Corporate social performance (CSP) has received tremendous attention in the past few decades with the realization that a company exists not only as an economic entity but also has other social responsibilities to various stakeholders and the environment. In addition, as the global economies become more and more inter-related due to expanded trade, multinational companies are not only corporate citizens of their domiciled country but also their corporate responsibilities now include cross-national issues as well. Various stakeholders are beginning to emphasize or expect more social responsibilities from companies. For example, more and more governments are requiring public companies to not only report their financial performance but also their performance on social and environmental matters (Franklin 2008). A McKinsey survey cited by Franklin (2008) shows that 95% of executives believe that society now has higher expectations of business in terms of taking on more public responsibilities than 5 years ago. In addition, investors are also starting to take interest in social responsibility investing. $1 out of every $8 under professional management in the United States is reported to be invested in funds related to corporate social responsibility (Laufer 2003). This has led to increased pressures for companies to address their corporate social responsibility (CSR) issues (Smith 2009).

However, the interpretation of what the domain of corporate social responsibility is and the implementation (measured by corporate social performance) may be different in each country due to cultural differences. Culture has been identified as one of the most important differentiators in cross-cultural ethics (Tan and Chow 2009; Donleavy et al. 2008; Scholtens and Dam 2007; Alas 2006; Weaver 2001; Hofstede 1985). Using a large scale global CSP database, this study empirically tests the impact of national culture on CSP. In addition, effects of differences of regions where companies are domiciled on CSP are examined (Hill et al. 2007; Alas 2006; Scholtens and Dam 2007).

While researchers recognize the importance of the relationship between CSP and culture, not much has been done to empirically test the nature of the relationship. There may be a few explanations for the dearth of research here. First, while CSP is recognized as a multidimensional concept, current research usually operationalizes only a component of it such as corporate philanthropy (McElroy and Siegfried 1985; Brammer and Millington 2006) and environmental performance such as pollution emissions (Stanwick and Stanwick 1998). Second, CSP is usually implemented and tested in a single industry, country, or geographical location (e.g., Brammer and Millington 2006; Stanwick and Stanwick 1998). This study attempts to test the nature of the relationship between CSP and culture by addressing the current limitations in this area. This will be achieved by testing CSP across different industry types with a large sample size of multinational companies from different countries and regions.

Literature Review and Hypothesis Development

Various researchers have attempted to define the scope and nature of CSP (Carroll 1979; Sethi 1979; Wartick and Cochran 1985; Ullman 1985; Wood 1991; O’Riordan and Firbrass 2008). For example, Carroll (1979) articulated that CSP involves an integration of three dimensions: (1) corporate social responsibility, (2) philosophy of corporate social responsiveness, and (3) social issues involved. This is the first study to build a paradigm for business and society that incorporates different competing perspectives (Wartick and Cochran 1985; Wood 1991). This paradigm asserts that the social responsibility of business “embodies the economic, legal, ethical, and discretionary categories of business performance” (Carroll 1979, p. 499). These four categories are not necessarily mutually exclusive; that is, they are neither cumulative nor additive. Corporate social responsiveness is the action phase of management responding to social issues. It can range from no response to a proactive response. Social issues refer to the operationalization of social responsiveness that business must address.

Wartick and Cochran (1985) extended Carroll’s (1979) model by addressing some of its weaknesses and then offering an extension that is based on the principles of social responsibility, the process of social responsiveness and the policies of issues management. The principles of social responsibility accept the existence and importance of both economic and social responsibility of companies; in other words, rather than being mutually exclusive, economic responsibility is a component of social responsibility. The process of social responsiveness states that both social responsibility and social responsiveness are complementary, and that they provide the means to achieving corporate social obligations. The policies of issues management refer to the operationalization of social responsiveness to minimize surprises and determine effective corporate social policies. While most researchers may differ on the definition, scope, and approach to CSP, most agree on the need for companies to identify and address social issues, develop processes of response and assess the outcome. Unfortunately, none of them has offered to identify specifically the types of social issues that are covered within CSP. Some researchers have attempted to develop a process to define what the social issues construct is but failed to identify what it encompasses (Husted 2000; Wartick and Mahon 1994). The present study classifies overall CSP into four subcomponents; i.e., those social issues that are related to the environmental, strategic governance, labor relations, and stakeholder management.

