MASTER’S THESIS
An Analysis of the Effects of Non-Profits’ Brand and Cause-Related
Transparency on Consumers’ Willingness to Pay for Charitable Products
Minke Kiers – 10678220 June 29th 2015
Under supervision of Drs. Anouar El Haji Second assessor Drs. Dr. Ed Peelen
Master of Science in Business Administration - Marketing Track Faculty of Economics and Business
2 STATEMENT OF ORIGINALITY
This document is written by student Minke Kiers who declares to take full responsibility for the contents of this document. I declare that the text and the work presented in this document are original and that no sources other than those mentioned in the text and its references have been used in creating it. The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.
3 TABLE OF CONTENTS
ABSTRACT ... 5
1. INTRODUCTION... 6
1.1. Research question ... 7
1.2. Theoretical and practical relevance ... 7
1.3. Thesis structure... 8
2. THEORETICAL FRAMEWORK ... 9
2.1. Non-profit marketing ... 9
2.2. Non-profit branding ... 13
2.3. Cause-related transparency for non-profit organizations ... 16
2.4. Combination of brand equity and cause-related transparency ... 20
3. METHODOLOGY ... 22 3.1. Research design ... 22 3.2. Research method ... 22 3.3. Product choice ... 23 3.4. Execution ... 24 3.5. Procedure ... 27 4. RESULTS ... 28 4.1. Sample characteristics ... 28 4.2. Data characteristics ... 28
4.3. Tests for normality ... 29
4.4. Testing the hypotheses ... 32
4.5. Additional findings ... 34
5. DISCUSSION AND CONCLUSION ... 37
5.1. Key findings ... 37
5.2. Contribution to theory and practice ... 38
5.3. Managerial implications ... 40
5.4. Limitations and suggestions for future research ... 40
6. REFERENCES ... 42
4
7.1. Appendix A: Experiment treatments ... 50
7.2.Appendix B: Survey ... 51
LIST OF TABLES AND FIGURES Table 1 ... 10 Table 2 ... 14 Table 3 ... 24 Table 4 ... 25 Table 5 ... 29 Table 6 ... 30 Table 7 ... 31 Table 8 ... 31 Table 9 ... 32 Figure 1... 15 Figure 2... 17 Figure 3... 22
5 ABSTRACT
This study focuses on examining the effects of non-profits’ brand disclosure and cause-related transparency on consumer’s willingness to pay for charitable products. Brand is a widely known concept in the business sector and is considered a tool for organizations to make themselves recognizable amongst consumers and make them distinctive from competitors. Besides, the influence of corporate transparency has been widely analyzed. Research demonstrates increased credibility and profits after firms disclosed information transparently. However, little is known about the effects of brands and the influence of cause-related transparency on non-profit organizations. Moreover, this study aims to fill the gap regarding the knowledge about branding and transparency between for-profits and non-profits. A controlled field-experiment is conducted in an online auction setting, in which consumers placed a bid on a product from a non-profit organization. The auction provided a way to analyze consumers’ true willingness to pay for charitable products. Results show a significant positive effect of the combination of the disclosed non-profit brand and transparency regarding the cause on consumers’ willingness to pay. Therefore it can be concluded that consumers are willing to pay a significant higher price if the non-profit’s logo and cause of the donation are disclosed, compared to a situation where this knowledge is not provided. However, the non-profit used for this study was highly familiar and trusted. Further research is necessary to analyze the effects on different non-profit organizations. This study contributes to the field of non-profit marketing and gives an understanding of non-profit product proposition.
Key words: non-profit branding, cause-related transparency, transparent marketing, charity marketing, charity auctions.
6 1. INTRODUCTION
Non-profit organizations are defined as companies who do not make or are not conducted primarily to make a profit (Oxford Dictionaries, 2015). They nowadays experience difficulties with their fundraising strategies. Consumers are difficult to persuade compared to the for-profit sector, and charity organizations suffer from a decrease in the amount of monetary donations compared to previous decades (McGann & Johnstone, 2005; Schuyt, Gouwenberg & Bekkers, 2013). The declining trend in willingness to donate can be explained by the financial crisis started in 2008, which made consumers more careful with their overall expenses (NRC, 2013; Centraal Bureau voor de Statistiek, 2009; Khan, 2008). However, non-profit organizations are highly dependent on charitable contributions, such as tax money from the local governments and donations from corporations and individuals (Brooks, 2004). Beside the financial crisis, negative publicity has influenced consumers’ perception of non-profit organizations and their genuine commitment to strive to the creation of a better world (Heck & Vennick, 2014). Consumers expect full transparency regarding expenditures and outstanding results with the provided money (Gaventa & McGee, 2013). Meanwhile, annually organizations spend approximately seven percent of the incoming money on marketing objectives, to ensure new sources of incomes, and to enhance brand image (Constandse, Plug & Ter Horst, 2014). This number is below the legally determined maximum of 25 percent in The Netherlands (Schmidt & Kooij, 2014). Another four percent of the income was spent on maintenance and administrative costs, leaving 89 percent of the money to be spent on the actual charitable cause. Donors tend to avoid charities with high overhead costs, consisting of fundraising and administrative costs (Gneezy, Keenan & Gneezy, 2014). Nevertheless, charity marketing is a necessity for non-profit organizations to achieve their goals and receive the desired resources, such as monetary donations (Macedo &
7 Pihno, 2006). But, since it takes noticeably more effort to receive enough resources for the non-profit sector compared to the for-profit sector (Macedo & Pihno, 2006), ideally the percentage of the income spent on marketing expenses should be minimized, to provide most part of the donations to the actual cause (Gneezy et al., 2014).
1.1. Research question
The aim of this research is to analyze the effects of the product proposition of a non-profit organization on consumers’ product valuation. The research question is formulated as follows: ‘To what extent do non-profit’s brand and cause-related transparency affect consumers’ willingness to pay for charitable products?’
In this context brand is considered as the non-profit’s brand name and brand logo (Bennett, 1988). Cause-related transparency is the disclosure of information regarding the non-profit’s cause towards the consumer (Behn, DeVries & Lin, 2010). Willingness to pay is defined as the maximum amount of money, and any price below this stated reference price, an individual is definitely willing to pay for a certain quantity of a product or service (Hanemann, 1991; Wertenbroch & Skiera, 2002; Varian, 1992). A charitable product is defined by the author as a product offered by a non-profit organization of which the total revenue is donated to a charitable cause.
