Abstract
Studies have shown that corporate social performance (CSP) is an antecedent of corporate reputation, acting as a signal that affects stakeholders’ perceptions and expectations about a firm’s future behavior. However, the perceptions, expectations, and interests of stakeholders may be affected by external factors, such as national culture, which shapes their beliefs about what role companies play in society. Drawing on institutional theory and Hofstede’s cultural dimensions, we analyze how stakeholders’ national culture moderates the relationship between CSP and corporate reputation. The results of the analysis of an international sample for the period 2010 to 2016 show that low individualism (i.e., collectivism), low masculinity (i.e., femininity), low power distance, and low uncertainty avoidance intensify the positive relationship between CSP and corporate reputation.
Introduction
Culture acts as a pattern in the way people perceive, relate to, and interpret information signals that influence both individual as well as group behavior (Goodenough, 1994). National culture affects both managers’ decisions and their actions (Thanetsunthorn, 2015) as well as the way in which different stakeholders perceive such actions (Maignan, 2001). Indeed, what role companies play in society is perceived and evaluated differently in different national contexts (Deephouse et al., 2016). The literature also shows that corporate social performance (CSP) has major consequences for a firm’s reputation (Brammer & Pavelin, 2006; Lai et al., 2010; Pérez-Cornejo et al., 2020; Rothenhoefer, 2019; Stanaland et al., 2011; Surroca et al., 2010). However, firms’ behaviors may be perceived differently and may have a different impact on corporate reputation in different national cultures. In this study, we extend and link these two lines of research. We posit that culture also plays an important role in moderating the effect of firms’ actions on stakeholders’ perceptions and expectations. Specifically, we analyze how national culture moderates the relationship between CSP and corporate reputation.
Corporate reputation has been acknowledged as one of firm’s most important assets (Hall, 1992) since it offers a source of sustainable advantage (Bergh et al., 2010; Roberts & Dowling, 2002). Evidence concerning how corporate reputation impacts company value (Fombrun & Shanley, 1990; Roberts & Dowling, 2002) has elicited growing interest in the antecedents of corporate reputation. Among the determinants of corporate reputation, CSP has been one of the most widely studied in prior research because it provides a platform to manage corporate reputation. CSP is a legitimating instrument (de Quevedo-Puente et al., 2007) that signals (Brammer & Pavelin, 2006; Fombrun & Shanley, 1990) a firm’s commitment to each and every stakeholder. Corporate reputation is the result of a legitimation process (Rao, 1994) in which different audiences observe the firm and build expectations about its most likely behaviors. However, the legitimation process may not be the same everywhere because people’s perceptions, expectations, and interests may differ across cultural contexts (Deephouse et al., 2016; Hofstede, 2001; Liu & Almor, 2016; Ren & Gray, 2009). Therefore, the influence of CSP on corporate reputation need not be the same in different national contexts.
Each national context is shaped by different institutional factors that influence the way in which companies act and in which stakeholders perceive such company behaviors (Ali et al., 2015; Brammer & Jackson, 2012; Deephouse et al., 2016; Gardberg & Fombrun, 2006; Hofstede, 2001; Katz et al., 2001; Maignan & Ralston, 2002; Scott, 1987, 1995) Culture is an important contextual factor (Breuer et al., 2018; Dikova et al., 2010; Hofstede et al., 2002; North, 1991) that provides standards for acting among those who are in the same geographical area (Schiffman et al., 2008; Triandis, 1996; West et al., 2016). In fact, prior research has already found evidence regarding the role of national culture as a factor that affects the design of companies’ social responsible policies and their implementation; in other words, their CSP (Aguilera et al., 2007; Halkos & Skouloudis, 2016; Ioannou & Serafeim, 2012; Ringov & Zollo, 2007; Waldman et al., 2006). Moreover, cultural values and norms also exert a powerful influence on individuals’ perceptions (Deephouse et al., 2016; Hofstede, 2001; Ren & Gray, 2009; West et al., 2016). Indeed, prior empirical research has shown that cultural characteristics affect stakeholders’ expectations about companies and their behaviors (Ali et al., 2015; Deephouse et al., 2016; Katz et al., 2001; Maignan, 2001). In this research, we analyze how national culture may influence the criteria stakeholders use to evaluate CSP and thus develop their perceptions and expectations about companies.
By adopting this focus, our study moves away from the predominant research approach, wherein the institutional context is considered to be an antecedent of CSP (Aguilera et al., 2007; Halkos & Skouloudis, 2016; Ioannou & Serafeim, 2012; Ringov & Zollo, 2007; Waldman et al., 2006). Instead, we focus on analyzing and studying how culture influences the way in which stakeholders from different cultural backgrounds perceive CSP. In other words, we analyze the moderating effect of national culture on the relationship between CSP and corporate reputation.
