Reputation of multinational companies: Corporate social responsibility and internationalization

Abstract

Purpose

The purpose of this paper is to use stakeholder theory as the theoretical reference framework to study the influence of internationalization (geographic international diversification) and social performance on multinational companies’ (MNCs) reputation.

Design/methodology/approach

The authors confirm the research hypotheses using a sample of 113 US MNCs in the chemical, energy and industrial machinery sectors during the period 2005-2010.

Findings

This study contributes to the literature in three ways. First, it incorporates literature on internationalization to study the possible connection between geographic international diversification and social performance in MNCs. Second, it sheds light on the debate between corporate social responsibility (CSR) and the reputation of MNCs in a very diverse transnational context in which MNCs must meet the needs of stakeholders at both local and global levels. Third, it incorporates the mediating role of social performance in the relationship between geographic international diversification and the firm’s reputation.

Originality/value

Prior studies have hardly analyzed this relationship, which becomes especially relevant for MNCs, since their implementation of advanced CSR practices in the different markets in which they operate will gain them a good reputation, not only in specific local contexts but also globally, benefitting the organization as a whole and enabling it to gain internal consistency (improvement in internal efficiency), transparency and legitimacy.

Citation

Aguilera-Caracuel, J., Guerrero-Villegas, J. and García-Sánchez, E. (2017), "Reputation of multinational companies: Corporate social responsibility and internationalization", European Journal of Management and Business Economics, Vol. 26 No. 3, pp. 329-346. https://doi.org/10.1108/EJMBE-10-2017-019

1. Introduction

Corporate social responsibility (CSR) has acquired great relevance in the academic world and in firm management in recent years (Barrena et al., 2016; Madorran and Garcia, 2016). Defined as organizations’ commitment to contribute to sustainable economic development, CSR includes issues such as employee labour conditions, improvements in products and services, progress that seeks to satisfy the needs of the local community and advances in environmental management, among others (World Business Council for Sustainable Development, 2004).

Since multinational companies (MNCs) are organizations that can incorporate advances and improvements in social issues in the different areas in which they operate, the literature increasingly considers them as precursors of economic and social progress (Porter and Kramer, 2011). One of the main reasons these firms invest in advanced CSR practices is to improve their reputation. While one definition of reputation is the perception that stakeholders have of the firm’s willingness and capacity to satisfy stakeholders’ interests (Fombrun, 1996), some studies argue against considering reputation as the aggregate perception of the set of stakeholders, primarily for two reasons (Walker, 2010). First, reputation depends on the issue to which it refers – for example, reputation with respect to behaviour in environmental, social, employee or corporate governance matters and product quality, among other issues. Second, reputation can vary for each specific interest group – for example, consumers, investors or the government (Lewellyn, 2002). Based on these assumptions, having an excellent reputation does not necessarily imply satisfying the different stakeholders; that would require determining what type of reputation we mean and for which specific group.

Previous studies have analyzed the relationship between CSR and the firm’s results, but findings are not conclusive (Madorran and Garcia, 2016) due to mediating and moderating variables such as the context (country and region) of stakeholders’ locations (Gardberg and Fombrun, 2006). The literature has paid little attention, however, to how MNCs, which must typically negotiate the needs of very heterogeneous stakeholders (both local and international), can increase their reputation by improving their social performance (Musteen et al., 2013). In fact, MNCs should manage relations in social issues with all stakeholders in the different contexts in which they operate. All of these issues can affect their reputation directly.

MNCs often operate in different countries and regions with varied institutional profiles (Kostova et al., 2008). Through internationalization, they can attend to the demands of stakeholders in the different contexts and markets in which they operate, there by achieving the image of responsible, legitimate, transparent entities committed to the environment (Christmann, 2004). Further, they can strengthen their internal organizational framework by extending their business model outside their boundaries (Aguilera-Caracuel et al., 2013), efficiently transferring their best practices, policies and business models (Hitt et al., 1997) by means of management standards (Aguilera-Caracuel et al., 2012). All of this activity can have very positive repercussions for both their social performance and significant improvement in their reputation driven by this social performance.

This study uses stakeholder theory as the theoretical reference framework to study the influence of internationalization (geographic international diversification) and social performance on MNCs’ reputation. We confirm the research hypotheses using a sample of 113 US MNCs in the chemical, energy and industrial machinery sectors during the period 2005-2010. The results show that the MNCs with the highest levels of social performance that can fulfil the expectations of stakeholders in both local and global contexts obtain a better reputation, as their stakeholders come to see them as responsible, consistent organizational entities. The results also support the relationship between geographic international diversification and reputation through the mediating role of social performance.

