Introduction

Many companies have come under criticism for their failures to respect the environment, the well-being of people, or human rights (e.g., Ikea, Nestlé, Nike, Spanghero). These criticisms raise ethical problems for consumers, who suffer lowered trust in retail companies (Lombart and Louis 2014). Although ethical research is a popular management topic in general (Brunk 2010), little attention centers on retail business ethics or the consumer’s perspective (Babin et al. 2004). Yet the economic crisis and its consequences represent an unparalleled opportunity for retailers to develop ethical practices and social policies for the good of society and their companies. Studies indicate a positive link between ethical perceptions and consumer responses, in terms of trust (Kenning 2008; Stanaland et al. 2011) and loyalty (Valenzuela et al. 2010). For example, noting consumers’ diminished purchasing power (OECD 2014), recent studies suggest discount prices or promotions as means to support social causes while also improving customer relationships (Andrews et al. 2014; Winterich and Barone 2011). Some retailers try to create discount spaces that offer a wider range of products at reduced prices, in an effort to help low-income consumers gain access to affordable products. Although a form of retail-oriented corporate social responsibility (CSR), such strategies have not been investigated previously, leaving unanswered the critical question: Do moral and socially responsible actions create value for retailers and consumers? Because such actions constitute a socialization of business practices (Babin et al. 2004; Lombart and Louis 2014), they might affect consumers’ trust in and loyalty toward the retailer; Kang and Hustvedt (2014) also emphasize the central role of trust, including consumer perceptions of transparency and CSR, for determining consumer loyalty behavior (e.g., purchase intentions).

Accordingly, with this study, we investigate whether a social policy to use discount spaces affects the relationships among retail consumers’ ethical perceptions, trust, and loyalty toward the retailer. Prior applied ethics literature considers multiple variables over time, but some of the rather mundane conclusions have not advanced theory or practice as far as is possible, because few studies examine the specifics of corporate policy. Noting this gap, we investigate the greatest societal problem affecting societies in economic crisis, using one corporate practice (social discount) as a potential moderator of relationships among already well-known variables (ethics, trust, reputation, CSR, and loyalty). Therefore, whereas previous research generally focuses on the direct effect of consumers’ ethical perceptions on loyalty (e.g., Valenzuela et al. 2010), we investigate not only these direct effects but also the potential mediating influence of consumer trust, the retailer’s CSR, and the retailer’s corporate reputation. Overlooking these possible mediation processes has limited the practical implications of prior research and left causality questions unaddressed. In addition, as a contribution to CSR research, we show that ethical perceptions affect consumer loyalty toward retailers that implement distinct social discount practices. To the best of our knowledge, no previous studies have addressed the moderating role of social discount practices, despite retailers’ need to understand their effectiveness as strategic tactics to increase sales in turbulent times, as well as improve their corporate image.

In the next section, we present our theoretical framework and main concepts, and then develop a series of research hypotheses based on previous studies. After we detail our research methodology, we present the results we derived from testing our measurement and structural models. Finally, we highlight the theoretical and practical implications of this research, as well as some limitations and directions for further studies.

Theoretical Framework and Research Hypotheses

Conceptual Framework

Few previous studies adopt consumers’ perspectives on business ethics (Barnett et al. 2005; Brunk 2012) or the related questions that increasingly confront retailers (e.g., Spanghero scandal).Footnote 1 According to Sherwin (1983), the term “ethics” refers to a set of moral norms, principles, or values that guide people’s behavior. In line with Barnett et al. (2005), Brunk (2012) contends that moral philosophy specifies two normative ethical theories: deontology and teleology. Deontology implies a non-consequentialist approach that guides evaluations; teleology represents a consequentialist view of moral judgment (Crane and Matten 2007). Specifically, deontology calls for rules based on moral norms, to guide actions in accordance with categorical imperatives of right or wrong. The morality of corporate behavior then can be defined only in legal terms (Clement 2006). A teleological evaluation instead is based on considerations of the possible outcomes of taking a particular action or some alternative route, including how much good or bad seemingly might result from either option (Brunk 2012).

We adopt the teleological approach and note that previous research has stressed the positive consequences of ethics (e.g., improving trust and reputation) for retail success (Lavorata and Pontier 2005; Tsalikis and Fritzsche 1989). For example, the food retailing sector has developed extensive ethics- and CSR-related strategies, such as fair trade, social equity, and cohesion measures. Such ethical practices, reflecting retailers’ CSR, seek to contribute to a better quality of life for employees, suppliers, and consumers (Lombart and Louis 2014). Yet even as retail chains implement various strategies based on social and human dimensions, especially related to employment practices (e.g., male–female parity, child labor), they still must maintain the demand of consumers who have grown increasingly sensitive to prices.