We test our hypotheses using a sample of multinational corporations included in Reputation Institute’s corporate reputation rankings. These enterprises are large enough and visible enough for their social behavior to be known by audiences in every country where they operate. This characteristic of the sample allows us to isolate the effect of culture so that we can analyze how the overall CSP of these global companies is evaluated by different stakeholders from different cultural backgrounds. We draw on Hofstede’s (1980) four original cultural dimensions (individualism-collectivism, power distance, masculinity-femininity, and uncertainty avoidance), using the cultural scores of Taras et al. (2012). These scores update those obtained by Hofstede in 1980 that have been criticized as out of date by prior literature arguing cultural changes along the time (Inglehart & Baker, 2000; Ralston et al., 1997). Addressing this issue, Taras et al. (2012) updated cultural scores with a meta-analysis of existing studies, which have employed Hofstede’s method to measure the four original dimensions. These scores have also been employed by prior research (Deephouse et al., 2016; Hauff et al., 2015; Steel et al., 2018).
Our article contributes theoretically and empirically to prior research in different fields. First, we contribute to CSP literature by providing further evidence concerning the role of CSP as an antecedent of corporate reputation from a multi-country approach. We also use a multilevel methodological approach, which allows us to control variables that have an effect at organizational or contextual levels. In addition, we explore how this relation between CSP and corporate reputation is not homogeneous across countries but rather depends on national cultural features. Second, from an institutional theoretical approach, this study suggests that culture is an institutional factor that affects stakeholders’ perceptions about company actions. These contributions may help practitioners to improve their management of corporate reputation as well as other intangible perception-based assets. Since companies increasingly operate in a more global environment, this article can also help managers understand the effect of their CSP proposals and thus strengthen their corporate reputation across multiple countries.
The remainder of this article is structured as follows. In the second section, we define the concepts of corporate reputation and CSP from the stakeholder perspective, discuss their relationships, and propose our first hypothesis. We then draw on Hofstede’s (1980) cultural framework to discuss how culture affects the relationship between CSP and corporate reputation to argue and propose the rest of our hypotheses. The fourth and fifth sections are devoted to the method and results, respectively. The article closes with the main conclusions and a discussion, also outlining the study’s contributions, limitations, future lines of research, and managerial implications.
Theoretical background and hypotheses
CSP and corporate reputation
CSP reflects a company’s proposal for value distribution among stakeholders through a series of principles, programs, and actions (Clarkson, 1995). These stakeholders interpret this value distribution as a signal (Fombrun & Shanley, 1990; Spence, 1974; Wood, 1991) of the degree to which the company will continue to meet their interests through its economic, social, and environmental actions (Barnett, 2007; Tang et al., 2015). Therefore, high CSP favors firm legitimacy and strengthens corporate reputation (Bansal & Clelland, 2004; Fombrun, 1996; Gardberg & Fombrun, 2006; Martín-de Castro et al., 2020).
There are many definitions of corporate reputation in prior literature (Lange et al., 2011). We adopt the definition of corporate reputation as being known for something by stakeholders (Fombrun, 2002; Lange et al., 2011; Wartick, 1992). This definition conceptualizes corporate reputation as the aggregation of the expectations of different stakeholders regarding a firm’s capacity to satisfy their interests (Fombrun, 2002; Wartick, 1992). Corporate reputation is thus the product of a legitimation process (Rao, 1994) in which different audiences observe a firm’s characteristics and past performance so as to then build expectations about likely future behavior (Breitinger & Bonardi, 2019).
High CSP may generate a legitimation process (Bansal & Clelland, 2004; Fombrun, 1996; Gardberg & Fombrun, 2006; Martín-de Castro et al., 2020; Rao, 1994) that results in a good corporate reputation (Logsdon & Wood, 2002). A strong CSP over time positively affects corporate reputation (de Quevedo-Puente et al., 2007; Logsdon & Wood, 2002) because a company’s stakeholders translate their perceptions of CSP into expectations about the company’s ability to meet their future interests. Conversely, if a company’s CSP does not meet the expectations generated by its corporate reputation, the company will lose the reputational capital that it has gradually accumulated over a long period of consistent CSP (de Quevedo-Puente et al., 2007). Although many studies have evidenced this relationship (Brammer & Pavelin, 2006; Lai et al., 2010; Pérez-Cornejo et al., 2020; Rothenhoefer, 2019; Stanaland et al., 2011; Surroca et al., 2010), we propose the following hypothesis for two reasons: first, it is the base for developing the hypotheses posited in the next section; second, we analyze this relation from a multi-country perspective, extending the bulk of the research, which has tended to test this effect from a single-country approach (Deephouse et al., 2016; Pérez-Cornejo et al., 2020).
H1: CSP is positively related to corporate reputation.
