The Reciprocity Effect: How Corporate Transparency Leads To Voluntary
Sharing Of Personal Data
Master Thesis
Author: Rosanne L.W. de Vos (5884411)
University of Amsterdam, Faculty of Economics and Business January 21, 2014
Table of contents
Abstract………...3
Introduction………...4
Theoretical framework………....6
Transparency………...7
The theory of reciprocity……….9
Reciprocity in company-consumer relationship………....10
Information source………11 Brand trust……….12 Information type………13 Privacy concerns………...15 Method...………..15 Results………...19 Discussion……….26
Limitations and future research……….29
ABSTRACT
Nowadays, transparency is the key to successful marketing. The more information a company has about a particular customer, the more value that company will be able to provide to that consumer. However, collecting consumer information has become increasingly difficult. As a result of several privacy scandals and misuse of data, consumers are more protective of their personal information more than ever before. Current research tries to identify conditions that increase consumer disclosure behavior. It includes (1) the theory of reciprocity to investigate whether corporate transparency can trigger increased willingness to disclose personal data and adds (2) the company’s motive for information collection. An online survey has been conducted, in which over 200 participants were asked to share their personal data with a large food producer. A total of 168 responded in a reliable way and their reactions were tested using advanced correlation analysis that included logistic regression. The results provided support for two alternative ways that allow companies to collect personal information without violating the relationship with the customer. These are being discussed with specific attention for brand trust and privacy concerns.
INTRODUCTION
Consumer information is the key asset of every 21st century business. The better one knows each customer, the higher the potential to improve customer value and firm performance (Moon, 2000). Indeed, the more information a company has about a particular customer, the more value that company will be able to provide to that consumer (Kauffman, Lee, Prosch and Steinbart, 2011). Collecting personal data has several advantages, including a more effective offering, more efficient communication with the consumer and enhanced profitability (Pitta, Franzak & Laric, 2003). Therefore, in order to initiate and maintain long-term relationships with preferred customers, acquiring information about them is inevitable (Hansen, 2003).
With the possibilities that transparency offers, challenges come as well. Consumer privacy and protection of personal information are among the forefront of emerging complex social issues (Glazer & Straus, 2014). Nowadays, more and more ‘smart products’ are being developed, that due to the application of information technology are able to collect, process and produce information on the user. From a consumer’s perspective, this provides several disadvantages. The bottom line is that consumers are frightened that their privacy is at stake as a result of the extensive information gathering. Indeed, research has shown that consumers are becoming increasingly protective of the information they disclose (Zimmer et al., 2010). In the last years, several privacy scandals have led to concern among consumers on how personal information will be used (Vijayan, 2004). People are worried about the misuse of their personal information and fear online identity theft (Cranor, 1999). In summary, consumers continue to express concerns about loss of privacy (Norberg & Horne, 2007). So, despite the growing demand for consumer information, consumers are increasingly aware of the benefits of their personal information and are more reluctant to freely disclose (Zimmer et al., 2010). More and more, consumers see potential risks of sharing their personal information (Hagel & Rayport, 1997). Understanding the value that individuals assign to the protection of their personal data is therefore of fundamental importance (Acquisti, John & Loewenstein, 2013). In fact, a firm that does not protect its customers’ private information jeopardizes its relationship with them (Kauffman et al., 2011).
With increasingly stringent legislation regarding information collection and the ever more conscious consumer, how can marketers ensure that consumer data remains available in the future? Of course, there is the opportunity to purchase personal data from so called ‘data brokers’, companies that gather, process and sell personal information. However, if society finds out an organization uses such a phenomenon this can permanently damage the corporate
reputation. Fortunately, there are several other opportunities that can reduce the barriers in self-disclosing personal information. Zimmer, Arsal, Al-Marzouq, Moore & Grover (2010) assume that organizations can create or implement conditions to increase disclosure behavior. Prior research has demonstrated that that consumer trust (Culnan & Armstrong, 1999) and a positive attitude towards the company (Jarvenpaa & Tractinisky, 1999) are able to increase the tendency to self-disclose. Current research provides two alternative conditions, namely the theory of reciprocity and the company’s motive for personal data collection.
The theory of reciprocity is added as a potential construct that could evoke self-disclosure. Reciprocity is a widely researched phenomenon in social psychology that can be defined as ‘the conditional behavior to return helpful and harmful acts’ (Stanca, 2009) and implies an exchange of information (Moon, 2000). It appears that behavior exchanged between individuals is expected to be of comparable value in order to maintain a norm of reciprocity (Gouldner, 1960). Reciprocity is incorporated to the research to investigate if it can work as a mechanism for consumer information disclosure. Can it be demonstrated that when organizations are transparent about their activities, consumers also tend to self-disclose? Furthermore, perceived privacy protection appears to decrease self-disclosure barriers. Metzger (2006) demonstrates that revealing information collection procedures and privacy statements may increase trust and therefore increase self-disclosure. Current research is trying to replicate these findings by disclosing the organization’s motive for data collection to the consumer. It is expected that disclosing the organization’s motive will decrease self-disclosure barriers.
When the proposed research question can be affirmatively answered, it would have an important managerial contribution. If organizational transparency leads to the tendency of the consumer to self-disclose, this mechanism can function as an instrument to gain more consumer knowledge. By being self-transparent, companies could make their consumers more transparent, resulting in numerous advantages. They would be able to customize and innovate their offering in a way that increasingly match consumer needs (Zimmer et al., 2010). This results in competitive advantage in today’s marketplace and is therefore highly valued. Besides managerial contributions, present research also tries to contribute to scientific knowledge about information disclosure in the area of transparency research. The inclusion of reciprocity from the discipline of social psychology is an interesting addition to the existing research on information privacy, which mainly occurs within the marketing division. More specific, by investigating the privacy dilemma from multiple disciplines, this research offers
potential marketing-related solutions (such as disclosing organizational motives) and potential psychological solutions (such as the principle of reciprocity).
Current research is constructed in the following way; first of all, a theoretical framework explains the variables included in the study, namely transparency, reciprocity, information type and information source. Additionally, their hypothesized relationships are outlined. Finally, the research method, the results and the discussion are displayed.
THEORETICAL FRAMEWORK
The growing interest in obtaining consumer information can be explained by three major developments in society, which, in the context of current research are interesting to mention.