Some retail organizations work with local associations to support charitable causes (i.e., cause marketing) or open stores in disadvantaged neighborhoods (Lavorata and Pontier 2005). Prior studies have investigated such practices by focusing on poor consumers (Arnold and Valentin 2013; Piacentini et al. 2000; Weidner et al. 2010), but we instead investigate the reactions of consumers when retail companies try to implement discount practices as part of their social policy. In practice, retail companies must determine how to integrate low-income consumers into their target markets without alienating or losing their traditional consumers when they implement ethical policy programs.

We argue that consumers’ perceptions of ethics might lead to greater loyalty to the retailer (Tsalikis and Fritzsche 1989). Loyalty usually is conceptualized as attitudinal or behavioral (Dick and Basu 1994), though more differentiated approaches also distinguish among cognitive, affective, conative, and action loyalty (Oliver 1999). In addition, loyalty is closely associated with other key constructs, such as satisfaction, brand equity, and performance (Lai et al. 2010). We contend that the relationship between consumer perceptions of companies’ ethics and loyalty is even more complex.

Specifically, we anticipate mediating effects of trust (Yoo and Donthu 2001), corporate reputation (Lai et al. 2010), and CSR (Ailawadi et al. 2014). We define trust as a customer’s personal assessment of the trustworthiness of a specific retailer (Rampl et al. 2012). In line with Lai et al. (2010) and Martenson (2007), we define the retailer’s corporate reputation as consumer perceptions that create favorable, long-term images for the retailer, in comparison with other retailers. Then CSR involves environmental friendliness, community support, the sale of local products, and fair treatment of employees (Ailawadi et al. 2014). Although some prior studies combine CSR with ethical perceptions, we regard them as separate concepts, such that ethical perceptions are antecedents of CSR (Arnold and Valentin 2013; Stanaland et al. 2011). Therefore, for this study, ethical perceptions refer to the codes, moral principles, standards, and values that guide behavior with respect to what is right or wrong, whereas CSR pertains to companies’ obligations to maximize their positive and minimize their negative impacts on society (Dimitriades 2007).

In turn, companies’ social practices (e.g., discounts for low-income consumers) affect the relationships among consumers’ ethical perceptions of the retailer, trust, and loyalty to the retailer. Winterich and Barone (2011) note the effects of the interplay of self-construal and a specific social identity on consumers’ preferences for two types of promotions: discount-based and donation-based. Recent studies underline a moderating role of price discount practices on the relationship between ethical values (e.g., cause marketing) and consumer responses in the form of purchases (Andrews et al. 2014). Such findings clearly reveal the theoretical relevance of investigating combinations of discount practices and social policies. Accordingly, in Fig. 1, we summarize our proposed conceptual framework, detailing the relationships among business ethics, social discount practices, and consumer loyalty in the retail sector.

Fig. 1
figure 1

Research model. We controlled for the effects of socio-demographic variables (age, gender, income, education) and consumer familiarity with the retailer, using concurrent models

Full size image

Hypotheses Development

Direct effects of Consumers’ Perceptions of Retail Business Ethics on Loyalty

Store loyalty is a central issue for retailers, because of the severe competition that marks the retail sector. Consumers are more likely to exhibit loyalty to a retailer if they perceive and receive value from it (Román and Ruiz 2005; Stanaland et al. 2011). However, retailers should consider ethics in their loyalty-building processes, because recent scandals, such as the Spanghero case, have affected customer confidence in manufacturers and retail chains’ ethical practices. Ethical behavior, based on ethical values, is good for retail business in the long run (Babin et al. 2004); Tsalikis and Fritzsche (1989) even note a positive effect of ethical business on profitability (i.e., organizational performance), such that ethical behavior results in a strong competitive advantage, long-run profits, and better relationships with customers. Prior empirical studies also establish a direct link between customers’ perceptions of a retailer’s ethical conduct and their level of loyalty (Valenzuela et al. 2010); customers are more willing to demonstrate attitudinal and behavioral loyalty to an ethical retailer or supplier. They also are more inclined to conduct business with retailers and repurchase their products if they judge those retailers favorably on both monetary and non-monetary aspects (Babin et al. 2004; Van Kenhove et al. 2003). Therefore:

H1

Consumers’ perceptions of retail business ethics have a positive influence on consumers’ loyalty to the retailer.