Culture as a moderator of the relationship between CSP and corporate reputation
Drawing on institutional theory, the literature has examined the role of institutional factors on business phenomena (Cabeza-García et al., 2019; Díez-Esteban et al., 2019; Fuentelsaz et al., 2019). Specifically, institutional factors have been found to affect stakeholders’ values and beliefs (Brammer & Jackson, 2012; Deephouse et al., 2016; Hofstede, 2001; Scott, 1987, 1995) and, therefore, their cognition (DiMaggio, 1997; North, 1991; Redding, 2005; Schutz & Luckmann, 1973; Scott, 1987, 1995) Culture is an institutional factor that has been closely linked to cognition (DiMaggio, 1997; Redding, 2005; Schutz & Luckmann, 1973). Culture consists of shared elements (knowledge, values, norms, beliefs, etc.) that provide standards for perceiving, interpreting, believing, assessing, communicating, and acting among those who share a language, history, and geographic location (Triandis, 1996). Goodenough (1994) argues that culture acts as a pattern in the way people perceive, relate to, and interpret information signals that affect individual and group behavior. Similarly, North (1991) argues that culture provides a conceptual framework based on language to encrypt and interpret information signals that the senses relay to the brain. Research has linked national cultural values in managers’ decision processes to corporate social engagement (Thanetsunthorn, 2015). Ample research has already provided evidence of the role played by national culture in pressuring companies to engage in CSP (Aguilera et al., 2007; Halkos & Skouloudis, 2016; Ioannou & Serafeim, 2012; Ringov & Zollo, 2007; Waldman et al., 2006).
However, national culture not only affects decision-making by managers but also influences the way audiences or stakeholders perceive and evaluate a firm’s behaviors, such that some are accepted while others are rejected by the national community (Deephouse et al., 2016; Gardberg & Fombrun, 2006; Ren & Gray, 2009). Stakeholders with different cultural backgrounds may, therefore, have different expectations and evaluations about how a company should behave toward its stakeholders. For instance, Katz et al. (2001) affirmed that cultural factors strongly influence national CSP expectations. Maignan (2001) also showed that customers from the United States, France, and Germany assigned different importance to a company’s social initiatives, depending on the country’s level of individualism. Similarly, De Mooij and Hofstede (2010) argued that cultural values define consumers’ personalities. In relation to this research, Deephouse et al. (2016) analyzed the direct impact of culture on corporate reputation and showed how the role of companies is perceived and appreciated differently in different cultures because national values influence individuals’ attitudes toward corporations. However, national culture influences more than just individuals’ expectations, perceptions, and judgment (DiMaggio, 1997; Hofstede, 2001; Liu & Almor, 2016; Triandis, 1996) of companies as institutions in society; culture also affects individuals’ expectations, perceptions, and evaluations of each and every social action undertaken by a given firm, rewarding each company with a different reputation in each national culture. El Ghoul et al. (2017) found support for the effect of the interaction between culture and CSP on firm value. Extending this line of research, we aim to analyze how national culture acts as a lens through which stakeholders evaluate CSP.
We draw on Hofstede’s cultural framework introduced in 1980, to analyze how national culture moderates the influence of CSP on corporate reputation. Specifically, we employ the updated scores that Taras et al. (2012) have developed based on the four original Hofstede’s (1980) cultural dimensions of power distance, individualism versus collectivism, uncertainty avoidance, and masculinity. Long- and short-term orientation and indulgence were added later to the original four Hofstede dimensions (Hofstede, 1980), but these dimensions have been less popular in cross-cultural research (Steel et al., 2018) and the four original dimensions are the most prominent in cross-cultural research (Hauff et al., 2015; Taras et al., 2010, 2012; Tung & Verbeke, 2010). Indeed, when Taras et al. (2012) conducted their research, not enough studies about the two new dimensions had been generated (Steel et al., 2018); therefore, Taras et al. (2012) do not provide updated information for these two dimensions.
Hofstede (2001) defines power distance as the degree to which a country accepts social inequalities. In countries with a low power distance, people are much less focused on class differences and social aspirations (Roth, 1995). One of the main values of such cultures is equality, and there is a struggle to minimize inequalities through the redistribution of power (Hofstede, 2001). In contrast, in high power distance contexts, importance is attached to prestige and wealth because they shape social hierarchies (Hofstede, 1984). Individuals in high power distance cultures consider that people have a rightful place in a social hierarchy, and seek their own satisfaction through power inequalities, which are both expected and desired (Hofstede, 2001). As argued earlier, socially responsible companies strive to satisfy every stakeholder’s expectations through a balanced distribution of value among all stakeholders. This democratic view of value distribution may be more valued in cultures with a low power distance than in those with a high power distance. Therefore, we expect CSP to exert a stronger positive impact on corporate reputation in low power distance countries than in high power distance countries. Accordingly, we propose the following hypothesis:
H2a: Power distance negatively moderates the influence of CSP on corporate reputation. In low power distance countries, the influence of CSP on corporate reputation is higher than in high power distance countries.