This study contributes to the literature in three ways. First, it incorporates literature on internationalization to study the possible connection between geographic international diversification and social performance in MNCs. Second, it sheds light on the debate between CSR and the reputation of MNCs in a very diverse transnational context in which MNCs must meet the needs of stakeholders at both local and global levels. Third, it incorporates the mediating role of social performance in the relationship between geographic international diversification and the firm’s reputation. Prior studies have hardly analyzed this relationship, which becomes especially relevant for MNCs, since their implementation of advanced CSR practices in the different markets in which they operate will gain them a good reputation, not only in specific local contexts but also globally, benefitting the organization as a whole and enabling it to gain internal consistency (improvement in internal efficiency), transparency and legitimacy.

2. Theoretical foundations

2.1 Stakeholder theory in MNCs

The term “stakeholder” refers to the individual or group that can affect or be affected by the firm’s action in pursuit of its objectives (Freeman, 1984). Stakeholder theory is especially useful for explaining why firms decide to put into practice socially responsible management, such as promotion of social progress in the local community, protection of disadvantaged groups, improvement of workers’ labour conditions or commercialization of ecological products and services. According to this theory, negative actions in social issues such as polluting the environment or abusing employees have a very negative effect on stakeholders’ perception of the firm (Freeman, 1984). For Porter and Kramer (2006), firms and society complement each other and should act jointly. Companies should thus make decisions related to the formulation and implementation of CSR initiatives in active and continuous dialogue between the company and its different interest groups (Christmann, 2004). Actions oriented to satisfying stakeholders’ needs should be a priority goal for firms, impacting them positively in the medium and long term (Valenzuela-Fernández et al., 2015). Whereas some studies assimilate the concept of the firm’s “social performance” with “stakeholders’ satisfaction level” (e.g. Clarkson, 1995; Post et al., 2002; Zagenczyk, 2004), others hold that the two terms are related but should not be confused. The complexity of the relationship between the firm and stakeholders grounds separation of the two concepts, due fundamentally to the diversity of the socially responsible initiatives and heterogeneity of the interests of each specific interest group. To determine the effect of social performance on stakeholder satisfaction, it is thus necessary to analyze the type and intensity of the firm’s social actions, as well as their impact on each specific group of stakeholders (Luna and Baraibar, 2011).

Stakeholder theory is especially relevant to MNCs, since they are in contact with a multitude of interest groups due to their operation in numerous countries and regions (Jamali, 2010). Each stakeholder has its own needs and expectations, some more specific and others more extendable to different contexts (Aguilera-Caracuel et al., 2015). Compared to local firms, MNCs are under more pressure because they face very relevant, diverse, strong interest groups in the local and global environments, groups that will grant them legitimacy and the power to act (Kang, 2013). For example, stakeholders in certain countries consider gender equality at work as essential, whereas those in others do not see it as a real priority (Connell, 2005). The same duality emerges on environmental issues, especially if we compare developing countries with developed ones (Becker and Henderson, 2000).

Finally, for the firm to benefit from relationships with its stakeholders, it is important to distinguish appropriately between them and to capture their needs. By responding appropriately to stakeholders’ expectations, firms can obtain licence to operate in foreign markets (Park et al., 2015).

2.2 Geographic international diversification and social performance in MNCs

We can define international diversification as “the number of markets in which the firm operates and their respective importance” (Hitt et al., 1997). International diversification of MNCs involves interacting with different cultures and levels of economic and legal development, including satisfying the needs of diverse interest groups – governments, political parties, unions, NGOs and customers, among others (Crane and Matten, 2010; Rodríguez et al., 2006). Although international diversification involves some risk due to operation in new markets with diverse institutional, political, environmental and cultural profiles (Aguilera-Caracuel et al., 2013), organizations can also benefit from economies of scale and overcome entry barriers in specific markets by having the right response to their stakeholders’ demands (Kang, 2013).