Indirect Effects of Ethical Perceptions on Loyalty Through Trust

To ensure long-term interactions with customers, companies seek to encourage their trust (Morgan and Hunt 1994). Conscientious behavior by companies builds trust, which can indirectly increase consumer loyalty (Kang and Hustvedt 2014). Moreover, Chaudhuri and Holbrook (2001) find that trust is one of the most significant predictors of loyalty. Because of its important effect on loyalty, trust has been widely studied in marketing literature, which shows that customer perceptions of ethical behavior influence trust directly (Valenzuela et al. 2010), and then customer loyalty through trust (Alrubaiee 2012). We posit that trust might be associated with a specific retailer (Kenning 2008) and be morally based (Stanaland et al. 2011). For example, prior studies indicate a positive relationship between ethical perceptions and trust in a retailer, which then enhances store loyalty (Kenning 2008; Stanaland et al. 2011). Ethically justifiable and moral behaviors help establish trust (Hosmer 1995), and greater trust in a company can be a means to improve customer relations and ultimately consumer loyalty (Alrubaiee 2012). On the basis of this reasoning, we postulate:

H2

(a) Consumers’ perceptions of retail business ethics have positive influences on consumer trust in the retailer, which (b) has a positive effect on consumer loyalty.

Indirect Effects of Ethical Perceptions on Trust Through Corporate Reputation and CSR

Ethical perceptions might have indirect effects on trust through the CSR policy. As we noted previously, CSR issues can involve ethical dimensions, but we regard a CSR policy as a consequence of business ethics (Román and Ruiz 2005). On the one hand, better perceptions of a retailer’s business ethics likely improves consumers’ evaluations of its CSR activities (Stanaland et al. 2011), such as environmental programs, ethical products, or social actions, that then contribute to better evaluations of the company (Öberseder et al. 2013). Respect for human rights is an element of ethical CSR, rather than of its instrumental or economic versions (Arnold and Valentin 2013). By institutionalizing and promoting ethical values, retailers seek to induce positive consumer perceptions (Mohr and Webb 2005). On the other hand, a socially responsible firm can benefit from engaging in initiatives, because doing so builds positive outcomes, such as consumer attitudes, corporate image, and a good reputation (Hur et al. 2013). A retailer’s CSR policy thus may have a positive, significant impact on consumer trust (e.g., Lin et al. 2011). Kang and Hustvedt (2014) also establish a positive relationship between CSR and trust. Therefore, the influence of ethical perceptions on consumer trust might be indirect, through CSR, and we postulate:

H3

(a) Consumers’ perceptions of retail business ethics have positive effects on consumer evaluations of retailer CSR, which (b) influence consumer trust in the retailer.

Ethical perceptions also might exert an indirect effect on trust through corporate reputation, which can increase a retailer’s competitive advantage, positive customer attitudes, and loyalty (Hur et al. 2013; Lai et al. 2010; Martenson 2007). Brunk (2010) notes the link between consumer expectations of business ethics and corporate ethical practices, and Gilani (2011) indicates that corporate ethics are positively associated with views of a corporate brand. A good retail reputation results from acting properly and doing the right things. If consumers perceive the retailer’s behaviors as just, moral, and right, they regard it as having a good reputation and adopt a positive attitude toward that company (Biong et al. 2010; Tsalikis and Fritzsche 1989). This favorable reputation then helps companies build trust and identification among customers (Keh and Xie 2009). From a social perspective, a strong reputation based on ethical behavior enables retailers to develop trust, signals their capabilities and reliability, reduces risk perceptions, and strengthens confidence (Fombrun 1996). Therefore, we anticipate that corporate reputation mediates the relationship between ethical perceptions and consumer trust and propose:

H4

(a) Consumers’ perceptions of retail business ethics have positive influences on corporate reputation, which (b) has a positive effect on consumer trust in the retailer.

Moderating Role of Social Discount Practices

Discount practices can serve an important function in companies’ social policy (Andrews et al. 2014; Winterich and Barone 2011). Weidner et al. (2010) stress the theoretical and managerial relevance of identifying the needs of and providing access to affordable consumption situations for low-income consumers. In hypercompetitive markets, consumers demand discounts but also expect firms to be caring (Andrews et al. 2014). Therefore, retailers might combine discount practices with cause marketing, with the belief that if they develop ethical practices toward low-income consumers, they can improve their own social and economic performance (Piacentini et al. 2000). Such combinations have not been studied previously though (Andrews et al. 2014).

In many European countries, discount food retailers (e.g., Aldi, Dia, Lidl) compete with traditional retail chains (supermarkets and hypermarkets), offering a relatively limited assortment of food products at very low prices (e.g., 20–30 % less than competitors). Traditional retailers (e.g., Auchan, Carrefour, Intermarché) have begun to respond by creating special discount spaces within their stores that mimic the characteristics of discount stores, in an effort to attract less wealthy consumers and increase their sales. This approach constitutes an integrated social discount (ISD) design, in that the discount space is integrated within the store and fully managed by the retailer. Social discount spaces offer means to emphasize the retailer’s CSR policy and convey its ethical values, especially if retailers collaborate with local charities to develop these spaces, in a type of cause marketing. These practices are ethically grounded, in that they likely improve the purchasing power of low-income consumers and reduce the negative effects of poverty (Lavorata and Pontier 2005). We refer to this approach as a collaborative social discount (CSD) design, because the discount space is managed in collaboration with a social partner. As we revealed in Fig. 1, such social discount practices may moderate the three relationships we propose in our research model.

