Ethically Minded Consumer Behavior, Retailers’ Commitment to Sustainable Development, and Store Equity in Hypermarkets

: The present paper analyzed the inﬂuence of ethically minded consumer behavior on retailer’s commitment to sustainable development as perceived by the consumer, and their contribution to the chain of relationships “store equity–loyalty–word of mouth communication”, in the context of retailing. Responses from 317 hypermarket customers in Ecuador were analyzed. The model was estimated using the partial least squares technique. The results show the sustainable development actions implemented by retailers, together with perceived value, contribute signiﬁcantly to store equity creation, which explains customer loyalty and, ultimately, positive word-of-mouth communication. However, the most concerned consumers of ethical aspects value the retailer’s e ﬀ orts towards sustainability to a lesser extent. The ﬁndings of this research allow developing a series of implications for the managers of these establishments.


Introduction
In the last decade, public concern for the environment has emerged as a problem of global importance [1,2]. This has led to great changes in consumer attitudes and behaviors, leading them towards patterns of ethical and sustainable consumption [3]. In response to these changes, organizations should examine their strategies with respect to sustainable development [4], as well as their trade policies [5], and companies in the commercial distribution sector are no exception. Specifically, regarding the hypermarket retail format, sustainable development is considered an important issue for both academia and the business sector [6].
In this context, various hypermarket chains have been carrying out actions to contribute to social and environmental sustainability through their activities [7], creating spaces for the development of sustainable management by retailers and ethical consumption. Therefore, investigating this commercial format is of interest in order to contribute to increasing knowledge regarding sustainability [8,9]. This present work focused on analyzing ethically minded consumer behaviour (hereinafter, EMCB), following the call for future research in this area by Sudbury-Riley and Kohlbacher [10].
EMCB implies the consideration of environmental principles and social causes in the purchase decision, beyond the mere satisfaction of needs [11], therefore tangible proof of the social, environmental, and economic commitment of retailers is expected [12]. In this sense, it is important to analyze the scope of the retailers' commitment to sustainable development as perceived by consumers (hereinafter, the consumer based on the costs and the benefit obtained from the quality and expected performance of the retail establishment [14,59].
In the retail sector, perceived value has been qualified as key to the success of organizations [60]. In research in the field of consumer behavior, perceived value is a widely studied variable, having been related to various constructs, such as loyalty and brand equity. According to Sweeney and Soutar [61], perceived value integrates aspects related to quality, value for money, emotional value, and social value, which include utilitarian and hedonic values, proving to be useful, parsimonious, and suitable for study in the retail sector [59]. The quality value dimension refers to the practical benefits that consumers can obtain with the use of the product. Value for money describes how a successful product compares to the cost, time, or effort spent in obtaining the product. The emotional value represents the benefit that consumers discern, which derives from the feelings that a product generates. Finally, social value refers to the collective utility that the consumption of the product transmits [62].
The consumer expects ethical actions, price, and quality based on the perceived value [63]. In this way, in the research applied to the hypermarket format, it was concluded that perceived value is an important factor to differentiate the consumer in regard to the perception that the latter has about the retailer's brand equity [64]. Studies have also indicated that perceived value plays a relevant role in the creation and improvement of store equity [15,55]. Based on these contributions in the literature, the second hypothesis was proposed: Hypothesis 2 (H2). Perceived value has a positive and significant effect on store equity.
In the retail context, several studies have concluded that the consumer positively relates to companies that they consider ethical [2], so hypermarkets should strive to take sustainable actions [51]. RCSD and store equity are constructs that are related in the context of the hypermarket commercial format, in the sense that sustainability initiatives have a positive impact on store equity [65]. Likewise, evidence has been found that identifies RCSD as a determinant of store equity [48,63,66,67]. Retailers face increasing pressure to integrate ethical values into their operations, communicate their commitment to sustainability, and have these values perceived by consumers. Various studies indicate that sustainable and ethical actions perceived by the consumer have a direct and positive impact on brand equity, e.g., [51]. In regard to all of the above, the third hypothesis was stated: Hypothesis 3 (H3). The retailers' commitment to sustainable development as perceived by consumers has a positive and significant effect on store equity.