Different reasons drive internationally diversified MNCs to develop socially responsible initiatives. First, greater visibility and exposure to these firms’ pressures from interest groups can make socially responsible behaviour a priority (Christmann, 2004; Crane and Matten, 2010; Kang, 2013; Rodríguez et al., 2006; Yang and Rivers, 2009). Greater visibility can motivate MNCs to be more proactive in social and environmental issues and ultimately strengthen relationships with the societies in which they operate (Garriga and Mele, 2004). The demands of the different stakeholders can vary, however, depending on the country in which the firm operates – demands on issues such as gender equality (Connell, 2005), corruption (Transparency International, 2004) or workers’ rights (Ahmadjian and Robinson, 2001). To achieve positive effects derived from these actions and to satisfy stakeholders, firms must make an effort to inform their stakeholders adequately of the social responsibility initiatives the firm performs (Sen et al., 2006). Strike et al. (2006) affirm that MNCs with a high level of international diversification can communicate more effectively with the different interest groups, capture their needs and play an active role in the design of national and international regulations in the social and environmental arenas. For example, MNCs can have stable, trust-based relationships with governments and public powers that grant them privileged access to subsidies and licences to act in different markets (Luo and Bhattacharya, 2009).

Second, response to social demands enables MNCs to reduce specific risks significantly (Deckop et al., 2006): failure to comply with legislation, pressures from firms in the same sector and from business associations, negative reactions from public opinion and consumers’ associations, problems with activists and NGOs and possible consumer boycotts. Along these lines, various studies show a positive relationship between high level of social responsibility and cost reduction. For example, some studies show that investing in environmental issues can help firms to avoid penalties (Aguilera-Caracuel et al., 2013; Hart, 1997) or pressure from civil and consumers’ associations (Henriques and Sadorsky, 1996; Russo and Fouts, 1997).

Third, in contrast to firms with little geographic diversification, MNCs that operate in diverse markets can redistribute the costs and benefits of investing in CSR among their subsidiaries, an economic incentive to invest in these topics (McWilliams and Siegel, 2001). For example, the MNC can use the positive image it derives from high involvement in socially responsible behaviour efficiently in different markets and cultures (Lichtenstein et al., 2004). Some studies show that MNCs standardize specific CSR activities that they consider as universal in such social responsibility issues as workplace health and safety, human rights, corruption and climate change (Bondy et al., 2012). MNCs can also adapt some CSR activities depending on the specific rules and values of the country or region in which they operate. Other studies, in contrast, hold that there are no absolute or universal CSR practices and that companies should fit these practices to each area (Ang and Maasingham, 2007; Mooij and Hofstede, 2010). Kakabadse et al. (2005) hold that CSR means different things in Europe, the USA and developing countries and that MNCs should adapt their levels of social responsibility to each area (Arthaud-Day, 2005; Wang and Juslin, 2009). MNCs in developing countries, for example, can orient specific actions to compensating for the inefficiency of public resources (Valente and Crane, 2010) and to making up for the lack of government resources to satisfy basic needs (Eweje, 2006) in areas such as environmental protection, labour rights and anticorruption and antidiscrimination policies (Barkemeyer, 2011).

In sum, the literature identifies various reasons that can drive MNCs with a high degree of geographic diversification to develop social responsibility initiatives. These firms are more visible and thus generally more subject to pressure from different interest groups. The right response to such pressures from these stakeholders can reduce the risks that affect the firm negatively (e.g. from consumers’ associations, activists, NGOs and customers). Further, MNCs can redistribute costs and their investments in social issues. Based on these arguments, we propose that greater geographic international diversity of MNCs will contribute to implementation and development of socially responsible initiatives that result in improvement in their social performance:

H1.

Geographic international diversification of MNCs is positively related to their social performance.

2.3 Social performance and reputation of MNCs

MNCs’ efforts to develop socially responsible initiatives have grown significantly in recent years (Park et al., 2014), and the literature in this field argues that these initiatives can enable firms to improve their reputation (Melo and Garrido-Morgado, 2012; Valenzuela-Fernández et al., 2015). As a result, reputation has become an extrinsic motivation to develop socially responsible activities (Fombrun, 2005). Having a good reputation helps to improve interest groups’ trust in the firm, while also enabling it to improve its competitive position (Fombrun, 1996). In contrast, damage to the reputation can seriously jeopardize the firm (Bertels and Peloza, 2008).

Some studies support viewing social initiatives as a strategic investment in maintaining and improving reputation (Valenzuela-Fernández et al., 2015). Along these lines, firms use CSR as a strategic tool to satisfy the demands of the different stakeholders – such as NGOs and consumers – in order to create a good corporate image (Jones, 2005) and achieve competitive advantage (Fombrun et al., 2000). According to Klein and Dawar (2004), developing social initiatives is a source of differentiation for firms and helps to create good brand image (Gardberg and Fombrun, 2006); consumers tend to reward firms that behave ethically by paying higher prices for their products (Creyer and Ross, 1997).