First, the relationship between consumers’ perceptions of retail business ethics and loyalty to the retailer should differ, depending on the retailer’s social discount practices. Socio-ethical perceptions generally have a positive impact on evaluations of the company and purchase intentions (Mohr and Webb 2005; Stanaland et al. 2011). Valenzuela et al. (2010) show that firm’s ethical level relates positively to loyalty to the firm, and it moderates the relationship between customer orientation and loyalty. However, Webb and Mohr (1998) also argue that consumers usually shop according to their price, quality, and convenience needs, rather than choosing retailers because of the social cause they support. Andrews et al. (2014) further show that price discounts moderate the relationship between cause marketing (based on ethical values) and consumer responses (purchases and sales revenues). Their field experiments affirm a positive influence of cause marketing on actual sales. Similarly, social discount practices should reinforce the link between consumers’ perceptions of retail business ethics and loyalty to the retailer. Therefore, we propose:

H5

The effect of consumers’ perceptions of retail business ethics on loyalty to the retailer is stronger when retailers create social discount space (integrated or collaborative) than when they do not.

Second, the relationship between consumers’ perceptions of retail business ethics and trust in the retailer should differ according to the retailer’s social discount practices. Trust tends to be stronger when firms communicate about their ethical and CSR initiatives, so retailers often use communications about their ethics “to improve a still negative image” (Lavorata and Pontier 2005, p. 3). Retail communications often focus on ethical activities, in an effort to improve consumer trust (Lombart and Louis 2014), especially in the course of the modern economic crisis. By developing discount spaces, retailers can enhance their social image, because they demonstrate that they offer accessible products for all consumers, such as through ISD. Although cause marketing, such as CSD, can benefit both the retailer and the charity, consumer skepticism also can invoke negative responses to such ethical actions (Singh et al. 2008; Webb and Mohr 1998). With deep discounts (e.g., 50 % off), cause marketing has only a limited influence on purchases, because customers experience reduced trust or warm-glow feelings (Andrews et al. 2014). Because CSD requires that retailers overcome any problems related to the charitable partner’s skills or legitimacy for managing products, the partnership might create complexity, and consumers could come to doubt firms’ motives or develop skepticism about their ethical actions (Seitanidi and Crane 2009). A partnership involving CSD spaces may be perceived as less positive and less credible, so we postulate:

H6

The effect of consumers’ perceptions of retail business ethics on trust in the retailer is stronger when it offers no or integrated social discount spaces rather than collaborative social discount spaces.

Third, the relationship between trust and loyalty may differ with social discount practices. Several studies confirm an important, positive effect of trust on loyalty, though this relationship also depends on the context (e.g., experience, atmosphere). Social identity theory (Tajfel 1981) explicates why people buy certain products, goods, or services to distinguish themselves from others, such that customer beliefs about a company and its values become self-defining. Customers define themselves by the same attributes that they believe define other customers of the company (Bhattacharya and Sen 2003). Because they target low-income consumers, social discount practices, whether integrated or collaborative, may provide a negative signal about product quality, which might lower consumers’ trust (Seitanidi and Crane 2009; Webb and Mohr 1998) and reduce loyalty to the retailer, compared with a situation without any discount space. Sen and Bhattacharya (2001) find that CSR initiatives can, in certain circumstances, decrease consumers’ intentions to buy a company’s products. We propose:

H7

The effect of trust in the retailer on loyalty to that retailer is stronger when it offers no social discount space rather than an integrated or collaborative social discount space.

Methodology

Data Collection and Research Design

We gathered empirical data using a self-administered survey of consumers of different retail companies in France. French consumers are an appropriate target, considering France’s long history of social preoccupation and its likely influence on business practices. However, French food retailers suffer from poor consumer confidence (58 %) with regard to their CSR policy (Lombart and Louis 2014), prompting increased interest in social and ethical issues. Most food retailers have adopted a CSR policy and committed to improving the environmental quality of their offerings and store designs, as well as reducing waste in their stores (Lombart and Louis 2014). According to Lavorata and Pontier (2005), specific ethical actions, including social conduct codes and adoptions of the ethical standard SA 8000, are spreading in France. For this study, we targeted a large city in northern France, where retailing is critical to the local economy, with the goal of investigating a homogeneous sample, in terms of consumption patterns and geographic area. The questionnaire was designed and administered online (Google docs), which helped ensure anonymity and reduce social desirability biases (Pirsch et al. 2007). Previous research has indicated the high risk of social desirability effects when investigating ethics and morality (Brunk 2012).