1.2. Theoretical and practical relevance
Within the field of branding, a gap is noticeable between the effects of brands for non-profit organizations and the effects of brands on commercial companies. Moreover, in current literature, a strong brand is assumed to have a positive effect on consumers’ willingness to pay for charitable products, similar to brand effects on commercial products. However, this assumption has not been confirmed with any research. Therefore this study contributes to the academic literature in closing the knowledge gap between for-profit and non-profit branding effects for charitable product in particular. Furthermore, this study develops an understanding
8 of the effects of cause-related transparency on product valuation. Knowledge on corporate transparency has been analyzed in the field of transparent marketing, mainly with a focus on the disclosure of financial information. Cause-related transparency includes mostly positive cause revealing information. If disclosed of the cause increases consumers’ willingness to pay is tested in this study. This study will altogether provide an example of best practices regarding product proposition, which can be adopted by non-profits in their marketing strategy. As such it contributes to the theoretical and practical development of the non-profit branding and transparent marketing approaches.
1.3. Thesis structure
The structure of this thesis is as follows. First, a theoretical framework about the key knowledge on the topic is presented and the hypotheses are described based on the existing literature. Next, in the methodology chapter an understanding of the execution of the experiment and data collection is provided. In the fourth chapter the findings of the experiment are analyzed and presented. Then, in the discussion and conclusion part, the results are evaluated and an answer to the research question is given. This chapter includes the academic contribution of the study and recommendations for further research.
9 2. THEORETICAL FRAMEWORK
Within this chapter the different concepts of the research question are reviewed from current existing literature and the hypotheses that are tested in this study are presented and substantiated.
2.1. Non-profit marketing
The social, non-profit or philanthropic sector is growing fast, but resembling for-profit business practices should be applied to make this sector more lucrative (Shani, 2015). However, effective marketing approaches used in the private, for-profit sector, differ substantially from the public, non-profit sector (Kotler, 1988).
Four key differences between for-profit and non-profit marketing strategies were found by Wright, Chew and Hines (2012). First of all, non-profits have non-economic objectives and do not measure success in terms of profit, in contrast with for-profit businesses. Secondly, non-profits mostly offer services, ideas and social behavior instead of physical products. This is supported by Andreasen (2012) who stated that the commercial sector mainly focuses on the principle of sales of products or services, while the non-commercial sector deals with behavior change, fundraising and corporate developments, aiming for social change or non-profit program. This corresponds with the third difference, representing that no direct payment-receipt linkage is found in the non-profit sector, meaning that an investment made by a consumer or a donor is not directly related to a tangible item (Wright et al., 2012). Last of all, one of the main differences between the commercial- and non-profit marketing approaches is the amount of stakeholders involved as secondary target audiences, whom all hold substantial expectations (see table 1). Andreasen (2012) supports the vision of different marketing approaches for non-profits and for-profit organizations. Non-profits often have to deal with minimal budget and the benefits for target audiences are
10 not directly visible as earlier indicated by Wright et al. (2012). Andreasen (2012) stresses that non-profits find themselves in the most complex context for marketing activities, compared to other types of businesses.
Table 1
Differences between commercial and social/non-profit marketing (Andreasen, 2012).
Feature Commercial Marketing Social/Non-profit marketing
Primary target audiences
Secondary target audiences
Expectations Budgets Tactical freedom Characteristics of key behaviors Limits of offerings Target audience benefits
Customers
Supply chain members Media
Modest Substantial Few limits
Often low involvement Audience indifferent or positive
Few Obvious
Immediate or near term
Downstream people with problem behaviors Upstream enablers Volunteers Donors Corporate partners Substantial Minimal
Close public scrutiny Often high involvement Audience indifferent or opposed
Often considerable Not obvious Far in the future
Nevertheless, charity marketing is a necessity for non-profit organizations to achieve their goals and receive the desired resources, such as monetary donations (Macedo & Pihno,
11 2006). Due to the dissimilarities with the for-profit sector, different and more challenging marketing strategies should be adapted by the non-profit sector (Andreasen, 2012). A comparison of non-profit marketing is made with sports marketing, given that they both focus on inspiration and unique stories (Shani, 2015). Nevertheless, the two are almost never seen together, since sports events require large financial investments, which cannot be justified by public organizations (Shani, 2015).
However, a more common used form of marketing in the non-profit sector is cause-related marketing (Liston-Heyes & Liu, 2013). Cause-cause-related marketing is a type of marketing activity where companies communicate towards the consumer that a certain percentage of the product’s or service’s retail price will be donated to charity (Varadarajan & Menon, 1988). For example, Proctor & Gamble started a collaboration with one of their brands, Pampers, and Unicef, with the aim to encourage consumers to buy Pampers diapers, for the reason that for every purchase a child vaccination was provided in a developing country (Vanhamme, Lindgreen, Reast & Popering, 2012). Another cause-related marketing strategy was started by a group of companies, including Coca Cola and American Express, to collectively collect money to fund the renovation of the Statue of Liberty in New-York City (Ross III, Stutts & Patterson, 2011). Again for purchases from this group of companies, a certain percentage of money was provided to the cause.
These types of partnerships have substantial evidence showing that firms will attract new customers, relations and build a positive brand identity, as the non-profits gain (short-term) funding and broader awareness for the cause (Kotler & Lee, 2005). Also, consumers indicate a more positive attitude towards both the cause as the linked company in cause-related marketing campaigns (Basil & Herr, 2003; Ross III et al, 2011). However, consumers may have different attitudes towards cause-related marketing, ranging from high appreciation to skepticism about the firm’s and/or non-profit’s true intentions (Webb & Mohr, 1998).
12 The downside of cause-related marketing is the degree of dependence that arises for the non-profit organization (Baur & Schmitz, 2012; Polonsky & Wood, 2001). The possibility exists that the partnering firm does not comply with the agreements, or damages the non-profit’s reputation in case the firm’s own reputation is harmed (Polonsky & Wood, 2001). It is also possible that the consumer holds a negative attitude about the organization, and therefore the partnership can harm the attitude towards the charity and cause (Basil & Herr, 2003). An imaginary example proposed by the authors could be a partnership between MADD (Mothers Against Drunk Driving) and an alcohol company to start campaigning against drunk driving. In essence both organizations have the ‘drunk driving’ factor in common. However, the alcohol company can be seen as a facilitator of drunk driving, whereas MADD is a preventer. A miss-match like this can cause a transfer of the negative attitude of the alcohol company onto MADD (Basil & Herr, 2003). On the other hand, cause-related marketing can positively influence consumers’ image in favor of a harmed company after bad publicity (Deshpande & Hitchon, 2002). And, in case the company-cause fit is high, bad attitudes towards a company are outvoted by the positive attitude of the cause and non-profit organization (Basil & Herr, 2003). However, in order to gain this positive effect, brand awareness from the consumer is necessary (Nan & Heo, 2007).