Primarily, the rise of the Internet and Social Media has led to multiplicity of information. Availability of and access to information is one of the more empowering and revolutionary forces of the Internet (Fournier & Avery, 2011). The Internet facilitates two-way conversations between organizations and consumers, shows in-depth information about consumer preferences and lifestyles and gives consumers power over companies through the millions of options that they can now choose from online. The advent of the Internet has led consumers to be able to share their experiences on social media, forums and rating and comparison websites (Granados, Gupta & Kauffman, 2009). As a result, consumers have become better educated, more sophisticated, demand more of companies and are more difficult to influence (Labreque, Esche, Mattwick, Novak & Hofacker, 2013). Taking these developments into account, the consumer of today demands reliable company information to ensure that what a company says is consistent with what a company does. Hence, the openness that the Internet has entailed makes it necessary for companies to present themselves transparently.
Besides the rise of Internet there appears to be another important trend in society, namely the increased valuation of ethics and sustainability. Bhaduri and Ha-Brookshire (2011) prove in their study on transparency in the apparel industry that consumers in general seem to support businesses that practice sustainability and ethics. Furthermore, they point at a shift in consumer preferences towards more sustainable products and services. Accordingly, Gray (1992) shows that the development of accountability and environment-centered approach increases the transparency of organizations: ‘it increases the number of things which are made visible, it increases the number of ways in which things are made visible and, in doing so encourages an increasing openness. The inside of the organization becomes more visible, that is, transparent’ (Gray, p 415). Langen, Grebitus and Hartmann (2010) support the
view of growing awareness among sustainability issues and go a step further by demonstrating that it starts to shape consumer-buying patterns: conscious consumers align their consumption with their morals on sustainability by preferring products that are made in an ethical and sustainable way over products that are not. According to the authors, this trend is clearly visible in the increased sales of fair trade products. In summary: Consumers increasingly care about the way products are made and therefore demand transparency among company practices.
Next to technological advancements and increased attention to sustainability, a commonly reported tendency is the emergence of consumer skepticism. The financial crisis in 2008 has awakened the public and the media (Kemper & Martin, 2010) and has led to increased risk perception within consumers. This, in turn, leads to more careful consumers and, moreover, has triggered consumer skepticism across companies and products. Fueled by the crisis, consumers have become more money minded. Mansoor and Jalal (2011) show in their paper that consumers are not only economically but also psychologically affected by the crisis. They propose that since the crisis, consumer decision-making processes have become more extensively and information gathering is done more thoroughly. This may cause increased demand for transparency in companies. Consistently, Kirby (2012) shows that high consumer skepticism leads to the need to engage in more transparent marketing.
TRANSPARENCY
First of all, it is important to define transparency. So far, there is no consensual definition in academic literature of information transparency. Dapko (2012) attributed a complete study to defining the construct. She developed a scale and a model of transparency and distinguished the construct with matching concepts in the academic literature like disclosure, communication, trust and honesty. Eventually she defined transparency as ‘the extent to which a stakeholder perceives a firm’s conduct is open and forthright regarding matters relevant to the stakeholder’. Bushman, Piotroski and Smith (2004) approach the construct from a more corporate perspective and define it as: ‘the availability of relevant, reliable information about the periodic performance, financial position, investment opportunities, governance, value, and risk of publicly traded firms’. Christensen (2002) on the other hand, describes the construct in its most broad definition, namely as ‘the public availability of relevant information’. In addition, Zhu (2004) approaches transparency from an economical perspective, and defines the construct as ‘the level of availability and accessibility of market information to its participants’. Hultman and Axelsson (2007), in turn, classify transparency
into four types: technological transparency, organizational transparency, supplies transparency and cost/price transparency. Besides the types of transparency, they also focus on the degree of transparency, the direction of transparency and the distribution of transparency.
Source: Hultman & Axelsson (2007)
Studies into the effects of transparency on consumer behavior show several interesting observations. Eisend (2006) found that consumers increasingly value transparency and they trust transparent companies more. Moreover, it appears that consumers consider transparency when making purchase decisions (Cohn & Wolfe, 2013). Furthermore, research has looked into the impact of transparency on customer satisfaction (Costa, Butler, Galletta, & Lopes, 2008), on brand equity and corporate reputation (Fournier & Avery, 2011), on purchase decisions (Cohn & Wolfe, 2013), on willingness to pay and premium pricing (Carter and Curry, 2010) and on trust and reciprocity (Kanagaretnam, Mestelman, Nainar & Shehata, 2012; Eisend, 2006). A recent study of Buell, Kim and Tsay (2014) finds that transparency introduces the possibility of reciprocal gains for both consumers and producers and state that transparency may enhance customer perceptions
In addition to what transparency exactly means, the content of the information, which information is being disclosed, is at least of equal importance. Granados et al. (2009) developed a schematic framework in which they enumerate various information elements; product information, price information, inventory information, cost information and process information. They developed several terms for proposing information, namely, by disclosing, distorting, biasing and concealing the information.
THE THEORY OF RECIPROCITY
Gouldner (1960) was the first who mentioned reciprocity, and he theorized it as the ‘rule of reciprocity’. He found out that behavior exchanged between individuals is expected to be of comparable value, in order to maintain a norm of reciprocity in mutual disclosure. Reciprocity stabilizes social systems, is visible in every interpersonal relationship and is applicable across every culture (Wu et al., 2006). The rule of reciprocity is a form of social exchange between partners. According to the social exchange theory, social exchange is a situation in which the actions of one person provide rewards or punishments for the actions of another person and vice versa in repeated interactions (Blau, 1964). Stanca (2009) demonstrated that reciprocity functions as a mechanism in the evolution of cooperation; if individuals interact repeatedly, reciprocating behavior can induce cooperative behavior.
There exist several definitions of reciprocity. Zucker (1986) defines reciprocity as ‘a system of diffused social norms, leading to mutual obligation and expectations of equitable treatment’. In addition, Bensaou (1997) refers to reciprocity as ‘the degree of fairness that the participating companies perceive about sharing risk, burdens and benefits’. Recently, Kanagaretnam et al. (2012) defined reciprocity as ‘an echo of trust by the party that was trusted’. Although there is no consensus on the definition of reciprocity, the underlying thought is the same. Reciprocity can be considered as the human need or tendency to want to give something back when something is received.
