Introduction

The U.S. company Johnson & Johnson (2011, p. 82) states in its corporate social responsibility (CSR) report: “We must be good citizens—support good works and charities and bear our fair share of taxes.” Anglo American, Xstrata, and other companies communicate their responsible corporate tax strategies (CTSs) similarly. These statements highlight that society is increasingly demanding companies to be good corporate citizens in tax-related matters, and companies are responding to these demands. At the same time, the international mass media are giving more and more attention to the topic of aggressive CTSs (Cunningham 2009; Doyle et al. 2009; Erle 2008). According to recent media reports, several well-known companies (e.g., Apple, Barclays, Diageo, General Electric, Google, Pfizer, Starbucks, and Vodafone) have been accused of engaging in aggressive CTSs. Whereas responsible CTSs try to insure that the company pays its fair share of taxes, aggressive CTSs aim to minimize corporate tax payments by all legal means possible, for example, by shifting profits to low-tax countries. Tax secrecy legislation, such as Sect. 6103 of the Internal Revenue Code in the USA and Sect. 30 of the German Fiscal Code, prevents public officials from disclosing tax information. However, due to the work of investigative journalists, consumers have become more informed about companies that practice aggressive CTSs.

What are the effects of media reports on CTSs on consumers? There is a large and growing body of evidence on the effects of positive and negative CSR activities on consumer attitudes and behavior (see Peloza and Shang 2011 for a review of prior research). However, prior research ignores the role of taxes in this context; analyzing consumer reactions to CTSs is crucial for several reasons. First, taxes are an important cost factor and have strong implications for the profitability of companies (Erle 2008). With statutory tax rates of 30 % in Germany and about 35 % in the USA, taxes are a matter of great importance to companies, even though effective tax rates fall slightly short of these statutory rates (Devereux et al. 2009). Second, the role of taxes within the CSR debate is becoming increasingly important because, in recent years, the power of national governments to prevent large multinational companies from avoiding taxes has decreased (Christensen and Murphy 2004; Doyle and Bendell 2009; Marsden and Andriof 1998 with respect to corporate regulation in general). This development has led to significant losses in tax revenue. Annual reductions of the tax base of 65 billion euros in Germany (Bundestag 2009) and a tax gap of 48 billion dollars in the USA (Internal Revenue Service 2011) due to corporate tax avoidance highlight the relevance of this topic.Footnote 1 Third, CTSs are a controversial CSR activity. Although many scholars and tax managers suspect that consumers consider aggressive CTSs to be a negative CSR activity (e.g., Erle 2008; Friese et al. 2008; Lanis and Richardson 2012; OECD 2009; Williams 2007), others disagree with this opinion (e.g., Freedman 2004; Gassner 2001; Timonen 2008). Consequently, there is neither a theoretical consensus nor are there empirical findings on potential consumer-related non-tax costs of aggressive CTSs. Given the importance of the customer to the value of companies (Ferrell 2004; Freeman 2010; Peloza and Shang 2011), researchers in the fields of accounting and taxes are calling for an analysis of the consumer-related effects of CTSs (Hanlon and Heitzman 2010). We have responded to this call by investigating the effects of media reports on an aggressive CTS and a responsible CTS on corporate success with consumers by means of two laboratory experiments (N = 150, 360).

We investigated the effects of CTSs on corporate success with consumers both on the corporate and on the product brand level. According to Bhattacharya and Sen’s (2004) CSR framework, CSR activities have several consumer-related outcomes from the perspective of companies: internal and external outcomes. First, there are outcomes “internal” to the consumer (e.g., awareness, attitudes, and attributions). Corresponding surveys show that almost any U.S. executive considers corporate reputation to be one of the most substantial measures of corporate success (Dunbar and Schwalbach 2001). Thus, for the purpose of our study, we focused on the effects of CTSs on corporate reputation as an internal outcome. Second, CSR activities have external or visible outcomes (e.g., purchase, price premium, loyalty, word of mouth recommendation) (Bhattacharya and Sen 2004). On the basis of these outcomes, we analyzed the effects of CTSs on customers’ purchase intention and willingness to pay. In doing so, we assessed changes in cash inflows and outflows on the sales side due to the CTSs. Given that higher tax payments due to renouncing an aggressive CTS lead to direct cash outflows, companies might be particularly interested in the impacts of CTSs on their cash flows.

The main contributions of our research are as follows. We offer a theoretical foundation for linking CTSs and CSR on the basis of Carroll’s (1991) Pyramid of CSR. Furthermore, we provide the first empirical evidence that CTSs affect corporate success with consumers. We show that aggressive CTSs have a negative effect on corporate reputation and purchase intention, whereas responsible CTSs have a positive effect. Nevertheless, consumers are not willing to pay a price premium for products sold by responsible tax-planning companies but rather punish aggressive tax-planning companies to some extent by a lower willingness to pay. Finally, we find that consumers with positive attitudes toward taxation (high tax morale, negative attitude toward tax avoidance) react in a stronger manner to media reports on CTSs.

The remainder of this article is organized as follows. First, we focus on the characteristics of CTSs and present a brief review of the relevant literature that deals with consumer reactions to CTSs. We then present our hypotheses about the impact of CTSs on measures of corporate success with consumers (i.e., corporate reputation, purchase intention, and willingness to pay). Next, we describe two empirical studies that we conducted to test our hypotheses. The article concludes with a summary of our findings, a discussion of their implications, and a description of the limitations and directions for further research.

Theoretical Background and Hypotheses

Definitions of Aggressive and Responsible CTSs

In the present study, we analyze two extremes of legal CTSs: aggressive and responsible CTSs. Aggressive CTSs can be defined as corporate efforts to minimize tax liabilities by all legal means possible. These efforts include artificial transactions conducted solely to avoid taxes (OECD 2008; Slemrod 2004). By aggressively interpreting tax laws and exploiting existing loopholes, a company that employs an aggressive CTS repudiates the intention of the legislature (OECD 2008). That is to say, this company achieves tax benefits that were not necessarily foreseen by the legislature. However, most companies that employ an aggressive CTS comply with the letter of the law and thus should be sharply distinguished from those companies that engage in illegal tax evasion (Merks 2006). In contrast, companies that utilize responsible CTSs comply with the intention of the legislature and do not pursue all legal possibilities for tax minimization. They intend to pay their fair share of taxes to insure that the government is adequately financed (Friese et al. 2008; Landolf and Symons 2008). This strategy adheres to the OECD Guidelines for Multinational Enterprises (OECD 2011, p. 58), which request that companies contribute to government finances by acting in accordance “with the letter and the spirit of the tax law.” To give an example, aggressive CTSs involve the creation of offshore subsidiaries in low-tax jurisdictions in order to shift income from high-tax jurisdictions to these subsidiaries by royalties, interest, and the strategic adjustment of transfer prices.Footnote 2 By means of these transactions the company (partially) avoids the payment of taxes in the country in which it derives its economic benefits. Another example of an aggressive CTS is international tax arbitrage,Footnote 3 for example, the objective to achieve a double deduction of the same tax losses through the use of dual-resident companies. In contrast, a company that pursues a responsible CTS would refuse to use such artificial corporate structures designed solely to avoid taxes.