Other studies show a positive effect of philanthropic initiatives on reputation (e.g. Lange et al., 2011). Fombrun and Shanley (1990) find that firms that have established a foundation and make more contributions to charity have better reputations. Analyzing determinants of reputation in large corporations in the UK, Brammer and Millington (2005) find that people perceive firms with more philanthropic behaviour as socially more responsible and that these firms enjoy a better reputation than firms that do not follow such behaviour. In general, this group of studies shows that greater involvement in the local community has a positive effect on reputation, suggesting that the different stakeholders expect the firm to behave well towards the community (Brammer and Pavelin, 2006). According to Gugler and Shi (2009), MNCs that operate in developing countries usually understand the concept of CSR in an ethical and philanthropic sense, for example, in terms of monetary contribution to the community. MNCs that operate in these markets sometimes develop ethical and philanthropic activities that enable improvement of sustainability and economic development in these countries (Kolk and van Tulder, 2010; Porter and Kramer, 2011). Through direct investment, MNCs can provide much-needed resources in the Third World, such as technology and work skills. Dickson (2003) holds that philanthropic initiatives are especially welcome when governments do not have sufficient resources to undertake social welfare projects. Through these activities, MNCs can drive sustainability and economic development in developing countries (Moon et al., 2005) to improve their reputation.

MNCs can also improve their reputation by developing socially responsible behaviour with their customers and employees (Park et al., 2014). Customers constitute an interest group that exerts significant pressure on the firm if it detects that the firm is behaving in a socially irresponsible way, as customers have instantaneous information and a multitude of alternatives (Lindgreen et al., 2009). To avoid loss of customers, some MNCs make an effort to develop CSR initiatives, taking into account the ethical standards of the markets in which they operate (Yang and Rivers, 2009). Further, firms with management practices oriented to encouraging workers’ well-being at work can simultaneously improve workers’ productivity and morale (McWilliams and Siegel, 2001), reduce the level of absenteeism (Berman et al., 1999) and generally have a more consistent workforce (Branco and Rodríguez, 2006). All of the foregoing enables the firm to have a comparative advantage over other firms in the same sector and geographic area (Berman et al., 1999; Heikkurinen, 2010), positively influencing its reputation (Melo and Garrido-Morgado, 2012).

Implementing CSR initiatives is especially important for MNCs’ reputation and public image (Fombrun, 1996). More specifically, loss of reputation for MNCs may be one of the greatest problems it can face (Levis, 2006), since MNCs are more visible and subject to greater stakeholder scrutiny (Christmann, 2004). Some studies suggest that institutional pressures have pushed MNCs to increase their social initiatives at both global and local level (Aguilera et al., 2007). Global CSR initiatives include establishing standards related to safe working conditions, protection of minorities (Jamali, 2010) and preservation of the natural environment (Aguilera-Caracuel et al., 2012). It is also important that MNCs’ social commitment be in accord with local stakeholders’ demands (Pedersen and Neergaard, 2009), that is, with the specific needs and circumstances of the territories in which they operate. For example, in South Africa it is essential to involve stakeholders in the fight against unemployment and AIDS (De Jongh, 2004). All such initiatives, at local and global level, enable stakeholders to protect and even improve their reputation (Kolk and van Tulder, 2010).

As a result, we can see reputation as an intangible resource (Heikkurinen, 2010) that organizations use to improve their image and reinforce their brand (Porter and Kramer, 2006). Improvement in reputation can result from implementing practices connected to CSR that can facilitate achievement of a sustainable competitive advantage (Melo and Garrido-Morgado, 2012). Based on the foregoing arguments, we propose that implementation and development of initiatives that improve MNCs’ social performance and that attend to the demands of the different stakeholders will contribute to improving MNCs’ reputation:

H2.

Social performance of MNCs is positively related to their reputation level.

2.4 Mediating role of the MNC’s social performance

As we argued above, MNCs with a high level of geographic international diversification operate in the presence of a broader group of stakeholders and thus suffer greater public scrutiny. Operating in many markets can create an opportunity to extend their social initiatives throughout their organizational framework (McWilliams and Siegel, 2001). Other studies show that one of the strategies most relevant for MNCs consists of developing ethical management practices linked to social development and very closely linked to obtaining a sustainable reputation – practices such as improving labour conditions, organizational climate and other measures outside the firm with repercussions in the areas of the environment and cultural and socioeconomic progress in specific regions (Park et al., 2015; Park and Ghauri, 2015). All of the foregoing can give these organizations strong credibility, especially in their social sphere, greatly benefitting their reputation.