Reciprocity is clearly reflected in sharing of personal information, also known as self-disclosure. Self-disclosure can be defined as ‘the act of revealing personal information to others’ (Archer, 1980, p183) and refers to revealing personal facts, thoughts and emotions to a partner (Altman & Taylor, 1973). In the early days, the work of Jourard (1964) showed that self-disclosure in a social environment often takes place in a dyadic manner, which means that disclosure of personal information is a result of the disclosure input from another person. According to Derlega, Harris and Chaikin (1973), the reciprocity effect is derived from the equity theory, proposed by Adams (1965). This theory states that people feel a degree of discomfort when they are placed in an inequitable relationship with someone else. Referring back to information disclosure, the recipient of information who fails to disclose in return, is at that moment in an inequitable relationship, which makes him feel uncomfortable. This feeling should disappear when the recipient discloses information in a reciprocal way. There is substantial evidence that people will engage in intimate self-disclosure if they first become recipients of such disclosures from their conversational partners (Gouldner, 1960; Zucker, 1986; Bensaou, 1997; Wu et al., 2006, Stanca, 2009; Kanagaretnam et al., 2012). Disclosure
seems to beget disclosure, such that people who receive intimate disclosures feel obligated to respond with a personal disclosure of equal intimacy (Berg & Derlega, 1987). The disclosure reciprocity effect can be can be defined as followed: ‘when one person discloses intimately, his or her partner tends to disclose intimately in return’ (Miller & Kenny, 1986). In summary, as one individual discloses more intimate information, his or her partner also discloses more information (Cozby, 1972).
Besides the reciprocity principal, self-disclosure is influenced by several factors. On the one hand, personal aspects such as gender (Dindia & Allen, 1992) and personality (Miller, Berg & Archer, 1983) drive the degree of self-disclosure. On the other hand, external aspects such as (lack of) trust and respect can result in increased or decreased self-disclosure (Tokic & Pecnik, 2011). Self-disclosure is often studied in the light of the development of close relationships (Laurenceau, Barrett & Pietromonaco, 1998). Apparently, people reveal information about themselves to create and maintain friendships (Reis & Patrick, 1996). Besides that, people tend to disclose more to people they like (Shelton, Trail, West & Bergsieker, 2010). The existing research on self-disclosure is primarily focused on disclosure between two people. Because current research addresses the disclosure between a company and a consumer, it is necessary to go deeper into this.
RECIPROCITY IN A COMPANY-CONSUMER RELATIONSHIP
The above-mentioned research about reciprocity is based on information disclosure between two or more persons. Now, we want to find out how this works in a context involving a company and a customer. To examine this, both sides of the relationship are highlighted.
First, we look at the role of the company in the company-consumer relationship. According to theory, in order to trigger reciprocity, a company would first have to reveal intimate information about itself to the consumer. Prior to making statements about reciprocity between a company and a consumer, there must first be more clarity about the corresponding relationship. The relationship between a company and a consumer forms the core of customer relationship management (CRM), a commonly heard term these days. CRM is a strategic approach that is concerned with creating improved shareholder value through the development of appropriate relationships with key customers and customer segments (Payne & Frow, 2005). It involves building and sustaining trust resulting in developing emotional and structural bonds with consumers (Iyer & Bejou, 2008). Nowadays, having a sincere and committed relationship with consumers is crucial and marketing managers invent ever-new actions to bind the consumers to the company. Hultman and Axelson (2007) also examined
the company-consumer relationship and indicate that trust is one of the core concepts. Consistently, Morgan and Hunt (1994) prove that trust and commitment are key elements of a relationship; having trust in a business partner you share information with is essential. Furthermore, Prahalad and Ramaswamy (2004) suggest that information transparency is necessary to create trust between institutions and individuals.
Second, we look at the role of the consumer in the company-consumer relationship. The prevailing view is that not only companies aim for increased transparency; consumers also value openness (Hultman & Axelson, 2007). In recent years, multiple studies have appeared on consumer information disclosure. For instance, Hansen (2003) proved that if the level of product specific knowledge among consumers increases, they become more involved in service delivery and as a result they will disclose more information about themselves. Besides that, recent research shows that disclosure of a firm’s information practices increases willingness to share personal information and reduces effects of privacy concerns among consumers (Kauffman et al., 2011). In addition, Moon (2000) conducted research into eliciting self-disclosure as a result of computer self disclosure. In her research, she created a conversation between people and a computer to investigate whether they would react in a reciprocal manner. By conducting two experiments, Moon (2000) proved that the tendency to self disclose was consistent with the norms of reciprocity and concluded that it is possible to use computers to elicit intimate information from consumers. If it can be demonstrated that the principle of reciprocity operates between computers and people, it can be assumed that the principle of reciprocity also works between companies and consumers.
It follows that:
H1: Organizational information disclosure leads to the reciprocal tendency of consumers to
self-disclose.
DISCLOSURE SOURCE
Besides the content of information, it appears to be relevant who discloses the information. Petty and Cacioppo (1981) set out the principles for research on disclosure source. In their study, they state that beliefs and attitudes can be influenced by several factors, including the message source. Sparks, Perkins & Buckley (2013) agree on the remark that the source of the message has some influence on customer perceptions and beliefs. Several aspects, including source similarity, source attractiveness, source credibility and source expertise influence information source (Wilson & Sherrel, 1993). According to Milburn (1991), the
trustworthiness of the information source is very important. It appears that when a source is seen as biased or as communicating the message for a purpose other than information, the credibility of the source is harmed (Hass, 1981).
When it comes to reciprocity, it is an assumption that the information disclosure is directly transferred from A to B. According to research (Pacheco, Traulsen, Ohtsuki, and Nowak, 2008) direct reciprocity relies on repeated encounters between the same two individuals. In other words, in order to maintain reciprocity, the information should come directly from the source, in this case the company. According to the reciprocity principle, it can be assumed that when information comes from the company itself, consumer information disclosure would be increased.
It follows that:
H2a: Participants are more likely to disclose information about themselves in the case when
the organization is the information source. BRAND TRUST
It appears that organizational transparency has several advantages. Rogers (1987) finds that it can improve honesty and trust. Consistently, Young-Ybarra and Wiersema (1999) found that open communication leads to more trust. It can be assumed that if being transparent leads to higher levels of honesty and trust, corporate transparency will have a similar effect on consumers’ levels of trust.
Furthermore, trust appears to be an important antecedent of people’s behavior. Milne and Boza (1999) for example, find out that trust mitigates concerns. Already in 1977, Wheeless and Grotz found out that a higher level of trust was associated with more consciously intended disclosure and a greater amount of disclosure. Apparently, willingness to disclose personal information is dependent upon the trust level of the person to whom the information is disclosed (Schoenbachler & Gordon, 2002). Consistently, Zimmer et al. (2010) show that trust in organizations is regarded as an important driver for Internet users’ intention and willingness to disclose their personal data for computer-mediated transactions. It appears that brand name and brand status are important predictors in this process. Indeed, Earp and Baumer (2003) demonstrated that people disclosed more personal and financial information to well-known companies compared to companies they didn't know.