Literature Review

To the best of our knowledge, there are no studies that analyze consumer reactions to CTSs. Since the present study addresses issues at the interface of taxation, marketing, and CSR, it touches on three different fields of research. Some conceptual and empirical literature questions whether there is any link between CTSs and CSR at all (1). Furthermore, little empirical literature focuses on the reactions of stakeholder groups to CTSs other than consumers, on those of investors for instance (2). Finally, consumers might perceive aggressive (responsible) CTSs as negative (positive) CSR activities. If so, the large body of literature that analyzes the effects of CSR activities on consumer attitudes and behavior can be considered relevant to our study (3). In the following section, we give an overview of the literature directly linked to CTSs. The literature on consumer reactions to CSR activities will be discussed briefly later on when we develop our hypotheses. We refer readers to Peloza and Shang (2011) for a broad review of the CSR literature.

The Link Between CTSs and CSR

The link between CTSs and CSR is contentious in conceptual papers. Although many scholars and tax managers suspect that aggressive CTSs are considered negative CSR activities (e.g., Erle 2008; Friese et al. 2008; OECD 2009; Williams 2007), others disagree with this opinion (e.g., Freedman 2004; Gassner 2001; Timonen 2008). Proponents of a link between CTSs and CSR (e.g., Christensen and Murphy 2004; Doyle and Bendell 2009) refer to the decreased power of national legislators to prevent multinational companies from pursuing aggressive CTSs. These aggressive CTSs significantly reduce national tax revenue that could otherwise be spent on public services, such as social services, healthcare, education, and infrastructure (e.g., Abdul Wahab and Holland 2012; Avi-Yonah 2008; Landolf and Symons 2008; Slemrod 2004). Moreover, aggressive CTSs transfer a substantial share of the tax burden from large multinational companies onto smaller companies and individual consumers. Small companies and consumers cannot exploit similar tax avoidance strategies because of their lack of the international presence and the different taxation regulations that they are subject to, respectively (e.g., Avi-Yonah 2000; Boyle 2005).

Opponents of a link between CTSs and CSR, such as Friedman (1970), argue that consumers might prefer a company to focus on maximizing its profits by minimizing its tax payments to insure its competitiveness and long-term viability. Consumers with a negative attitude toward taxation might sympathize with a company that pursues an aggressive CTS (Freedman 2004). As for a responsible CTS, consumers might oppose this tax strategy because they assume that the company shifts the higher tax costs to other market participants (e.g., consumers) in the form of higher prices for its products (Friedman 1970; Friese et al. 2008; Gassner 2001; Slemrod 2004). Freedman (2004) and Timonen (2008) even point out that aggressive CTSs constitute a legislative rather than a CSR issue because legislators provide the loopholes in tax laws that enable companies to avoid taxes legally.

Besides conceptual papers, there is little empirical literature that addresses the association between CTSs and CSR. Preuss (2010) found a relationship between the code of conduct of companies and their engagement in tax havens. In almost all cases the commitments by companies based in tax havens vis-à-vis key stakeholders fell short of those made by the sample of U.S. firms. In two papers, Lanis and Richardson (2011, 2012) observed a negative association between the level of corporate tax aggressiveness and the responsiveness of the companies’ board of directors to CSR as well as their level of CSR activities. Overall, these three empirical studies indicate that socially responsible companies are less tax-aggressive.

Other Stakeholder Groups’ Reactions to CTSs

Prior tax accounting literature provides some empirical evidence on investor reactions to aggressive CTSs by means of archival data (e.g., Abdul Wahab and Holland 2012; Cloyd et al. 2003; Desai and Dharmapala 2009; Desai and Hines 2002; Hanlon and Slemrod 2009; Seida and Wempe 2004; see Table 1 for an overview of prior empirical studies). Hanlon and Slemrod (2009) conducted an event study that analyzed stock price reactions to news about aggressive CTSs. They found that, on average, the price of corporations’ stock declines when news spreads about aggressive CTSs. Other researchers analyzed stock price reactions to announcements of corporate inversions (i.e., when a corporation is reorganized such that the parent corporation is located in a low-tax jurisdiction). Cloyd et al. (2003) observed nonsignificant or negative stock price reactions to the announcement of a corporate inversion, whereas Desai and Hines (2002) showed positive stock price reactions. Desai and Dharmapala (2009) explored the effect of tax avoidance on firm value by means of a panel data analysis. Using book-tax differences as a proxy for tax avoidance, they found nonsignificant effects on firm value for a sample of U.S. corporations. In contrast, Abdul Wahab and Holland (2012) even showed a negative effect of aggressive CTSs on firm value in the United Kingdom. Overall, except for Desai and Hines (2002), analyses of the effect of aggressive CTSs on investors yielded negative or nonsignificant results. The authors explained these results by the non-tax costs of aggressive CTSs. One type of non-tax cost is the political cost of being labeled socially irresponsible (e.g., Cloyd et al. 2003; Hanlon and Slemrod 2009). According to these studies, investors seem to expect negative consumer reactions to aggressive CTSs. By conducting an event study, Hanlon and Slemrod (2009) supported this assumption. They showed that stock price declines due to media reports on aggressive CTSs of companies in consumer-oriented industries are more pronounced than for companies in other industries. However, the authors’ study neither isolates the pure effect of aggressive CTSs on consumers nor confirms whether the more negative stock price reactions in these industries can really be attributed to anticipated consumer reactions.

Table 1 Overview of previous empirical studies addressing the effects of aggressive CTSs on investors
Full size table

Conceptual Framework

For the present study, we consider aggressive (responsible) CTSs to be negative (positive) CSR activities. Theoretically, this classification can be derived from Carroll’s Pyramid of CSR (Carroll 1991). According to Carroll, one of the most prestigious scholars in CSR research (Garriga and Melé 2004), “the social responsibility of business encompasses the economic, legal, ethical, and discretionary expectations that society has of organizations at a given point in time” (Carroll 1979, p. 500). These four domains of CSR are incorporated in his Pyramid of CSR (Carroll 1991). The economic domain requires companies to be as profitable as possible (Carroll 1991). The legal responsibility is defined as obeying or complying with the law (Carroll 1991). The ethical domain of CSR entails “any activities or practices that are expected by members of society although not codified in law, i.e., the responsibilities which embody those standards, norms, and expectations that reflect a concern for what consumers, employees, shareholders, and the community regard as fair, just, or in keeping with the respect or protection of stakeholders’ moral rights” (Carroll 1991, pp. 40–42). Finally, the discretionary domain encompasses the desire of the company’s stakeholder groups for the company to act as a good corporate citizen, e.g., by making financial contributions to the arts (Carroll 1991). In an alternative approach, Schwartz and Carroll (2003) suggested integrating this latter category into the economic and/or ethical domain. Consequently, we neglect the discretionary domain in further considerations. According to Carroll (1991), compliance with the economic and the legal responsibilities is required by the company’s stakeholder groups, whereas being ethical is expected.