To test the moderating effects of social discount practices, we created scenarios (see Mohr and Webb 2005) to assess their effects on the relationships among consumers’ perceptions of business ethics, consumer trust, and loyalty toward the retailer. Becheur and Valette-Florence (2014) use a similar scenario-based approach to investigate the effects of negative emotions in public health communication. In line with our research purpose, we established a control group (no reference to social discount space) and two scenarios (ISD and CSD spaces), as described in Appendix 1. In each condition, we asked participants to imagine that they were considering the social discount space offered by the retailer where they shopped most often.

Participants in the survey were at least 18 years of age and fully or partially in charge of their household purchases of food products and groceries. They indicated the retail chain where they shopped most often, which ensured their familiarity with the retailer. To increase external validity, we avoided student sampling and instead aimed to achieve a realistic sample of food customers. However, we limited the sample to shoppers at grocery mass retailers, to ensure homogeneity in terms of the CSR policy and social discount practices. Overall, we collected 689 usable questionnaires, divided among the control group (N = 332), the ISD sample (N = 200), and the CSD sample (N = 257). The participants reported shopping quite frequently each month (28 % two–three times, 54 % more than four times). They also indicated their long relationships with the retailer (20 % 3–5 years, 40 % more than 5 years). On average, they spent 191€ each month at the retailer’s outlets. These values imply strong loyalty behavior by the participants toward their respective retail chains. Overall, respondents were well distributed across socio-demographic categories (i.e., age, gender, income, and education), as the sample profiles in Table 1 reveal.

Table 1 Demographic characteristics of sample (N = 689)
Full size table

Variable Measurements

We developed the survey instrument on the basis of a comprehensive review of relevant literature. Other than demographic information and trust, we measured all the variables using Likert scales (1 = “strongly disagree” to 5 = “strongly agree”), operationalized with well-established items. For consumers’ perceptions of business ethics, we used three items from Brunk (2012), in line with our conceptualization of the construct, which referred to whether the retailer respected moral norms, adhered to the law, and was a good company. The retailer’s CSR was measured with five items adapted from Lai et al. (2010), involving respect for the environment, respect for customers, and philanthropic actions (Lombart and Louis 2014). For retailer corporate reputation, we measured three items (overall experience perception, relative perceptions, and future of the retailer), adapted from Lai et al. (2010). Trust in the retailer used two items adapted from Rampl et al. (2012); this five-point scale ranged from “low” to “high.” We captured this measure at both the beginning and the end of the questionnaire, to assess both affective and cognitive aspects (Kenning 2008). For loyalty to the retailer, we relied on four items (favorable attitudes and behaviors resulting in repurchases), adapted from Yoo and Donthu (2001) and Lai et al. (2010). We also included variables pertaining to consumer shopping, such as familiarity with the retailer (i.e., frequency of store visits, length of store use; adapted from Diallo et al. 2013), average expenditures in the retailer’s stores, and socio-demographic variables (age, gender, household income, and education), measured according to a categorization, as we show in Appendix Table 4.

Data Analysis Approach

The data analysis was based on partial least squares (PLS) path modeling, which is appropriate for this study for several reasons. Compared with covariance-based structural equation modeling (CB-SEM), PLS is pertinent if the primary research goal involves predicting the dependent variable (Reinartz et al. 2009). In this sense, PLS is especially suitable for exploratory studies (Hair et al. 2011) that seek to extend existing structural theory, as we do in this study. Therefore, we employ PLS to explore the dependence across consumers’ ethical perceptions and their loyalty behavior, while assessing potential mediators (trust, CSR, and reputation) and moderation by social discount practices. In addition, some of our subsample sizes (N = 200) are smaller than recommended for traditional SEM but satisfy the “ten times rule of thumb” (Barclay et al. 1995; Marcoulides et al. 2010), so PLS provides more robust, accurate inferences. Finally, our data are not multinormal (Mardia test, c.r. = 33). In this setting, PLS is more appropriate than CB-SEM because it does not make distributional assumptions. For a general discussion of the advantages and boundary conditions for PLS, see Henseler et al. (2014).

Analysis and Results

We analyzed our results using a stepwise approach: measurement tools, research hypotheses, and then extended analyses.