Evidently, trust functions as a heuristic in information disclosure situations (Scholz & Lubell, 1998). According to Norberg, Horne & Horne (2007), the assessment of an
organization in terms of trustworthiness may function as a cue for consumers in considering how much and what type information will be given to that particular organization. So, the higher the levels of trust in the company, the higher the willingness to provide personal info to that company (Hoffman, Novak & Peralta, 1999; Schoenbachler & Gordon, 2002).
Since organizational transparency leads to higher consumer trust, and that trust in turn leads to increased information sharing, it can be assumed that trust in the brand is a mediator in the relationship between organizational and consumer information disclosure.
It follows that:
H2b: Participants are more likely to disclose information about themselves in the case when
the organization is the information source (H2a), this is mediated by brand trust. INFORMATION TYPE
The more information a company has about a particular consumer, the more value that company will be able to provide that consumer (Kauffman et al., 2011). As a result, there has been a remarkable growth in the amount of personal information that firms collect about their consumers, often through the use of computer technology (Moon, 2000). Obtaining consumer data is done in different ways, either in a conscious or an unconscious manner. Research shows that 86% of commercial web sites provide no information of any kind on how the collected data will be used, or even whether data is being collected (Landesberg, Levin, Curtin & Lev, 1998). This realization penetrates more and more into the brains of the consumer. Not surprisingly, information privacy of consumers has evolved into an important matter in today’s society (Kauffman et al., 2011) and can be defined as ‘the control of transactions between persons and others, the ultimate aim of which is to enhance autonomy and to minimize vulnerability’ (Margulis, 1977). Multiple studies into the area of information privacy demonstrate that consumers are more and more worried about the misuse of their personal information (Cranor, 1999; Vijayan, 2004). They have become increasingly aware of the value that their personal information provides to companies and are therefore ever more reticent in the sharing of their information (Hagel & Rayport, 1997). In the event that the consumer is willing to share his or her personal information with the organization and the organization uses the data in an unauthorized manner it can lead to a loss of trust in the organization (Kauffman et al., 2011).
In summary, there is a potential downside to the collection and use of greater amounts of detailed personal information (Culnan & Armstrong, 1999). Apparently, consumers find it
highly important how their personal information is collected in online settings, the way that information is used, handled and protected (Collier & Bienstock, 2006). Therefore, the challenge for organizations is to find a balance in the benefits that the use of information offers and the potential risks of privacy loss in their dealings with customers.
An opportunity to reduce the possible costs of self-disclosure is to show extensive privacy policies that claim to protect consumer privacy (Andrade, Kaltcheva & Weitz, 2002). When consumers believe in the adequacy of legal protection of the personal data, this could increase disclosure (Beldad, van der Geest, de Jong & Steehouder, 2012). According to Culnan and Armstrong (1999), organizations should view the collection of personal information as a social contract in order to create willingness in consumers to disclose personal information. Therefore, organizations have to develop information practices that address the perceived risks of information disclosure. Culnan and Armstrong (1999) state that companies that establish fair information practices and disclose these before collecting personal information reduce the perceived risks and subsequent negative consequences.
Thus, when consumers know that information collection is carried out safely, they tend to share their personal information. Would this also work when a company identifies the motive for information gathering? According to Zimmer et al. (2010) and Moon (2000) it does. They incorporated ‘reasoned dyadic information’ in their studies on reciprocity in information disclosure. Briefly, this means that information is given on why the information collected is done and how the information will be used. Their goal with this reasoned dyadic condition is to reassure people who already intend to share information and to attempt to persuade people who may not be willing to share information to reconsider that decision by giving them a motive for the information collection. It can be assumed that if the motive of the company for collecting consumer data leads to consumer benefits, such as better-adapted and improved products, consumers are willing to share their information more easily. Current research looks at whether having a motive to collect consumer information can also elicit information disclosure.
It follows that:
H3a: Participants are more likely to disclose information about themselves when
PRIVACY CONCERNS
Online privacy concern can be broadly viewed as consumer concerns about the use of their revealed information for marketing purposes, beyond its intended purpose (Mothersbaugh, Foxx, Beatty & Wang, 2012). An important aspect in privacy concerns is the control consumers have on their personal information. Research shows that marketing practices that give consumers little input or control evoke the highest levels of privacy concerns (Nowak & Phelps, 1992). It could be argued that by adding the motive for information collection, a company gives consumers some degree of control over the data. When sharing the motive for information collection, consumers are aware of the purpose of their personal data. When the specific motive for information collection is displayed, this could lead to a decreased risk perception and decreased privacy concerns among consumers.
Furthermore, perceptions of the risks involved in sharing personal data online can strongly discourage people from entering transactions that lead to the sharing of such data (Beldad et al., 2012). It appears that increased online privacy concerns decreases willingness to disclose personal information (Mothersbaugh et al, 2012). When turning this around, decreased online privacy concerns should increase willingness to disclose personal information. It follows that if consumer awareness and agreement of what will happen to the data were achieved, this would increase willingness to disclose information (Beldad et al, 2012).
When companies show why they gather information, it can be assumed that this could lead to decreased privacy concerns. In turn, low privacy concerns are expected to increase information disclosure. Therefore, it is expected that privacy concerns function as a mediator in the relationship between organizational and consumer information disclosure.
It follows that:
H3b: Participants are more likely to disclose information about themselves when
organizations disclose their motive for information collection (H3a). This effect is mediated by privacy concerns.
METHOD Participants
Initially 214 people participated in the study. Of those, a total of 203 completed the survey and therefore only their data was used. Yet, participants who answered the manipulation check question on text source incorrectly were deleted from the sample. After excluding these participants, 168 participants remained. Of those, 64 were male, 104 were female. Participants
were on average 27.01 years old. The average height of participants was 174.2 cm; the average weight was 71.5 kilo’s. The main part of participants has completed a Bsc (48.8%) or an Msc (24.4%), and therefore it can be said that the participants were highly educated. Of the participants, the largest part lived in Amsterdam (54.1%). The major part of the participants reported to not have children living at home (88.7%). Most of the participants (73.2%) indicated to do their grocery shopping at Albert Hein, and the majority spent on average between €25 and €49 per week on grocery shopping. 60.7% of the participants earned less than €10.000, followed by 19.6% that earned less than €30.000 yearly. When looking at the figures, it can be concluded that the greater part of the participants consisted of students. Of the 168 participants, 115 wanted to share their information with Kraft foods and 53 did not. Design
The five hypotheses were tested using an experimental design. The experiment was a 2 (information type: disclosure vs general) x 2 (information source: Kraft foods vs UvA) + 1 (control group) between subjects design. The dichotomous dependent variable was the willingness to disclose personal information (yes/no).
