and if they are willing to pay for the product.

These effects were influenced by the mediation effect of manipulative intent and anticipated warm glow. I expected that when the company passed on the cost of the donation to the consumers, the consumer would infer motives of manipulative intent compared to when the company pays for the donation. As well as, that when manipulative intent was inferred it would negatively affect the willingness to pay and the brand likeability. The results confirmed that when the consumer pays for the donation, people inferred the motives more towards manipulative intent and thereby lowered in their willingness to pay for the product in the same line in which they lowered in how much they liked the brand. A participant explained: “I do not like the fact that they inflated the original price and then try to act like they are the ones doing something good. I would have felt differently about this brand, had they not inflated the price and actually donated their own money.” Another participant noted: “The company is making the same amount of profit on the product, and is not actually contributing to the cause itself. I am tired of these types of advertisements and try to stay away from making such purchases.” These quotes are in line with H2 and H3.

Expected was that when the consumer pays for the donation, the consumer would feel a stronger anticipated warm glow. This would mediate the effect of the payment structure on willingness to pay and brand likeability in a positive way. Contradictory from what I expected, the anticipated warm glow did not increase when the consumer pays for the donation. In fact, participants experienced more anticipated warm glow when the company

pays for the donation amount. This is the opposite of what I assumed in H4 and H5.

Thus, the two mediators do not have an effect in opposite directions, but they rather complement one another. The results of the sequential mediation analysis indicated that consumers who infer manipulative intent lower in their feeling of anticipated warm glow. As well as, when a consumer experienced less anticipated warm glow, the consumer will infer more manipulative intent. The anticipated warm glow did, however, increase the willingness to pay and brand likeability as predicted.

the conditions to the condition not running a CM campaign, I found that when a company communicated explicitly that the company is paying for the donation, people were more positive. And when a company communicated that the consumer is paying for the donation, people react less positive. So, transparency influences consumer response, but does not seem to moderate the effect of who is paying for the donation.

Theoretical implications

This thesis makes a theoretical contribution to the literature on the approach of cause-related marketing and the corporate outcomes of these different approaches. The present study demonstrates contrasting findings (i.e. consumers assume that the company pays for a donation during a CM campaign and react much less positive when they are made aware of the fact that they pay for the donation), relative to previous studies (Habel et al, 2016, Koschate-Fischer, Huber and Hoyer, 2015; Krishna and Rajan, 2009), in which it is assumed that the consumer pays for the donation amount during a CM campaign and that the consumer is aware of this fact. However, the current findings are in line with and build on other past research (Barone, Miyazaki and Taylor, 2000; Subrahmanyan, 2004). The likelihood of purchasing does not only decreases when the price of the product of the CM brand increases to its alternatives (Subrahmanyan, 2004), but also decreases when comparing the original product price to the increased product price (i.e. consumer pays) of the same product during a CM campaign. When the consumer is required to make such a trade-off in exchange for selecting the brand that is running the CM campaign, the campaign will be less effective (Barone, Miyazaki and Taylor, 2000) because it will negatively influence the likeability of the brand and willing to pay for the product.

Further, the findings are consistent with theory about the inference of underlying motives of a company on consumer responses (Habel et al, 2016; Campbell, 1999; Gneezy et al, 2012). When underlying motives of a company are inferred as negative (e.g. cause exploitive when running a CM campaign), the behavior of the company may be perceived as unfair and lead to negative responses and attributions (Habel et al, 2016; Campbell, 1999). The current study extends this work by adding the payment structure in which the consumer pays for the CM donation as a scenario to be a cause of inferring negative underlying motives. It can be explained as taking advantage of the consumer with the goal to generate sales (cf., Gneezy et al, 2012) and this negative inference lead to less willingness to pay and consumers questioning the likeability of the brand.

(Habel et al, 2016; Koschate-Fischer, Stefan and Hoyer, 2012; Gneezy et al, 2012). I demonstrated that this result could be explained by the sequential effect of inference of manipulative intent on warm glow, also known as a domino-effect (Erisen, 2010). When people infer the motives of a company as negative (i.e. manipulative intent) it dampens the potential warm glow feeling and when the motives of a company are inferred as positive it heightens the anticipated warm glow feeling. The influence of the inference motives of manipulative intent on the warm glow feeling can be added to the theory of the effect of positive emotions derived from warm glow on the outcomes of a CM campaign (Giebelhausen et al, 2017; Habel et al, 2016). So, the experience of anticipated warm glow is affected by the emotions derived from the inference of manipulative intent. Thereby, it has not only an effect on the outcomes of the CM campaign but also on the effect of the warm glow feeling on the outcomes.

