Influence of Corporate Social Responsibility on Loyalty and Valuation of Services

Transcription

1 Journal of Business Ethics (2005) 61: Ó Springer 2005 DOI /s Influence of Corporate Social Responsibility on Loyalty and Valuation of Services Ma del Mar García de los Salmones Angel Herrero Crespo Ignacio Rodríguez del Bosque ABSTRACT. The study of corporate social responsibility has been the object of much research in recent decades, although there is a need to continue investigating its benefits as a marketing tool. In the current work we adopt a multidimensional perspective of social responsibility, and we carry out market research to determine the perceptions of users of mobile telephone services about economic, legal, ethical and social aspects of their operating companies. With these data we determine the structure and components of the concept of social responsibility. Subsequently, this is related with the overall evaluation of the service and loyalty by means of a model of structural equations, in order to determine the influence of corporate social responsibility on these concepts, and hence its benefits as a commercial tool. KEY WORDS: consumer behaviour, citizenship corporate, corporate social responsibility, loyalty, mobile telephone, overall valuation of services Dr. Ma del Mar García de los Salmones is Professor of Marketing at the University of Cantabria (Spain). Her current research interests include corporate social responsibility, brand image and consumer behaviour. She received her Ph.D. in Business Administration in 2002 with her work Corporate Image as a Key Variable in Firm Choice: Identity and Positioning of Mobile Phone Business. Dr. Angel Herrero Crespo is Professor of Marketing at the University of Cantabria (Spain). His current research interests include consumer behaviour, e-commerce and corporate social responsibility. He received his Ph.D. from the University of Cantabria in 2002 for his work on The Process of Adoption of a New Sales System: An Application to B2C e-commerce. Dr. Ignacio Rodríguez del Bosque is Professor of Marketing at the University of Cantabria (Spain). His areas of research include business communication, relationship marketing and distribution channels. His works have previously been published in the several international journals, like Journal of Retailing and Consumer Services, Tourism Management and Industrial Marketing Management. Vitae Introduction The concept of corporate social responsibility has been the object of much research in recent decades, and researchers have adopted varying perspectives. Initially, the concept was exclusively associated with economic aspects, understood as the firm s obligation to maximise shareholder value (Zenisek, 1979). In this perspective we include Friedman s (1962) position, for whom corporate responsibility is simply carrying out actions that increase profits while respecting the rules of the game. However, Mcguire (1963) defends the idea that firms have responsibilities towards their environment that go beyond their legal and economic obligations. Another research stream thus appears relating the concept with marketing activities with a social dimension (Davis and Blomstrong, 1975; Drumwright, 1996; Handelmann and Arnold, 1999), in areas such as environment protection, community development, resource conservation and philantropic giving (Quazi and O Brien, 2000). These approaches to social responsibility are quite narrow, however, since they are limited to particular aspects of the concept and do not take its multi-dimensional nature into account (Maignan and Ferrell, 2001). Corporate social responsibility should be understood as a broad concept, since it takes in the whole set philosophical and normative issues relating to the role of business in society (Maignan and Ferrell, 2001), all the moral obligations that maximise the positive impact of the firm on its social environment and minimise the negative impact (Pride and Ferrell, 1997). These definitions have arisen in the management literature and more specifically from the research into corporate social actions (Maignan

2 370 Ma del Mar García de los Salmones et al. et al., 1999). Of particular note is the Carroll s (1979, 1991) framework, one of the most widely accepted and used proposals to explain the construct (Wartick and Cochran, 1985; Burton et al., 2000). According to this model, firms have economic, legal, ethical and philanthropic obligations towards their environment, and these four dimensions make up corporate social responsibility. Within the branch of management, stakeholder theory has arisen most recently; this also enters in the field of social responsibility. Without centring on the content of corporate social behaviour, stakeholder theory defines its field of application sustaining the idea that firms do not have responsibilities towards society in general, but rather they should only be concerned about individuals or groups that may be directly or indirectly affected by their activity (Clarkson, 1995; Donaldson and Preston, 1995). Specifically, employees, shareholders or investors, customers, suppliers, the government and the community in which the firm operates are included within the group of primary stakeholders; while the communications media and interest groups are regarded as secondary stakeholders, since they are not essential for the firm s survival (Clarkson, 1995). Taking both these theories into account, and from an integrative perspective, Maignan et al. (1999) propose to define social responsibility as the degree to which firms assume economic, legal, ethical and discretionary responsibilities towards their stakeholders. With regards the empirical research on corporate social responsibility, initially it centred on measuring the potential outcome from corporate behaviour (Wood, 1991), relating social behaviour with the financial performance (Moskowitz, 1972; Vance, 1975; Alexander and Buchholz, 1978; Aupperle et al., 1985). Other authors have studied the degree to which corporate social responsibility is applied in firms (Abbot and Monsen, 1979; Lewin et al., 1995; Clarkson, 1995; Maignan and Ferrell, 2000; Joyner and Payne, 2002). More recently scholars have tried to measure the relation between social performance and employer attractiveness (Greening and Turban, 2000; Backhaus et al., 2002), the orientation towards social behaviour of business students of different nationalities (Beggs and Lane, 1989; Ibrahim and Angelidis, 1993; Mukherji and Mukherji, 2002) and the