In sum, existing literature describes the requirement of non-profits to adapt to different marketing approaches compared to the for-profit sector. Still, too little best practices are presented that are applicable for non-profits. Partnerships in the sense of cause-related marketing seem to lead to positive outcomes, but limit the non-profit organizations in carrying out their own vision and create higher dependency which causes risks, such as financial dependence and imposed limitations for other collaborations. Instead of focusing on partnerships and external sources of income, non-profits have the ability to make profits
13 themselves. In the next paragraph the importance of non-profit’s own branding and brand value are explained.
2.2. Non-profit branding
Brand is defined by Bennett (1988) as ‘a name, term, design, symbol or any other feature that identifies one seller’s good or service as distinct from those of other sellers’. Because of the high variety of brand definitions in different academic papers, De Chernatony and Dall'Olmo Riley (1998) categorized the brand concept through a meta-analysis in twelve themes; (1) legal instrument, (2) logo, (3) company, (4) shorthand, (5) risk reducer, (6) identity system, (7) image in consumer’s minds, (8) value system, (9) personality, (10) relationship, (11) adding value, and (12) evolving entity. Although overlaps occur between the different elements, following the branding literature, these twelve keywords have been identified best to provide a representation of the branding proposition (Bastos & Levy, 2012).
Brands are not only developed to create familiarity, they can offer a significant value both for the organization as for the consumer (Kamakura & Russell, 1993). The added value caused by the combination of brand awareness, brand loyalty, perceived quality and brand associations is labeled as brand equity (Aaker, 1991).
Brand valuation is considered one of the two main perspectives of brand equity (Feldwick, 1996), mainly used from an economic point of view based on price and profit. The second perspective is brand strength, which consists of the degree of consumers’ attachment to the brand (Feldwick, 1996). A possible third perspective added by Wood (2000) is consumers’ image about the brand, based on associations and beliefs. Image is of importance to organizations in creating differentiating features from competitors (Wood, 2000). Also, the intention to give money to charity is for 31 percent explained by brand image (Michel & Rieunier, 2012).
14 Aaker (1996) created the brand equity ten as a measurement tool to analyze brand equity (see table 2). This measurement tool can be considered to focus on the first perspective introduced by Feldwick (1996), with a financial approach. Three out of the ten elements of Aaker’s brand equity ten are focused on direct profits (price premium, market share and price and distribution indices).
Table 2
The Brand Equity Ten (Aaker, 1996).
Loyalty measures 1. Price premium
2. Satisfaction/loyalty Perceived quality/leadership measures 3. Perceived quality
4. Leadership Associations/differentiation measures 5. Perceived value
6. Brand personality
7. Organizational associations
Awareness measures 8. Brand awareness
Market behavior measures 9. Market share
10. Price and distribution indices
Next to Aaker's brand equity ten, Keller (1993) presents a consumer-based brand equity model, focusing specifically on individual consumer behavior, the second perspective by Feldwick (1996). Keller describes consumer-based brand equity as the differential effect of brand knowledge on consumer response to the marketing of the brand. Brand name is a strong predictor for brand preference, brand usage and brand recognition, and thus for purchase intention (Maison, Greenwald & Bruin, 2004). Keller (1993) states that consumers
15 show different reactions towards a product or service with the same marketing mix, based on the attributed brand (familiar- versus fictional- or no brand). Following this analysis consumer-based brand equity develops if an individual consumer holds favorable, strong and unique associations towards the brand in memory. This means that a familiar brand name is essential to encourage consumer-based brand equity (Lassar, Mittal & Sharma, 1995). Furthermore, high consumer-based brand equity can be translated into financial benefits for the company (Kim, Gon Kim & An, 2003; Stahl, Heitmann, Lehmann & Neslin, 2012).
Keller’s findings are structured in the consumer-based brand equity model (see figure 1). Compared to the Aaker’s brand equity ten, this model focuses mainly on individual attitudes defined as the second perspective by Feldwick (1996).
Figure 1
Consumer Based Brand Equity model (Keller, 2009).
As discussed, consumer-based brand equity is mainly focused on individuals’ brand associations. French and Smith (2013) developed a new measurement tool to analyze the strength of associations. They emphasize that both negative and positive associations should
16 be mapped for brands, together with the strength of these associations, in order to understand the autonomous effects on brand perception.
Much is currently known about the influence of brand equity on organizations’ success, though mainly regarding for-profit organizations. Stride and Lee (2007) explain the equal importance of creating brand equity in for-profit and non-profit organizations. However, the researchers strengthen the idea that non-profit organizations might encounter more difficulties and challenges in creating a strong brand. Consumers are skeptical towards charities and the transactions are usually one-sided. People have difficulties in the valuation of the brand and organization, because the profits for the consumer are difficult to experience. Hence, non-profit organizations have great interest in a compelling marketing strategy, to receive the resources needed to reach their ideological goals (Macedo & Pinho, 2006).
Knowledge about the effects of non-profit brands lags behind in the field of branding and non-profit marketing. Specifically clarification for these types of businesses can provide an understanding of the possibly positive or negative effects of non-profit brands on consumers’ willingness to pay. The following hypothesis is created based on the assumption that brand disclosure will have a positive effect for non-profits on consumer’s willingness to pay as suggested by the literature on for-profit organizations.
Hypothesis 1: Non-profits’ brand disclosure leads to higher consumers’ willingness to pay for charitable products, compared to a context in which the brand is not disclosed.
2.3. Cause-related transparency for non-profit organizations
Corporate transparency is determined as the disclosure of financial-, governmental-, and political corporate information (Bushman, Piotroski & Smith, 2004). Ball (2009) indicates that the definition of transparency consists of three metaphors; (1) transparency can be
17 defined as a public value, embraced by society to counter corruption, (2) transparency is equal to open decision-making by governments and non-profits, (3) transparency is a complex tool of good governance in programs, policies, organizations and nations. Altogether transparency provides accountability, openness, efficiency and effectiveness for an organization. Nowadays transparency is requested by the public and sometimes mandatory by the law (Loewenstein, Sunstein, & Golman, 2013).
In the field of marketing management, Hultman and Axelsson (2007) created a typology of transparency (see figure 2).
Figure 2
Typology of transparency for marketing management research (Hultman & Axelssson, 2007).
In the transparency typology, the degree, the direction and the distribution of transparency lead to four main types of transparency; (1) technological transparency, also
Degree of transparency: High/low Increasing/decreasing Direction of transparency: Uni-/Bidirectional Upstream/downstream Distribution of transparency: Direct/indirect Horizontal/vertical Types of transparency: Technological Organizational Supply Cost/price
18 called attribute transparency, includes information of the quality and features of a product or service, (2) organizational transparency is for example the disclosure of organizational structure or information of the employees, (3) supply transparency, includes information of the supply chain, (4) cost/price transparency, focuses on the distribution of revenues. For non-profit marketing an element is missing in this typology of transparency; cause-related transparency. That is a form of transparency where a non-profit discloses the cause to which the money is donated. It closely relates to both supply transparency and cost/price transparency, but mainly focuses on the final goal of the cash flow.