Individualism-collectivism refers to the degree to which individuals are integrated into groups in society (Hofstede, 2001). In countries with a high degree of collectivism, there is a strong sense of belonging to the society. Individuals trust and are loyal to the group (Hofstede, 2001), and people value goals that favor society as a whole (Deephouse et al., 2016). In such countries, all stakeholders may give more positive evaluations of companies that are responsible to all their stakeholders because they feel they are part of the same community and expect the company to seek a balance in its treatment of different stakeholders. In contrast, in countries with high levels of individualism, people look out for their own and their closest relatives’ interests and satisfaction, without considering the interests of the rest of society (Hofstede, 2001). In collectivist countries, stakeholders not only appreciate when their own interests are met, but also value companies’ attempts to meet the interests of other stakeholders. By contrast, in individualistic contexts, each stakeholder may only care whether their own interests are met. Because high CSP implies that the company considers the interests of all stakeholders, CSP would fit better with collectivist cultures and may have a greater impact on corporate reputation in such cultures. Consistent with these arguments, Maignan (2001) showed that more individualistic consumers attach less importance to corporate social behaviors. Therefore, we expect CSP to have a greater impact on corporate reputation in more collectivist countries. Accordingly, we propose the following hypothesis:
H2b: Individualism negatively moderates the influence of CSP on corporate reputation. In culturally collectivist countries, the influence of CSP on corporate reputation is greater than in culturally individualistic countries.
Masculine societies are characterized by values based on success, recognition, earnings, assertiveness, aggressiveness, and achievement (Hofstede, 2001; Newburry & Yakova, 2006). According to Hofstede (2001), masculine countries favor economic growth over environmental protection. In contrast, feminine societies value looking after others, quality of life, and relationships (Hofstede, 2001). Steensma et al. (2000) suggest that more masculine societies have a lower appreciation of cooperation strategies than more feminine societies. High CSP implies a balanced distribution of value among stakeholders. Therefore, we expect that in culturally feminine countries, where relationships among agents are important, stakeholders will appreciate CSP more than in culturally masculine countries. We also expect the impact of CSP on corporate reputation to be greater in feminine countries than in masculine countries. Therefore, we propose the following hypothesis:
H2c: Cultural masculinity negatively moderates the influence of CSP on corporate reputation. In culturally feminine countries, the influence of CSP on corporate reputation is greater than in culturally masculine countries.
Hofstede defines the cultural dimension of uncertainty avoidance as “the extent to which the members of a culture feel threatened by uncertain or unknown situations” (Hofstede, 2001, p. 161). Corporate social activities strengthen a firm’s relationships with stakeholders and enhance their trust in the company (Barnett, 2007; Tang et al., 2015). These activities therefore reduce uncertainty between the company and its stakeholders. Furthermore, CSP is an important element for achieving sustainable development, implying that companies with high levels of CSP behave without compromising future generations. Blodgett et al. (2001) found that uncertainty avoidance cultures heighten ethical sensitivity toward various stakeholders. Therefore, the values of high uncertainty avoidance countries are better aligned with CSP. Accordingly, we expect CSP to have a greater impact on corporate reputation in high uncertainty avoidance contexts than in low uncertainty avoidance contexts. We propose the following hypothesis:
H2d: Uncertainty avoidance positively moderates the influence of CSP on corporate reputation. In high uncertainty avoidance countries, the influence of CSP on corporate reputation is higher than in low uncertainty avoidance countries.
Method
Sample
Our analysis was based on 2,554 observations for 16 countries (Argentina, Germany, Belgium, Brazil, Canada, Chile, China, Spain, Netherlands, Mexico, Peru, South Africa, Sweden, Switzerland, the United Kingdom, and the United States) covering the period 2010–2016. 1 The sample covered all companies with data in the two databases used to measure the dependent and independent variables in this study (RepTrak® and Thomson Reuters Eikon™) and which operate in countries included in the cultural scores provided by Taras et al. (2012). 2
The final sample is particularly well-suited to testing our hypotheses because it consisted of observations for 681 multinational corporations from multiple sectors (basic materials, consumer cyclicals, consumer non-cyclicals, energy, financials, healthcare, industrials, technology, telecommunication services, and utilities). These companies are also large and visible enough to be known by the stakeholders in every country they operate in. This characteristic of these companies allows us to isolate the effect of culture by analyzing how the overall CSP of these global companies (based on data from Thomson Reuters Eikon™) is assessed by different stakeholders in different cultural contexts.
Variables
Corporate reputation
We obtained data on corporate reputation from all RepTrak Pulse reports for the sampled countries until December 2016. These reports were available on Reputation Institute’s website (www.reputationinstitute.com). This index provides scores for the most reputable companies across 55 countries. Several studies have validated RepTrak Pulse methods (Fombrun et al., 2015; Ponzi et al., 2011; Sarstedt et al., 2013; Sánchez-Torné et al., 2020), and its scores have been employed in previous research (Deephouse et al., 2016; Fombrun & Pan, 2006; Pérez-Cornejo et al., 2020). Using online surveys each year, Reputation Institute measures the reputation of thousands of the world’s most prestigious companies across seven key rational dimensions of reputation: products and services, innovation, workplace, governance, citizenship, leadership, and performance. This corporate reputation measure is based on four elements: admiration and respect, recognized reputation, good impression, and trust. Respondents come from a random sample of the population, stratified by age and gender for each year and country. The final corporate reputation score averages responses from 100 or more people who are familiar with the company. Therefore, this rater panel composition avoids the financial halo effect 3 (Ponzi et al., 2011). Unlike studies of Brown and Perry (1994) and Roberts and Dowling (2002), in this research, correlation between financial performance and corporate reputation is low. This suggests that there is no financial halo in our data. Furthermore, we followed the methodology by Roberts and Dowling (2002) and we did not detect financial halo problem in our measure of corporate reputation. Finally, our finding is consistent with prior research that points out this feature as a strength of Reptrack Pulse (Ponzi et al., 2011). To facilitate international, inter-industry comparisons, Reputation Institute publishes statistically adjusted data and the overall score for each firm’s reputation is on a scale of up to 100 points.