In addition to the development of a theory or framework for CSP, many researchers have attempted to measure the impact of CSP. One stream of research has looked at the relationship between CSP and financial performance (e.g., Orlitzky et al. 2003; Hill et al. 2007; Laufer 2003). There is growing evidence that there is a positive correlation between CSP and financial performance. For example, in a meta-analysis of 52 studies representing a total sample size of 33,878 observations, Orlitzky et al. (2003) found that there is a positive association between CSP and corporate financial performance across industries and across study contexts. A second stream of research uses intermediate performance indicators; for example, it includes (1) perception and awareness of various stakeholders of a firm’s CSR affecting their beliefs and attitudes toward the firm (Sen et al. 2006), (2) consumers’ perception of a firm’s CSR affecting their trust toward the firm (Vlachos et al. 2009), (3) impact of CSR on customer satisfaction (Luo and Bhattacharya 2006), and (4) brand equity and reputation (Brown and Dacin 1997; Wang 2010).

Another stream of research explores cross-national differences related to CSP. For example, Maignan (2001) examined consumers from the U.S., Germany, and France in terms of their readiness to support socially responsible organizations and found that both French and German consumers were more actively willing to support socially responsible organizations than the US consumers. Jurgens et al. (2010) argue that the difference can be fundamentally explained by cultural and philosophical differences between these two regions in the way they look at stakeholders, and how relations with each group is managed. The European stakeholder model (also known as Northern European or “Rhineland”) allows for the recognition and inclusion of important stakeholder groups into the process of policy and decision making. This may include board representation reserved for employee stakeholders and financial institutions. On the other hand, the US firms reserve board representatives for institutional investors or major shareholders. While the US model attempts to maximize shareholder value, the European model tries to meet the needs of those stakeholders with vested interest in the company’s welfare by ensuring board representations from those groups. Other researchers have also found differences in investing returns between various regions. Hill et al. (2007) found that financial return of mutual funds that invest in socially responsible investing provided better return in the short term for European funds while in the long term, both the European and US funds outperformed the larger equity markets. No significant difference was observed for Asian portfolios for the same period. In summary, the literature suggests a wide variation of ethics not only across culture but also regional boundaries (Tan and Chow 2009; Donleavy et al. 2008). Therefore, the first research hypotheses are stated as follows:

H1a

European firms will have higher CSP than North American firms.

H1b

North American firms will have higher CSP than Asian firms.

Hofstede’s Cultural Dimensions

Culture is recognized as one of the fundamental determinants of differences between individuals from different cultural backgrounds. This is attributed mostly to the different value and belief system, tradition, customs, etc. This is not only true for individuals but also for organizations (Hofstede 1985). Various researchers have found that cultural background affects many aspects of ethics. For example, research has found that culture affects ethical attitudes (Franke and Nadler 2008), ethical sensitivity (Simga-Mugan et al. 2005), value system (Ford et al. 2005), ethics judgments (Whipple and Swords 1992), ethical decision making (Srnka 2004), and ethical perception (Vitell and Paolillo 2004), among others.

Even though there may be differences on the surface, Hofstede’s research has found four cultural dimensions that appear to account for much of the variability. These dimensions have been utilized extensively in disciplines that look at culture and have not only been validated by a number of other studies (e.g., Hofstede and Bond 1984) but it has also been shown that each dimension will have different impacts on ethics issues (e.g., Lu et al. 1999; Scholtens and Dam 2007; Sims and Gegez 2004). Below is a brief definition of the four dimensions.Footnote 1

Power Distance Index

Power distance refers to the extent to which the less powerful members (within family, organization, and institutions) accept and expect that power is distributed unequally. This is not only endorsed by the followers but as much by the leaders also. Individuals who are considered to have high Power Distance Index (PDI) (such as Malaysia) tend to accept this inequality as natural and believe that superiors are entitled to such privilege. Those who have low PDI (such as Australia and New Zealand) are less likely to tolerate such inequality and more likely to disagree with superiors. Research has found that people from countries that score high on PDI are more likely to accept questionable business practices (Cohen et al. 1996); therefore, one would expect companies from countries that are high on PDI would have lower CSP. The hypothesis is stated as follows:

H2a

The higher the PDI (more likely to tolerate inequalities), the lower the CSP.