influence of the concept on consumer behaviour (Brown and Dacin, 1997; Lafferty and Goldsmith, 1999; Handelmann y Arnold, 1999; Maignan et al., 1999; Maignan, 2001). With respect to this last group of studies, it could be said that in spite of its recent proliferation, the empirical evidence about the relation between corporate social responsibility and consumer attitudes is still limited. The idea that socially responsible behaviour is good market practice and that it generates positive consequences for the business is increasingly accepted, but there is still a lack of research on the potential benefits of social responsibility from the marketing perspective (Maignan et al., 1999), as well as on its effects on the consumer segment (Sen and Battacharya, 2001). Taking these ideas into account, the current work aims to advance the scientific study of the influence of social responsibility on consumer behaviour. With this in mind, we collect the perceptions of users of mobile telephone services in the Spanish market about the economic, legal, ethical and discretionary or philanthropic behaviour of their service providers. The first objective of this research is exploratory, and it attempts to identify the dimensions of social responsibility from the consumer s point of view, as well as the weight of each dimension within the global construct of social responsibility. We subsequently study its implications or consequences, using a model of structural equations to analyse the potential influence of the construct on the consumer s overall valuation of the service and loyalty towards their operators. There are various reasons to justify our choice of the services market in general and the mobile telephone sector in particular. First, we wish to study the benefits of social responsibility as a marketing tool in a real context i.e., directly asking consumers for their perceptions about the firms and their products. We take into account the fact that it is difficult for consumers to acquire and memorise information about firms social responsibility (Mohr et al., 2001). This is largely due to its complexity and multi-dimensional nature (Maignan and Ferrell, 2001), a characteristic which can inhibit individuals sensitivity towards these activities and explain why consumers do not take them very much into account when assessing a firm or choosing a product. In order to get around this problem, we decided to focus the research on a market in which the firms were very well known in society, where companies actually

3 Influence of Corporate Social Responsibility 371 invested in social actions and with which consumers to some extent felt involved. The strongest relationship between the consumer and the firm clearly arises in the services market. Because of the characteristics of intangibility, inseparability, heterogeneity and perishability of services compared to tangible goods (Grönroos, 1994; Zeinthaml and Bitner, 2000) the user not only approaches purchase in a different way, but also establishes a stronger and more direct relationship with the service providers. In view of this deeper involvement, consumers perception of socially responsible behaviour can take on great importance and have direct consequences in, for example, their valuation of the service and the perceived quality (Sureschandar et al., 2001). Second, among the services the mobile telephone sector is notable for its high penetration (92.3% of the Spanish population in December 2003) 1 so this service has become a necessary good of common use nowadays. Our review of the secondary information about the three operators active in the market Telefónica Móviles, Vodafone and Amena 2 allows us to observe their active behaviour in terms of social responsibility and constant references to their social orientation, search for generation of value and concern for the customer. Indeed, Telefónica Móviles belongs to the Global Reporting Initiative, an international initiative whose aim is to formulate codes and declarations of principles concerning social responsibility and aspects relating to the involvement of firms in questions of sustainability. Thus, we consider it to be an attractive market from the point of view of marketing, in which at least from the firms there is a notable effort to behave responsibly towards the environment in general and towards the service users in particular. We shall try to describe the consumers real perceptions about the social responsibility of the firms and the specific implications thereof in this present work. Corporate social responsibility as a marketing tool: literature review and research hypotheses Studies of corporate social responsibility oriented to the consumers have adopted various perspectives of analysis. Thus, many authors propose a limited conception of the concept, not taking its multidimensional nature into account (Maignan and Ferrell, 2000). Within this group, Smith and Alcorn (1991), Ross et al. (1992) and Eller et al. (2000) analyse the impact of cause-related marketing actions, observing the favourable predisposition of consumers towards firms that engage in these activities. Brown and Dacin (1997) study the effect of social responsibility on the overall valuation of a firm and its product in greater depth. Initially, these authors measure the concept using two social variables, specifically the donations made by the firm to worthy causes and its involvement in the community, to which concern for the environment is added in a second study. The same dimensions are included in Lafferty and Goldsmith s (1999) research, while Handelmann and Arnold (1999) consider that commitment to the family, the community and the nation are indicative of the actions of marketing with a social dimension. Other works have focused more on ethical aspects. For example, Creyer and Ross s (1997) research applied in a primary school tries to measure the response of parents to the school s ethical/ non-ethical behaviour. For their part, Sen and Battacharya (2001) consider the two items from Brown and Dacin s (1997) study, along with others relating to the firm s behaviour towards ethnic minorities, disabled minorities, women s issues and position on gay and lesbian issues, and relate these aspects with the image of the firm and purchase intention. Compared to this unidimensional perspective of corporate social responsibility, other authors propose a multi-dimensional structure for the concept. This point of view leads us to the Carroll s (1979, 1991) model, one of the most complete proposals. According to this model, social responsibility includes all the economic, legal, ethical and philanthropic expectations that society has vis-à-vis a firm at any given time. The economic dimension the most important refers to the firm s obligation to be productive, profitable and to maintain economic wealth. Firms legal responsibility refers to carrying out their activities within the confines of legal requirements. Their ethical responsibility refers to them having ethical codes, norms going beyond mere legal frameworks, and being honest in their relationships with their customers and their own employees. Finally, the discretionary component includes voluntary or philantropic activities aiming