Turning to the responsibilities of businesses with regard to taxes, aggressive CTSs lead to lower tax liabilities and thus they are consistent with the economic responsibility. Furthermore, this strategy constitutes compliant tax behavior and is in line with the legal responsibility. However, Schwartz and Carroll (2003, p. 510) would classify aggressive CTSs as “opportunistic compliance” as companies take advantage of loopholes in legislation (e.g., tax arbitrage) or choose to operate in jurisdictions because of weaker legal standards (e.g., tax havens). As outlined above, aggressive CTSs reduce national tax revenue and transfer a substantial share of the tax burden from large multinational companies onto smaller companies and consumers (e.g., Abdul Wahab and Holland 2012; Avi-Yonah 2000, 2008; Landolf and Symons 2008; Slemrod 2004). Consequently, aggressive CTSs are inconsistent with the ethical domain of CSR, whereas responsible CTSs are in line with this domain. These considerations correspond to the generally accepted (Dahlsrud 2008) definition of CSR according to the Commission of the European Union (2001, p. 4): “Being socially responsible means not only fulfilling legal expectations, but also going beyond compliance and investing ‘more’ into human capital, the environment and the relations with stakeholders.”

Effects of the CTSs on Corporate Reputation

Corporate reputation represents the knowledge and feelings held by individuals about a corporation (Hall 1992; Schwaiger 2004). A strong reputation is an appropriate tool to achieve strategic competitive advantages (Schwaiger 2004). Drawing on legitimacy theory, companies that fail to fulfill societal expectations lose social legitimacy and develop a negative reputation (Aguilera et al. 2007). Scholars define legitimacy as “a generalized perception or assumption that the actions of an entity are desirable, proper, or appropriate within some socially constructed system of norms, values, beliefs, and definitions” (Suchman 1995, p. 574). Following Carroll’s (1991) argumentation, in contrast to responsible CTSs, aggressive CTSs conflict with stakeholder expectations. Consequently, aggressive CTSs lead to a loss of legitimacy and a negative reputation, whereas responsible CTSs produce a positive reputation. Brown and Dacin (1997) presented evidence suggesting that corporate reputation can be damaged by negative CSR activities. Related research showed that positive CSR activities can have positive effects on corporate reputation (Eberl and Schwaiger 2006; Fombrun et al. 2000). Hence, we suggest that aggressive (responsible) CTSs negatively (positively) affect corporate reputation.

Effects of the CTSs on Purchase Intention

Prior experimental studies have also found effects of CSR activities at the product brand level. According to Brown and Dacin (1997), one major influencing factor in consumers’ brand choices is their perception of CSR. Earlier research provided evidence of the detrimental effect of negative CSR activities on consumer purchasing behavior (Brown and Dacin 1997; Carrigan and Attalla 2001; Creyer and Ross 1997; Folkes and Kamins 1999; Mohr and Webb 2005). Conversely, positive CSR activities enhanced consumer purchase intention (Auger et al. 2008; David et al. 2005; Du et al. 2007; Mohr and Webb 2005; Sen and Bhattacharya 2001). Mohr and Webb (2005) explained the effects of CSR activities on consumers’ intent to purchase by arguing that CSR activities add value to a product. Positive (negative) CSR activities might bolster (diminish) the value of the product. Turning to CTSs, responsible (aggressive) CTSs might be value-creating (-destroying) activities from the consumer perspective. Consequently, we hypothesize that aggressive (responsible) CTSs have a negative (positive) effect on consumer purchase intentions.

Effects of the CTSs on Willingness to Pay

With regard to consumers’ willingness to pay, prior research has yielded equivocal evidence. In a study of Bhattacharya and Sen (2004), most consumers were unwilling to pay a price premium for the products of a socially responsible company. Creyer and Ross (1996) found that, compared to a control group that received only general information, respondents were not willing to pay more for the products by the socially responsible firm, but they did demand lower prices on products sold by the socially irresponsible firm. In contrast, in a further experiment, Trudel and Cotte (2009) found that consumers were both willing to pay more for products of socially responsible companies compared to a control group and wanted to pay less for products of socially irresponsible companies. Nevertheless, the socially irresponsible scenario had twice the impact of the positive scenario on consumers’ willingness to pay. Auger et al. (2008) also showed that consumers expressed their willingness to pay more for products made ethically. Overall, the effects of the CTSs on consumers’ willingness to pay are not easy to predict, especially with regard to responsible CTSs. According to Bhattacharya and Sen (2004), positive CSR activities are most likely to result in consumers’ willingness to pay a price premium when the additional money is clearly earmarked for specific CSR activities. When it comes to CTSs, consumers cannot be sure that their willingness to pay for the product directly influences the amount of corporate and trade taxes paid by the company. However, following Mohr and Webb (2005), we argue that aggressive CTSs diminish the value of the products sold by the company. Hence, we predict that consumers tend to punish (reward) aggressive (responsible) CTSs by a diminished (enhanced) willingness to pay. Overall, we hypothesize the following:

H 1

Compared with a neutral control condition, an aggressive CTS has a negative effect on corporate success (i.e., (a) corporate reputation, (b) consumer purchase intention, and (c) willingness to pay).

H 2

Compared with a neutral control condition, a responsible CTS has a positive effect on corporate success (i.e., (a) corporate reputation, (b) consumer purchase intention, and (c) willingness to pay).

Moderating Effects

Sen and Bhattacharya (2001) propose that consumers’ reactions to CSR are contingent on the amount of congruence they perceive between the company’s character, as revealed by its CSR activities, and their own (i.e., consumer-company congruence). With regard to consumer reactions to CTSs, we expect the consumers’ tax mentality, i.e., their attitude toward taxation (Lewis 1979), to determine the perceived consumer–company congruence. Among the several dimensions of tax mentality, tax morale and one’s attitude toward taxation in particular might have an impact. When consumers personally support the social issues that the company targets, they are likely to see greater congruence between themselves and the company and are more likely to respond favorably to its CSR activities (Sen and Bhattacharya 2001). When it comes to CTSs, consumers that disapprove of tax evasion and tax avoidance perceive a strong congruence between their own character and that of a responsible tax-planning company, i.e., a company that renounces tax avoidance. These consumers are likewise aware of a lack of congruence between their character and that of an aggressive tax-planning company, i.e., a company that avoids taxes by all legal means possible.