Loyalty
Loyalty is defined as "a deeply held commitment to rebuy or repatronize a preferred product/service consistently in the future, thereby causing repetitive same-brand or same brand-set purchasing, despite situational influences and marketing efforts having the potential to cause switching behaviour" [68].
In the retail literature, loyalty is a construct that includes two perspectives: (a) behavioral, emphasizing past experience; and (b) attitudinal, which is based on future actions, considering both purchasing attitudes and behavior, repeat purchases [59,69], and recommendations by the retailer [70]. Loyalty is the consumer's strong commitment to repurchase a product from a store, despite any contextual influence or marketing efforts of the competition [71]; consequently, the positive sentiment that the consumer holds towards a particular brand is likely to impede the search for alternatives and subsequently favor customer loyalty to the brand [72]. In this sense, several studies have linked brand equity with loyalty in the retail sector, concluding that store equity contributes to generating consumer loyalty to the brand [73][74][75]. Likewise, there has been a discussion in the literature as to whether loyalty is a dimension or consequence of store equity [15,53]. To this regard, Gil-Saura et al. [15] concluded that loyalty works best as a dimension of store equity. Based on these findings, the fourth hypothesis was proposed: Hypothesis 4 (H4). Store equity has a significant and positive effect on loyalty.

Word-of-Mouth (WOM)
When customers are loyal to a brand, they are likely to convey their positive feelings towards that brand to other people [72]. The word-of-mouth communication construct encompasses informal conversations about the brand, the organization, and/or its products or services [76]. The loyalty and word-of-mouth communication constructs have been extensively related in marketing research, specifically in the retail sector. Due to the importance of the quality of relationships between customers and companies in retailing, several authors have found that customer loyalty positively influences word-of-mouth communication [77].
Along the same lines, several authors concluded that loyalty is a significant antecedent of word-of-mouth communication [78][79][80], highlighting the willingness of consumers to share information as an outcome of their loyalty intentions and behavior.
In line with these findings, the fifth hypothesis was put forward: Hypothesis 5 (H5). Loyalty has a significant and positive effect on word-of-mouth communication.

Methodology
To corroborate the proposed hypotheses, quantitative research is proposed. The information was collected with the administration of an ad hoc, structured, closed-ended questionnaire. Using the questionnaire, information was collected on EMCB, RCSD, perceived value, store equity, consumer loyalty towards the store, and word-of-mouth recommendations. To measure EMCB, the scale proposed by Sudbury-Riley and Kohlbacher [10], was considered. To measure RCSD, the scale from the work of Lavorata [12] was adapted. To measure perceived value, the Sweeney and Soutar [61] scale was taken as a reference. Store equity was measured using the Shen scale [52], loyalty was measured with the scale proposed for the retail sector in the research carried out by Arnett et al. [81], and the items for the scale to measure word-of-mouth communication were adapted from Chaudhuri and Ligas [82] and Zeithaml et al. [83]. In all cases, the items (see Appendix A) were measured with five-point Likert-type scales, ranging from 1 (strongly disagree) to 5 (strongly agreeing). We have used 5-point Likert scales based on the evidence provided by empirical research, in the sense that these scales require the shortest reaction time, compared to 7-point scales, and when considering the data showing potential acquiescence bias and extreme response bias, five is the optimal number of anchors [84]. Additionally, in order to characterize the sample obtained, a series of classification questions was included.
The fieldwork was set in Ecuador, a middle-income country in South America where, in spite of some pro-environmental measures implemented by the government in the last decades and the active role of civil society in strengthening local healthy and sustainable food systems [85], the need to raise awareness about environmental issues among the population has been highlighted, e.g., [86]. The questionnaire was administered following a probability sampling method, surveying one out of every five customers over 18 years of age in hypermarkets located in commercial areas in the main cities of Ecuador in terms of population according to the Ecuadorian Institute of Statistics [87], i.e., Quito, Guayaquil, Cuenca, and Machala, with more than 2,800,000 inhabitants older than 18 altogether, from March to August 2017. The questionnaire included a control question to guarantee that the respondent personally buys from that hypermarket. As a result, 317 valid questionnaires were collected (sampling error: +/−5.5% for a confidence level of 95%). Table 1 shows the distribution of the sample based on the sociodemographic variables and their purchasing behavior patterns. From the collected data, for the analysis of the results obtained, we proceeded in different phases. In the first phase, an exploratory factor analysis was performed using the SPSS software, with the objective of studying the dimensionality of the different constructs analyzed. Next, a confirmatory factor analysis was carried out and the psychometric properties of the measurement model were confirmed, and then the proposed model was estimated using the partial least squares (PLS) method with the SmartPLS software.