Based on these arguments, and considering the hypotheses presented above, we propose that social performance plays a mediating role between geographic international diversification and the MNC’s reputation. That is, social performance is the way in which highly diversified MNCs can improve their reputations in the different markets in which they operate:

H3.

MNCs’ social performance mediates the relationship between geographic international diversification and reputation level.

3. Sample and variables

3.1 Sample

We tested the hypotheses on a sample of US MNCs from the chemical (SIC 28), energy (SIC 29) and industrial machinery (SIC 37) sectors. These three activity sectors typically have great environmental impact, with significant repercussions in the social arena. In fact, they are activity sectors relevant for analyzing the impact of CSR practices on organizations with international presence (Christmann, 2004). The choice of firms with headquarters in the USA is especially relevant in this study because we analyze firms that usually have a significant impact on social, environmental and international issues.

Starting from the information available in Standard & Poor’s (Capital IQ) database, we perform simple random sampling to select a total of 100 firms from the chemical sector, 100 from the energy sector and 100 from the industrial machinery sector. The final sample included a total of 113 MNEs and 672 observations over a period of six years (2005-2010). We also obtained the financial information from Standard & Poor’s (Capital IQ) and draw the information on social performance from the KLD database. Table I presents the methodology used in the study.

3.2 Variables

Social performance of the firm

We obtained the different CSR policies and practices in the sample of firms in this study from the KLD database, created by the firm Kinder, Lydenberg, Domini. KLD provides a ranking of firms based on evaluation of a series of dimensions in the social arena. This study includes the social dimensions that play an essential role in establishing relationships with the different interest groups (McWilliams and Siegel, 2001). Based on the proposal of Waddock and Graves (1997), we used the following indicators: relationships with the local community, relationships with women and disadvantaged groups, relationships with employees, impact on the natural environment and socially responsible characteristics of the products that the organizations offer. KLD assigns each of the five indicators of CSR a point value ranging from +2 to −2, where +2 is evident strength, 0 a neutral position and −2 evident weakness. They then calculate the overall value of the variable “social performance of the firm” using the arithmetic means of the values of the five practices for each observation in the sample. A broad range of prior studies use KLD (Servaes and Tamayo, 2013; Waddock and Graves, 1997). KLD uses a great variety of sources to measure and evaluate the socially responsible behaviour of the firm. On the one hand, it includes annual information from the firm drawn from a questionnaire on the firm’s social responsibility practices. It also gathers information on annual accounts, quarterly reports and other reports related to the responsible initiatives that the firm develops. Further, KLD obtains information from external sources, such as articles from the economics and business press (Fortune, Business Week, Wall Street Journal, among others), surveys and internet. Based on this information, we believe that KLD’s measure does not evaluate firms starting from either stakeholders’ or managers’ perceptions but uses internal and external sources to increase objectivity.

International diversification

Prior studies conducted in the area of international business use the regional entropy index (REI) to measure the degree to which a firm operates in different markets (Hitt et al., 1997). The following equation defines this index:

REI j = ∑ i = 1 n P i , j × ( Ln1 P i , j )

where P_i,j refers to the percentage of sales of firm j in a region i, and Ln (1/P_i,j) represents the weight assigned to each region. The recent literature on international diversification widely accepts this measure from Hitt and colleagues (e.g. Yeoh, 2004).

Reputation of the MNC

To measure the MNC’s reputation, we used the “Fortune Corporate Reputation Index”. We obtained the data on reputation from the survey Fortune “American’s Most Admired Corporations”, based on the responses of approximately 10,000 executives, directors and financial analysts who evaluate firms in their sector in eight dimensions – use of assets, financial solvency, respect for environment and community, people development, degree of innovation, value of investment, management quality and product quality. A broad range of previous studies use Fortune’s measure of reputation (e.g. Philippe and Durand, 2011).

In addition, we incorporated three control variables – activity sector, firm size and financial performance.

Activity sector

To consider the possible effect of industry type on the sample of firms, we incorporated two dichotomous variables for two of the three activity sectors – industrial machinery and chemical.

Size

Size can positively influence reputation (Deephouse and Carter, 2005), visibility and relationship to the environment (Deephouse, 1996). We use the figure for total sales (operating revenue) from each MNC, including all of its business units.

Financial performance of the MNC

Implementing initiatives on social issues can affect the firm’s reputation not only directly but also indirectly; good reputation can drive good financial performance (e.g. Berman et al., 1999; Berrone et al., 2007). We use the ratio of profitability over total assets (Bansal, 2005).