According to Schmölders (2006), we define tax morale as a taxpayer’s attitude toward the notion that people are obligated to satisfy their individual tax liabilities. That is to say, tax morale represents one’s attitude toward illegal tax evasion. In contrast, we define the attitude toward tax avoidance as one’s attitude toward practices that aim to legally minimize tax payments. Consumers with a high tax morale and a negative attitude toward tax avoidance (i.e., a positive attitude toward taxation) might exhibit strong support for a responsible CTS. Consequently, companies that practice responsible CTSs might enhance corporate reputation, purchase intention, and the willingness to pay among those consumers in particular. By the same token, we assume that these consumers will penalize a company that pursues an aggressive CTS more than consumers with a low tax morale and a positive attitude toward tax avoidance (i.e., a negative attitude toward taxation).

Earlier empirical research indicated that consumers’ reactions to CSR activities depend on their personal attitude toward the particular CSR activity (e.g., Bhattacharya and Sen 2004; Creyer and Ross 1997; Mohr and Webb 2005; Sen and Bhattacharya 2001). For example, Bhattacharya and Sen (2004) found that consumers who were willing to spend more money in the case of socially responsible companies tended to be strong supporters of the company’s CSR activity. Sen and Bhattacharya (2001) revealed that consumers who strongly support a company’s CSR activity evaluate that company in a more positive light than consumers whose support for this activity is low.

We expect tax morale and the attitude toward tax avoidance to moderate the effects of the CTSs on measures of corporate success. Therefore, we put forth the following hypotheses:

H 3

Compared with a low tax morale, a high tax morale strengthens the effect of a CTS on corporate success (i.e., (a) corporate reputation, (b) consumer purchase intention, and (c) willingness to pay).

H 4

Compared with a positive attitude toward tax avoidance, a negative attitude toward tax avoidance strengthens the effect of a CTS on corporate success (i.e., (a) corporate reputation, (b) consumer purchase intention, and (c) willingness to pay).

The conceptual framework is depicted in Fig. 1.

Fig. 1
figure 1

Conceptual framework

Full size image

Empirical Study 1

Design

We organized our investigation in the form of a laboratory experiment. In study 1, we investigated the effects of aggressive and responsible CTSs on corporate success with consumers using a three (CTS: aggressive vs. responsible vs. control conditions) × one (industry: groceries) between-subjects design (see Fig. 2). Within the experiment, we systematically varied the levels of CTSs for a fictitious company within the fast-moving consumer goods (FMCGs) sector. We analyzed the effects of the CTSs on FMCGs because this is a key sector in marketing research and a reference point for marketing studies in other sectors (Holzmüller 2005). We aimed at investigating an industry that is relevant to our participants. Given that the groceries industry is among the most important industries from the consumer perspective within the FMCGs sector (Axel Springer 2009), we chose this industry for the purpose of study 1.

Fig. 2
figure 2

Set-up of study 1 and 2

Full size image

We used a fictitious company to control for prior learning and threats to internal validity (Boush and Loken 1991; Brown and Dacin 1997). To test the effects of the CTSs, we created a German company called Brandl from the groceries industry and the corresponding product brand Delizioso Pizza. This kind of food is a favorite among German consumers (Financial Times 2007) and thus was highly relevant to our participants.

Stimuli

Newspaper cover pages of the well-known German business paper Handelsblatt served as test stimuli for the aggressive and the responsible CTS conditions. Under the control condition, participants did not receive newspaper articles. Our study focuses on media reports on CTSs because, due to tax secrecy legislation, tax-related information is primarily published by investigative journalists, who therefore represent the main information source for consumers. Furthermore, the media play a key role in bringing ethical issues to the attention of the public (Dyck et al. 2010; Miller 2006); we reasoned that the use of a business paper would create an objective and reliable source compared to a more popular paper. The cover page featured a fictitious article that reported on either a responsible or an aggressive CTS on the part of Brandl. In addition, each newspaper article provided information on the corporate brand, the product brand, the industry, and Brandl’s German headquarters.

The article on the aggressive CTS was structured in the following way. First, the article stated that journalists have discovered that Brandl has been using an aggressive CTS. Second, the article provided a definition of this strategy as well as an explicit reference to its legality and its objective (avoiding German corporate and trade taxes in order to enhance the company’s competiveness). Third, examples of the aggressive tax behavior (shifting profits to captive finance companies, sales companies, and research and development companies in low-tax countries, as well as tax arbitrage) were given. Finally, the article outlined the societal consequences of aggressive CTSs (reduction in tax revenues, transfer of the tax burden to smaller companies and households).

The article that was part of the responsible CTS condition contained the following information. At the beginning, the article revealed that Brandl was commended for its responsible CTS by a nongovernmental organization. Then, the article outlined the objective of the CTS. After that, it was specified that — in contrast to Brandl — several other companies are known to use aggressive CTSs. Similarly to the article that was part of the aggressive CTS condition, a definition of this strategy, an explicit reference to its legality, as well as the above-mentioned examples and societal consequences of aggressive CTSs were given.

Procedure and Measures

One hundred and fifty undergraduate students from a major German university participated in the study 1 (M age = 24.53, 50.00 % female) in exchange for participating in a lottery for Amazon vouchers. Half of the students were enrolled in a business administration program (50.00 %). Students who were enrolled in an economics program or industrial engineering studies represented 9.30 and 11.30 %, respectively, of the sample. The remaining 29.40 % were enrolled in miscellaneous programs.

The validity and generalizability of student samples have been questioned in earlier research because the student population does not represent the general population of consumers. However, the use of student respondents was not a drawback for the present study. First, the respondents were a relevant consumer segment for the tested product (groceries). Second, about 57 % of the students lived in households with a monthly income after tax of more than 1,000 euros. Thus, we assume that at the very least members of their households have already paid income tax. In addition, like all consumers, our respondents pay value-added tax and consumption taxes. Third, for theory-testing research, a student sample has been deemed acceptable and even desirable because a maximally homogeneous sample (e.g., a student sample) has important advantages for theory validation research (Calder et al. 1981).

The participants were recruited outside class time on campus by research assistants and guided to the behavioral laboratory facility. In the behavioral laboratory facility, we randomly assigned our subjects to the experimental conditions. Participants under the aggressive (responsible) CTS condition were provided with the test stimuli in the form of a newspaper report on Brandl’s aggressive (responsible) CTS. Subjects under the control condition were not presented with reading material prior to testing.