Measurement Scales’ Psychometric Properties

To begin, we assessed the measurement (outer) models with confirmatory factor analysis using SmartPLS 2.0, before moving to the structural (inner) model tests (Gerbing and Anderson 1988). Table 2 contains the construct measurement scales (all reflective). We used four criteria to assess the convergent validity and internal consistency of the constructs: item loading, communality (R 2), reliability indicators (ρ and α), and average variance extracted (AVE). All the loadings between an indicator and its underlying construct factor were greater than .5 (DeVellis 2011), suggesting that the variables were salient. The reliability value of all constructs exceeded the .7 criterion, and the AVE values were above the recommended threshold of .5, in support of convergent validity (Fornell and Larcker 1981). Appendix Table 4 contains the details of the test of discriminant validity; because the AVE for each construct was greater than the square of the interconstruct correlations, all constructs fulfilled the requirements for discriminant validity (Fornell and Larcker 1981). We also assessed discriminant validity at the indicator level and found it satisfactory, because the loading of each indicator was greater than all of its cross-loadings (Chin 1998). Thus, all five latent variables and their indicators were distinctive. Finally, we checked for measurement invariance to ensure that we could compare the path coefficients meaningfully (Steenkamp and Baumgartner, 1998). First, we evaluated model fit in each group; it was satisfactory (significant item loadings at p < .01; R 2 and goodness of fit [GoF] values homogeneous across groups). Second, we inspected differences between the item loadings across the three groups and found them invariant. Third, following Becheur and Valette-Florence (2014), we constrained the measurement parameters to equality across the three groups, to avoid any error due to measurement variance.

Table 2 Measurement properties (N = 689)
Full size table

Results and Hypotheses Testing

Because the measurement was adequate, we moved on to the structural (inner) model and tested the direct effects hypotheses by examining the path coefficients (γ) and their significance levels. A bootstrapping estimation verified the statistical significance of each path coefficient. The variance explained (R 2) in the endogenous latent variables and p-values of the regression coefficients (t test) indicated the explanatory power of the model. The GoF values were appropriate and suggested correct model fit: GoFabsolute = .48 and GoFrelative = .96. In Fig. 2, we show that consumers’ perceptions of retail business ethics had positive effects on loyalty to the retailer (γ = .27, p < .01), in support of H1. For the indirect effects, we computed the product of the direct effects using bootstrapping (Zhao et al. 2010). Because SmartPLS does not reveal the significance of indirect effects with bootstrap intervals, we also used a Monte Carlo method to assess mediation (MacKinnon et al. 2004).Footnote 2 With this approach, we uncovered a significant effect of consumers’ perceptions of retail business ethics on trust in the retailer (γ = .27, p < .01), which had a significant influence on loyalty to the retailer (γ = .36, p < .01). The product of these effects was significant (γ = .10, p < .01), indicating a mediating role of trust in the retailer, as we predicted in H2. Consumers’ perceptions of retail business ethics also exerted a significant effect on retailer CSR (γ = .61, p < .01), which significantly influenced trust in the retailer (γ = .17, p < .01). The product of these effects also was significant (γ = .10, p < .01), in support of H3 and the predicted mediating effect of retailer CSR. Finally, consumers’ perceptions of retail business ethics had a significant effect on retailer corporate reputation (γ = .60, p < .01), which influenced trust in the retailer (γ = .32, p < .01), and the product was significant (γ = .19, p < .01). We thus found support for H4 regarding the proposed mediation by retailer corporate reputation.

Fig. 2
figure 2

Direct and indirect effects of consumers’ perceptions of retail business ethics. All coefficients are significant at p < .01

Full size image

To assess the moderating effects of social discount practices, we performed a multiple group analysis (structural equation modeling approach) with SmartPLS 2.0 (Henseler and Fassott 2010).Footnote 3 Classical analyses of variance are not appropriate, because we need to assess differences in paths coefficients among three groups. The results of our estimation based on the bootstrapping estimates (Table 3) showed that social discount practices did not moderate the relationship between consumers’ perceptions of retail business ethics and loyalty to the retailer. The coefficients were in the predicted directions, but we did not find any statistical differences across social discount practices, so we must reject H5. For the relationship between consumers’ perceptions of retail business ethics and trust in the retailer, we found significant differences between the control condition with no social discount space (γ = .31, p < .01) and the ISD (γ = .30, p < .01) and CSD (γ = .17, p < .05) conditions, in support of H6. Significant differences also arose in the relationship between trust and loyalty toward the retailer, highlighting the distinction between the no social discount space (γ = .50, p < .01) and the CSD (γ = .30, p < .05) and ISD (γ = .38, p < .01) conditions, in support of H7.

Table 3 Moderation effects of social discount practices
Full size table

Extended Analysis: Assessment of Competitive Models

To assess the robustness of the results and explicate the relationships, we tested two competitive models. First, we included consumer characteristics (age, gender, income, education) as control variables to determine their effects on our results. Adding the covariates did not change the main results or improve the explanatory power of the model (R 2 values remained the same). No socio-demographic variables affected the dependant variables either. Therefore, the relationships between retailer socio-ethical perceptions and loyalty to the retailer were unlikely to be affected by consumers’ personal characteristics.