CSP measure
CSP data were obtained from Thomson Reuters Eikon™, which covers over 6,000 public companies from several countries across more than 400 environmental, social, and governance (ESG) metrics. These scores have been employed in recent research (Duque-Grisales & Aguilera-Caracuel, 2019; Pérez-Cornejo et al., 2020), and they enhance and replace the ASSET4 ratings widely used in previous research (Luo et al., 2015; Maniora, 2017; Nardella et al., 2019). More than 150 analysts process the measures manually for each company to standardize the information and ensure that it is comparable across the entire range of companies. These scores are based on CSP worldwide—that is, in all the contexts where the company operates. Our CSP measure has two pillars: social and environmental. The social performance score combines four indicators provided by Thomson Reuters Eikon™ ESG scores (workforce, human rights, community, and product responsibility). Calculation of the environmental performance score is based on three categories (environmental resource use, emissions, and innovation). The final measure for CSP is the average of the two performance measures. It ranges from 0 to 1, where values close to 0 mean a low CSP and values close to 1 a high CSP.
Culture
Management research has drawn on several cultural frameworks, including those of Hofstede (1980, 2001), Schwartz (1994, 2006), GLOBE (House et al., 2004), Trompenaars (1993), and the World Values Survey (Inglehart, 1990, 1997) However, Hofstede’s framework is the most cited in social sciences (Beugelsdijk et al., 2015; Lewellyn & Muller-Kahle, 2019). Although Hofstede’s framework covers the largest number of countries, several authors have criticized Hofstede’s scores on the grounds that cultures have changed (Inglehart & Baker, 2000; Ralston et al., 1997). For this reason, Taras et al. (2012) developed measures from a meta-analysis using Hofstede’s dimensions of power distance, individualism, masculinity, and uncertainty avoidance. We, therefore, tested our analysis with scores provided by Taras et al. (2012). This study offers an updated set of national cultural scores for Hofstede’s cultural dimensions provided in 1980 that have been employed in prior research (Deephouse et al., 2016; Hauff et al., 2015; Taras et al., 2010). In particular, the scores we use in our research are based on a meta-analysis of studies conducted between 2000 and 2010, and “that have reported cultural values of their participants measured using models and methodology comparable with those devised by Hofstede” (Taras et al., 2012, p. 331). The cultural scores of our database are constant for each country for the period analyzed. This approach is common in studies with this type of methodology and based on cultural indices (Deephouse et al., 2016) since, although not static, cultural values are felt to evolve slowly over time, unlike other types of variables at the macro level (Beugelsdijk et al., 2015).
Following Aiken and West (1991), we calculated the moderating effects between CSP and culture by multiplying the standardized values of power distance, individualism, masculinity, and uncertainty avoidance by the standardized value of CSP to minimize the effects of multicollinearity.
Control variables
We used eight control variables. First, company size, company age, return on equity (ROE), and leverage as control variables at the company level. We also included three contextual variables, national CSP, institutional development, and industry. Finally, we also controlled by year. We measured firm size as the natural logarithm of total assets. Ample empirical evidence suggests that larger firms have better corporate reputations (Cordeiro & Sambharya, 1997; Deephouse, 1997). As they are more visible in markets (Walker et al., 2018), larger firms are expected to come under closer scrutiny from different audiences and are, therefore, expected to exhibit a balanced value distribution among their stakeholders, evidencing highly responsible behavior in an effort to build and sustain a good corporate reputation. Smaller companies, which may go unnoticed in the market, are expected to be less scrutinized and, therefore, less careful in the distribution of firm value, thus forfeiting their corporate reputation. We also included ROE as a control variable because research has frequently analyzed the influence of returns on corporate reputation (Dunbar & Schwalbach, 2000; Inglis et al., 2006; Rose & Thomsen, 2004). Stakeholders use previous firm performance to build future expectations about a firm because its ability to satisfy stakeholders’ future demands is greater when the value it creates is high. We included leverage, measured as the debt-to-equity ratio, because high leverage may threaten future returns and, thus, corporate reputation (Delgado-García et al., 2013; Pérez-Cornejo et al., 2020; Walker et al., 2018). Although empirical studies provide ambiguous findings regarding the influence of firm age on corporate reputation (Rao, 1994; Schultz et al., 2001), we controlled for age (measured as the difference between the year of analysis and the year the company was founded) because companies that have stayed in business through long periods of market supervision can be expected to have maintained stakeholders’ satisfaction. Therefore, stakeholders should extrapolate previous behaviors to generate expectations of future behaviors (von Weizsacker, 1980). We also included context control variables because expectations about companies may be affected by the context in which they operate. We also included national CSP as a control variable. National CSP was calculated as the average CSP of the companies with corporate reputation scores in each country in our sample. We used the variable institutional development because standards of living may affect how stakeholders expect a firm to behave toward society, depending on the national context (Deephouse et al., 2016). We measured this using the Global Competitiveness Index (GCI) developed by the World Economic Forum and employed in prior research (Chung & Beamish, 2012; Demirbag et al., 2009; Fang et al., 2013; Shaner & Maznevski, 2011). Finally, we used dummy variables to control for industry and year. We obtained data on company age from an Internet search. Company size, ROE, leverage, and industry data were gathered from the Thomson Reuters Eikon™ dataset.