Individualism

Individualistic cultures (such as the U.S.) attach more importance on personal self-interest and expression. Ties between individuals are loose, and membership in groups can change when needed. Individuals who are high on individualism (IDV) tend to value personal time, freedom, and independence; they believe that personal interests are more important than the group interests. Individualists stress primary responsibility for their own actions. On the other hand, individuals who are low on IDV (Latin American and Asian countries) are considered as “collectivists” who subordinate themselves to the group and thus place the interest and welfare of the group ahead of themselves. Akaah (1990) found workers from individualistic organizations were less ethical than those from collectivistic organizations. As such, one would expect collectivist societies to emphasize more of a concern about the impact of business on society; therefore, the research hypothesis is stated as follows:

H2b

The higher the IDV (individualists), the lower the CSP.

Masculinity

Masculinity (MAS) refers to the distribution of roles between the genders. This dimension focuses on the relative importance of assertiveness, materialism/material success, self-centeredness, power, strength, and individual achievements (masculine values) versus the spirit of modest, caring, helpfulness and social support (feminine value). Societies that are considered masculine (such as Japan) describe men that are assertive, aggressive, ambitious, competitive and materialistic while women play more of the nurturer role. Femininity describes a society (such as Sweden) where the social roles of men and women overlap—with neither genders exhibiting competitive behavior. Research has found that individuals from masculine countries have a lower appreciation for cooperative behavior. In addition, some of the most frequently cited reasons for unethical behaviors among masculine individuals were greed and competitiveness (Vitell and Festervand 1987) that lead to personal financial gains. Given that masculine societies emphasize the need for competitiveness, success, and individual achievements, therefore, the research hypothesis is stated as follows:

H2c

The higher the MAS (masculine), the lower the CSP.

Uncertainty Avoidance

Uncertainty avoidance (UAI) deals with the extent a culture’s tolerance for uncertainty and ambiguity is, i.e., how an individual is socialized to feel comfortable or uncomfortable in unstructured situations. Individuals high on UAI (uncertainty avoidance; e.g., Japan) prefer a structured environment, such as clear hierarchy, strict laws, and rules to minimize the uncertainty. Individuals low on UAI (uncertainty accepting; e.g., the U.S.) tend to be more tolerant of different opinions and try to have as few rules as possible. Previous research has found that uncertainty accepting individuals are more likely to take risks (Hofstede 1984), and this risk taking is highly correlated with unethical actions (Rallapalli et al. 1994). Given that uncertainty avoiding cultures have more strict rules and regulations to minimize uncertainty, therefore, the research hypothesis is stated as follows:

H2d

The higher the UAI (uncertainty avoiding), the higher the CSP.

Impact of Economic Development

To test the effects of cultural variables in cross-cultural studies, Hofstede (2001) recommends controlling for economic development of the countries studied. The main reason is that if economic, biological, and technological variables can predict a country variable better, then cultural indexes become redundant. Srnka (2004) argues that economic systems and levels of economic development, along with national cultures and other elements, are part of a broader multidimensional concept. Therefore, economic development becomes a relevant factor to test the effects of culture on ethics. For example, Harrison and Huntington (2000) have found that less-developed countries that lack financial resources are less likely to provide the legal safeguards needed for economic growth; therefore, the research hypothesis is stated as follows:

H3

The higher the country’s level of economic development, the higher the CSP.

Methodology

CSP Measure

Historically, researchers have used a number of approaches to measure CSP. Reputation indices and databases including The Kinder, Lydenberg, and Domini (KLD) Database, the Fortune Index, and Canadian Social Investment Database (CSID) are among the most widely used methods (Turker 2009). KLD rates the US public companies on eight attributes, such as community relations, employee relations, environment, etc. using a five-point rating scale. The Fortune Index is based on a large scale survey of executives, outside directors, and corporate analysts, rating the largest 10 firms in their industry on eight dimensions, such as long term investment value, quality of management, wise use of corporate assets, etc. CSID measures the average performance of a firm’s net strength and weakness on seven dimensions, such as community, diversity, and environment. However, each method has its significant limitations (Simerly and Li 2000). KLD assigns equal weight to each of the dimensions, which inhibits the capture of incremental differences across the dimensions. The Fortune Index, on the other hand, has been criticized as biased toward financial performance instead of social performance. CSID details only public companies on the Canadian stock exchange, which limits its geographic scope. Another method uses single- and multiple-issue indicators, such as the pollution control performance, reported by the Council of Economic Priorities (Freedman and Jaggi 1982). However, its unidimensionality is a significant limitation.