After reading through the newspaper report, the participants under each of the experimental conditions were given a paper-and-pencil questionnaire. In the questionnaires, fictitious logos of Brandl and its product brand were shown. We created logos that consisted of the company’s name and a chef’s hat. Consequently, participants were aware of the link between the company and the product.

First, the subjects had to evaluate the dependent measures corporate reputation (Homer 1995), purchase intention (MacKenzie et al. 1986), and willingness to pay (Völckner 2006). Given the equivocal link between CTSs and CSR in the literature, we asked for their perception of Brandl’s CSR (Wagner et al. 2009) to ascertain if CTSs really affected the participants’ perceptions of CSR. Second, we measured the two potential moderating variables, tax morale (Körner and Strotmann 2006) and attitude toward tax avoidance (Lewis 1979). We drew on prior research as the source of measures for all of the constructs. Corporate reputation and purchase intention were measured using seven-point multi-item bipolar scales that consisted of 1 (i.e., the first statement) and 7 (i.e., the second statement). Except for willingness to pay, all remaining variables were obtained using seven-point multi-item Likert-type scales, which ranged from 1 (“disagree completely”) to 7 (“agree completely”). We determined willingness to pay by asking the participants directly for the amount they were willing to pay in order to receive the product. Table 2 summarizes the variables, scales, and measures of their reliability and validity. At the end of the questionnaire, except for the control condition, we assessed whether the subjects perceived the test stimuli as intended (manipulation check) and whether the subjects expected the company to utilize the respective CTS. Furthermore, the participants had to rate the trustworthiness of the test stimuli on a seven-point Likert scale. In doing so, we wanted to investigate whether the participants perceived the fictitious articles as believable and whether there were any significant differences in trustworthiness created by the different treatments. Finally, the participants had to complete demographic questions. After the experiment, we debriefed the subjects in a written form. Everyone received a note that contained information on the purpose of the study and the fictitiousness of the newspaper article.

Table 2 Measures and properties
Full size table

Results

Participants and Measures

The experimental conditions did not differ with respect to the participants’ age (F(2, 147) = 1.53, p = .220), household income (F(2, 147) = 1.17, p = .312), and gender (χ 2(2, N = 150) = 1.12, p = .571). Given Nunnally’s (1978) suggestion of .70, the multi-item measures showed good levels of reliability with respect to the dependent measures (α = .94–.97) and the attitude toward tax avoidance (α = .73). An exploratory factor analysis proved that the multi-item scales exhibited an acceptable level of convergent validity. For the dependent measures, the degree to which the variation in an item was explained by its factor exceeded .84 (minimum according to Fornell and Larcker 1981: .50). All of the items loaded on a single factor for each construct. Table 2 displays all reliability and validity measures.

Manipulation Check and Controls

A manipulation check provided evidence to suggest that the subjects perceived the test stimuli as intended. In asking for the pre-study expectations regarding Brandl’s CTS, we found no significant difference between the aggressive and responsible CTS conditions (M aggr = 3.18 vs. M resp = 3.38, t(98) = .67, p = .501). Similarly, there were no differences with respect to the trustworthiness of the articles under both conditions (M aggr = 4.60 vs. M resp = 4.74, t(98) = .45, p = .654). In addition, we performed an analysis of variance (ANOVA) to evaluate if the CTSs really affected the perception of CSR. The effect was highly significant for Brandl (F(2, 147) = 152.18, p < .001). The planned contrasts indicated that the aggressive (responsible) CTS decreased (increased) the perception of CSR significantly more than the control condition (M aggr = 1.57 vs. M control = 3.69 vs. M resp = 4.97, p aggr vs. control < .001, p resp vs. control < .001, p aggr vs. resp < .001).

Effects on the Dependent Measures

We performed ANOVAs to evaluate the effects of the CTSs on corporate success with consumers. Post-hoc tests based on the Games-Howell procedure should insure the significance of all pairwise comparisons. To evaluate the size of the effects, we computed Cohen’s d (Cohen 1988). Our results revealed that the CTSs had a highly significant impact on corporate reputation (H1a,2a) (F(2, 147) = 59.44, p < .001). The planned contrasts showed that the aggressive CTS decreased the reputation of Brandl significantly more than the control condition, whereas the responsible CTS significantly increased the reputation (M aggr = 2.80 vs. M control = 4.32 vs. M resp = 4.93, p aggr vs. control < .001, p resp vs. control < .001, p aggr vs. resp < .001). The sizes of the effects indicated a large impact on reputation (all d values ≥ 0.80). Furthermore, there was a significant primary effect of the CTSs on the consumers’ intention to purchase products from the product brand (H1b,2b) (F(2, 147) = 21.02, p < .001). The planned contrasts suggested that the responsible CTS condition produced greater intent to purchase products related to Delizioso than the control and the aggressive CTS conditions (M resp = 4.37 vs. M control = 3.64 vs. M aggr = 2.67, p aggr vs. control = .001, p resp vs. control = .002, p aggr vs. resp < .001). The effect sizes indicated a medium effect (0.80 > all d values ≥ 0.50). In addition, the results showed that the CTSs had a significant impact on consumers’ willingness to pay for Delizioso (H1c,2c) (F(2, 147) = 6.12, p = .002). The planned contrasts revealed that the willingness to pay was lower if the company employed an aggressive CTS than under the control condition (M aggr = 1.57 vs. M control = 2.27, p aggr vs. control = .016) and the responsible CTS condition (M resp = 2.37, p aggr vs. resp = .002). Cohen’s d indicated a medium effect. Although the willingness to pay was higher under the responsible CTS condition than under the control condition, the planned contrasts were nonsignificant (p resp vs. control = .460). Overall, we could confirm H1 and H2. Tables 3 and 4 depict the results of the analyses.

Table 3 Means and standard deviations for CTSs and the three dependent variables
Full size table
Table 4 Summary of one-way analysis of variance evaluating the effects of CTSs on the three dependent variables
Full size table

Moderating Effects

We tested H3 and H4 by moderation analysis (Aiken and West 1991). We found a significant interaction effect of tax morale on the impact of the CTSs on corporate reputation (H3a, β = .35, F(1, 146) = 3.58, p = .031). Contrary to H3, we found no significant interaction effect between the CTSs and tax morale with respect to consumers’ purchase intentions toward Delizioso (H3b, β = .12, F(1, 146) = 3.05, p = .291) and their willingness to pay for Delizioso (H3c, β = .02, F(1, 147) = .04, p = .422). For H3a–c, the main effects of the CTSs remained robust. Contrary to our assumptions, the interaction effects of the CTSs and the attitude toward tax avoidance on the corporate reputation (H4a, β = .22, F(1, 146) = 3.01, p = .079) and the subjects’ purchase intention toward Delizioso (H4b, β = −.05, F(1, 146) = 3.06, p = .404) were nonsignificant. According to our hypothesis, the willingness to pay for Delizioso was moderated by the attitude toward tax avoidance (H4c, β = .39, F(1, 146) = 3.98, p = .024). For H4a–c, the main effects of the CTSs remained robust.