The finding that experiencing anticipated warm glow positively affects willingness to pay is in line with previous studies (Giebelhausen et al, 2017; Habel et al, 2016; Koschate-Fischer, Stefan and Hoyer, 2012). These studies explained that positive emotions derived from warm glow make people willing to pay more and are more satisfied with the company than those experiencing negative emotions (Giebelhausen et al, 2017; Habel et al, 2016; Heussler et al, 2009) and the present study adds brand likeability to these desirable outcomes. Finally, the experiment provides a new inside into the relationship between transparency during a CM campaign and consumer response, however transparency does not seem to moderate the “who pays” effect. When a company explicit communicates that the company is paying for the donation, people are more positive compared to when the company is not running a CM campaign. In this scenario the motives of the company are inferred towards not manipulative and therefore the anticipated warm glow increased. This could be due to the fact that the communication is credible and has a strong sense of justice (Lau and Lee, 1999) especially compared to the scenario in which the consumer pays for the donation. Therefore, consumers find out through this direct communication that a company is competent and has no hidden motives (Campbell and Kirmani, 2000), which could lower the perception of manipulative intent. As an effect, consumers may be willing to rely on that brand (cf., Lau and Lee, 1999). This finding could contribute to the literature on consumers’ trust in a brand.

Managerial implications

themselves rather than increasing the price of the product. Although increasing the price seems like the easiest way to go for a company (e.g. they do not loose profit, they expect higher sales and they think they will create a prosocial brand image), this study shows that people are reluctant to accept this approach and it might backfire. For example, Shell introduced a cause-related marketing campaign in the Netherlands in 2019 which implies that consumers can choose to pay an extra €0.01 per liter gasoline. This extra money is then used to plant trees in Indonesia and Peru, to compensate for the CO2 emissions. Shell uses the slogan “go well” for it, and in its advertisement tries to suggest that the organization cares about CO2 emissions. However, if a large and wealthy company as Shell cares about such a cause, the company should not let the consumer pay for it, but should invest money in it themselves (Huisman, 2019). Therefore many people are skeptic about this campaign.

When a company wants to run a CM campaign, the best option is to go with a payment structure in which the company pays and communicates explicitly that the company is paying for the donation. More generally, this study suggests that when implementing a CM campaign marketers should be really careful with the approach they choose. A CM campaign could easily go down the wrong way and could have negative consequences for the number of sales and hurt the brand name. However, when choosing the right approach the campaign could have a positive effect. Especially in the long term positive attitude toward a company’s CSR engagement can be achieved. This can be compared to gaining consumers’ trust. It takes time to gain the trust of the consumer, however once earned, it is hard to maintain and it can be easily violated.

Limitations and future research

the consumer pays for the donation amount. All in all, this study could be seen as a conservative test.

Second, overall there was a positive effect on warm glow when the company was paying. However, some participants responded differently “I do not feel that I made a contribution. That is, the contribution is solely Softsoap and not due to my desire to contribute.” This could be due to differences in personality traits, which I did not include in this study. For future research it would be interesting to include personality variables that could influence manipulative intent and anticipated warm glow. The amount of warm glow one receives from helping others can be affected by for instance the extent to which a person is self-oriented, empathic or by the internal affective state of a person (Erlandsson, Jungstrand and Västfjäll, 2016; Ferguson, Taylor, Keatley, et al, 2012; Basil, Ridgway and Basil, 2008).

Third, I distinguished two payment structures, the company pays for the donation amount or the consumer pays for the donation amount. I choose these two options, because I wanted to be able to disentangle the effects of when company pays and when consumer pays. However, a third payment structure is also a possibility; the company pays part of the donation amount and the consumer pays the other part. This way the company shows that they are committed to the cause as well (Koschate-Fischer, Huber and Hoyer, 2015). There are multiple options how the donation amount can be divided between the company and the consumer. For future research, varying the degree to which the company and consumer contribute to the donation can be an interesting line of research. Will this payment structure be linear to the other two payment structures (company pays vs. consumer pays) or do people mind it less when they have to pay just a part of the donation amount? Koschate-Fischer, Huber and Hoyer, (2015) found that when a company tries to cover part of the costs by implementing a price increase it is possible that consumers under certain conditions will perceive this increase as fair and will not react negatively.

reduce their guilt feelings (Strahilevitz and Myers, 1998; Basil, Ridgway and Basil, 2008). This stimulus is missing when people intend to buy a utilitarian product.