Another recent measure of CSP is based on the approach of Intangible Value Assessment (IVA) from an independent evaluation agency called Innovest that provides financial and sustainability-based investment relevant research.Footnote 2 The Innovest’s IVA approach links long-term corporate out-performance to superior management of environmental and social risks and opportunities. It looks to environmental, social, and governance (ESG) criteria as leading indicators for management quality and long-term financial performance, not as commentaries on the intrinsic ethical worth of the companies. Innovest’s approach combines more than 120 performance factors that are broken down into four distinct components: stakeholder capital, human capital, strategic governance, and environment (see Appendix for main issues assessed under each of these four main headings). Overall, Innovest research seeks to address two objectives: (1) identify key ESG risks and opportunities applicable to each sector, and (2) assess if companies have strategies to capture potential opportunities in the ESG space.

Innovest’s IVA approach offers some advantages to address some of the limitations from other methods. First, the Innovest database covers more than 2,000 international companies. The nationality of a company is based on where its corporate headquarter is domiciled. For multinational enterprises and those with foreign subsidiaries, the corporate parent is evaluated with all overseas branches and subsidiaries taken into account. In other words, even though foreign branches and subsidiaries are not rated separately from their corporate parents; however, their ESG performance is factored into the overall score. This allows researchers to compare companies from different countries, regions, or cultures. Second, at least two officials from two different departments from each company are interviewed. The interview method allows for interaction and probing which may be lacking in a questionnaire-driven methodology. Third, the company interview information is then augmented with other information from other sources, such as corporate documents, government data, industry sources and non-governmental organizations (NGOs). This addresses the weakness of relying too much on financial indicators of performance rather than a true social performance evaluation. Each company gets a weighted composite CSP rating on a scale from the worst (CCC) to the best (AAA). These evaluations are forward-looking and take into account a company’s environmental and social risks, its structural capacity and strategy to manage them, and its ability to profit from these features in the future. For example, companies that receive overall “AAA” ratings have a strong ability to manage social and environmental risks and are well positioned to capitalize on socially and environmentally driven opportunities, whereas “CCC”-rated companies possess a questionable ability to handle risks and liabilities and suffer from a poor position for capitalizing on profit opportunities. For the CSP variable used in the analysis, the seven-grade categories were converted into a numerical scale, with 1 as the worst (CCC) and 7 the best (AAA). In addition, each company gets a score for each of the four components that make up the overall CSP index. The subcomponent scores range from 1 to 10. The empirical analysis for this study includes the composite CSP score and the four subcomponent ratings from 2003 to 2008 on companies covered in the Innovest database. The data were augmented with the cultural dimensions from Hofstede’s framework. The resulting sample comprised 3,680 observations with companies from 49 developed and developing countries across regions of North America, Europe, and Asia Pacific.

Control Variables

A set of firm and industry specific variables were employed in the analysis to control for their effects on CSP. For firm specific variables, firm size, prior financial performance, growth rate, leverage, R&D, and marketing-related expenses were included. In addition, the study also controlled for industry fixed effects and the degree of competitive intensity. These data were all derived from COMPUSTAT database.

The log of the number of employees was used to measure firm size. Return on asset (ROA) in the prior period was used to measure prior financial performance. As the level of ROA differs across countries, this measure was further adjusted according to the respective country average. Firm growth rate was measured by the percentage change in sales in the previous 1 year. Leverage was measured by the total debt divided by total asset. This variable served as a proxy for the riskiness of a firm (Waddock and Graves 1997). General selling and administration (GSA) expenses were used as a proxy to measure a firm’s marketing-related expenses. In COMPUSTAT, there were many missing values of R&D and GSA expenses. Therefore, to control for the systematic biases that may arise between the groups that reported these expenses and those who did not, two dummy variables were created, which indicate the availability of R&D and GSA data (1 = available and 0 = missing). The R&D (GSA) intensity variable was constructed as the ratio of R&D (GSA) expense relative to total sales.

As for the industry controls, the Global Industry Classification Standard (GICS) at the two-digit level was used as the basis to identify the main industry group that a company belongs to and to create the industry-specific dummy variables for each company. In addition, to measure industry competitive intensity, Herfindahl–Hirschman Index (HHI) at the two-digit industry level of GICS was used and the number rescaled by 10,000 in the analysis. The HHI is a measure of the amount of competition in the industry. It is calculated as the sum of the squared market shares of all the firms in an industry. Increases in the HHI index generally indicate a decrease in competition and an increase of market power. A small index indicates a competitive industry with no dominant players. The HHI increases both as the number of firms in the market decreases and as the disparity in size between those firms increases. For example, a HHI index below 1,000 indicates a highly competitive industry; a HHI index between 1,000 and 1,800 indicates moderate concentration; and, a HHI index above 1,800 indicates high concentration.