Discussion

The results of our study provide evidence that CTSs impact corporate success with consumers. Aggressive CTSs diminish corporate success with consumers, whereas responsible CTSs enhance it. We could confirm our hypotheses with regard to corporate reputation and purchase intention. Although the primary effect of the CTSs on the consumers’ willingness to pay was significant, the planned contrasts between the responsible and the control conditions were nonsignificant. Our hypotheses about the moderating effects could be partially supported. Consumers’ tax morale moderated the effects of the CTSs on corporate reputation, whereas their attitude toward tax avoidance moderated the effects on consumers’ willingness to pay.

Before drawing conclusions from our findings, we have to address a limitation of the present study. The use of hypothetical brands enhances the internal validity of an experiment; however, it represents a serious threat to the external validity of the results (Boush and Loken 1991). Consequently, we organized a second study in order to overcome these limitations and add to the external validity of our findings.

Empirical Study 2

Design

The purpose of study 2 was to replicate the design of study 1, but to use two real companies in order to enhance the generalizability of the results. This time, we analyzed further industries within the FMCGs sector that are among the most important industries from the consumer perspective (Axel Springer 2009): toiletries and beverages. We created a three (CTS: aggressive vs. responsible vs. control conditions) × two (industries: beverages vs. toiletries) between-subjects design (see Fig. 2). In accordance with Brown and Dacin (1997), we tested well-known companies. Therefore, we chose the three companies headquartered in Germany with the highest category sales within each industry (Welt 2009). For toiletries, we selected Henkel, Beiersdorf, and DM, and for the beverages industry Radeberger KG, Bitburger GmbH, and InBev.Footnote 4 Then, we conducted a pretest (N = 387) to identify the company with the highest company awareness ratings in each industry. We selected Henkel (with the product brand Schwarzkopf) because this company had higher company awareness ratings than Beiersdorf (Nivea) and DM (Balea) (AwHenkel = .99, AwBeiersdorf = .97, AwDM = .69). Additionally, we chose Radeberger KG (Radeberger) over Bitburger GmbH (Bitburger) and InBev (Becks) (AwRadeberger = .82, AwBitburger = .77, AwInBev = .37).

Procedure, Stimuli, and Measures

As in the first study, undergraduate students from a major German university were approached outside class time by research assistants and guided to the behavioral laboratory facility. Once again, the participants could take part in a lottery for Amazon vouchers. Nearly half of the students were enrolled in a business administration program (45.80 %). Students who were enrolled in an economics program or industrial engineering studies represented 13.30 and 11.10 %, respectively, of the sample. The remaining 29.80 % were enrolled in miscellaneous programs. In the behavioral laboratory facility, we randomly assigned 360 undergraduate students (M age = 23.73, 53.33 % female) to one of the six conditions. Except for the two control conditions, participants were provided with a fictitious newspaper report on an aggressive (a responsible) CTS of the respective company (Henkel or Radeberger). Subjects under the control conditions were not presented with reading material prior to testing. The fictitious newspaper articles were similar to those used in study 1 except that we changed the corporate and product brand and adapted the text to the respective industry for Henkel and Radeberger.

After reading through the newspaper report, the participants were given a paper-and-pencil questionnaire under each of the experimental conditions. In the questionnaires, the actual logos of the companies and their product brand were shown. Thus, participants were aware of the link between the respective company and the product. All dependent and moderating variables were the same as those used in the first study. In addition, we asked similar manipulation check and demographic questions. After completing the questionnaire, we debriefed the subjects in written form. Specifically, we informed our subjects about the fictitiousness of the newspaper articles and the purpose of the study.

Results

Participants and Measures

All of the conditions had a homogeneous group structure with respect to age (F pooled sample(2, 356) = .16, p = .855), household income (F pooled sample(2, 357) = 1.07, p = .344), and gender (χ 2pooled sample (2, N = 359) = 1.43, p = .489). An exploratory factor analysis indicated a good level of convergent validity. The multi-item measures showed good levels of reliability with respect to the dependent measures (α = .91–.97) and the attitude toward tax avoidance (α Radeberger = .63, α Henkel = .74). All constructs loaded on a single factor. Table 2 includes all reliability and validity measures.

Manipulation Check and Controls

The manipulation check suggested that the subjects perceived the test stimuli as intended. The perceived trustworthiness of the stimuli was equal under all of the conditions (Henkel: M aggr = 4.88 vs. M resp = 4.57, t(118) = 1.18, p = .239; Radeberger: M aggr = 5.20 vs. M resp = 5.03, t(118) = .66, p = .255). Furthermore, the subjects’ expectations toward Henkel’s and Radeberger’s CTS prior to the study did not differ under the different conditions (Henkel: M aggr = 3.60 vs. M resp = 3.43, t(118) = .54, p = .593; Radeberger: M aggr = 3.28 vs. M resp = 3.43, t(118) = .57, p = .568). In the following sections, we present the results of the pooled sample. We analyzed the effects of the CTSs on consumer perception of the companies’ CSR by means of an ANOVA. We detected a significant primary effect of the CTSs (F(2, 357) = 300.31, p < .001). The planned contrasts showed that the CTSs affected the participants’ perceptions of CSR as we had assumed (M aggr = 1.95 vs. M control = 3.62 vs. M resp = 5.00, p aggr vs. control < .001, p resp vs. control < .001, p aggr vs. resp < .001).

Effects on the Dependent Measures

As predicted in H1a and H2a, the CTSs had a significant impact on corporate reputation (F(2, 357) = 57.27, p < .001). The planned contrasts revealed that an aggressive CTS decreased the corporate reputation significantly more than the control condition, whereas the responsible CTS condition enhanced the corporate reputation significantly more than the control condition (M aggr = 3.58 vs. M control = 4.41 vs. M resp = 5.14, p aggr vs. control < .001, p resp vs. control < .001, p aggr vs. resp < .001). The impact of the CTSs on corporate reputation was mainly medium or large. Furthermore, we detected a significant primary effect of the CTSs on consumers’ intentions to purchase products from the investigated product brands (H1b,2b) (F(2, 357) = 11.07, p < .001). The responsible CTS condition (M resp = 4.56) generated higher purchase intentions than the control condition (M control = 3.86) and the aggressive CTS condition (M aggr = 3.38). The planned contrasts indicated that all of the comparisons were statistically significant (p resp vs. control = .003, p aggr vs. resp < .001), except for the comparison between the aggressive CTS condition and the control condition (p aggr vs. control = .142). However, those means were in accordance with the hypothesized direction of the effect. The CTSs mainly had a small (0.50 > d ≥ 0.20) or medium effect on consumer intentions to purchase products from the brands. Contrary to our assumption, the CTSs had a nonsignificant primary effect on consumers’ willingness to pay a premium for the product brand (H1c,2c) (F(2, 351) = 2.12, p = .061).Footnote 5 While the means were in accordance with our hypotheses (M aggr = 1.68 vs. M control = 1.87 vs. M resp = 1.97), the planned contrasts revealed only statistically significant differences in the willingness to pay between the responsible and the aggressive CTS condition (p aggr vs. control = .197, p resp vs. control = .384, p aggr vs. resp = .045).