Stimuli and materials
In the survey, combinations of different questions and items measured several variables. Self-disclosure. ‘Self-disclosure’ was measured with a single question whether consumers wanted to share their previously collected personal information with Kraft Foods. This information contained personal characteristics, such as weight, gender, age, education and monthly income.
Health and sustainability preferences. ‘Health and sustainability preferences’ were measured by adapting seven items of the 36-item scale of Food Choice Questionnaire (Steptoe, Pollard and Wardle, 1995) to suit the specific context of this study. The health and sustainability preferences were added into the survey to create the impression that the information asked was valuable and credible. The participants rated their agreement on the items on a 5-point scale (1=disagree to 5=agree). A sample item of the variable ‘health and sustainability’ was: ‘it is very important to me that the food I eat on a typical day keeps me healthy’. (M = 3.12, α = .716).
Brand trust. ‘Brand trust’ was measured using the four-item index of the brand turst scale, developed by Chaudhuri and Holbrook (2001). The participants rated their agreement
on the items on a 7-point scale (1=totally disagree to 7=totally agree). A sample item of the variable ‘brand trust’ was: ‘I trust Kraft foods’. (M = 3.97, α= .856).
Privacy concerns. ‘Privacy concerns’ were measured by adapting six items of the 16-item scale of privacy concerns (Smith, Milberg & Burke, 1996) to suit the specific context of this study. The participants rated their agreement on the items on a 7-point scale (1=totally disagree to 7=totally agree). A sample item of the variable ‘privacy concerns’ was: ‘I am concerned that Kraft foods will share my personal information with other parties’. (M= 3.40, α= .850).
Procedure
The manipulations and measures were administered in an online survey. Each participant received the survey sent online via the researcher. First, a general informed consent was shown to the participants, with the purpose of the research. By clicking trough, the participant gave permission to proceed with the survey. Next, all participants filled in the self-disclosure questionnaire and the health and sustainability preference questionnaire. Afterwards, the participants were randomly assigned to one of the five conditions, as table 1 reveals.
Table 1: Five possible manipulations
Brand source UvA source
No disclosure 1. No disclosure Information
disclosure (source)
2. Information disclosure Kraft Foods source
3. Information disclosure UvA source
Motive for
information
4. Motive for information Kraft Foods source
5. Motive for information UvA source
After the manipulation, the participants received an overview of their answers of the self-disclosure and health questionnaire. The participants were asked if they were willing to share this information with Kraft Foods. Finally, the participants filled in the brand-trust and privacy concerns questionnaire. After completing the questionnaire, the participants were thanked for their participation.
Manipulations
Information source. Information source was manipulated by a different background and text in the survey.
UvA condition. In the UvA condition, the UvA logo was clearly visible. In addition, the following text was added to point out the information source to the participants:
‘The University of Amsterdam collaborates with several national and international companies. One of these companies is Kraft Foods. In a study on the performance of food manufacturers with regards to sustainability, the University of Amsterdam reports the following on the performance of Kraft Foods in the area of sustainability’.
Brand condition. In the brand condition, the Kraft Foods logo was clearly visible. In addition, the following text was added to point out the information source to the participants: ‘The University of Amsterdam collaborates with several national and international companies. One of these companies is Kraft Foods. In the context of this study, Kraft Foods wants to share some information on its performance in the area of sustainability with you as a participant’.
Information type. Information type was manipulated by a different text in the survey. In the general information disclosure condition, neutral information was disclosed on the practices of Kraft foods. It contained information about environmental footprint, greenhouse gas emission, the use of plastic and waster waste.
In the specific explanation condition, neutral information was disclosed on the motive of Kraft foods to collect consumer information. It contained information about why and how Kraft Foods collects data. Also, it was explained that consumer information is never shared with third parties and stored in a secure manner.
Pretest. A pretest was conducted to determine the positivity and relevance of the information and the reputation of Kraft Foods. This was tested because an extremely positive or negative text could perhaps bias the test results. The relevance of the text was tested to ensure that the text was meaningful and credible. Twenty participants filled in a survey containing eight items. Participants rated their agreement on the items on positivity and relevance a 7-point scale (1=strongly disagree to 7=strongly agree) and the item on reputation on a 5-point scale (1=disagree to 5=agree). Based on their results, it appears that the general disclosure was somewhat positive (M = 5.18) and relevant (M = 5.63). Also the specific disclosure was somewhat positive (M = 5.18) and relevant (M = 5.49). In addition, the reputation of Kraft foods appeared to be somewhat positive (M = 2.92).
Manipulation checks. Information source was measured with a manipulation check at the end of the survey. This consisted of a question in which participants were asked about the source of the information they had just seen.
RESULTS
First, a multiple logistic regression analysis was performed to test if information source (H2a) and information type (H3a) had a significant effect on the decision whether or not to disclose information. A logistic regression analysis was chosen because of the dichotomous nature of the dependent variable. The control group was excluded from the regression, since they didn’t see any text or text source. Results are presented in table 3. In line with H2a, the model shows that information source had a significant influence (β -1.44, p .032) on the probability of someone sharing information versus not sharing information. This means that, when the value of information source increases, so if the source becomes ‘more UvA’, the degree of information sharing decreases. In other words, when one was exposed to information given by Kraft Foods, he or she was more willing to share information compared to when one was exposed to information given by the UvA. Hypothesis 2a was confirmed. Against the expectation of H3a, there was no significant direct effect of information type, indicating that the probability of disclosing information was the same for participants in the general disclosure condition and participants in the explanation disclosure condition. Hypothesis 3a was not confirmed. The interaction effect of information source and information type appeared to be not significant, however, a not significant trend (β 1.37, p .16) was detected.