4 372 Ma del Mar García de los Salmones et al. to raise the well-being and development of society as a whole. Hence there is no full consensus about the measurement of social responsibility, since in some cases only the social dimension is included, while in others a broader point of view is adopted. Following Maignan and Ferrell (2000), Maignan (2001) and Seifert et al. (2003), the best proposal is the one which considers the multi-dimensionality of corporate social responsibility. Within the group of work oriented towards consumers, only Maignan (2001) has adopted the Carroll s proposal. Specifically, this author analyses European and American buyers support for socially responsible firms, as well as the possible differences between individuals of different countries when evaluating each of these dimensions. However, there is a need to examine how consumers actually perceive the socially responsible behaviour of firms and how many dimensions they include in the construct. Maignan (2001), for example, wonders if consumers consider firms economic responsibilities to be part of corporate social responsibility. Similarly, we must identify which of the dimensions is perceived to have more weight in the overall valuation of social responsibility (Maignan and Ferrell, 2001). Thus, we propose as the first general research objective of this current work: to confirm the multidimensional nature of corporate social responsibility from the perspective of the consumer and to determine the weight of the various dimensions in the global construct of social responsibility. Corporate social responsibility and valuation of services It is important to advance in research that analyses how the perception of responsible behaviour affects the consumer s behaviour and attitude towards the firm and its products. For example, Maignan and Ferrell (2001) theoretically propose that corporate social responsibility has a positive influence on consumers in their valuation of a firm s products. This view is confirmed empirically by Brown and Dacin (1997), who observe that social responsibility associations influence product evaluations through the overall corporate image. On the other hand, Sen and Battacharya (2001) consider that the effect of the social associations of a firm on the valuation of its products is moderated by the consumer s support for these actions. In the services market the result of the overall valuation has traditionally been conditioned by the perceived quality (Alet, 1994). Studies of perceived quality basically differentiate between two dimensions: the quality centred on the output and on the process (Parasuraman et al., 1988; Lehtinen and Lehtinen, 1991), the technical and functional quality (Grönroos, 1994) or the contractual and relational quality (McDougall and Levesque, 2000). The first dimension includes all aspects relating to the quality of the service provided, while the second refers to interpersonal relationships and how the service is delivered. However, as Parasuraman et al. (1994) and Anderson et al. (1994) establish, consumers valuation of a service depends not only on its quality but also on its price, such that those users who perceive that they are receiving value for their money are more satisfied then those who do not (Zeinthaml, 1988). Taking this into account, the models that only include the basic quality dimensions are offering an incomplete picture of the determinants of satisfaction or overall evaluation of the service (McDougall and Levesque, 2000). In studies focusing on tangible products the influence of social responsibility on product evaluation has been approached in an indirect way (Brown and Dacin, 1997), but in the services market social responsibility can take on a much clearer role. We should remember that the characteristics of intangibility, inseparability, heterogeneity and perishability of services compared to tangible goods (Grönroos, 1994; Zeinthaml and Bitner, 2000) mean that the consumer approaches purchase in a different way and establishes stronger and more direct relationships with the service provider. In this sense, Sureshchander et al. (2001, 2002) consider that social responsibility, understood as the firm s ethical behaviour, is another component of service quality. According to these authors, a firm whose behaviour is perceived to be ethically correct transmits trust to users and influences their evaluation of the overall quality of the service received. Therefore, we can consider the possible existence of a direct relation between perceptions of social responsibility and service valuation. The corporate citizenship, understood in a broad sense, generates trust among users and these actions can

5 Influence of Corporate Social Responsibility 373 influence consumers when evaluating the service that the firm provides to them. Thus, we propose the following hypothesis: H 1 The perception of socially responsible behaviour has a direct and positive influence on consumers overall valuation of the service received. Corporate social responsibility and loyalty towards a service In the services market loyalty tends to be defined as an observed behaviour (Liljander and Strandvik, 1995). However, measures of loyalty based on behaviour for example the sequence of repeatpurchase are very limited (Day, 1969), since a low level of repeat-purchase of a service can be due to various external circumstances, such as unavailability, barriers to change or lack of alternatives (Andreassen and Lindestad, 1998; Bloemer et al., 1999). In this respect, a second current of research has arisen that considers loyalty as an attitudinal construct (Biong, 1993), which includes for example positive word-of-mouth (Selnes, 1993). In this regard, Delgado and Munuera (2001) speak of commitment to a brand, defined as trust, esteem or