The disaggregate results, i.e., the individual results for the companies Henkel and Radeberger, were very similar to the pooled results with respect to the corporate reputation, purchase intention, and willingness to pay. However, the planned contrasts at the disaggregate level were nonsignificant in the case of purchase intention between the responsible and the control conditions for Henkel. In contrast, at the disaggregate level, the effects of the CTSs on the willingness to pay were significant for Radeberger (F(2, 177) = 3.00, p = .027). Tables 3 and 4 depict the pooled and disaggregate results.

Moderating Effects

As H3a suggested, in the pooled sample, we found a significant interaction between the CTSs and tax morale on the corporate reputation (H3a, β = .29, F(1, 356) = 4.96, p = .014). The primary effect of the CTSs on corporate reputation remained significant. However, there were nonsignificant effects on purchase intention (H3b, β = .13, F(1, 356) = .78, p = .190) and on consumers’ willingness to pay (H3c, β = −.01, F(1, 351) = .01, p = .462). Our analysis of the moderating impact of the attitude toward tax avoidance (H4) yielded significant results in the pooled sample for all dependent variables. Given that the interaction between the CTSs and the attitude toward tax avoidance had a significant effect on the corporate reputation (H4a, β = .30, F(1, 356) = 5.66, p = .009), purchase intention (H4b, β = .28, F(1, 356) = 5.66, p = .023), and willingness to pay (H4c, β = .24, F(1, 356) = 2.76, p = .049), a more negative attitude toward tax avoidance, i.e., higher values on the scale, enhanced the influence of the CTSs on the three dependent measures. For H4a–c, the main effects of the CTSs on the dependent variables were significant.

Discussion

Study 2 confirmed the findings of study 1, namely that aggressive (responsible) CTSs significantly diminish (enhance) corporate reputation and consumers’ purchase intentions. However, this time the main effect of CTSs on the willingness to pay was only significant for the brand Radeberger. Overall, study 2 strengthens our conclusions because we were able to replicate the findings of study 1 based on two real companies.

Nevertheless, all effect sizes were stronger in study 1 compared to study 2. These different effect sizes may be due to testing a hypothetical brand in study 1. Klink and Smith (2001) argue that hypothetical scenarios that provide the experimental subjects with restricted information overestimate the impact of the information given them. The marketplace is commonly considered information-rich. The use of a hypothetical brand restricts the amount of information that is available to subjects in the real marketplace. When subjects have limited information, they rely only on available cues (i.e., aggressive or responsible CTS) to make inferences. Conversely, the impact of a single cue diminishes as other relevant cues become available (Klink and Smith 2001). With regard to our research, we can conclude that study 1 might enhance the impact of CTSs on consumers by omitting information on other company and product attributes. In study 2, as in the real market place, subjects also base their decision on prior knowledge and experience with the brands. Under these circumstances, the impact of CTSs weakens.

General Discussion

Summary

In an interdisciplinary approach that links taxation, marketing, and CSR, we provide the first empirical evidence that a company’s tax strategy can influence corporate success with consumers. We found that a media report on an aggressive CTS damages corporate reputation and lowers brand purchase intention when compared to a neutral control group. Conversely, a media report on a responsible CTS generates the opposite effect. However, the effects of the CTSs on the consumers’ willingness to pay yielded mixed results. The subjects were not willing to reward responsible tax-planning companies with a higher willingness to pay a premium for those companies’ products, but they did tend to want to pay less for the products of aggressive tax-planning companies. Clearly, consumers are reluctant to pay higher prices for products in order to induce companies to pay their fair share of taxes. On the one hand, this might be due to the fact that households already carry a heavy tax burden, e.g., in the form of payroll tax and value-added tax. In Germany, for example, value-added tax and payroll tax are the largest source of tax revenue and represent about 57 % of total tax revenue (Destatis 2012). On the other hand, consumers cannot be sure that paying a price premium actually influences the amount of corporate and trade taxes paid by the company. This uncertainty might inhibit consumers from paying higher prices. The extant research underlines this hypothesis. Bhattacharya and Sen (2004) suggest that positive CSR activities are most likely to result in consumers’ willingness to pay a price premium when the additional money is clearly earmarked for specific CSR activities. According to a global survey by Nielsen (2011), many consumers reported a personal preference for eco-friendly goods, but only 22 % said they would pay more for such a product. Large percentages of respondents reported setting their preference aside and buying whichever product is cheapest. Altogether, there seems to be a gap between the consumers’ preferences for CSR and their willingness to pay a price premium. This gap might explain why our hypothesis about purchase intention was supported, whereas our hypothesis about willingness to pay was (partially) nonsignificant.

Consumer tax morale moderated the effects of the CTSs on corporate reputation exclusively. Given that the measure of tax morale assesses one’s attitude toward illegal tax evasion, consumers with a high tax morale oppose tax evasion, but do not necessarily reject aggressive (i.e., legal) CTSs. The ability to distinguish between an aggressive CTS that operates within the confines of legal tax behavior and one that allows for illegal tax evasion might have caused tax morale to often have nonsignificant moderating effects on the CTSs. Finally, we showed that the consumers’ attitude toward tax avoidance moderated the effects of the CTSs with the exception of corporate reputation and purchase intention in study 1. The nonsignificant moderating effect on the relationship between the CTSs and these two variables might be due to the use of a hypothetical brand in the first study. According to Boush and Loken (1991), the use of hypothetical brands represents a serious threat to the external validity of the results. However, because the consumers’ attitude toward tax avoidance significantly moderated the effects of the CTSs on all measures of corporate success in study 2, the results of this study appear to capture market behavior more accurately (i.e., in terms of higher external validity).

Implications

Theoretical Implications

Given the lack of a theoretical consensus on the link between CTSs and CSR in the prior literature, our study offers a theoretical foundation based on Carroll’s (1991) Pyramid of CSR for linking CTSs and CSR. Due to reductions in national tax revenue and a redistribution of the tax burden to fall on smaller companies and households, we argue that aggressive CTSs conflict with the ethical domain of CSR. In addition, we provide empirical evidence that CTSs affect consumer perceptions of CSR. The majority of consumers seem to hold companies responsible for paying taxes. Consequently, we argue that CTSs should be considered a subdomain of CSR.