Table II shows the variables used in the empirical analysis.

4. Results

We performed static analysis of the panel data. This analysis takes into account unobservable heterogeneity, determining whether to include fixed or random effects in the model. Whereas the fixed effects estimator assumes that unobservable individual effects are fixed parameters to calculate using correlations with regressors, the random effects model considers the firms chosen to constitute a representative sample, incorporating the unobservable individual effects as stochastic and not correlated to the regressors included in the error term (Hausman, 1978).

We performed the Hausman test to determine whether to apply fixed or random effects. The null hypothesis supports the conclusion that there is no difference between the fixed and random estimators. Cases that reject the null hypothesis thus use fixed effects (Hausman, 1978). Based on this logic and our rejection of the null hypothesis, we chose the fixed effects model.

Table III presents the descriptive statistics and correlations among all variables used in the analysis.

Finally, Table IV shows the regression analysis using the fixed effects model. The resulting variance inflation factors (VIF) below 5 indicate that there are no problems of multicollinearity among the variables used. We standardized the values of the variables to facilitate the analysis (Hair et al., 2009). The model shows good fit, which the R²within value and the F-statistic support.

As we observe in Table IV, international diversification has a positive and significant effect on the firm’s reputation in Model 1(b=0.31, SE=0.07; p<0.01). This direct effect does not, however, consider one of the key aspects analyzed in this study: the mediating role of social performance.

Model 2 shows that greater geographic international diversification improves the firm’s social performance, fulfilling H1 (b=0.31, SE=0.07; p<0.001).

Model 3 shows that the firm’s social performance has a positive effect on improving reputation levels, likewise confirming H2 (b=0.42, SE=0.09; p<0.001).

Finally, H3 predicts that social performance mediates the relationship between geographic international diversification and firm reputation. Model 4 confirms this mediating effect through the PROCESS macro, installed in the programme SPSS (Hayes, 2013). We can thus determine the direct effect (geographic international diversification – reputation) and the mediating effect of our model (geographic international diversification – social performance – reputation) independently. The results support mediation if the indirect effect is significant. To determine this, we use the Sobel test under conditions of normality. The direct effect is 0.10 (b=0.10, SE=0.10, p<0.055), and the indirect effect is significant, yielding the value 0.21 (b=0.21, SE=0.10, p<0.01). The total effect is thus 0.31 (see Model 1: b=0.31, SE=0.07; p<0.01).

As to the indirect effect, we observe that the parameters BootLLCI (0.05) and BootULCI (0.23) do not include the value 0 in this interval, indicating the significance of the other indicator that supports the indirect effect (Preacher and Hayes, 2004, 2008).

Based on these data, we can conclude that social performance mediates the relationship between geographic international diversification and reputation. In Model 4, however, the direct relationship between international diversification and reputation continues to be significant, although to a lesser degree (b=0.10, SE=0.10, p<0.055). Finally, social performance partially (not totally) mediates the direct relationship mentioned. Using the procedure developed by Frazier et al. (2004, p. 231), we see that the firm’s social performance mediates approximately 67.7 per cent of the total effect of international diversification on reputation.

Figure 1 provides a summary of the mediation model explained above.

5. Conclusions

Using stakeholder theory, this study analyses whether a high degree of international diversification in the MNC (in terms of presence in different countries and regions) encourages improvement in its social performance. The study also sheds light on the debate concerning CSR and MNCs’ reputation. Finally, we analyze the effect of international diversification on reputation, taking into account the mediating role that the firm’s social performance plays.

First, we find that a high degree of geographic international diversification enables the MNC to improve its social performance. Improvement in its visibility through contact with a heterogeneous group of stakeholders with diverse needs located in markets with different institutional profiles motivates such diversification (Aguilera-Caracuel et al., 2013). MNCs must take this fact into account when performing management practices that bring credibility and respect from stakeholders (Park et al., 2015). Second, the MNC will be able to decrease risks that derive from failure to comply with legislation and from conflicts with associations and firms in the sector (Deckop et al., 2006) and, ultimately, to improve its corporate image, a result that translates to the other organizational units (headquarters and subsidiaries) (McWilliams and Siegel, 2001).