Furthermore, with regard to non-profits, transparency is of importance due to consumers’ donation decision-making (Behn et al., 2010). Information accessibility is a key determinant in the efficiency of the allocation of resources and economic growth. Koschate-Fischer, Stefan and Hoyer (2012) show that consumers’ willingness to pay increases for products if a link is disclosed between the product’s company and a good cause. This does not apply when a low company-cause fit occurs or in case solely a low percentage of money is donated to the cause. Elfenbein and McManus (2010) support this theory and found that in an auction situation, bids placed on products for charity are on average six percent higher compared to the same product but not marketed as for charity. These results are confirmed by Leszczyc, Qiu, Li & Rothkopf (2013), who analyzed if a premium price would be paid if a certain percentage was donated to charity in an auction setting. Remarkably, the bidding price is only affected if a certain percentage is donated to a charitable cause, in the situation where a fixed amount of money is donated, consumers do not place higher bids (Popkowski & Rothkopf, 2010).
These findings indicate the probability that people are willing to pay more if it is disclosed that (a certain percentage of) the money paid by the consumer is given to a charitable cause. However, these studies all focus on products that are offered by commercial
19 companies, who donate a certain percentage to charity in the sense of the earlier discussed cause-related marketing (Varaderan & Mennon, 1988). Prices do increase if the companies donate a percentage to charity. This increase can be assumed to take place for products from charity organizations itself as well. But, there is no evidence for this effect on non-profit organizations.
Previously it is discovered that consumers prefer to donate money to a charity with low overhead cost, since this will leave more money available for the charitable cause (Gneezy et al., 2014) Elaborating on this, Wong and Ortmann (2013) describe the relevance of a price-information trade-off, where the non-profit displays the ‘charity price’, based on fundraising and administrative costs, to the consumer. Charities who communicate openly about overhead costs are demonstrated to receive more and higher donations from consumers (Wong & Ortmann, 2013). This is explained by the fact that donors or other stakeholders put significant (financial) efforts in non-profits and therefore they want to know if their efforts are used efficiently and spent on the right cause (Behn et al., 2010). Following Behn et al. (2010) transparency is primarily maintained by non-profits in cases where a non-profit has more dept, is a larger organization, has a larger amount of stakeholders that are involved, has a higher compensation expense ratio, and is classified as a higher education institute. It is suggested that this last determinant appeared in the outcomes of the research, because higher education institutes are familiar with a higher level of scrutiny. In case lobbying expenses appeared to be high, non-profits were more reluctant in disclosing (financial) information. This is in contrast with Saxton, Kuo and Ho (2012) who found that transparency in the non-profit sector is more likely to be carried out by smaller organizations, with lower debt. Also, their results show that disclosure of (financial) information is not considered as the most effective means to promote public accountability, as requested by the consumers.
20 Different points of view on transparency in the non-profit sector are displayed. However, no consensus has been formed. Therefore, a gap in knowledge on transparency in the non-profit sector is visible. Also, a more comprehensive typology of transparency should be created to include the effects of cause-related transparency. Based on the available information at this point, the next hypothesis is defined.
Hypothesis 2: Non-profits’ cause-related transparency leads to higher consumers’ willingness to pay for charitable products, compared to a context in which the cause is not disclosed.
2.4. Combination of brand equity and cause-related transparency
Due to technological innovations, such as the internet, combined with the creation of large online social network platforms, nowadays information is widely available for a large group of people (Bakshy, Rosenn, Marlow & Adamic, 2012; Castells, 2011; Fournier & Avery, 2011). The source of the provided information plays a significant role in whether people believe the content of the message. Besides, the true intentions behind the disclosure of information are important (Stone, Woodcock & Wilson, 1996). For example, governmental institutions can obligate organizations to disclose certain information (Loewenstein et al., 2013) or organization can decide to disclose information with reference to the firm themselves (Stone et al., 1996). If organizations decide to disclose information themselves, it is better respected by consumers and possible purchase intentions increase (Stone et al., 1996). In addition, research indicates that organizations who disclose negative information during a crisis about themselves are considered more credible and purchase intentions from the consumers increase (Arpan & Roskos-Ewoldsen, 2005). But, Fournier and Avery (2011) describe a paradox created about companies who actively interact with their customers. Pushing too much information can create aversion towards the brand, which can lead to a
21 decrease in consumers’ purchase intention. On the other hand, does consumers’ willingness to pay increase if people sense a personal relationship with the cause or organization (Van der Westen, 2006). This could mean that consumers are even willing to pay an extra premium if the product is offered by a charity and if they support the cause. It is expected that if a non-profit discloses both its brand logo and transparently communicates about the cause, this will lead to a higher willingness to pay compared to a situation where no information is disclosed. This assumption is formulated as the third hypothesis.
Hypothesis 3: The combination of non-profits’ brand disclosure and non-profits’ cause-related transparency leads to higher consumers’ willingness to pay for charitable products, compared to a context in which the brand and the cause are not disclosed.
How the above introduced hypotheses are tested in this study, is described in the methodology chapter.
22 3. METHODOLOGY
The research question ‘To what extent do non-profit’s brand and cause-related transparency affect consumers’ willingness to pay for charitable products?’ was tested with a controlled field experiment. In this chapter the different aspects, such as the research design, method, execution and procedure of the experiment are described.
3.1. Research design
This study has a 2 (brand: non-disclosed versus disclosed) x 2 (cause-related transparency: non-disclosed versus disclosed) factorial design. The dependent variable is consumers’ willingness to pay. Figure 3 shows the conceptual model which includes the three hypotheses.
Figure 3
Conceptual model of research.
3.2. Research method
This study was conducted with a controlled field experiment. A sealed second-price Vickrey auction was used as a measurement tool to examine consumers’ purchase behavior and their
Brand Cause-related transparency Bran + Cause-related transparency + H1 + H3
23 maximum willingness to pay (Vickrey, 1961). The auction was held in a real-life setting, since people really have to pay if they win the auction, but was executed online. This ‘real-life’ type of data collection provides insight in consumers’ actual behavior instead of solely intentional behavior. The Vickrey auction is designed in such a way that the participant with the highest bid will eventually pays the second, or lower, highest bid, and is aware of this before placing the bid. Extensive research on the effectiveness and reliability of a Vickrey auction demonstrates that offering the second or lower highest bid to consumers, leads to a situation where customers indicate their absolute maximum willingness to pay (Myerson, 1981; Shogren, Hayes, Kliebenstein & Fox, 1994). A Vickery auction provides insight in consumers’ purchase behavior and emphasizes the influence of the provided knowledge towards customers about not paying their actual bid.