Second, we tested a concurrent model that included familiarity with the retail chain (i.e., frequency of store visits and length of relationship) as a potential determinant of the main constructs. Although familiarity can improve trust, enhance satisfaction, and ultimately breed customer loyalty (Ha and Perks 2005), prior retailing research has not widely addressed its effects in relation to socio-ethical perceptions. Our results showed that familiarity with the retail chain affected consumer loyalty toward the retailer (γ = .11, p < .01), trust in the retailer (γ = .07, p < .05), and consumers’ perceptions of retail business ethics (γ = .10, p < .05). These findings support the nomological validity of our research model and underline the need for researchers to address consumer familiarity more appropriately.

Conclusions and Discussion

Theoretical Implications

Our research supports academic findings that indicate consumers expect companies to make broader contributions to society (Brunk 2012; Sen and Bhattacharya 2001; Van Kenhove et al. 2003). As a contribution to literature pertaining to consumer perceptions of retail companies, we empirically examine both the direct and indirect effects of consumers’ perceptions of retail business ethics on loyalty to the retailer. By focusing on consumers’ perceptions of retail companies’ socio-ethical behavior, we extend and complement prior studies that take a company perspective (Biong et al. 2010). To the best of our knowledge, the current study is the first to investigate moderation by social discount practices, which retailers increasingly use to cope with dire economic situations. In turn, this study offers three main theoretical contributions.

First, we show that consumers’ socio-ethical perceptions lead directly to loyalty to the retail chain. Both consumers’ perceptions of retail business ethics and retailer CSR have positive effects on consumer loyalty to the retailer, in accordance with prior studies (Pirsch et al. 2007; Valenzuela et al. 2010). Nevertheless, though ethical perceptions and CSR are related, they differ significantly in their consequences, confirming the need to distinguish the two concepts carefully (Dimitriades 2007). Consumers’ ethical perceptions are more important for building trust and creating loyalty than is retail CSR, at least in France. Furthermore, consumers’ perceptions of retail business ethics have indirect effects on consumer loyalty, through the mediation of consumer trust. These results emphasize the importance of ethical perceptions for building loyalty and thus extend prior studies on business ethics (Lai et al. 2010; Van Kenhove et al. 2003). We also clarify the indirect effects of ethical perceptions on consumer trust, through the mediation of retailer CSR and retailer corporate reputation. Overall, these results affirm that good ethics constitute a good retailing practice (Babin et al. 2004); they show that the relationship between consumers’ ethical perceptions and loyalty to the retailer does not depend on socio-demographic variables. This finding confirms assertions about the limited relevance of personal characteristics for explaining ethical consumption behavior (De Pelsmacker et al. 2005).

Second, by adopting an approach focused on consumers’ reactions to retailers’ efforts to integrate the poor through social policies (social discount), we emphasize positive, negative, and neutral effects stemming from the use of social discount practices designed for the poor. The integration of the poor in retail strategies through such practices is more complicated than might be assumed and exerts distinct effects, depending on the outcome variable. The relationship between consumers’ perceptions of retail business ethics and trust in the retailer is stronger when the retailer does not offer a social discount space or provides an ISD rather than a CSD. These results indicate the positive consequences of ISD spaces in the relationship between consumer ethical perceptions and trust in the retailer. They also confirm prior studies indicating the effectiveness of a moderate level of discount practices (similar to ISD) (Andrews et al. 2014), whereas a deep discount associated with cause marketing (similar to CSD) decreases purchase incidence. Yet we find a moderating effect of discount practices on the relationship between consumer trust and loyalty, such that we reveal a stronger effect in the absence of a social discount space than in its presence (both CSD and ISD), indicating a negative consequence of social discounts. In some settings, CSR initiatives can fail to invoke positive responses (Sen and Bhattacharya 2001). Signaling theory similarly stresses the negative signal that social policy practices can send when companies try to integrate the poor into their business strategies (Bhattacharya and Sen 2003; Seitanidi and Crane 2009; Webb and Mohr 1998). Finally, we find no moderating effects in the relationship between ethical perceptions of retailers and loyalty, suggesting that some social discount practices are neutral and do not increase or decrease these relationships (see also Singh 2006).

Third, CSD seems less effective than ISD for efforts to integrate poor consumers in retail settings. We offer two possible explanations. Perhaps CSD reduces consumers’ trust in retailers, because they believe charities are not qualified to manage products in retail outlets. Special care should be taken with partner selection when designing and implementing social discount spaces. Seitanidi and Crane (2009) arrive at a similar conclusion in their study of two partnerships (Earthwatch–Rio Tinto and Prince’s Trust–Royal Bank of Scotland). The reduced trust in CSD spaces also might stem from the company’s insufficient involvement in CSR activities, based solely on discount spaces designed for poor consumers. Vlachos et al. (2009) warn that consumers need to believe that social policies exist because the company believes in the cause, not for economic or tax benefits. In line with these findings, (Winterich and Barone 2011) find that the effects of self-construal are attenuated when donation-based promotions do not involve an identity-congruent charity.