Analysis of Results
First, an exploratory factor analysis was performed to identify the dimensions of the different constructs analyzed. In this way, it was verified that the items to measure the EMCB construct were grouped into a single factor (KMO: 0.893; determinant: 2.56 × 10 −011 ; Bartlett's test of sphericity sign. level: 0.000), which explained 89.5% of the variability of this construct.
By replicating the exploratory analysis on the scale used to measure RCSD, nine items were retained, which were grouped into three factors which have been named economic RCSD, social RCSD, and environmental RCSD, which together explained 89.0% of the variability of the phenomenon, providing a satisfactory fit (KMO: 0.874; determinant: 5.19 × 10 −009 ; Bartlett's test of sphericity sign. level: 0.000).
The exploratory analysis was repeated on the scale used to measure perceived value, retaining eleven items, from which two factors emerged that have been named perceived value-quality, and perceived value-social, both together explaining 88.3% of the phenomenon, providing a satisfactory fit (KMO: 0.888; determinant: 1.60 × 10 −009 ; Bartlett's test of sphericity sign. level: 0.000).
Finally, exploratory factor analyses were carried out for store equity, loyalty, and word-of-mouth communication, observing that the items for these constructs were grouped into a single factor, which explained 76.8% (KMO: 0.788; determinant: 0.000; Bartlett's test of sphericity sign. level: 0.000), 89.4% (KMO: 0.846; determinant: 0.006; Bartlett's test of sphericity sign. level: 0.000), and 87.0% (KMO: 0.500; determinant: 0.453; Bartlett's test of sphericity sign. level: 0.000), respectively. Therefore, once the multidimensionality of the RCSD and perceived value constructs, on the one hand, and the unidimensionality of the EMCB, store equity, loyalty, and word-of-mouth communication constructs were confirmed, the final model, depicted in Figure 1, included two second-order formative constructs, i.e., RCSD and perceived value, which were configured from 3 and 2 first-order reflective constructs, respectively. once the multidimensionality of the RCSD and perceived value constructs, on the one hand, and the unidimensionality of the EMCB, store equity, loyalty, and word-of-mouth communication constructs were confirmed, the final model, depicted in Figure 1, included two second-order formative constructs, i.e., RCSD and perceived value, which were configured from 3 and 2 first-order reflective constructs, respectively.

Measurement Instrument Validation Process
Firstly, in order to contrast the proposed model, the psychometric properties of the measurement model were verified using confirmatory factor analysis to subsequently estimate the structural model, using the partial least squares (PLS) regression method [88], which permits the incorporation of both formative and reflective constructs into the model to be estimated [89]. Items 8, 9, and 11 of RCSD and items 2 and 3 of PV were eliminated as the individual loads were