3.3. Product choice
For this study a collaboration with non-profit organization Oxfam Novib was formed and an existing product was auctioned. Oxfam Novib is the Dutch department of the world-wide operating organization Oxfam (Oxfam Novib, 2015). The organization strives to empower people to protect themselves against starvation. They try to find practical an innovative ways for these people to ensure themselves of enough food now and for the future. The product that was offered is a plate made of bamboo with a design of the different life-styles around the world, which can be used to serve food. It is named ‘Bamboe Wereldbord’. The product is available since November 2014 in the web shop at a fixed selling price of 10.00 Euros. The visualization of the product in the auction of this study closely resembles the way it is presented in the actual web shop. This offers the chance to analyze the effectiveness of the current product’s proposition.
24 3.4. Execution
Willingness to pay can be used as a measurement tool for product valuation and determines individuals’ perceived value for a certain product (Miller, Hofstetter & Zhang, 2011). This dependent variable was tested in the online environment of Veylinx, an auction platform for (market) research, developed at the University of Amsterdam (Veylinx, 2015). Veylinx is launched with the aim to provide a better understanding of customers’ valuation of products and services. In the experiment, the product was tested in four different treatments with different applied manipulations, which resulted in a certain willingness to pay based on a bid amount in eurocents. All different treatments have a picture of the bamboo plate, the product’s name and its features, such as; ‘the plate symbolizes the different lifestyles in the world’ and ‘the plate has a diameter of 21 centimeters and is dishwasher proof’ (see appendix A, T1: control condition). The manipulations that were applied for testing variables brand and cause-related transparency are visible in table 3.
Table 3
Applied treatment manipulations to test the independent variables on the dependent variable.
Variables Dimensions Disclosed Non
-disclosed
Brand Brand logo -
Cause-related transparency
Disclosed cause The profits of this plate are provided to give people the chance to ensure themselves of sufficient food and own income.
25 To test the non-profit’s brand effects, the brand logo of Oxfam Novib is displayed in the auction. Oxfam Novib uses different forms of its logo with minor differences, such as with or without the organizational slogan; ‘ambassadors of doing it yourself’. For this study, in order to check forthe autonomous effects of the brand and cause-related transparency on willingness to pay, it was decided to present the logo including brand name without the accompanying slogan. This decision was made, because the additional line discloses part of the Oxfam Novib’s cause, which retains the possibility to check for independent relationships of brand and cause-related transparency. The effect of cause-related transparency was tested in the auction by disclosing the additional text: ‘the profits of this plate are provided to give people the chance to ensure themselves of sufficient food and own income’. This text provided information to the consumer about what cause the paid money was spent on. An overview of the four different experimental treatments is displayed in table 4.
Table 4
An overview of the different treatments.
Treatment Brand Cause-related
transparency
T1: control condition - -
T2: disclosed brand + -
T3: disclosed cause - +
T4: combination + +
In the control condition (T1), the product was displayed without any further manipulations and is therefore labeled as the control condition. Treatment 2 (T2) contains the
26 general information plus the Oxfam Novib logo and is called the disclosed brand treatment. On the contrary, treatment 3 (T3) has no logo, but includes the additional text, which is named the disclosed cause treatment. Ultimately, the fourth treatment (T4) is composed of a combination of the Oxfam Novib logo and the additional cause disclosing text. Therefore it is labeled as the combination treatment.
In order to check whether the actual effects are being measured, five additional questions were asked to the participants after they had placed their bid (see appendix B.). For the variable willingness to pay the question, ‘are you familiar with the retail price of the ‘Bamboe Wereldbord’ (answer options: yes/no), was posed to control if the bid amount is affected by this knowledge because a reference price can influence consumer’s true willingness to pay either negative or positive (Gneezy, Gneezy, Nelson & Brown, 2010).
In addition, based on the theoretical framework of brand equity (Aaker, 1996), two questions were asked to control for participants’ brand awareness on Oxfam Novib; ‘are you familiar with Oxfam Novib?’ (answer options: yes/no) and brand attitude on Oxfam Novib; ‘I expect my money to be spent on the right cause by Oxfam Novib’ (5-point Likert scale answer option, varying from 1 = totally agree to 5 = totally disagree). The image the participants hold towards Oxfam Novib might influence the dependent variable willingness to pay.
The last statement was likewise used for checking on cause-related transparency regarding the trustworthiness of Oxfam Novib, because trust is an important element of transparency (Kanagaretnam et al., 2010; Kirby, 2012). Secondly, the statement ‘I think it is essential to know beforehand to what cause the money is provided’ (5-point Likert scale answer option, varying from 1 = totally agree to 5 = totally disagree) was presented to the participants. Transparency is considered to be of great importance to the consumers and increases organizational profit (McKinney & Benson, 2013; Peppers & Rogers, 2012).
27 Finally an extra question was asked about the frequency with which people donate money to charity (answer options: daily/weekly/bi-weekly/monthly/annually/hardly or never). Based on the theory of planned behavior, it was expected that people who donate to charity more regularly are more eager to place a bid higher than 0 in this auction, because they are familiar with this charitable behavior (Ajzen, 1985).
3.5. Procedure
Through the platform, subscribed people were notified and invited by e-mail to participate in an online auction. They were asked to place one sealed bid in the online auction within a time limit of six minutes. Participants in the experiment were randomly allocated to one of the four different treatments of the experiment. People were encouraged to place a bid of 0 Euros in case they were not interested in the product. However, it was emphasized that they engage in a purchase obligation from the moment they bid over 0 Euros and win. Participants were not able to see bids placed by others in the auction. After placing the bid, the participants were asked to answer a 5-question survey, where all questions were displayed at the same time.
Once the auction was closed and the final selling price was determined, all participants were informed about whether they won the auction or not, and how much the winners actually needed to pay for the product, which was always less than their initial bid. The winners were asked to pay within 48 hours via the Veylinx platform, and to provide their personal details in order to ship the product. They all paid in time. The product was sent to the people shortly afterwards.
For this study, the auction started on Monday the second of June 2015 at 09.00 in the morning, at this time people were notified by e-mail, and the auction was closed on Tuesday the third of June 2015 at 21.00 in the evening. During a period of 36 hours people had the ability to place a bid. In the next chapter the sample of the study is discussed among others.
28 4. RESULTS
The results chapter focuses on the outcomes of the experiment. First, the sample and data characteristics are described. After which a test for normality specifies the way the data set is distributed. With this knowledge an one-way ANOVA and regression analysis are applied to the data set to test the three hypotheses. Finally, short conclusions are formed based on the results.