Practical Implications

This study has practical implications for retail chains. First, they should enhance socio-ethical perceptions, because customers’ trust and loyalty stem not exclusively from retailer characteristics but also from intangible resources, such as ethical perceptions, CSR, and corporate reputations. The economic crisis has had negative consequences for retailers, largely due to the decline in consumers’ purchasing power, and these effects have been magnified by customer doubt and skepticism in response to ethical scandals. If retailers want to reestablish consumer trust and build long-term relationships, which ultimately lead to loyalty, they must behave ethically (i.e., respect for moral norms and adherence to laws). In addition to the positive impacts on trust and loyalty, consumers’ perceptions of retail business ethics have strong positive effects on retailer corporate reputation and CSR. Therefore, more transparency and justice is needed across all retailer actions (e.g., pricing, product quality, store management). Socio-ethical practices cannot be simply a communication tool; they should reflect the retailer’s overall strategy. Although consumers usually choose retailers for price, quality, and convenience reasons, they are less likely to shop in outlets that engage in unethical practices.

Second, this research reveals that retailers must recognize the specific ways in which their social discount practices can create or destroy value. We provide guidance along these lines by identifying boundary conditions of the efficiency of social discount spaces. The use of social discount spaces designed for poor consumers is generally a welcome practice, but they must be managed carefully and in line with the retailer’s ethical values. If the retailer mainly targets high-income consumers, it should not provide a discount space, because its customers will perceive social partners managing products in a CSD with skepticism. However, such a retailer could establish a separate, isolated CSD space to avoid any confusion. Integrated social discounts are appropriate for retailers whose customers have weak purchasing power, though these retailers still must provide proof that their ISD actions reflect their existing ethical values. Otherwise, consumers regard the discount space as a tactic to maximize the retailer’s profit at the expense of the poorest segments of society. Rather, CSD spaces are most appropriate for segments of consumers with strong ethical orientations. Regardless of their economic status, these consumers do not worry about “mixing” with other (poor) consumers. With regard to the partners to include in CSD, retailers should establish careful selection and implementation policies. Working with social partners in the store requires that the parties share ethical values and communicate appropriately to send consistent signals to customers. Our findings thus reveal that retailers must clearly state the purpose of their discount spaces if they hope to create social legitimacy. Furthermore, they should use these spaces in ways consistent with their targets’ characteristics, not just as marketing tools to signal their attention to ethical issues.

Limitations and Directions for Research

Our results expand current knowledge of business ethics, but some limitations also offer possibilities for research. Although we distinguished CSR from ethical perceptions, we treated it as a whole construct, not a composite one. Similar to Lai et al. (2010), we thus cannot specify the effects of different dimensions of CSR activities (e.g., economic, social, environmental) on consumer trust and loyalty. We explore the relationship between business ethics and CSR (Brunk 2012); additional studies should explore the conditions in which usual norms or other facets prevail.

Other limitations arise from our research context, design, and methodology. France has a long history of social connection (e.g., welfare state), so its people tend to be more demanding of firms’ ethical practices and social roles. An extension might replicate this study in other European countries and thus provide a cross-cultural investigation (Singh et al. 2008). Although we have appropriate sample sizes, studying only one city in one country is not enough to understand the entire phenomenon. Retailers such as Carrefour and Auchan in France and Tesco in the United Kingdom are developing ISD, but they need to know how to adapt these spaces for the different European countries in which they operate. In addition, our experimental conditions did not provide information about the charities involved in the CSD scenarios; more research should pursue greater insights into the effect of consumers’ awareness of retailers’ social practices on consumer trust and loyalty. We measured loyalty using self-reported scales. Further studies could operationalize it differently, to determine how business ethics and social discount practices relate to other loyalty metrics (e.g., actual purchasing behavior).

Finally, a potential way to clarify the effects of retailer socio-ethical perceptions on the dependant variables (trust and loyalty) and the moderating role of social discount practices further would be to examine the influences of other characteristics, such as price sensitivity and self-perceptions of poverty. The ethical values of discount stores might be clearer; they mainly seek to appeal to price-sensitive consumers. Discounters and mass retailers that offer discount spaces then could exploit different or similar sources of value, reflecting different dimensions. Investigations of other sustainable values related to ethics and CSR (e.g., altruism), across industries and nations, might provide new insights into how they affect retail companies’ perceived value.