Measurement Instrument Validation Process
Firstly, in order to contrast the proposed model, the psychometric properties of the measurement model were verified using confirmatory factor analysis to subsequently estimate the structural model, using the partial least squares (PLS) regression method [88], which permits the incorporation of both formative and reflective constructs into the model to be estimated [89]. Items 8,9, and 11 of RCSD and items 2 and 3 of PV were eliminated as the individual loads were less than 0.60 [90]. From the results of the confirmatory factor analysis (see Table 2), the reliability of all the scales could be confirmed, since both the values for composite reliability (CR) and for Cronbach's alpha coefficient (CA) were higher than 0.8 [91] for all first order constructs. The standardized root mean square residual (SRMR) was 0.059; in CB-SEM, a value for SRMR below 0.08 shows a good adjustment, but no reference thresholds for goodness-of-fit measures have been provided for PLS-SEM [92,93]. To assess the convergent validity, it was verified that the standardized loadings were higher than 0.7 and statistically significant for all the items, and it was found that all the values for the average variance extracted (AVE) exceeded the critical level of 0.5 [94]. All this allowed us to confirm the reliability of the scales and the convergent validity of the proposed constructs (see Table 2).
To analyze the discriminant validity in the measurement instrument, firstly, it was found that the estimated correlations between two factors were less than the square root of the average variance extracted of each factor [94]. Secondly, all the HTMT ratios showed values below 0.90 [95], and when comparing the factor loadings across the columns, the item loadings for their own construct were in all cases greater than their cross-loadings with other constructs. Therefore, from Table 3 it can be seen that this condition was met in all cases. Regarding the second-order formative constructs, it was evident that the first-order reflective constructs make a statistically significantly contribution to their corresponding second-order formative constructs (see Table 4). Moreover, the values of the variance inflation factors (VIF) were less than 5 for all the first-order factors, which allowed us to rule out potential collinearity problems [96]. Next, the structural equation model was estimated using the partial least squares (PLS) regression method (Table 5), and the explanatory power of the structural model was verified through the coefficient of determination ( Table 6). As can be seen in Table 6, the R 2 values exceeded the reference value of 0.5 rated as moderate, with the exception of word-of-mouth communication with a 0.39, which is considered weak [97]. The Q2 were greater than 0, therefore, the model presented an adequate explanatory and predictive value and allowed us to assess the significance of the previously hypothesized causal relationships. SRMR of the estimated model was 0.103.  The results of the study allowed us to support the assertion that RCSD has a positive and significant effect on store equity, hypothesis H3, as indicated by previous works, e.g., [2,48,51]. Similarly, the perceived value positively influenced the generation of store equity, in line with other studies [15,67,78], corroborating hypothesis H2. In turn, store equity had a positive and significant effect on loyalty, in support to hypothesis H4, along the lines of works such as that of Frasquet et al. [73] and Su and Chang [74]. In correlation with the results, loyalty had a positive and significant effect on word-of-mouth communication and, therefore, hypothesis H5 was accepted, corroborating the findings of Harris and Khatami [79] and Markovic et al. [72].
The estimation of the model, however, did not support the existence of a relationship between EMCB and RCSD in the case of hypermarkets, so there was no support for the H1 hypothesis. Furthermore, EMCB negatively influenced RCSD, which can be explained by the fact that it is the most ethical consumers who are most critical of the actions of retailers in the field of sustainability, in the sense that the actions taken by retailers are interpreted as "greenwashing" tactics and not as actions really aimed at sustainability. "Greenwashing" is the process of conveying a false impression or providing misleading information about how a company's products are more environmentally sound [98]