Second, the results show that improvement in the MNC’s social performance positively affects its reputation. The literature shows that MNCs, in contrast to local firms, are more sensitive to socially responsible and irresponsible behaviour (Mahmood and Humphrey, 2013). MNCs sometimes receive accusations of opportunistic behaviour for locating their factories in countries with questionable respect for human rights and for exploiting the natural resources of the most disadvantaged countries and those with laxer regulations. Yet others consider MNCs as pioneers in establishing ethical and moral standards (Engle, 2007). The current study highlights the mutual benefits involved in implementing CSR activities, both for MNCs and for the communities in which they operate, since the MNCs can improve their reputation and public image, while the communities obtain support for sustainable development of their territory.

Third, we conclude that internationalization of the MNC, which we understand as the number of countries and regions in which the MNC has a presence, positively influences the MNC’s reputation, to a large extent through the mediating role that social performance exerts. In other words, the primary way for MNCs to gain in reputation is through local and/or global social initiatives in the different markets in which they operate and by interacting with the various stakeholders. This gain in reputation occurs primarily because the MNC can meet the expectations on social issues of numerous stakeholders from different contexts (local, national and global) and can thus both attend to the needs of a specific local community and create its own standards, which it can extend to different contexts (Jamali, 2010). While they can definitely encounter difficulty in initiating relationship with specific stakeholders (Strike et al., 2006), MNCs can minimize all of these efforts by managing the different relationships correctly, fostering the MNC’s communication capability (Hah and Freeman, 2014). The MNC’s internal organizational structure may be quite complex in many cases, especially if it has numerous organizational units located in markets with very different institutional profiles. The MNC can overcome this issue, however, if it is sufficiently flexible to adapt, when necessary, to specific local contexts and specific social demands. Finally, considering that MNCs increasingly receive more pressure from stakeholders to perform responsible behaviours, especially in the social and environmental spheres (Kang, 2013), a high degree of geographic international diversification establishes an ideal scenario to enable reflection of their social initiatives in significant increase in their reputation, in both local and transnational contexts. In sum, social performance constitutes the main vehicle by which highly diversified MNCs can improve their reputations in the different markets in which they operate.

Our paper has implications for management and public powers. Managers of MNCs should pay attention to capturing the different needs and expectations of the stakeholders in the different markets in which the organization operates. They should distinguish the needs of more specific local environments rather than more global needs that they can satisfy by creating standards internal to the MNC itself (Aguilera-Caracuel et al., 2013). They should also particularly stress the social and environmental spheres, as MNCs can play a significant role in this domain, perhaps becoming precursor-drivers of economic and sustainable development for the regions (Valente and Crane, 2010). Finally, they should both make an effort to ensure that all organizational levels and workers at the MNC develop communication capabilities that permit a fluid relationship with stakeholders and that can obtain a licence to operate in foreign markets (Rugman and Verbeke, 1998) to overcome institutional barriers (legal and cultural, among others).

Public and regulatory powers should be able to create certain common game rules for all organizations in the different countries, independently of their degree of economic, legal and cultural development (Aguilera-Caracuel et al., 2013). All of the foregoing is very useful when avoiding opportunistic behaviour from organizations, particularly international ones, in the social and natural environments. In addition to the more specific ongoing local needs in some individual communities, other communities tend increasingly to become standardized (e.g. relative to human and minority rights or minimal international environmental standards that protect against abusive practices), making it easier to satisfy these requirements at the global level. By way of example, we would highlight the United Nations Global Compact, based on ten international principles on issues of human rights, labour regulations, environment and anticorruption, to which over 2,500 firms adhere.

This study has a series of limitations. First, all MNCs in our sample have their headquarters in the USA. Since some of the directives from company headquarters influence some MNCs’ policies and management practices (Kostova et al., 2008), it would be interesting to incorporate MNCs that have sites in other countries to take this possible effect into account. Second, we measure social performance using the KLD database, widely used in the prior academic literature (McWilliams and Siegel, 2001). Future research could complement and contrast our these measures with questionnaires and surveys addressed to managers and workers in the MNCs. Research could also consider objective measures of CSR, such as those developed by the United Nations World Treaty based on four fundamental pillars – human rights, labour regulations, environment and fighting corruption. Third, we measure the firm’s reputation using the Fortune database, based on the perceptions that managers and other expert sector agents have of the firm’s visibility and importance in society. Future research can complement this measure with construction of an index that includes the number and type of news items that firms in our sample received in order to capture the repercussions of this news in society and to strengthen the indicators of reputation (Carter, 2006; Philippe and Durand, 2011).