Data Analysis and Results

Table 1 shows the descriptive statistics and correlations of CSP (and its subcomponents) with Hofstede’s cultural dimensions. The results show that there is a high item-to-total correlation between CSP and its corresponding components. Also, the high correlations among the four components suggest a high internal consistency in measuring the aggregate CSP. In addition, CSP is significantly correlated (albeit small) with all the four Hofstede’s cultural dimensions.

Table 1 Descriptive statistics and correlations
Full size table

To test for regional and level of economic development differences, firms were classified based on the region and the level of economic development of their respective countries where their head offices are domiciled. In particular, firms were classified into four categories: developed economies located in (1) North America, (2) Europe, (3) Asia, and (4) those from developing economies in general. The average CSP ratings and culture scores were compared using ANOVA across the four groups as shown in Table 2. The results show that firms from the developed European countries in general have significantly higher ratings in overall CSP and the four subcomponents. On the other hand, firms from the developing countries on average have significantly lower ratings across the board. Of interest, firms from developed economies of North American or Europe share close similarities in culture based on the Hofstede scores; however, companies in North America consistently have lower performance than their European counterparts. Another interesting result is that developed Asian countries have higher CSP than North American countries. The cross-regional/economic comparison implies that factors pertaining to region or economic development contribute to differences in CSP. Other researchers have found similar differences between European and North American (Sotorrio and Sanchez 2008). Therefore, the study found support for H1a and H3.

Table 2 Cross region-economy analysis
Full size table

Before testing the main hypotheses pertaining to the association between cultural dimensions and CSP, the endogeneity of the culture variables needs to be addressed. It is likely that there are unobserved variables correlated with both cultural dimensions and CSP. To deal with the endogeneity problem, Garcia-Castro et al. (2010) suggest the estimation approach that involves the use of an instrumental variable that correlates with the endogenous variable (culture dimensions in our case), but uncorrelated with the dependent variable (CSP). For the cultural dimensions, this study employs the instrumental variable based on the linguistic practice of pronoun drop in languages. The data originates from the study of Kashima and Kashima (1998). This instrumental variable was also used by Licht et al. (2007) and Tabellini (2008) in studying the effects of culture on socioeconomic outcomes.

Using a two-stage least square (2SLS) estimation method, a series of instrumental variables regression analyses was conducted on the overall CSP using different sets of explanatory variables. Model I includes industry and firm control variables only. Model II adds two dummy variables that indicate whether firms are from developed countries in Europe and those from developing economies. Models III–VI include the four cultural variables (instrumented by pronoun drop practices), respectively. As observed in Table 3, the geographic factor contributes a 7.4% increase in the R 2 value from Model I to Model II. The results show that after controlling for firm and industry-specific effects, the dummy variable of European region explains a significant variation in CSP. Also, the dummy variable of the developing countries is significant at 10% level. This shows further support for H1a and H3. Model fit is improved from 8.6% to 9.2% to 9.4% in Models III–VI after the culture dimensions were included, respectively. In particular, PDI, MAS, and UAI have positive effects on CSP whereas IDV has a negative effect. Therefore, based on the results from Models III–VI, this study found support for the linkages between all the four Hofstede’s cultural dimensions and CSP; that is, higher CSP is significantly related to cultures that are collectivist societies (H2b) and tend to prefer a structured environment to minimize uncertainty (H2d). Even though both PDI (H2a) in Model III and MAS (H2c) in Model V are also significant, both hypotheses were not supported because the result shows an opposite direction than that hypothesized. The positive associations are likely driven by the relationships between cultural dimensions and the specific areas of CSP. Additional regression analyses of the subcomponents of CSP on the two dimensions of MAS and PDI reveal that while environmental performance is positively linked to the two factors, their linkages with human capital are negative. It is likely that in countries with high PDI and MAS cultures, there are possibly more environmental regulations and strict enforcement. The negative associations with human capital partially confirm the hypotheses H2a and H2c.