Conclusions, Limitations, and Future Research Lines
This work has addressed the study of ethical consumer behavior and its relationship with RCSD, as well as the analysis of its contribution to the creation of store equity and its consequences in the context of hypermarkets. The results obtained show the positive and significant effect of RCSD and perceived value on store equity, consistent with the evidence obtained from previous studies in other geographical contexts [15,55,67,78]. Specifically, there was a positive relationship between the environmental, social, and economic RCSD in its triple bottom line, and the retailer's brand equity, in a manner consistent with the evidence obtained from previous studies [2,51].
Similarly, examining the positive and significant effects of store equity on consumer loyalty, it was observed that when customers perceived the sustainability actions of the hypermarket, the more highly they rated the store equity and attributed this perceived value, which creates a long-term relationship with the retailer, translating into preference and loyalty, in line with the existing literature [15,72,74]. This in turn positively influences the generation of positive word-of-mouth communication [72].
On the other hand, a negative relationship was observed between EMCB and RCSD. This result could be explained by the differences between the cultures of retail consumers in the countries where this research has been carried out and those of Ecuador, where the issue of ethical consumption is only just beginning to emerge, e.g., [86], and the most conscientious consumers may be skeptical in regard to the retailer's commitment to sustainability, believing these actions to be little more than mere advertising claims. This evidence contrasts with the results of studies that, in other geographical contexts, have highlighted the importance of proving sustainability actions to customers due to the fact that this creates an emotional connection between the customer and the retailer, which translates into purchase, repeat purchase, and recommendation [70], positively influencing consumer perception [99].
All in all, the results reveal that, although there are ethical consumers, and hypermarkets communicate their sustainability activities to all of their target markets, the concepts of ethical consumption and sustainability may not be sufficiently established amongst consumers in Ecuador. In fact, these are issues that are only just starting to gain awareness amongst Ecuadorian consumers today. Despite this, it is evident that consumers are aware that, with their purchase they can promote the wellbeing of the environment and society and reward the sustainability actions that the hypermarket executes and communicates, given that their perceptions of these actions influence their beliefs in regard to perceived value, store equity, and loyalty, which leads the consumer to communicate their positive experiences to others. With this, to the extent that the retailer influences the consumer with the development of sustainability actions, integrating them into their culture as a commitment and a responsibility, they will be rewarded with better perceptions of store equity and positive results in terms of customer loyalty and word-of-mouth communication. Therefore, it is considered that this work contributes to the existing literature by providing evidence regarding the usefulness of implementing RCSD actions in hypermarkets and their effect on consumers with an ethical mentality.
In the managerial and academic fields, these results hold a range of implications for management. Given the increasing competition, it is necessary for managers to understand how actions that benefit society, the environment, and their workers and suppliers influence consumer perception. In this way, a combination of marketing strategies, sustainable development management, and superior value delivery will lead to better positioning through increased store equity that allows them to differentiate themselves from the competition. Efforts should be made to demonstrate that the retailer's commitment to sustainable development is real, through the retailer's various communication channels with consumers. Additionally, it is necessary to take care of both the functional and emotional aspects that can contribute to strengthening the perceived value of the consumer and store equity to achieve greater consumer loyalty towards the hypermarket, leading to positive word-of-mouth communication.
To conclude, we can highlight some limitations that can also be considered as possible future lines of research. On the one hand, the restricted geographical scope of application of the study (i.e., four of the most populated cities in Ecuador) raises concerns about the sample representativeness, so that it is necessary to expand its scope in order to compare results with the perceptions of consumers from other countries. In addition, a larger sample could allow the results to be further refined through multi-sample analysis to verify the moderating effect of different variables, such as gender or age of the consumer, on the relationship between EMCB and RCSD.
On the other hand, the study could be applied to other retail formats that have already incorporated sustainable development strategies into their management. Consumer perceptions of the retailer's commitment to sustainable development are assumed to be holistic, based on the actions of the retailer. In our study, we focused on hypermarkets, which sell basically food, but also include other nonfood products in their assortment. Thus, given the peculiarities of the agri-food market, further research is required to analyze differences in consumer perceptions of the retailer's commitment to sustainable development across different types of products and retailers.
Finally, it is worth highlighting the opportunity to advance this line of research with the in-depth analysis of the mediated relationships between the constructs included in the proposed model as well as the introduction of new relevant variables, such as image, satisfaction, perceived quality, and information and communication technologies (ICT), amongst others.

Conflicts of Interest:
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.