the consumer s desire to maintain the relationship or to acquire the same brand. Finally, loyalty also has a cognitive component (Lee and Zeiss, 1980), which includes tolerance to pay a higher price for the product (Anderson, 1996; Fornell et al., 1996). A clear antecedent of loyalty is the perceived service quality (Dick and Basu, 1994). Thus, in many cases authors report the existence of a direct relation between this concept and loyalty (Spreng et al., 1995; Zeinthaml et al., 1996; Bloemer et al., 1998; Jones et al., 2000; Athanassopoulos et al., 2001), so we put forward the following hypothesis: H 2 The overall valuation of the service has a direct and positive influence on consumers loyalty towards the firm. Until this point we have considered only aspects relating to the commercial strategy of the firm as antecedents of loyalty. But corporate social responsibility can also influence loyalty (Sureshchandar et al., 2001, 2002; Maignan and Ferrell, 2001). Social responsibility associations may be useful in strengthening trust and the consumers union with the company (Aaker, 1996; Maignan et al., 1999), since, as Delgado and Munuera (2001) establish, trust is based on the individual s experience of the brand the result not only of the direct relations derived from the use of the service but also of other indirect contacts coming from advertising and reputation. Taking this idea into account, the perception of socially responsible behaviour can strengthen commitment towards a brand, since it transmits a character (Keller and Aaker, 1992; Brown and Dacin, 1997), a system of differentiating values (Turban and Greening, 1997), respect for the customer and concern to serve them correctly, as well as to comply with the quality standards. This explains why consumers reward these efforts with loyalty towards the firm (Maignan et al., 1999). On the other hand, some authors have observed that a large number of consumers claimed to be more willing to buy products from companies involved in social causes (Ross et al., , 1992; Jones, 1997). It seems, therefore, that individuals value companies efforts in programmes of donations, energy conservation or sponsorship of local events, among other actions, and that this greater support may materialise in stronger loyalty towards the firm (Maignan et al., 1999). These authors empirically demonstrate this idea in a study based on the perceptions of company executives. Thus, we put forward the following research hypothesis: H 3 The perception of socially responsible behaviour has a direct and positive influence on consumers loyalty towards the firm. Research methodology and development of measures In order to comply with our objectives and test our research hypotheses, we designed a study based on a personal survey of mobile telephone users. This is a sector in which three firms that are strongly involved in social responsibility compete, so that it is interesting to determine how their efforts are perceived by the users, and the consequences for consumer behaviour. The data collection was carried out in the period 7 21 April We obtained 689 valid responses, with a sample error of 3.7% for a confidence level of 95% and the most unfavourable case

6 374 Ma del Mar García de los Salmones et al. (p = q = 0.5). The sample was stratified in proportion to the firms market share and consumers gender and age. The data were treated using the statistics programs SPSS v.11 and EQS 5.7b. With regards the scale for social responsibility, we considered variables used in several studies. Thus, for the philanthropic aspects, we started from the proposal of Brown and Dacin (1997), adding a reference to sponsorship activities, as in the social commitment scale of Rust et al. (2000). For the economic, legal and ethical questions, we considered items of the battery developed in Maignan s (2000) research, although we had to adapt them in order to make them more suitable for the specific requirements of our research. We should note that we asked interviewees not to value their perceptions about what the firm should do, but rather what the firms actually do with regards social responsibility. In view of this fact, it was not possible to consider the strategies of each company in great depth, since in a previous qualitative stage 3 we noted the consumers ignorance of these questions. Additionally, in the initial scale we included statements concerning the firm s responsibility towards its employees, but the pretests carried out demonstrated that the users had difficulty in evaluating these questions. Thus, we decided to eliminate them from the questionnaire. As far as measuring the overall valuation of the service is concerned, we based on the works of Lee et al. (2001) and Olsen and Johnson (2003), and we considered the dimensions relating to satisfaction with the service in itself (technical quality), the provision of the service (functional quality) and the price (perceived value). Finally, loyalty was measured including behavioural, attitudinal and cognitive questions (Bloemer et al., 1999), which were taken from the commitment scale of Delgado and Munuera (2001) and the scale of Zeithaml et al. (1996). All the items considered are detailed in the appendix. Results In order to fulfil the first objective of our research, we first carried out a principal components factor analysis. The initial values obtained 4 demonstrate the existence of a correlation between the attributes and allowed us to continue with the analysis. We used varimax rotation, which resulted in three factors explaining % of the total variance (Table I), with all the factor loadings significant, since they exceed 0.5 (Hair et al., 1999). With regards the scale for social responsibility, we considered variables used in several studies. Thus, for the philanthropic aspects, we started from the proposal of Brown and Dacin (1997), adding a reference to sponsorship activities, as in the social commitment scale of Rust et al. (2000). For the economic, legal and ethical questions, we considered items of the battery developed in Maignan s (2000) research, although we had to adapt them in order to make them more suitable for the specific requirements of our research. We should note that we asked interviewees not to value their perceptions about what the firm should do, but rather what the firms actually do with regards social responsibility. In view of this fact, it was not possible to consider the strategies of each company in great depth, since in a previous qualitative stage we