Analysis method
We used a multilevel regression approach to test our hypotheses (Hofmann, 1997). This analytical method enables simultaneous assessment of the effect of multilevel data such as organizational information and contextual factors (Koos, 2012; Snijders & Bosker, 1999). This approach addresses and corrects issues of biases in parameter estimation at different levels of analysis by apportioning the variance to the organizational level or the contextual level (Guo & Zhao, 2000). Organizational characteristics and performance lie on the first level. To avoid endogeneity problems, independent and control variables at this level are measured in period t−1. At the second level, we have the four cultural variables: power distance, individualism-collectivism, masculinity, and uncertainty avoidance. Given this nested multilevel data structure, we employed hierarchical linear modeling (Harrison & Raudenbush, 2006; Hofmann, 1997).
This statistical approach proceeded in two steps. First, models with no independent variables (“empty models”) were estimated to assess how much of the total variance can be attributed to the contextual level. Next, organizational and contextual variables were included in the linear random intercept models. These models enabled an assessment of which variables (at both levels) affect the dependent variable and how much of the variation in the dependent variables between countries can be explained by organizational characteristics and country culture factors.
Results
Table 1 presents the correlation matrix and descriptive statistics for the sample. The values of the variance inflation factor range between 1.02 and 5.23, and also show no multicollinearity problems.
Descriptive statistics and correlation matrix.
CSP: corporate social performance; GCI: Global Competitiveness Index; ROE: return on equity.*p = .10; **p = .05; ***p = .01.
Table 2 summarizes the cultural characteristics of the sample. The sample of countries used can be seen to cover a wide diversity of cultural profiles. While the Netherlands presents the lowest power distance profile, Mexico has the highest rate of power distance. We also see that Mexico has the lowest rate of individualism as opposed to Sweden, which exhibits the most individualistic profile in the sample. Regarding masculinity, our sample covers countries such as Sweden and Belgium, which have the highest feminine profiles, in contrast to the United Kingdom and the United States, which exhibits the highest masculine profiles. Finally, our sample also contains national contexts that display high levels of uncertainty avoidance, such as Spain, which is the country with the highest rate. In contrast, Sweden shows the lowest uncertainty avoidance cultural profile.
Culture profile of the sample based on the scores by Taras et al. (2012).
CSP: corporate social performance; GCI: Global Competitiveness Index; PD: power distance; IND: individualism: MAS: masculinity; UA: uncertainty avoidance.
To test our hypotheses, we employed the multilevel regression approach. A Durbin–Wu–Hausman test (Davidson & MacKinnon, 1993) shows an absence of endogeneity problems. This multilevel regression approach was used to explain the integration of companies in different national cultural contexts. First, we evaluated how much of the total variance can be attributed to the contextual level (country). For this step, an empty model, with no independent variables, was estimated. Table 3 reports the results for the multilevel modeling. The results for the empty model reveal there is a significant amount of unexplained (residual) variance at the contextual level (r2 = 53.82; p < .01). The empty model residual variance was used to evaluate how much of the unexplained variance at the contextual level is reduced when variables are included in the models. The estimation of the residual intraclass correlation shows that roughly 20% of the variance is attributable to the contextual level. While this is only a rough variance estimate, it justifies the use of this multilevel regression approach (Snijders & Bosker, 1999).
Results from linear multilevel regression analysis.
CSP: corporate social performance; ROE: return on equity; GCI: Global Competitiveness Index.
Number of countries: 16.
Number of observations: 2,554.
p = .10; **p = .05; ***p = .01.