4.1. Sample characteristics
A total of approximately 5000 people were notified about the auction by an e-mail sent from Veylinx of which 647 people participated in the auction (a response rate of approximately 13 percent). The total sample consisted of 307 (47 percent) male and 340 (53 percent) female persons, with an average age of 39.54 years (SD = 13.60). The youngest participant had an age of 15, the oldest participant was 86 years old. Before conducting the analyses, the data set was checked for missing values. All participants provided their gender and year of birth. Furthermore, all participants placed a bid amount. However, a total of 70 participants did not fill out the survey after the auction, and 4 persons partially completed the survey. The missing values are not considered as a concern. All participants placed a bid amount and the number of participants who completed the survey is high enough for valid analyses (Field, 2013).
4.2. Data characteristics
Since different treatments were compared in this study, an overview of the data within the different treatments and the total sample is given in table 5. As shown, the data of the variable bid amount fluctuated between 0 and 3000 eurocents. The average bid was 185.25 (SD = 315.59). Except for the control condition, all other treatment group means were above the mean of the total sample.
29 Table 5
Overview of the number of participants, average bid amount, and minimum and maximum values in eurocents within the different treatment groups.
Treatment N Min Max M SD
T1: control condition 160 0 1250 119.21 227.03
T2: disclosed brand 166 0 3000 195.92 367.02
T3: disclosed cause 160 0 1550 211.22 299.38
T4: combination 161 0 2500 214.08 341.93
Total 647 0 3000 185.25 315.59
To control for biased outcomes, it was asked if people knew the retail price of the bamboo plate. A total of 16 people indicated to be aware of this price, which is 2 percent of the total sample (3 valid percent, controlled for missing values). This group of people placed an average bid of 209.37 (SD = 346.52). This mean did not significantly differ from the mean (M = 183.62, SD = 297.87) of the sample without this price knowledge (p > .05). Therefore all bid amounts within this experiment could be considered as equally reliable and unbiased (Field, 2013).
4.3. Tests for normality
The data set needed to be normally distributed to be used for accurate parametric analyses in SPSS. A normal distribution of the data should not be flat nor peaked (Field, 2013). Different tests demonstrated the distribution of this study’s data set. First, the total bidding behavior was visualized in a nominal distribution hologram. The graphical observation illustrated the skewedness of the data to the right, which indicated a positive skew, meaning the median was possibly greater than the mean of the data and therefore the data seemed non-normally
30 distributed (Field, 2013). Further, a numerical interpretation was done with the Kolomogorov-Smirnov and Shapiro-Wilk test to check for a significant deviation from normality (see table 6). Accordingly the data set was non-normal distributed; it deviated significantly from normality (p < .001).
Table 6
Tests of normality of variable bid amount.
Kolomogorov-Smirnova Shapiro-Wilk
Statistic Df Sig Statistic Df Sig
Bid amount (N = 647) .279 647 .000 .638 647 .000
a. Lillefors Significance Correction
The non-normal distribution of the data could be explained by the participants’ bidding behavior. A total of 338 participants (52 percent of the total sample) placed a bid of 0 eurocents, which was the lowest possible value. These bidders showed no interest in the product by offering 0 eurocents. However this large amount of 0 values strongly affected the direction of the data set and the analyses.
Before further statistical analyses a natural logarithm was applied to dependent variable bid amount, where the bid amounts of 0 were disregarded and only values above the median were taken into account. This transformed the collected information into a relatively more balanced and normal distributed data set. The hologram of the normal distribution of the variable bid amount, after the natural logarithm was applied, showed a slightly more balanced normal distribution. On the other hand, the Kolomogorov-Smirnov and Shapiro-Wilk of the normal logarithm bid amount indicated a significant difference between the data set and a normal distribution (p = < .001). Nevertheless, the data was less divergent following
31 the Skweness and Kurtosis test, compared to the total data set (table 7). Finally, it could be concluded that the data was highly skewed and kurtotic and it differed significantly from normality.
Table 7
Tests of normality of variable natural logarithm bid amounts above median.
Kolomogorov-Smirnova Shapiro-Wilk
Statistic Df Sig Statistic Df Sig
Nat. log. bid amount (N = 272)
.148 272 .000 .815 272 .000
a. Lillefors Significance Correction
Table 8
Skewness and Kutosis test on variable bid amount and natural logarithm bid amount.
Skewness Kurtosis
Statistic Std. Error Statistic Std. Error Bid amount
(N = 647)
3.087 .096 16.006 .192
Nat. log bid amount (N = 272)
-2.162 .148 7.730 .294
In table 9 an overview of the data within the different treatments and the total sample, where the natural logarithm was applied, is given. Now, the data fluctuated between 0 and 7.82, because the data is displayed in natural logarithm instead of eurocents. The average bid
32 amount of the total sample was 5.543 (SD = 1.180). In this situation both the control condition (T1) and the disclosed brand treatment (T2) were below the total mean. The highest bid was placed in the combination treatment (T4) and the lowest maximum value was located in the control condition (T1).
Table 9
Overview of the number of participants, average bid amount, and minimum and maximum values in natural logarithm within the different treatment groups above median.
Treatment N Min Max M SD
T1: control condition 65 0 7.13 5.152 1.262
T2: disclosed brand 63 0 7.44 5.507 1.438
T3: disclosed cause 73 .69 7.35 5.708 1.038
T4: combination 71 2.30 7.82 5.762 .879
Total 272 0 7.82 5.543 1.180
4.4. Testing the hypotheses
Based on the outcomes of the tests for normality, the data set to which a natural logarithm was applied and only the values above the median were taken into account was best to use for further analyses.
An one-way analysis of variance (ANOVA) was conducted to check whether there was any type of mean variety between the different treatment groups (F(3, 271) = 3.800, p = .011). The results showed a significant difference between the means of the bid amounts in the treatment groups. Next, a post-hoc multiple comparison Bonferroni test offered insight in where the differences exactly lie.
33 The mean difference was significant (p < .05) between the control condition (T1) and the combination treatment (T4), where both the brand and the cause were disclosed. With a mean difference of .610, participants in the combination treatment group, who have been exposed to both the Oxfam Novib logo and the cause, were willing to pay a significant higher amount of money compared to the control condition. The mean difference between the disclosed brand treatment (T2) and the combination treatment (T4) was also significant (p < .05), which indicates that the additional text disclosing the cause significantly increased consumers’ willingness to pay over the disclosed logo. The means of the control condition (T1) and the disclosed cause treatment (T3) did not differ significantly from each other (p = 1.00). Therefore the independent positive effect of cause-related transparency on willingness to pay could not be regarded as highly causal.