noted the consumers ignorance of these questions. Additionally, in the initial scale we included statements concerning the firm s responsibility towards its employees, but the pre-tests carried out demonstrated that the users had difficulty in evaluating these questions. Thus, we decided to eliminate them from the questionnaire. As far as measuring the overall valuation of the service is concerned, we based on the works of Lee et al. (2001) and Olsen and Johnson (2003), and we considered the dimensions relating to satisfaction with the service in itself (technical quality), the provision of the service (functional quality) and the price (perceived value). Finally, loyalty was measured including behavioural, attitudinal and cognitive questions (Bloemer et al., 1999), which were taken from the commitment scale of Delgado and Munuera (2001) and the scale of Zeithaml et al. (1996). All the items considered are detailed in the appendix. The first factor includes all the economic questions, referring to consumers perceptions of firms orientation to profits, success and good performance, so that we labelled this factor economic responsibility. Second, we observed that the legal and ethical dimensions fused into a single factor, which we labelled Ethical legal responsibility. Included in this factor are statements relating to firms compliance with the law and with obligations vis-à-vis the various agents, honesty in relationships with their customers and the priority of

7 Influence of Corporate Social Responsibility 375 TABLE I Dimensions of corporate social responsibility Factor 1 Economic responsibility Factor loadings ER1 Tries to obtain maximum profit from its activity ER2 Tries to obtain maximum long-term success ER3 Always tries to improve its economic performance Eigenvalue % variance explained Factor 2 Ethical legal responsibility Factor loadings ELR1 Always respects the norms defined in the law when carrying out its activities ELR2 Is concerned to fulfil its obligations vis-à-vis its shareholders, suppliers, distributors and other agents with whom it deals ELR3 Behaves ethically/honestly with its customers ELR4 Respecting ethical principles in its relationships has priority over achieving superior economic performance Eigenvalue % variance explained Factor 3 Philanthropic responsibility Factor loadings PR1 Is concerned to respect and protect natural environment PR2 Actively sponsors or finances social events (sport, music...) PR3 Directs part of its budget to donations and social works favouring the disadvantaged PR4 Is concerned to improve general well-being of society Eigenvalue % variance explained ethical compared to economic behaviour. Finally, the social aspects, referring to the environment, involvement in social events, corporate giving or community development, are included in philanthropic responsibility. The evaluation of the pshychometric properties of the scale (reliability and validity) was made following Churchill s (1979) methodological suggestions. Specifically, we carried out a confirmatory factor analysis with maximum likelihood estimation using EQS 5.7b (Bentler, 1995) and evaluated in the first place the combined model applying absolute, incremental and parsimonious measures of fit. These measures provided support for our acceptance of the proposed model (Hair et al., 1999; Del Barrio and Luque, 2000). We also examined the measurement model by individual factors, confirming the statistical significance of each loading obtained between the indicator and the construct, as well as the reliability of the measure of each dimension. With these measures we also analysed the convergent and discriminant validity of the proposed model. The analysis carried out with all the variables demonstrated the need to eliminate one of the items of the scale proposed to measure philanthropic responsibility, the statement relating to sponsorship of events (PR2). The standardised lambda coefficient of this variable is less than the minimum required value of 0.5, and the goodness of fit of the total model loses significance. If this item is eliminated from the scale, we obtain an adequate specification of the proposed factor structure (Table II). Specifically, the statistics BBNFI, GFI and AGFI in this case reach values higher than 0.9 and the BBNNFI is close to this value, while the RMSEA is at exactly On the other hand, although the v 2 coefficient is significant, it cannot be regarded as a reliable indicator of the goodness of fit of the confirmatory

8 376 Ma del Mar García de los Salmones et al. TABLE II Confirmatory factor analysis of corporate social responsibility Latent variable Measured variable Standar lambda R 2 Cronbach s a Composie reliability Confidence intervals for the correlations Goodness of fit indices Economic R. (ER) Ethical legal R. (ELR) Philanthropic R. (PR) ER1 ER RE-REL ( ) ER ELR REL-RF ELR ( ) ELR REL PR RE-RF )0.045 PR ()0.548 to )0.352) PR S-Bv 2 = (p=0.0000), BBNFI=0.904, BBNNFI=0.885, GFI=0.944, AGFI=0.904, RMSEA=0.08 model in view of its sensitivity to samples exceeding 200 units (Bollen, 1989). The reliability of the measuring scales proposed was evaluated using the Cronbach s alpha and composite reliability coefficients (Bagozzi and Yi, 1988). The values of these statistics exceed the minimum recommended value of 0.7 (Hair et al., 1999), which confirms the internal reliability of the proposed constructs. In addition, all the items are significant to a confidence level of 95% and their standardised lambda coefficients exceed 0.5 (Steemkamp and Van Trijp, 1991), confirming the convergent validity of the proposed model. Finally, in order to confirm the discriminant validity we followed the procedure described by Anderson and Gerbing (1998), in which the confidence intervals for the correlation of the constructs are estimated and compared with unity. In none of the cases did the intervals contain the value 1. Hence the measurement model proposed is correct. From the results obtained in these initial analyses we can draw various conclusions. First, we find that the consumers in principle perceive three levels of corporate behaviour: economic, ethical legal and philanthropic. As Carroll (1979) establishes, the ethical dimension is difficult to define, although it can be understood as society s expectations that firms behaviour is in accordance with the law and indeed goes further than legal frameworks. Carroll considers that the legal factor also includes ethical aspects, a fact that may explain why both these responsibilities