Model 1, in Table 3, includes variables related to the contextual level of the company. Including these variables in the model reduced the unexplained intercept variance by 28% (r2 = 39.00; p < .01). When companies’ performance and characteristics were included in Model 2, the unexplained intercept variance decreased by 29% (r2 = 37.32; p < .01) compared to the empty model. In Model 3, the interaction effects between the cultural dimensions and CSP were introduced (r2 = 37.02; p < .01). Models 2 and 3 confirm the significant effect of CSP on corporate reputation, thereby supporting our first hypothesis (H1). The results for Model 3 show a negative and significant effect of the interaction between CSP and power distance (p < .01), thereby supporting Hypothesis 2a (the lower the power distance of the country, the higher the impact of CSP on corporate reputation). Model 3 also confirms that the coefficient of the interaction between CSP and individualism is significant and negative (p < .01). This result supports Hypothesis 2b (the lower the level of individualism in the country, the higher the impact of CSP on corporate reputation). Model 3 also shows a negative effect of the interaction between masculinity and CSP (p < .10), thereby supporting Hypothesis 2c (the lower the level of masculinity in the country, the higher the impact of CSP on corporate reputation). Finally, Model 3 shows a negative and significant effect of the interaction between the cultural dimension of uncertainty avoidance and CSP. This effect is not consistent with Hypothesis 2d. This finding suggests that the lower the level of uncertainty avoidance in the country, the higher the impact of CSP on corporate reputation will be. Therefore, we cannot confirm Hypothesis 2d.
We also analyzed the direct effects of national cultural dimensions on corporate reputation. The results reveal a negative and significant effect of power distance and individualism on corporate reputation. However, the results do not show a significant effect of masculinity and uncertainty avoidance on corporate reputation. The results for the control variables reveal a negative and significant effect of company leverage on corporate reputation, a significant and positive effect of company age on corporate reputation, and a significant effect of the average CSP level on corporate reputation in a given country. Furthermore, the level of institutional development measured by the GCI has a significant and positive effect on corporate reputation.
Discussion and conclusions
This research aims to analyze how national culture affects stakeholders’ perceptions about companies’ actions. Based on an analysis of an international sample of 2,554 firm-observations over a 7-year period (2010–2016), our results confirm that national culture moderates the relationship of CSP on corporate reputation. Specifically, low power distance, collectivism, femininity, and low uncertainty avoidance intensify the relationship between CSP and corporate reputation. For instance, Germany fits in more with these national features of culture compared to the rest of the countries. In fact, if we look into our sample, the average CSP level of firms in Germany is lower in comparison to Argentina or China, although the average level of company reputation in Germany is higher than in those countries. These results support prior research that has highlighted the role of national culture on cognition (DiMaggio, 1997; Hofstede, 2001; Triandis, 1996), confirming the conclusions by Katz et al. (2001) concerning the importance of national culture values in terms of shaping national expectations of CSP.
Regarding each of the cultural dimensions, our results first suggest that CSP is better aligned with values of low power distance than high power distance because CSP promotes a balanced value distribution among all stakeholders that is aligned with the values of equality found in low power distance cultures. Conversely, CSP is less well-aligned with high power distance cultures, in which people accept the privileges of a small group and the inequalities between individuals (Hofstede, 2001). This finding is consistent with the study by G. Williams and Zinkin (2008), which supports the notion that countries with a lower power distance profile have a greater propensity to punish irresponsible company behaviors than those with high power distance profiles.
Second, with regard to individualism, our results also suggest that CSP is better aligned with values of collectivism. In collectivist cultures, individuals have a strong feeling of taking part in society (Hofstede, 2001) and value goals that favor society as a whole (Deephouse et al., 2016). Therefore, stakeholders may demand that companies act more responsibly toward all stakeholders. Accordingly, they may be more appreciative of a high CSP than in individualistic countries, where people focus on their own and their closest relatives’ interests (Hofstede, 2001). This finding is in line with prior research, such as Maignan’s (2001) findings which reveal that costumers from less individualistic countries are more concerned about businesses’ social responsiveness, or Hur and Kim (2017), who find that collectivism is positively related to consumers’ CSP perceptions.
Third, masculinity negatively moderates the relationship between CSP and corporate reputation. Our results suggest that because the core values of feminine societies, such as cooperation, taking care of others, preserving the environment, and human welfare (Hofstede, 2001), are better aligned with the essence of CSP, firms’ social actions have a stronger impact on corporate reputation in countries that display a more feminine profile. This result is consistent with Hur and Kim (2017), who confirm that feminine values are positively related with consumers’ CSP perceptions.
Finally, we also find that uncertainty avoidance negatively moderates the relationship between CSP and corporate reputation. This result is not consistent with our hypothesis, and may be explained by the fact that the values of low uncertainty avoidance cultures may be better aligned with CSP, with such cultures being open to change (Hofstede, 2001). CSP involves a shift in the role of companies in society (O. F. Williams, 2014), increasing the chances for innovation. Therefore, although having a high level of CSP may be a way of reducing uncertainty because it develops stakeholders’ trust in the company (Barnett, 2007; Tang et al., 2015), it also poses a challenge by presenting a source of change. These results are in line with G. Williams and Zinkin (2008), who indicate that customers from low uncertainty avoidance backgrounds are more willing to punish irresponsible CSP.