Furthermore, our results support the negativity bias found in prior CSR research (Bhattacharya and Sen 2004; Creyer and Ross 1997; Folkes and Kamins 1999; Mohr and Webb 2005; Sen and Bhattacharya 2001). The effects of the aggressive CTSs were often larger than the effects of the responsible CTSs, especially with regard to what consumers are willing to pay. Hence, in line with Bhattacharya and Sen (2004), we can confirm that “doing bad” hurts more than “doing good” helps. Finally, our findings support the assumptions of prior empirical studies that analyze the effect of aggressive CTSs on investors (e.g., Hanlon and Slemrod 2009). These studies explain the rather negative results as being to some extent based on anticipated negative consumer-related outcomes of an aggressive CTS.

Managerial Implications

Our results are important for marketers, because—ideally—positive CSR activities (here: responsible CTSs) eventually affect corporate profit positively. Especially the opposite scenario, negative CSR activities (here: aggressive CTSs), should be taken into consideration since these activities can eventually decrease corporate profit. The objective of an effective CTS is to maximize after-tax returns (Scholes et al. 2009). An exclusive focus on tax minimization might introduce significant costs along non-tax dimensions. Given the fact that consumers constitute an important stakeholder group (Ferrell 2004; Freeman 2010; Peloza and Shang 2011), consumer-related non-tax costs and gains of CTSs covered in the media ought to be integrated into tax-related decision-making processes. Companies face a tradeoff in implementing their CTS: Aggressive CTSs are only in the financial interest of the company if the corresponding tax savings exceed the negative consumer-related effects.Footnote 6 In contrast, responsible CTSs pay off if the corresponding higher tax payments fall below the consumer-related benefits. By using the sizes of the effects as proxies for a quantification of the observed consumer-related effects, companies can expect strong consumer reactions with regard to corporate reputation, but only medium or small effects on consumer intent to purchase. In addition, aggressive tax-planning companies risk being punished by a slightly lower willingness to pay among their consumers, whereas responsible tax-planning companies cannot expect to achieve a price premium. Hence, within a short-term perspective, implementing a responsible CTS will require up-front costs in the form of higher tax payments that might not be compensated by higher consumer-related financial payoffs. These findings are in line with prior CSR research. Typically, financial payoffs of CSR activities are not guaranteed and will take time to materialize (Mohr and Webb 2005).

In addition, this research has implications for marketing departments. The positive consumer-related effects of a responsible CTS covered by the media indicate that a responsible CTS can be integrated successfully into corporate communication activities. Companies that inform their consumers about their responsible CTS by means of their CSR reports (e.g., Anglo American, Diageo, Johnson & Johnson, Vodafone) or advertising campaigns (e.g., Liqui Moly in Germany) might enhance their reputation. By integrating information about the responsible CTS into the commercialization of the companies’ product brands, marketing managers might increase their consumers’ purchase intention. These findings might encourage companies to proactively disclose their tax strategy. However, two caveats must be expressed. First, this research suggests that consumers are unwilling to pay a price premium for products sold by responsible tax-planning companies. Consequently, a responsible CTS is not an appropriate justification for higher prices than competing products. Therefore, responsible tax-planning companies have to be careful that their consumers do not expect their products to be more expensive than competing products. Second, companies have to insure that their tax department actually follows a responsible approach to tax. Prior research suggests that inconsistent CSR information elicits strongly negative reactions from consumers (Wagner et al. 2009). These findings suggest that companies will suffer a negative effect on their success with consumers if the communication of a responsible CTS is followed by the revelation of an aggressive CTS.

This research suggests that the strength of consumer reactions depends on the consumers’ attitudes toward taxation (tax morale and attitude toward tax avoidance). Prior research (e.g., Alm and Torgler 2006; Körner and Strotmann 2006; Riahi-Belkaoui 2004) shows differences in attitudes toward taxation among countries and individuals depending on the socioeconomic characteristics of the people (e.g., sex, religion, educational background, income, and patriotism). Consequently, companies might face different consumer reactions depending on their consumers’ socioeconomic characteristics. These findings might prompt companies to gain insight into their consumers’ attitudes toward taxation in order to adapt and communicate their CTS accordingly.

Finally, our findings are relevant to the organization and the performance measurement of corporate tax departments. Recent surveys among large companies show that the tax departments are to a large extent independent in determining the CTS (Erle 2008; Vera 2001). Since tax departments consist solely of tax specialists, who might not value CSR and marketing issues, organizational changes are necessary to improve the interface between the tax department and the marketing and CSR departments. These improvements could comprise the inclusion of CSR and marketing specialists in determining the CTS or training regarding CSR- and marketing-related issues for the tax department. In addition, the board of directors should define a code of conduct for tax issues to dictate corporate ethical standards with respect to the CTS (Erle 2008). As for the performance measurement of corporate tax departments, our findings underline that tax-related measures such as effective tax rates fail to measure the overall performance of the department. Performance measures that neglect the consumer-related effects of a CTS might encourage the tax department to adopt a CTS that is too aggressive.

Limitations

Our study has several limitations that are inherent to laboratory experiments. First of all, we cannot rule out the possibility that the observed effects are influenced by a social desirability bias. In addition, we have no evidence for the sustainability of the consumer-related effects. Future studies should determine the dynamic effects of CTSs using longitudinal or panel data. We artificially induced consumer awareness of both the companies’ CTSs and the link between the company and its product. A field study by Sen et al. (2006) showed that consumer awareness of CSR activities constitutes a crucial point in the real-world marketplace. Conducting an event study on media reports on CTSs might overcome these limitations. Furthermore, the present research suffers from a sample selection (student sample) that cannot claim to be representative of the larger population of consumers. Many of our participants might not have paid income taxes themselves in a significant way and thus are not able to compare their own tax burden with the company’s tax payments. In addition, about 70 % of the students were enrolled in business, economics, or industrial engineering classes. Compared to the average consumer, these students might be more appreciative of corporate efforts to minimize tax payments in order to maximize their profits.

Finally, there are limitations related to the test stimuli (business paper), the selected industries (FMCGs), and the national character of the sample. Additional research could enable the generalization of our findings to other product categories. Previous literature showed that ethical considerations generally play a more substantial role in the case of high-involvement products (Devinney et al. 2006). Furthermore, future research needs to be extended to an international context. There might be different effects in low-tax countries compared to a high-tax country such as Germany. Since the attitude toward taxation differs from country to country (e.g., Alm and Torgler 2006; Körner and Strotmann 2006; Riahi-Belkaoui 2004), the moderating impact of consumers’ tax morale and attitude toward tax avoidance would be different among consumers from different countries.

We are confident that our findings will stimulate future research that addresses the interface between taxation, CSR, and marketing.