The first hypothesis was stated as, ‘non-profits’ brand disclosure leads to higher consumers’ willingness to pay for charitable products, compared to a context in which the brand is not disclosed’. The non-profit’s brand effect was tested by disclosing or not disclosing the Oxfam Novib logo. The logo was displayed both in the disclosed brand treatment (T2) as in the combination treatment (T4). The first hypothesis was tested in the disclosed brand treatment (T2) and was compared to the control condition (T1). Since consumers in the disclosed brand treatment (T2) were not willing to pay significantly more than consumers in the control condition (T1), hypothesis 1 was rejected. Thus, the influence of the Oxfam Novib logo on willingness to pay did not significantly differ from the control condition, where no logo was adjusted to the product (p = 1.00).
Next, the second hypothesis, ‘non-profits’ cause-related transparency leads to higher consumers’ willingness to pay for charitable products, compared to a context in which the cause is not disclosed’ was measured with analyzing the difference between the control condition (T1) and the disclosed cause treatment (T3). This hypothesis was rejected as well,
34 since consumers in the disclosed cause treatment (T3) were not willing to pay significantly more than consumers who place a bid in the control condition (T1). There seems to be no difference in consumers’ willingness to pay between disclosing the cause to the consumer compared to the control condition.
On the contrary, the third and last hypothesis, ‘the combination of non-profits’ brand disclosure and non-profits’ cause-related transparency leads to higher consumers’ willingness to pay for charitable products, compared to a context in which the brand and the cause are not disclosed’ was confirmed. This hypothesis was tested by comparing the control condition (T1) to the combination treatment (T4) and encompassed both the Oxfam Novib brand name and logo and revealed the charitable cause. This outcome confirmed the positive influence of disclosing both the brand and the cause on willingness to pay for charitable products. People who were exposed to these conditions placed a significantly higher bid amount (p = .015).
4.5. Additional findings
Next to the data generated from the bid amounts, conclusions were drawn based on the additional survey questions asked at the end of the experiment. First, familiarity with Oxfam Novib was tested with one of the survey questions. The results showed that 14 percent of the total sample (16 valid percent, controlled for missing values) was not familiar with Oxfam Novib at the time of the auction. Thus, 75 percent (84 valid percent) did know the organization beforehand. The participants who did not know the organization (N = 91) placed an average bid of 114.46 (SD = 249.042), while the people who knew the organization (N = 484) offered a mean bid amount of 198.13 (SD = 305.961). People who were familiar with Oxfam Novib placed significantly higher bid amounts compared to the unknowns (p < .05).
The answer options for the other survey questions were between 1 (totally agree) and 5 (totally disagree). For analysis it was more logical to reverse code the data to have a higher
35 value answer, linked to a more positive attitude. All 5-point Likert scale questions were reversed coded into new variables.
Next, with a regression analysis, the effects of the control questions about the degree of trust, the valuation of knowledge about the cause, and previous charitable behavior, were checked on consumers’ willingness to pay. In this regression analysis control variables gender and age were also included.
First, the effect of Oxfam Novib’s trustworthiness, whether it was expected by the participant that Oxfam Novib spend donated money to the right cause, was checked on bid amount. The biggest group of the sample (42 percent, 47 valid percent) reported that they agreed or totally agreed on this statement and therefore expected Oxfam Novib to spend the money to the right cause. The regression analysis showed a significant relationship between the degree of trust and bid amount in natural logarithm above median (β = .274, t(269) = 4.454, p < .01). Meaning, that the stronger the degree of trust, the higher the bid amounts were, and thus the more money people were willing to pay for the charitable product.
Next to the degree of trust, it was asked how people value the importance of cause disclosure. The substantial majority of the sample (73 percent, 82 valid percent) agreed or totally agreed on the statement that it is important to know beforehand to what cause the money is provided. A small minority (6 percent, 7 valid percent) disagreed or totally disagreed on the importance of cause-related transparency. The results of the regression analysis demonstrated a non-significant effect (β = -.008, t(269) = -.112, p > .05). Therefore the degree of importance consumers attach to transparency of the cause did not significantly cause change in consumers’ willingness to pay.
Lastly, participants were checked on their charitable behavior. In total 135 people (21 percent, 23 valid percent) indicated to rarely or never donate money to charity. The largest group (32 percent, 36 valid percent) answered to donate money on a monthly basis. Merely 9
36 percent pointed out to donate more often than monthly. The mean bid amount of the group who rarely or never donated to charity was 94.13 (SD = .414). The participants who donated more than monthly (bi-weekly or weekly) had an average bid amount of 215.07 (SD = 339.843). The outcomes of the regression analysis showed that the frequency of making charitable donations did not lead to a significant difference in consumers’ willingness to pay (β = .067, t(269) = 1.229, p > .05).
The data was checked for the effects of control variables gender and age. The variables gender (β = .173, t(269) = 1.203, p > .05) and age (β = .173, t(269) = 1.203, p > .05) did not significantly affect consumer’s willingness to pay. Consequently, gender and age did not affect the relationships measured of the other items on consumers’ willingness to pay.
37 5. DISCUSSION AND CONCLUSION
After analyzing the results of this study, in this chapter a general discussion and conclusion is provided regarding the hypotheses. Besides, the implications of the research and recommendations for further studies are specified.
5.1. Key findings
The question, whether non-profits’ brand and cause-related transparency affect consumers’ willingness to pay for charitable products, was positively confirmed. Although, the two independent variables, disclosed brand and cause-related transparency, did not significantly affect consumers’ willingness to pay. Only the combination of them shows a significant difference compared to the control condition. Besides, a significant difference was found between the disclosed brand treatment and the combination treatment, where both the brand as the cause were disclosed. This outcome revealed the strength of cause-related transparency on top of brand disclosure. Still the independent effect of cause-related transparency did not result in a significant difference from the control condition. Thus, the measured effect between the disclosed brand treatment and the combination treatment remains questionable. Also, brand familiarity resulted into higher bid amounts, and the amount of money people were willing to pay depended significantly on the degree of trust the consumers’ held towards Oxfam Novib. Moreover, it can be stated that the participants did notably thought it was essential to know beforehand to what cause the money is provided by the non-profit. However, no effect was found between the different treatment groups and the level of importance people indicated. This means that people who were allocated in a non-disclosed cause treatment, did not place significant different bid amounts compared to those who were exposed to the cause. Lastly, previous bidding behavior, gender and age were considered as of no influence on willingness to pay within this data set.
38 5.2. Contribution to theory and practice
The participants who indicated to be familiar with Oxfam Novib were willing to pay a significant higher amount of money. This is in line with the studies of Lassar et al. (1995) and Maison et al. (2004), who emphasize the positive influence of brand familiarity on consumer purchase intentions. However, the bid amounts people placed were not affected by the disclosure of brand name and logo, which is in contrast with the literature. An explanation can be provided by the size of the group. The people who did not know the brand was quiet small, consequently the outcome of the comparison may be biased. Also, the degree of familiarity was only tested with a yes or no question, and no measurement scale on familiarity was conducted. Measurement based on a certain scale would have provided a broader perspective on the degree of knowledge individuals have and how the degree of knowledge affected the ultimate willingness to pay.