Finally, performing this study opens a path to new and interesting lines of research with greater utility for the academic community, managers and businesspeople, as well as public institutions. First, it would be interesting to undertake studies using theories from social psychology as the reference framework. These theories have great capability to explain why specific stakeholders become involved in socially responsible initiatives, the motivational structures that foster socially responsible and irresponsible behaviour and analysis of the processes by which CSR shifts from being a set of organizational practices that develop as the result of pressure from interest groups to become a set of practices that represent values that the firms and their stakeholders share (Aguinis and Glavas, 2013; Rupp et al., 2013). Second, it could be important to take into account the role that MNCs perform in developing countries and regions (Hitt et al., 1997) and the differing impact of these roles on both local and global reputation (that of the entire MNC as a whole). Third, future studies could take into consideration in the context of MNCs the internal management of human resources in the different locations and their influence on improvement of the MNCS’ levels of social performance and reputation.

Figures

Figure 1

Mediator model

Table I

Summary of methodology

Methodology	Panel data: linear regression (fixed effects model)
Country of firm headquarters	USA
Activity sectors	3 sectors: chemical, energy and industrial machinery
Study sample	113 MNCs and 672 observations
Distribution of MNCS by sector	53 MNCs from the industrial machinery sector (279 observations) 40 MNCs from the chemical sector (260 observations) 20 MNCs from the energy sector (113 observations)
Period of analysis	2005-2010 (6 years)
Information sources	Financial and sales information by region: Standard & Poor’s (Capital IQ) database Information on corporate social responsibility (social performance): KLD database

Table II

Description of variables used

Variable name	Measures
Control variables
Industrial machinery sector	Dichotomous variable (0=does not belong to sector; 1=belongs to sector)
Chemical sector	Dichotomous variable (0=does not belong to sector; 1=belongs to sector)
Firm size	Sales volume (billing) of MNC
Financial performance	Profitability over total assets (Bansal, 2005)
Independent variables
Social performance of the MNC	Indicators obtained from the KLD database, considering criterion proposed by Waddock and Graves (1997):  Relations with the local community  Relations with women and disadvantaged groups  Relations with workers  Impact on the environment  Socially responsible characteristics of the products the firm offers
Geographic international diversification	Regional entropy index (Hitt et al., 1997)
Dependent variable
Reputation	Overall point value assigned by the Fortune database

Table III

Descriptive statistics and correlations

	Mean	SD	1	2	3	4	5	6
1. Social performance of MNC	0.11	0.5	1
2. Industrial machinery sector	0.43	0.50	−0.01	1
3. Chemical sector	0.41	0.49	0.01	−0.72***	1
4. Firm size	21.45	45.34	0.05	−0.04	−0.01	1
5. Financial performance	0.24	0.36	0.06*	−0.10**	0.10**	0.10**	1
6. Geographic international diversification	0.6	0.43	0.26***	0.03	0.03	0.02	0.15***	1
7. Reputation	0.29	0.39	0.32***	0.02	0.03	0.02	0.17**	0.35***

Notes: Number of observations (n)=672; number of groups (MNCs)=113. *p<0.055; **p<0.01; ***p<0.001

Table IV

Results of the fixed effects linear regression model

Dependent variables (horizontal)	Model 1 Reputation (direct effect)	Model 2 Social performance (H1)	Model 3 Reputation (H2)	Model 4 Reputation (H3)
Constant	1.57 (2.23)	1.24 (1.80)	1.11 (1.93)	1.21 (2.11)
Industrial machinery sector	2.46 (4.24)	1.21 (3.22)	2.13 (3.36)	2.23 (3.12)
Chemical sector	−16.17 (9.56)	−18.37 (9.67)	−12.11 (9.87)	−13.67 (9.97)
Firm size	−0.24 (0.30)	−0.14 (0.30)	−0.17 (0.28)	−0.14 (0.20)
Financial performance	0.09* (0.07)	0.08* (0.07)	0.09 (0.07)	0.09* (0.07)
Geographic international diversification (indep. var.)	0.31** (0.08)	0.42*** (0.07)		0.10* (0.08)
Social performance (indep. var.)			0.29*** (0.06)
Social performance (mediating var.)				0.21** (0.08)
Hausman Test	9.55*	10.05*	9.55*	9.55*
Number of observations	672	672	672	672
Number of groups (MNCs)	113	113	113	113
R2 within	0.18	0.27	0.19	0.24
F	10.71**	12.36**	10.85**	11.34**

Notes: The table includes coefficients of the regression model (estimators); Standard deviations are included in parentheses. *p<0.055; **p<0.01; ***p<0.001