Furthermore, our results show a positive effect of CSP on corporate reputation, confirming that CSP is a tool to satisfy the interests of different stakeholder groups and can lead to positive expectations about a firm. These findings provide evidence from a multi-country perspective, and support the results of several studies that mainly focus on a single-country analysis (Brammer & Pavelin, 2006; Lai et al., 2010; Rothenhoefer, 2019; Stanaland et al., 2011; Surroca et al., 2010),
Regarding the control variables, our results are in line with many prior studies and exhibit a negative effect of leverage on corporate reputation (Delgado-García et al., 2013; Pérez-Cornejo et al., 2020; Walker et al., 2018), supporting the notion that high leverage may be perceived as a threat to the company’s future. However, in this research, we find no effect of economic performance on corporate reputation, thereby reinforcing the evidence about the lack of a financial halo problem. Furthermore, our research supports the idea that company age affects corporate reputation (Rao, 1994), supporting that older firms have had more time to build corporate reputation. Although we expected a positive influence of company size on corporate reputation, our results do not find any such support. This may be due to the fact that all the companies in our sample are large enough to be recognized. As for contextual control variables, our research shows a positive influence of national CSP and institutional development on corporate reputation, suggesting that standards of performance and living affect stakeholder expectations (Deephouse et al., 2016).
This article makes three main contributions to the literature. First, it contributes to the literature on CSP and corporate reputation by providing new evidence of the positive relationship between CSP and corporate reputation from an international approach. Most previous studies have only analyzed this phenomenon in a single country (Deephouse et al., 2016; Pérez-Cornejo et al., 2020). Moreover, we use a multilevel methodological approach that allows us to control the influence of variables from different levels, such as the organizational level and the contextual level, and which affect corporate reputation. Second, this research confirms that this relationship is not equal across national cultures. Power distance, individualism, masculinity, and uncertainty avoidance affect the intensity of the relationship between CSP and corporate reputation. Third, the study provides new evidence that institutional factors (specifically, culture) condition stakeholder’s perceptions and expectations about companies’ behaviors. Our approach complements prior research in which institutional factors such as culture were linked to CSP as antecedents of engagement in social responsibility initiatives (Aguilera et al., 2007; Halkos & Skouloudis, 2016; Ioannou & Serafeim, 2012; Ringov & Zollo, 2007; Waldman et al., 2006). We offer a complementary view of the institutional effect by describing the impact of national culture as a catalyst of stakeholder’s perceptions of corporate behavior.
Although our article adds fresh insights to research on corporate reputation, it still has certain limitations. First, all the companies in the sample were listed firms. Therefore, the results may not apply to small or unlisted firms. Second, we are aware that a company’s CSP may differ across the countries in which it operates. However, as explained in the “Method” section, this effect is impossible to isolate because of the global context in which these companies operate. Nevertheless, in an effort to control this effect at least partially, we included each county’s average level of CSP as a control variable. Furthermore, we study a specific range of cultural dimensions. Future research may approach this problem from other cultural frameworks ( House et al., 2004; Inglehart, 1990, 1997; Trompenaars, 1993). Although the sample is made up of countries with very different cultural features, we are conscious of the fact that, due to limitations in data availability, most of the countries analyzed are from America and Europe. Therefore, future research might examine whether our findings can be generalized to other countries and continents.
Our findings suggest possibilities for new lines of research to address the relationship between cultural context and corporate reputation. First, researchers may consider separate dimensions of CSP and different stakeholder groups. Second, other alternative cultural frameworks are also available, and future research could extend our analysis by using them. Third, it would be interesting to analyze how national culture moderates the influence of other company signals on corporate reputation. Fourth, future research could also study how corporate reputation is affected by the cultural distance between the home country and the host country. This distance may have an impact on how companies’ behavior fits in with the host country stakeholders’ expectations. Finally, future research might explore whether these effects could condition managerial decisions in internationalization processes, particularly in terms of host country selection.
This study also has relevant implications for managers. Our results suggest several recommendations that managers should consider when managing their CSP and corporate reputation. First, practitioners should be aware that CSP offers a strategic tool to reinforce a company’s relationship with stakeholders and to manage corporate reputation. Another important implication of our article is that it shows how culture is a relevant institutional factor that affects how stakeholders perceive companies’ social performance and, therefore, affects CSP’s influence on corporate reputation in different country settings. In fact, national values act as lenses through which stakeholders observe companies’ actions and they also affect the way stakeholders assess and judge company behaviors. Therefore, international business managers should consider the national culture of the countries where they operate to understand what impact their firm’s CSP has on corporate reputation. Such knowledge may help managers to design company policies so that they fit in with the national values of the country. For example, companies that operate in countries such as Germany may be particularly concerned about their CSP because it has strong implications for their corporate reputation due to the expectations and values of German citizens. Furthermore, managers should be aware that companies are affected by different levels of environments, which implies the need to combine different strategies and policies to fit in with national and global company contexts. Multinationals’ actions are perceived by a wide range of stakeholders from different national backgrounds who do not have the same expectations about companies’ responsibilities. Corporations must, therefore, deal with these issues by reinforcing some behaviors in certain contexts to achieve better levels of reputation.
Footnotes
Declaration of conflicting interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