Table 2. Results of the logistic regression analysis for information type and information source on consumer
information disclosure N=97
95% CI for Odds Ratio
B (SE) P Lower Odds Ratio Upper Included
Constant 1.66 (.55)
Information type (0 = general disclosure, 1 = specific disclosure)
-.50 (.75) .508 -1.96 .61 .97
Information source (0 = brand, 1 = UvA) -1.44* (.67) .032 -2.75 .24 -.12
Interaction type*source 1.37 (.97) .156 -.52 3.94 3.26
Note: R2 = . Model χ2 (3) = 5.61, p > .01. * p < .05.
A second logistic regression analysis was performed to test whether the different conditions had a significant effect on the decision whether or not to disclose information. The results of the logistic regression analysis are presented in Table 3. In line with H2a, it appeared that condition two, brand general disclosure (β 1.58, p .020) significantly influenced
consumer information disclosure. This means that, compared to the control condition, information disclosure when it came from Kraft foods significantly elicited more consumer information disclosure. Condition four, brand specific disclosure, appeared to be not significant, however, a not significant trend (β 1.08, p .096) was detected. Similarly, condition five, UvA specific disclosure, appeared to be not significant, however, a not significant trend (β 1.10 p .100) was detected.
Table 3. Results of the logistic regression analysis for condition on information disclosure, N = 97 95% CI for Odds Ratio
B (SE) P Lower Upper
Included
Constant (control group) -.08(.40)
2. Brand, general disclosure 1.58* (.68) .020 .25 2.91 3. UvA, general disclosure .14 (.56) .797 -.95 1.24 4. Brand, specific disclosure 1.08 (.65) .096 -.19 2.36 5. UvA, specific disclosure 1.10 (.62) .100 -.19 2.23 Note: R2 = .056. Model χ2(6) = 11.62, p > .05. * p < .05.
Next, a cross tabulation was conducted between the conditions and the dichotomous dependent variable information sharing. Table 4 and figure 1 provide an overview of the number of participants per condition who wanted to share personal information and who didn’t want to share their personal information.
Table 4: Cross tabulation between the conditions and the dichotomous dependent variable
Conditions Information sharing
Yes No Total
1.Control 12 13 25
2.Brand, general disclosure 4 21 25
3.UvA, general disclosure 12 15 27
4.Brand, explanation 5 16 21
5.UvA, explanation 6 18 24
Figure 1: Overview of the participants in the conditions and the dependent variable.
Furthermore, the five conditions were compared with each other, visible in table 5, to investigate if it made a significant difference in what condition the participants were when it came to information sharing. To test reciprocity (H1) in current study, condition two (Kraft Foods discloses company information itself) and condition three (University of Amsterdam discloses information on Kraft Foods) were compared. It was expected that only in condition two the principle of reciprocity should apply: only in condition two the participants should significantly share more information with the company. This proved to be correct. It appeared that condition two and three significantly (χ2 = 4.60, p .03 with df =1) differed from each other. In line with H1 and H2a, when Kraft Foods itself disclosed general information, it led to significantly more information sharing by the consumer, compared to when the university disclosed general information. In other words, when a company itself discloses the information, voluntary consumer disclosure increases significantly compared to the situation in which another information source discloses information. This finding is consistent with hypothesis 1 and hypothesis 2a.
Furthermore, condition two and four significantly differed from each other (χ2 = 0.44, p .05 with df =1). Apparently, when the company disclosed general information, it led to significantly more information sharing by the consumer, compared to when the company disclosed specific information. This goes against the hypothesized direction and hypothesis 3a is therefore not confirmed.
A trend toward significance was detected between condition three and four (χ2 = 2.14, p .14 with df =1). Also, a trend toward significance was detected between condition three and five (χ2 = 2.06, p .15 with df =1). 52% 84% 56% 76% 75% 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%
Control Brand, general disclosure UvA, general disclosure Brand, specific disclosure UvA, specific disclosure
There was no significant difference between condition four and condition five, indicating that it makes no difference in voluntary information disclosure if a company shares the motive for information collection or another information source does, such as the UvA. This goes against the expectation, and hypothesis 3a is not confirmed.
Table 5: Logistical regressions between conditions 2 Brand, general disclosure 3. UvA, general disclosure 4. Brand, specific disclosure 5. UvA, specific disclosure 2 - 3 χ2 = 4.60, p .03* - 4 χ2 = 0.44, p .05* χ2= 2.14, p .14 ** - 5 χ2 = 0.6, p .44 χ2 = 2.06, p.15 ** χ2 = 0.01, p .93 - Df (1), * p < .05, ** p < .15
Subsequently, a multiple mediated logistic regression analysis was performed to test if brand trust (H2b) and privacy concerns (H3b) significantly mediated the decision whether or not to disclose information. The four causal-steps test of Baron & Kenny (1986) was used to check the effects of those mediators.
First, the difference between the control group on the one hand and condition 4 (specific disclosure from Kraft Foods) and condition 5 (specific disclosure from UvA) on the other hand was examined, shown in figure 2. This was investigated to see if privacy concerns or brand trust may explain the large difference between these conditions. According to hypothesis 3b, it is expected that privacy concerns act as a mediator in the relationship between information source and information disclosure. In other words, disclosing the company’s motive for information collection decreases privacy concerns, which in turn would increase information disclosure.
Figure 2: Schematic overview of the comparison of the control group with condition four, and five.
Model 1a: Schematic overview of stepwise regression (Baron & Kenny, 1986), control condition versus brand
specific disclosure + UvA specific disclosure.
To begin with, the direct effect of grouping membership on information disclosure was tested (step 1), which appeared to be significant (β 1.05, p. 048). This indicates that it did matter if participants were in the control condition or in the brand or UvA specific disclosure condition. Second, the effect of grouping membership on brand trust was measured (step 2a), which appeared to be significant (β .701, p. 010). Third, the effect of grouping membership on privacy concerns was measured (step 2b) and appeared to be significant (β -.881, p .014). These results indicate that the levels of both brand trust and privacy concerns were predicted by grouping membership.
Furthermore, the effect of brand trust on information disclosure was measured (step 3a), which appeared to be significant (β 1.13, p .000). Also, the effect of privacy concern on information disclosure was checked (step 3b), which appeared to be significant (β -1.11, p .000). This indicates that both brand trust and privacy concern directly influenced the probability of voluntary information disclosure.