are perceived from the consumers perspective to be in the same dimension. Second, the sponsorship of events is not included within philanthropic responsibility. This is not surprising, since sponsorship is not really considered to be a social activity in many of the studies on corporate social responsibility (Brown and Dacin, 1997; Lafferty and Goldsmith, 1999; Handelmann and Arnold, 1999; Maignan, 2001), due to its commercial and non-philanthropic nature. Following D Astous and Bitz (1995), philanthropic activities are associated with patronage i.e., financing cultural events or supporting social and humanitarian causes, voluntary actions in which the role of patron is less prominent and the firm does not expect an immediate return. In contrast, sponsorship has a strongly commercial nature, since it aims to link the firm with some massive event in order for it to benefit commercially, in the form of greater brand awareness, support for the image and higher sales. These initial conclusions lead on to the secondorder factor analysis which allow us to confirm the multi-dimensionality of social responsibility from the consumer s perspective and to determine the weight of each factor in the global construct. The most important conclusion drawn from this model is that the economic behaviour of firms is not perceived to be a component of corporate social responsibility by consumers. The low correlation between the economic and non-economic dimensions observed in the first-order model led us to suspect at this early stage

9 Influence of Corporate Social Responsibility 377 D ELR 0,473 R 2 = 0,776 0,692* ELR1 0,722 EELR1 R 2 = 0,479 0,881* ELR 0,661* 0,725* 0,601* ELR2 ELR3 0,750 EELR2 R 2 = 0,437 0,689 EELR3 R 2 = 0,525 CSR ELR4 0,799 EELR4 R 2 = 0,362 0,847* 0,532 PR D PR R 2 = 0,565 0,674* 0,653* 0,752* PR1 PR2 PR3 0,738 EPR1 R 2 = 0,455 0,757 EPR2 R 2 = 0,427 0,659 EPR3 R 2 = 0,565 Figure 1. Second-order factor analysis of corporate social responsibility. the weak association between economic responsibility and the global construct of social responsibility. Thus, only firms ethical legal and philanthropic behaviour are included in this concept (Figure 1). The factor loadings between the second-order factor and these dimensions (REL and RF) are statistically significant (0.881 and 0.847, respectively), and the goodness of fit statistics also indicate that the model fits the data well. With regards this result, Aupperle et al. (1985) conclude in their work oriented to firms that social responsibility can be measured in terms of the importance of the three non-economic components compared to the economic component. They observe that the economic dimension is inversely correlated with the other three, a result which leads them to consider two categories: concern for society, which includes the ethical, legal and philanthropic dimensions, and concern for economic performance. For these authors this differentiation is logical, because social responsibility has generally been associated more with ethical, legal and philanthropic actions. When consumers perceptions about firms responsible behaviour are examined, this result is much clearer. Maignan and Ferrell (2001) already wondered if consumers would include firms economic responsibility within the concept of social responsibility, and going by the results obtained in the model, they do not. As has been confirmed in the confirmatory models, the ethical, legal and philanthropic actions are what really make up corporate social responsibility. In addition, both have a similar importance within the construct. The economic dimension, or concern for economic performance according to Aupperle et al. (1985), is an independent factor of corporate behaviour. We now test the research hypotheses. In this respect we propose a theoretical model of structural equations that directly relates corporate social responsibility with the overall valuation of the service (Hypothesis 1) and with loyalty (Hypothesis 3). Loyalty is also conditioned by the users overall valuation of the service (Hypothesis 2). Following Anderson and Gerbing (1998), we first confirmed the reliability and validity of all the concepts by means of a confirmatory analysis. Previously we had calculated second-order factors with the constructs overall valuation of service and loyalty. We proposed that overall valuation is made up of the dimensions technical quality, functional quality and perceived price, which is confirmed since all the dimensions converge appropriately on the factor. As far as loyalty is concerned, we initially proposed that this construct was made up of commitment to the brand, repeat-purchase behaviour and price tolerance. However, the model had to be redefined: we obtained better results of convergence

10 378 Ma del Mar García de los Salmones et al. and global goodness of fit when only two dimensions were considered: commitment-behaviour and price tolerance. This result is accepted, since what is really evaluated is loyalty towards the service in a sector in which the providers are contracted for the medium to long term, so that intentions to change operators are measured in a long timescale and implicitly indicate users commitment to the brand. The results of the confirmatory factor model with all the constructs of the final model are reported in Table III. The model thus proposed is correct, since all the second-order latent variables have internal consistency, as is demonstrated by the Cronbach s alpha and composite reliability, which exceed the 0.7 threshold. The exception is in the loyalty construct, which achieves a lower value in the composite reliability coefficient, because the price tolerance dimension has a low, although representative, loading in this factor. The standardised coefficients are all significant and exceed 0.5, guaranteeing the convergent validity of the dimensions. We observe that in the overall valuation of the service the dimensions functional quality and perceived price have a greater weight, while the more technical aspects have a somewhat lower loading. In loyalty, commitment to the brand takes on a much greater weight than price tolerance. Finally, and with regards social responsibility, in this confirmatory model the weight of ethical legal behaviour takes on more importance within the global construct than that of the philanthropic dimension. The discriminant validity of the model is evaluated using the confidence intervals for pairs of latent concepts. As can be seen in Table IV, none of the