Alongside the degree of brand familiarity, Keller (1993) describes the importance of associations that individuals’ hold towards the brand. The outcomes of this study suggested that the majority trusted the organization and expected their money to be spent on the right cause. This can be concluded as positive associations of consumers towards Oxfam Novib. This confirms Keller’s own findings and indicate positive purchase outcomes. This study confirms the equal importance of brands for non-profits as for for-profits, in line with the statement made by Stride and Lee (2007). But the frequency of charitable giving did not significantly predict people’s willingness to pay in this study, contrasting the theory of planned behavior (Ajzen, 1985). Also, Leszczyc et al. (2013) showed a predictable consistency in the bidding behavior of bidders in future auctions. On the contrary, Johnson, Peck and Schweidel (2014) recently found that organizations are not able to identify future donors, based on consumers’ past purchase information. This study contributes to the
39 outcomes of Johnson et al. (2014). It is suggested to prolong analyzing purchase behavior for non-profits in the future, since different outcomes are reported.
Subsequently, cause-related transparency did not seem to have a significant effect on consumer’s willingness to pay, which contrasts the positive effects demonstrated in existing literature on corporate transparency (Bushman et al., 2004). Nevertheless, this type of transparency differs from corporate transparency. While corporate transparency mainly focuses on financial accountability and disclose possible negative aspects of the company, cause-related transparency shows the positive side of the non-profits. It is suggested that further definition and conceptualization is needed to reach a sufficient understanding of cause-related transparency. A possible understanding of the lack of effects can be supported by the study of Van der Westen (2006) suggesting that people need to relate to the organization and/or cause to develop a purchase intention. No questions were asked about the perceived relationship between the participants and the cause. Also, Zagefka, Noor & Brown (2013) explain the importance of area knowledge of the cause. Oxfam Novib focuses on developing countries, where people deal with starvation. For Dutch people, this is far from home and usually an unknown area and thus difficult for the Dutch consumer to relate to. Different outcomes can be expected if consumers are familiar with the area of the cause (Zagefka et al., 2013).
As a final point, the combination effect of brand and cause disclosure on consumers’ willingness to pay was expected to be positive, due to the fact that the non-profit organization provides additional information to the consumer (Stone et al., 1996). The findings of this study reject the paradox, introduced by Fournier and Avery (2011), that giving too much information would create aversion. The study contributes to the outcomes described by Arpan and Roskos-Ewoldsen (2005) that organization’s credibility not only increases in situations where negative information is disclosed, but also in cases of positive information.
40 5.3. Managerial implications
E-commerce already enhances non-profit organizations to find and notify the right people to donate (Lee, Berneiker & Van Wyhe, 2006). The online advertisement possibilities lower marketing expenses. With this study additional knowledge is provided on how non-profits should position their products and therefore, on how to establish a stronger marketing strategy. It is suggested that if a non-profit has a relatively strong brand (familiar and trustworthy), it is best to disclose the brand’s name and logo in the product advertisement. But, not only the brand should be disclosed, also the cause should be included. The combination created synergetic value in this study. This direct improvement might lead to higher revenues for non-profits in the present and decreased marketing expenses in the future.
5.4. Limitations and suggestions for future research
Although this research shows significant effects, the research was limited to a five-question survey and the usable data set only consisted of 272 participants. Whether it is representative for the Dutch population as a whole is questionable. In addition, a pre-test to check for preliminary attitudes towards different Dutch non-profit organizations would have been supportive to this study. After this study an understanding is formed of the reputation of Oxfam Novib, which seems positive. However, it is not known how people look towards other non-profits and if the Oxfam Novib study is generalizable for the non-profit sector as a whole. Also, multiple participants indicated that they placed a bid of 0 Euros due to the fact that they do not see the point of buying one separate plate. These people suggested that they would have placed a higher bid if the plates were offered as a set of 2, 4 or 6 plates. Therefore, this might have negatively influenced the outcomes.
Altogether, this study provides a first step in reaching an understanding of the importance of brands and cause-related transparency for non-profit organizations. However, further research should be conducted to resolve remaining questions. For example, this study
41 is executed in collaboration with Oxfam Novib, focusing on humanitarian aid. It is possible the outcomes would have been different in case the cause involved an environmental issue (Bennett, 2003). Moreover, complying with Lee and Chang (2007) and following Hibbert and Horne (1996), personal characteristics may have a predictive function to which cause people prefer to donate money to and how this decision-making process is different for each individual. This research did not focus on personal characteristics; however this could be valuable to understand individual bidding differences in the future. These results were controlled for gender differences. However, against the expectations of Nelson, Brunel, Supphellen and Manchanda (2006), no differences were found. Furthermore, it is proven that product type influences purchase behavior (Strahilevitz & Meyers, 1998). Overall, sustainable, hedonic or even symbolic products seem to lead to the highest purchase intention against the highest price (Briers, 2007; Strahilevitz & Meyers, 1998; Tully & Winer, 2014).
42 6. REFERENCES
Aaker, D. A. (1991). Managing brand equity: Capitalizing on the value of a brand name. Free Press: New York.
Aaker, D. A. (1996). Measuring brand equity across products and markets. California management review, 38(3), 102-120.
Ajzen, I. (1985). From intentions to actions: A theory of planned behavior. Springer: Berlin. Arpan, L. M., & Roskos-Ewoldsen, D. R. (2005). Stealing thunder: Analysis of the effects of
proactive disclosure of crisis information. Public Relations Review, 31(3), 425-433. Bakshy, E., Rosenn, I., Marlow, C., & Adamic, L. (2012). The role of social networks in
information diffusion. Proceedings of the 21st international conference on World Wide Web, 21(1), 519-528.
Ball, C. (2009). What is transparency? Public Integrity, 11(4), 293-308.
Basil, D. Z., & Herr, P. M. (2003). Dangerous donations? The effects of cause-related
marketing on charity attitude. Journal of Non-profit & Public Sector Marketing, 11(1), 59-76.
Bastos, W., & Levy, S. J. (2012). A history of the concept of branding: practice and theory. Journal of Historical Research in Marketing, 4(3), 347-368.
Baur, D., & Schmitz, H. P. (2012). Corporations and NGOs: When accountability leads to co-optation. Journal of Business Ethics, 106(1), 9-21.
Behn, B. K., DeVries, D. D., & Lin, J. (2010). The determinants of transparency in non-profit organizations: An exploratory study. Advances in Accounting, 26(1), 6-12. Bennett, P. D. (1988). Dictionary of marketing terms. The American Marketing Association:
Chicago.