Furthermore, I mentioned in the study explicitly the original price of the product in the introduction text. Therefore, people were aware of the price increase when the consumer pays for the donation, even when the company was not transparent about it. However, one might argue that under actual market conditions the consumer may not always notice a price increase and it is only when consumers are aware of the price increase that any potential negative effects will occur (Koschate-Fischer, Huber and Hoyer, 2015). Especially, when the company is not transparent about the payment structure and the original price is not mentioned, there is a possibility that people simply do not notice the price increase and thereby have a different attitude towards the brand than when they are aware of the price increase. An interesting future study would be to study people’s change in organizational attitude when they find out that the CM campaign the organization was running is actually letting the consumer pay for the donation. Will they react in a more extreme way and what will be the impact on the for-profit organization? When a company proactive communicates their CSR campaign and the company’s CSR statements precede conflicting observed behavior it leads to higher levels of perceived hypocrisy (Wagner, Lutz and Weitz, 2009). As well as that it could violate the trust of a customer in a company. Especially if a company has a history of unfavorable behaviors it could consequently suffer damage to their corporate reputation (Zou et al, 2015).

REFERENCES

Andreoni, J. (1989). ‘Giving with Impure Altruism: Applications to Charity and Ricardian Equivalence.’ Journal of Political Economy, 97(6), pp. 1447-1458.

Andreoni, J. (1990). ‘Impure Altruism and Donations to Public Goods: A Theory of Warm-Glow Giving.’ The Economic Journal, 100(401), pp. 464-477.

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, 28(2), pp. 248-262.

Basil, D. Z., Ridgway, N. M. and Basil, M. D. (2008). ‘Guilt and Giving: A Process Model of Empathy and Efficacy.’ Psychology & Marketing, 25(1), pp. 1–23.

Becker-Olsen, K.L., Cudmore, A.B. and Hill, R.P. (2006). ‘The impact of perceived corporate social responsibility on consumer behavior.’ Journal of Business

Research, 59(1), pp. 46-53.

Boenigk, S. and Schuchardt, V. (2013). ‘Cause-Related Marketing Campaigns with Luxury Firms: An Experimental Study of Campaign Characteristics, Attitudes, and Donations.’ International Journal of Nonprofit and Voluntary Sector Marketing, 18(2), pp. 101–121.

Campbell, M.C. (1999). ‘Perception of Price Unfairness: Antecedents and Consequences.’

Journal of Marketing Research, 36(2), pp.187-199.

Erisen, Cengiz. (2010). ‘Sequential Mediation Analysis in Experimental Research.’ Paper

Presented in Panel 8-13 at the 106th Annual Meeting of the American Political Science Association. Available at:

https://www.researchgate.net/publication/228162340_Sequential_Mediation_Analysis _in_Experimental_Research (Accessed: 14 June 2019).

Erlandsson, A., Jungstrand, A.Å., and Västfjäll, D. (2016). ‘ Anticipated Guilt for Not Helping and Anticipated Warm Glow for Helping Are Differently Impacted by Personal Responsibility to Help.’ Frontiers of Psychology, 7(1475).

Ferguson, E., Taylor, M., Keatley, D., et al. (2012). ‘Blood donors ’helping behavior is driven by warm glow: more evidence for the blood donor benevolence hypothesis.’

Transfusion, 52, pp. 2189-200.

Forehand, M.R. and Grier, S. (2003). ‘When is honesty the best policy? The effect of stated intent on consumer skepticism.’ Journal of Consumer Psychology, 13(3), pp. 349-56. Giebelhausen, M., Lawrence, B., Chun, H. E. H. and Hsu, L. (2017). ‘The Warm Glow of

Restaurant Checkout Charity.’ Cornell Hospitality Quarterly, 58(4), pp. 329-341. Gneezy, A., Imas, A., Brown, A., Nelson, L.D. and Norton, M.I. (2012). ‘ Paying to Be Nice:

Consistency and Costly Prosocial Behavior.’ Management Science, 58(1), pp. 179-187.