When all predictors were included in the model (step 4), the model appeared to be significant: χ2 (3): 32.46 (p = .000), Nagelkerke R square = .517. The model showed that only privacy concerns had a significant influence on the probability of someone sharing information versus not sharing information. This means that, the more people are worried about their privacy, the less they are inclined to share information (odds ratio -.32). Furthermore, when privacy concerns as a predictor was added in the multiple regression, the effect between group membership and information disclosure disappeared. Therefore, in line with hypothesis 3b, privacy concerns acts as a mediator on the relationship between group membership and information sharing. It can be concluded that expressing the company’s motive for data collection decreases privacy concerns, which in turn increases the tendency of disclosing personal information.
Model 1b: Results of stepwise regression (Baron & Kenny, 1986)
Second, the difference between the control group and condition three (general disclosure from UvA) on the one hand and condition two (general disclosure from Kraft Foods) on the other hand was examined. This is investigated to see if privacy concerns or brand trust may explain the large difference between these conditions. According to hypothesis 2b, it is expected that brand trust acts as a mediator in the relationship between information type and voluntary information disclosure. In other words, when general information is disclosed to the consumer, this would increase brand trust, which in turn would increase information disclosure.
Figure 3: Schematic overview of the comparison of the control group and condition three with condition two.
Model 2a: Schematic overview of stepwise regression (Baron & Kenny, 1986), control group + brand general
disclosure versus UvA general disclosure
To begin with, the direct effect of grouping membership on information disclosure was measured (step 1) which appeared to be significant (β 1.5, p.014). This means that it does indeed matter in which condition the participant was considering the degree of information disclosure. Second, the effects of grouping membership on brand trust was measured (step 2a), which appeared to be marginally significant (p .063). Third, the effect of grouping membership on privacy concerns was measured (step 2b), which was not significant (p .158). These results indicate that neither brand trust, nor privacy concerns, was predicted by grouping membership. In other words, it didn’t matter in which group participants were for their levels of brand trust or privacy concerns. Furthermore, the effect of brand trust on
information disclosure (step 3a) was measured, which appeared to be significant (β 1.13, p.000). Also, the effect of privacy concerns on information disclosure (step 3b) was measured, which appeared to be significant (β -1.11, p.000). These results indicate that both brand trust and privacy concerns influence the degree of information disclosure.
When all predictors were included in the logistical regression (step 4), the model appeared to be significant: χ2(3): 36.65, p = .000, Nagelkerke R square = .519. The model shows that brand trust and privacy concerns significantly influenced the probability of someone sharing information versus not sharing information. This means that, the more trust people have in a brand; the more they are inclined to share information (odds ratio 3.01). Furthermore, the more people are worried about their privacy, the less they are inclined to share information (odds ratio -.41). When brand trust and privacy concerns were added in the regression, the effect between group membership and information disclosure disappeared. Therefore, it depends on brand trust and privacy concerns if people disclose information. Model 2b: Results of stepwise regression (Baron & Kenny, 1986)
DISCUSSION
In the light of ever-increasing concerns of privacy, a growing loss of control and the dilemma of being transparent, the present study attempts to identify methods companies can deploy to collect personal data without undermining consumer trust. By including (1) the theory of reciprocity and (2) the company’s motive for consumer data collection, this research investigates whether this increases the tendency of consumers to disclose personal data. An additional source of information, the University of Amsterdam (3) is added to the research in order to ensure that the observed effects actually are the result of the manipulation and not
simply from the overall effect of information. Furthermore, brand trust and privacy concerns are included in current study to investigate their relevance in terms of information disclosure. In effort to identify the effects of reciprocity and including the company’s motive for data collection, an online survey was conducted which contained several manipulations.
The primary finding of current research is the demonstration of the reciprocity effect in a company-consumer relationship. By including the theory of reciprocity, current research contributes to the debate on transparency and privacy. Although it is somewhat premature to attribute the increased information disclosure directly to the effects of reciprocity, current research provides a step in the right direction. In line with the expectations, the willingness to share personal information increases when a company is transparent on its practices. The difference in consumer information disclosure between participants who were exposed to information directly from Kraft Foods and participants who were exposed to information from the University of Amsterdam is of such magnitude that the influence of reciprocity cannot be denied. As the literature proposes, the principle of reciprocity should only apply when there is a direct exchange between A and B. Indeed, reciprocity relies on encounters between the same two parties (Pacheco et al., 2008). According to the equity theory (Adams, 1965) people feel a degree of discomfort when they fail to disclose information in return. This could also apply in a consumer-company relationship. Since people cannot tolerate such disparity within their personal relationship with other people, perhaps this is also the case when it comes to their relationship with organizations. Nowadays, relationships between companies and customers are more important than ever (Payne & Frow, 2005). They involve building and sustaining trust and result in the development of emotional ties (Iyer & Bejou, 2008). It can be argued that company-consumer relationships are increasingly starting to resemble personal relationships. People and businesses are moving closer together. Therefore, the proven reciprocal relationship between companies and consumers is, in our opinion, logical rather than far-fetched.
The second finding of the research was against expectations. Indeed, no significant main effect was found in the motive for information sharing. This indicates that expressing the motive for collecting information has no effect on the willingness to disclose personal information. However, an additional analysis was made regarding the company’s motive for information collection. Surprisingly, when comparing the conditions in which the company’s motive for information collection was expressed to the control group, there was a significant difference in the probability of information sharing. More specifically, the tendency to share information increased significantly when the company’s motive for information collection
was communicated to the consumers. In consonance with Zimmer et al. (2010), current research demonstrates that consumers can indeed be persuaded to share their personal data when the company’s motive for data collection is expressed. In addition to the existing studies (Andrade et al, 2002; Beldad et al, 2012) that focus on showing extensive privacy policies to consumers, our research reveals an alternative way for increasing personal data disclosure. It can be argued that communicating the company's motive provides the consumer with a certain degree of control. After all, consumers are aware of what will happen to their personal data. Given the fact that control is an important aspect in privacy concerns (Nowak & Phelps, 1992), this could explain the positive relationship between company’s motive and willingness to share personal data. Besides that, it appeared that privacy concerns had a significant influence on the probability of someone sharing information. Naturally, the more worried people are about their privacy, the less inclined they are to share information with the company. It turned out that privacy concerns act as a mediator on the relationship between the company’s motive for information disclosure and consumer information disclosure. Specifically, privacy concerns clarifies the nature of the relationship between expressing the company’s motive and personal information disclosure. Showing the company’s motive for data collection decreases privacy concerns, which in turn increases the tendency of disclosing personal information.