confidence intervals calculated for the pair-wise comparisons of the model variables include unity. Subsequently, we estimated the structural model. The indices of goodness of fit of the model to the data are correct, since the statistics exceed or come close to optimal values (Table V). As far as testing the hypotheses goes, we observe that corporate social responsibility has a significant influence on consumers valuation of the service, so that Hypothesis 1 cannot be rejected. With regards the antecedents of loyalty, we confirm the important weight of the overall valuation of the service (Hypothesis 2 not rejected), although the proposed direct relation between social responsibility and loyalty is not significant (Hypothesis 3 rejected). TABLE III Confirmatory factor analysis of the final model Latent variable Measured variable Standar lambda R 2 Cronbach s a Composie reliability Goodness of fit indices Corporate Social Responsibility (CSR) ELR S-Bv 2 = PR (p = ), Overall Valuation of the Service (OVS) TQ BBNFI = 0.89, FQ BBNNFI = 0.90, PP GFI = 0.90, Loyalty (LOYAL) CC AGFI = 0.88, PT RMSEA = 0.06 TABLE IV Confidence interval for the correlations among pairs of latent variables Correlation Confidence interval Corporate Social Responsibility (CSR) Overall Valuation Service (OVS) Corporate Social Responsibility (CSR) Loyalty (LOYAL) Overall Valuation Service (OVS) Loyalty (LOYAL)

11 Influence of Corporate Social Responsibility 379 TABLE V Structural model estimation Dependent variables Overall valuation Loyalty Independent variables Overall valuation of service (7.557) Corporate Social Responsibility (7.595) n.s. R 2 = R 2 = S-Bv 2 = (p = ), BBNFI = 0.889, BBNNFI = 0.900, GFI = 0.902, AGFI = 0.877, RMSEA = Corporate social responsibility does have an effect on loyalty, but it only acts indirectly by means of service valuation (Figure 2). Hence, social responsibility becomes a clear marketing tool within the services market. Specifically, through our test we have observed that in the segment of service users the efforts in corporate social responsibility have commercial rewards, because the perceptions of good ethical and philanthropic corporate behaviour affect the overall evaluation of service. On one hand a socially responsible company transmits the impression of being responsible with its product, and on the other hand the honesty and ethic of its performance generates confidence and safety to the consumers. On the whole, the effect on evaluation of service is positive. As far as loyalty is concerned, we find that the overall valuation of the service has an important weight when explaining users behaviour and attitude towards a brand. With 79% of the variance explained, it is a very strong antecedent of loyalty, unlike the case of socially responsible behaviour. OVS CSR 0,94 0,69 LOYAL Hypothesis accepted Hypothesis rejected Figure 2. Structural model estimation. The perceptions of a social nature are not significant or determinant in the maintenance of the relationship with a firm: this is more conditioned by aspects relating to the quality of the service and the perceived price. However, the influence of social responsibility on loyalty is indirect, since as we have seen, it does have an influence on the valuation of the commercial aspects relating to the provision of the service and the service in itself. Conclusions, limitations and directions for future research The current work has tried to deepen understanding about the concept of corporate social responsibility from the consumers perspective, as well as its benefits as a commercial tool. With this in mind, we reviewed the extant literature on the relation between social responsibility and consumer behaviour, which brought to light two questions. First, we observed that various measures of the concept have been adopted in previous work: in some cases authors consider only social or philanthropic aspects relating to the role the firm plays in social well-being; while in others, the number of dimensions is greater. Specifically, one of the most complete and widely used proposals in the study of social responsibility is that of Carroll (1979, 1991), in which the concept is considered to be composed of four components: economic, legal, and philanthropic. The four dimensions of this model have been of use to us in developing this current study. Specifically, we carried out market research based on personal surveys of users of mobile telephone services in the Spanish market, collecting the consumers direct perceptions about the behaviour of the operating firms in terms of

12 380 Ma del Mar García de los Salmones et al. these issues. By means of a principal components factor analysis, we confirmed that there are three dimensions of corporate behaviour from the perspective of the consumers: the economic, Ethical legal and philanthropic aspects. Validating the scale by first-order confirmatory analysis obliged us to eliminate the concept referring to sponsorship of social events from the philanthropic dimension. This activity is really related more with commercial aspects the search for short-term returns in terms of sales and image so that consumers logically do not include it within the philanthropic or altruistic component of social responsibility. On the other hand, we have validated the multi-dimensionality of corporate social responsibility by means of a second-order confirmatory analysis, in which we observed that from the consumers perspective the economic dimension is not a component of social responsibility. For service users, firms search for profits and success in their activity is a totally independent dimension of corporate behaviour. With regards the second question, from the literature review it became clear that studies oriented towards the consumers need to be developed even more, since the consequences of socially responsible actions in this segment are still unclear. In this respect, we proposed the research hypothesis that firms socially responsible behaviour has a direct and positive influence on users in their overall valuation of the service (measured by means of technical quality, functional quality and perceived value), as well as in their loyalty towards the firm (measured by means of behaviour-commitment and price tolerance). We used a model of structural equations to demonstrate the direct relation between social responsibility and the overall valuation of the service, which confirms that responsible behaviour has commercial benefits for firms. We did not observe, however, any