Grau, S. L., Garretson, J. and Pirsch, J. (2007). ‘Cause-Related Marketing: An Exploratory Study of Campaign Donation Structures Issues.’ Journal of Nonprofit & Public

Sector Marketing, 18(2), pp. 69–91.

Guerreiro, J., Rita, P. and Trigueiros, D. (2015). “Attention, Emotions and Cause-Related Marketing Effectiveness.” European Journal of Marketing, 49(11-12),

Habel, J., Schons, M.L., Alavi, S. and Wieseke, J. (2016). ‘Warm Glow or Extra Charge? The Ambivalent Effect of Corporate Social Responsibility Activities on Customers’ Perceived Price Fairness.’ Journal of Marketing, 80, pp. 84-105.

Haruvy, E. and Popkowski Leszczyc, P. T. L. (2009). “Bidder Motives in Cause-Related Auctions.” International Journal of Research in Marketing, 26(4), pp. 324–331. Hayes, A. F. (2013). ‘Introduction to mediation, moderation, and conditional process

analysis : a regression-based approach.’ New York, NY: Guilford Press

(Methodology in the social sciences).

Available at: https://ebookcentral.proquest.com/lib/rug/reader.action?docID=1186800. (Accessed: 12 June 2019).

Heussler, T., Huber, F., Meyer, F., Vollhardt, K. and Ahlert, D. (2009). ‘Moderating Effects of Emotion on the Perceived Fairness of Price Increases.’ Advances in Consumer

Research, 36, pp. 332- 338.

Huisman, C. (2019). ‘Zijn shell-klanten bereid 1 cent extra per liter te betalen, vijf menigen aan de tank.’ De Volkskrant, 17-04. Available at: https://www.volkskrant.nl/nieuws-achtergrond/zijn-shell-klanten-bereid-1-cent-extra-per-liter-te-betalen-vijf-meningen-aan-de-tank~baf0473a/ (Accessed: 6 June 2019). Kang, I.H. and Leliveld, M.C. (2019). ‘When consumers do good and companies take the

credits: studying the effect of who is paying for donations in cause-related marketing campaigns.’ Working paper.

Kang, Leliveld and Ferraro (2019). Face valence titel. Under review.

Koschate-Fischer, N., Huber, I.V. and Hoyer, W.D. (2015). ‘When will price increases associated with company donations to charity be perceived as fair?’ Journal of the

Koschate-Fischer, N., Stefan, I.V. and Hoyer, W.D. (2012). ‘Willingness to pay for Cause-related Marketing: The Impact of Donation Amount and Moderating Affects.’

Journal of Marketing Research, 49(6), pp. 910-927.

Krishna, A., and Rajan, U. (2009). ‘Cause marketing: spillover effects of cause-related products in a product portfolio.’ Management Science, 55(9), pp. 1469–1485.

Lau, G. T. and Lee, S. H. (1999). ‘Consumers' Trust in a Brand and the Link to Brand Loyalty.’ Journal of Market-Focused Management, 4(4), pp. 341–370.

Liu, W. and Aaker, J. (2008). ‘The happiness of giving: The time-ask effect.’ Journal of

Consumer Research, 35(3), pp. 543-557.

Marín, L., Cuestas, P.J. and Román, S. (2016). ‘Determinants of Consumer Attributions of Corporate Social responsibility.’ Journal of Business Ethics, 138, pp. 247-260.

Melero, I. and Montaner, T. (2016). “Cause-Related Marketing: An Experimental Study About How the Product Type and the Perceived Fit May Influence the Consumer Response,” European Journal of Management and Business Economics, 25(3),

pp. 161–167.

Müller, S.S., Mazar, N. and Fries, A.J. (2016). ‘The Cause Matters! How Cause Marketing Campaigns Can Increase the Demand for Conventional over Green Products.’

Journal of the Association for Consumer Research, 1(4), pp. 540-554.

Nguyen, B., Ekinci, Y., Simkin, L. and Melewar, T.C. (2015). ‘The brand likeability scale An exploratory study of likeability in firm-level brands.’ International Journal of

Market Research, 57(5), pp. 777- 800.