The third finding of current study is that being transparent on the company’s motive only works if the company discloses the information itself. Indeed, the findings show a main effect of information source, which means that it indeed mattered whether the information came directly from the company in comparison to when the information came from the UvA. When the company itself communicated information, willingness to share the personal information was significantly higher compared to when the University of Amsterdam gave exactly the same information.
Finally, current study found that the degree of disclosing personal data depends on the amount of brand trust and privacy concerns people have. Indeed, privacy concerns and brand trust did directly influence the tendency of consumers to disclose information. Moreover, when brand trust and privacy concerns were considered in the analyses, the direct effect between information source and information disclosure disappeared, indicating that the difference in information sharing is explained by the corresponding variables. On the one hand, this means that the more trust people have in a company; the more they are inclined to share information with that company. This finding is in agreement with prior research (Wheeless & Grotz, 1977), which demonstrated that having trust in a brand results in
consciously intended disclosure and a greater amount of disclosure. Indeed, willingness to disclose personal information is dependent upon the trust level of the person to whom the information is disclosed (Schoenbachler & Gordon, 2002). On the other hand, the findings demonstrate that the less people are worried about their privacy, the more inclined they are to share information with the company. In line with research of Mothersbaugh et al. (2012) our findings show that increased privacy concerns reduce willingness to disclose personal information.
The study results have some important implications for the marketing literature.
Although a majority of studies are interested in obtaining consumer data, there is little research on how companies can increase voluntary information sharing without damaging the consumer relationship. With this study, we’re the first to give meaning to this gap. We show that businesses can deploy either transparency in their practices or communicate the motive for information collection to increase voluntary information sharing. Although other researchers already demonstrated the reciprocity effect between people and objects, like Moon (2000) did with people and computers, this is the first study that takes companies into account.
Furthermore, this study provides some concrete guidelines for companies with regard to obtaining consumer data in the most optimal way. First of all, the results show that companies should behave in a transparent manner in order to collect and use personal data of their consumers without jeopardizing their relationship with them. In that way, at the same time they build rather than undermine consumer trust. Moreover, current study finds that in order to work, direct communication must be facilitated between the organization and the consumer. Therefore, when a party other than the company shares company information, transparency has no effect. This implies that when a secondary party expresses company information, this would not result in increased tendency to share personal data. Next, communicating the motive for information collection to consumers increases willingness to share personal data.
Limitations and future research
Although choosing a survey in the present investigation was most obvious, there are a number of disadvantages to this kind of investigation. First of all, the survey was self-reported; therefore, the information obtained might be inaccurate or incomplete. In addition, the survey answer options might be interpreted differently by participants and may therefore have led to unclear data. Therefore, it would be interesting to replicate present findings in a study that is
more exploratory in nature. This study could also investigate the consumer’s motive to act in a reciprocal manner to provide more clarity there.
Regarding the results, untested variables such as gender or age may account for the impacts of the findings. It is a possibility that there is a difference in gender related to the tendency to share personal information. In addition, one can expect that an elderly population might be less likely to share data online, because they are often more resistant to sharing personal information. Since the sample of current research mainly included students (the average age was 27.01 years old), the results of the study might be slightly biased. It is possible that if older participants are used, the observed effects won’t be replicated. A study that takes age and gender into account might elucidate such assumptions.
Another limitation of the present research is the design of the stimuli. It appeared that many participants were not consciously aware of the information source, despite the textual and visual manipulation. Therefore, the participants who answered the manipulation question wrongly were removed from the dataset. As a consequence, some of the conditions contained less than 25 participants, which is actually a requirement when performing logistic regression. Future research should therefore focus more on the manipulation of the stimuli.
Furthermore, current research does not address the content of the information that is shared with the consumer, although this appears to be very important (Granados et al., 2009). A suggestion for future research is to investigate the influence of content on the reciprocity effect. At which kinds of transparency is the information sharing reciprocal? There are endless possibilities in this area, which can vary from begin transparent on pricing structure to production chain but also annual reports or employee information.
Despite the limitations, current research extends prior studies by demonstrating that the theory of reciprocity also applies within a company-consumer relationship, which makes this research the first of its kind. Hence, it is expected that new research lines will be developed based on the present insight, which will help companies to develop strategies to maximize the consumers’ participation in sharing data that will be beneficial to both consumer and company (win-win situation).
REFERENCES
Adams, J. (1965). Inequity in social exchange. In L. Berkowitz, Advances in experimental social psychology, 2, New York: Academic Press.
Altman, I., & Taylor, D., A. (1973). Social penetration: The development of interpersonal relationships. New York: Holt, Rinehart & Winston.
Andrade, E., Kaltcheva, V., & Weitz, B. (2002). Self-disclosure on the web: The impact of privacy policy, reard and company reputation. Advances in Consumer Research, 29, 350-353.
Archer J.,L. (1980). Self-disclosure. In The Self in Social Psychology, Wegner D, Vallacher R (eds). Oxford University Press: London; 183-204.
Acquisti, A., John, L., & Loewenstein, G. (2013). What is privacy worth? The Journal of Legal Studies, 42(2), 249-274.
Beldad, A., Van Der Geest, T., De Jong, M., & Steehouder, M. (2012). Shall I tell you where I live and who I am? Factors influencing the behavioral intention to disclose personal data for online government transactions. International Journal of Human-Computer
Interaction, 28(3), 163-177.
Bensaou, M., (1997). Interorganizational coorperation: The role of information technology an empirical comparison of U.S. and Japanese Supplier relations. Information Systems Research, 8(2), 107-124.
Berg, J., & Derlega, V. (1987). Themes in the study of self disclosure, Self-Disclosure: Theory, Research and Therapy, New York: Plenum, 1-8.
Bhaduri, G. & Ha-Brookshire, J. (2011). Do transparent business practices pay? Exploration of transparency and consumer purchase intention. Clothing and Textiles Research Journal, 29, 135-149.
Blau, P., (1964). Exchange and Power in Social Life. New York: Wiley.
Buell, R., Kim, T., & Tsay, C. (2014). Creating reciprocal value through operational
transparency. Harvard Business School Technology & Operations Mgt. Unit Working Paper, 14-115.
Bushman, R., Piotroski, J., & Smith, A. (2004). What determines corporate transparency? Journal of Accounting Research, 42(2), 207-252.
Carter, R. E. & Curry, D. J. (2010). Transparent pricing: theory, tests, and implications for marketing practice. Journal of the Academy of Marketing Science, 38(6), 759–774. Christensen, L. (2002). Corporate communication: The challenge of transparency. Corporate