Table 3 Instrumental variable regression analysis results
Full size table

The firm size effect is found to be significant and positive, indicating that larger firms, in general, tend to perform better on CSP than smaller ones. Other researchers have also found similar results when firm size is taken into account (Orlitzky 2001; Ringov and Zollo 2007). Larger firms, because of their visibility, access to resources, and scale of operation, are subject to more scrutiny by various stakeholders and, therefore, may be more motivated to participate in CSP (Udayasankar 2007). In addition, firm leverage is found to be negatively associated with CSP. The negative association is consistent with the result of the study of Waddock and Graves (1997). This implies that firms with higher leverage may be less able to make discretionary investments in corporate social activities.

Conclusion and Discussion

The results of this study indicate that differences in CSP appear to be linked to national culture, geographic region, and level of economic development. European companies overall perform much better than North American companies with regard to average CSP score. Asian companies tend to lag behind their European and North American counterparts, but still ahead of developing countries. This seems to be consistent with other researchers who have found that European countries have higher awareness of and perform better on CSP than those from other continents (Hill et al. 2007; Maignan and Ferrell 2003). There may be a couple of reasons for such a difference. It has been shown that national culture defines a nation’s value system which in turn influences people’s attitudes (Sirmon and Lane 2004). These attitudes in turn determine how individuals perceive and respond to such issues as corporate responsibility and performance. For example, Asian societies, by their collective nature, may be less likely to pass judgment or question what senior executives in the company are doing. Western societies, on the other hand, being more individualistic, tend to focus more on equality and are more likely to bring any indiscretions to the fore. A second reason may be governmental regulations. European countries tend to be more regulated while the United States tends to favor self-regulation. A third reason may be the different stakeholders that companies respond to. Clarkson (1995) and Donaldson and Preston (1995) argue that companies have many stakeholders that can affect and be affected by corporate activities. Empirical evidence shows that consumers from the US and European countries have different priorities with regard to which stakeholder’s goals should be emphasized and rewarded (Maignan and Ferrell 2003).

Higher CSP is significantly associated with cultures characterized by higher power distance (less likely to tolerate questionable business practice), more collectivist societies, more masculine, and more uncertainty avoidance (prefer a structured environment). The association between culture and CSP is also documented in Ringov and Zollo’s (2007) study. However, this study differs from theirs with regard to the significance and direction of associations between cultural dimensions and CSP. While their study found only two dimensions (power distance and masculinity) are significantly related to CSP, this study found support for the relationship between CSP and the four cultural dimensions. An interesting difference is while they did not find support for UAI, they hypothesized that countries with higher UAI should exhibit lower CSP, whereas this study finds a significant relationship where higher UAI leads to higher CSP. Another difference is that their study found a negative relationship between MAS and CSP whereas this study found a positive relationship. It is possible that the difference could be caused by whether endogeneity issue of the culture dimensions is taken into account and the way each study employs and control variables. Endogeneity issues, if not accounted for, could lead to bias in the inference about the relationship between culture and CSP in terms of magnitude and direction (Garcia-Castro et al. 2010). While they included other explanatory variables, such as GLOBE cultural dimensions (House et al. 2004) and firm variables, this study includes geographic region and the level of economic development in the current analyses.

Limitations and Future Research

While this study tested and found that cultural and geographic environments have significant impacts on general CSP, the results may be moderated by a few factors, which may lead to potential fruitful future research. First, the sample in the empirical analysis is limited to public companies. It is possible that the environmental effects may be different in smaller or non-public firms because of lesser regulations and stakeholder groups. Also, the study uses only one macro variable that reflects the economic development of a country. Other finer-grained variables, such as income and foreign investments measured at country and/or firm level, may better explain the nature of the linkage between the cultural variables and CSP (Husted 2005; Park et al. 2007). Furthermore, the lack of time variance of the cultural variables may be a matter of concern (cf. Aguilera and Jackson 2010 for a more detailed discussion). For example, Holden (2002) argues against the “time-invariant” assumption of Hofstede’s cultural dimensions. It assumes that cultures remain relative stable over time; therefore, it becomes difficult to use culture to explain organizational change. Other researchers, e.g., McSweeney (2002) argue that Hofstede takes no account of within-country variance in his dataset, which assumes that there is a uniform national culture in each case.

For future research, it may be interesting to see other important firm-specific factors that moderate their effects on CSP. For example, the effects may differ between consumer goods and industrial companies. As consumer awareness and sensitivity depend on cultural context, the effects of culture may be strengthened in the case of consumer goods companies. Another interesting future research project would be to look at the mediating role of corporate culture in the linkage between national culture and corporate social performance. Domestic companies may have a stronger emphasis on core stakeholders, such as customers and shareholders and bottom-line performances, than their multinational counterparts.