direct relation between the perception of social responsibility and consumers loyalty towards the firm. In this sense, the weight of service valuation in loyalty is very strong, such that it becomes its maximum antecedent. Nevertheless, corporate social responsibility does influence loyalty indirectly via the overall valuation the user makes of the service received. The results obtained in this research have allowed us to detect certain commercial implications from investment in corporate social responsibility. The fact is that profit making companies have economic responsibilities related to long term success and maximization profit and share value. Nevertheless, the end doesn t justify the means and the economic orientation must be closely linked to social orientation. We can say that a company with well defined ethical codes and a clear philosophy of social commitment, respect for the environment and honesty in its relationships with the stakeholders, will be able to achieve better economic results. The first step to improve the socially responsible activities of a firm is to understand the meaning and implications of this philosophy. In this sense, it will be important to subscribe to international codes and initiatives in order to formalize the policy with regard to this matter. Besides, we have to take into account that the responsible corporate behaviour has to be perceived in order to get results. The influence on consumers attitudes will take effect only when the users perceive that the firms do indeed carry out such actions, so defining ethical codes is not enough, the important thing is setting them in motion. It will be important to involve all the employees in this philosophy and to target this internal public with a responsible policy because corporate social responsibility begins inside the company. Furthermore, a company should communicate intelligently its social orientation and its environmental involvement. It can use internet and its web page in order to inform about its activities with a social dimension. Taking into account that the mass media reaches many people, companies can also refer to what they do in their advertising or with publicity. Besides, it s possible to involve users by organizing some activities that encourage their active participation. For example, in the market considered the operators could favour recycling campaigns of mobile phones. There are various limitations of this work, which in turn lead us to propose future lines of research. On the one hand, the hypotheses have been tested in a specific sector in the services market, so that in order to generalise the results, and taking into account the scarcity of empirical work on this particular case, we consider it necessary to broaden the study to other environments. On the other hand, we have worked with global constructs of social responsibility, overall valuation and loyalty, and not

13 Influence of Corporate Social Responsibility 381 with each dimension individually. In a future study we propose to analyse the relations between the variables with the dimensions considered separately, in order to determine possible differences in consumer behaviour. For example, we might expect Ethical legal behaviour to have more impact than philanthropic behaviour on functional quality, since it includes variables relating to honesty in dealings with customers. It s also possible to include new constructs in the model. The evaluation of service can be seen as an antecedent of satisfaction and it would be interesting to check the impact of switching brand costs on loyalty. Similarly, we could include the economic dimension of corporate responsibility, to see how this affects consumers attitudes. In addition, there is a need to further develop the measuring scale of social responsibility from the consumers perspective. We started from Carroll s model and their four dimensions, although as Maignan (2001) establishes, consumers may not see social responsibility in this way, nor include the same obligations in the construct. In the previous qualitative phase we carried out for the current research we detected the users lack of awareness about firms social actions, which led us to eliminate from the scale statements concerning treatment of employees and other ethical aspects, such as firms positioning with respect to minorities or disadvantaged groups. We suggest, then, to advance in the study of social responsibility with more research allowing us to deepen understanding of how consumers understand social responsibility. In the same sense, we should consider economic responsibility. Traditionally, this been measured considering aspects closely relating to the search for profits and enhanced performance i.e., with a perspective highly oriented towards the firm. This fact may explain why consumers do not consider this dimension to form part of the global construct of corporate responsibility. We feel it is necessary to work more in the presentation of this component to the consumers, since from their point of view it may be explained better by other economic aspects such as the generation of wealth in the region, value added or job creation. It may be that these factors measured as such do form part of social responsibility. Finally, we related the perception of responsible behaviour with loyalty, without taking into account, as Sen and Battacharya (2001) do, the support consumers give to these measures or whether these actions are coherent with the personality or the concept users have of themselves. Not considering this aspect has meant that social responsibility does not directly affect loyalty towards the firm. If we had taken various consumer groups as reference in function of this aspect, our finding may have been different. Taking into account the scarcity of studies oriented towards consumers, the current work has attempted to provide new ideas about the components of social responsibility and its benefits for service users. At the same time, it should stimulate new research that can deepen our understanding of the commercial implications of companies efforts and investment in socially responsible behaviour. APPENDIX Corporate Social Responsibility (7 point Likert scales: 1, total disagreement; 7, total agreement) I believe my operating company... Tries to obtain maximum profit from its activity Tries to obtain maximum long-term success Always tries to improve its economic performance Always respects the norms defined in the law when carrying out its activities