Taking social responsibilities overseas? The learning paths and performance consequences of Chinese multinationals

Abstract

Corporate social responsibility (CSR) has been seen as an effective strategy that emerging market multinational enterprises increasingly adopt in order to cope with the liability of origin, yet much less attention has been paid to the micro-foundations of such a strategy. It remains less explored as to how CSR is leveraged by individuals—that is, expatriates. Addressing this research gap, a total of 150 valid survey responses from 38 Chinese multinationals were tested using the structural equation modeling technique. The results show that expatriates’ cross-cultural competence and engagement in CSR facilitate the learning efficiency of subsidiaries. The findings also suggest that expatriates’ effort positively mediates the relationship between CSR and performance outcomes (at both individual level and organizational level). Embracing the micro-foundations perspective in CSR research, this study offers both theoretical and practical implications for future international management research.

Keywords

Chinese multinationals corporate social responsibility cross-cultural competence expatriates micro-foundations

Introduction

Corporate social responsibility (CSR) is defined as a voluntary business action for supporting a firm’s sustainable development in economic growth, social cohesion, and environmental protection (Jenkins, 2004; Tang et al., 2012; Tashman et al., 2019). In recent years, CSR activities have been increasingly adopted by emerging market multinational enterprises (EMNEs), particularly Chinese multinationals, in foreign countries (Luo et al., 2017; Miska et al., 2016; Yin and Zhang, 2012). For instance, to sustainably cooperate with Sri Lankan local government and partners, China Communications Construction Company assigned expatriates to make a proposed US$800 million investment in building a road network for Sri Lanka in 2018 that is expected to benefit housing, hospitals, and schools for the next 30 years (Reuters, 2018). Most recently, the Chinese e-commerce giant Alibaba and its subsidiaries donated quantities of medical supplies, such as personal protective equipment, to the United States and other Asian countries immediately after the outbreak of the COVID-19 crisis (Forbes, 2020).

Since China’s “going global” policy and the aggressive propagation of the Belt and Road Initiative (BRI) (Zhu et al., 2020), Chinese multinationals have been encouraged to engage in global competition and adopt CSR practices for improving their global image and reputation (Miska et al., 2016; Tashman et al., 2019; Zhang et al., 2020). Existing studies explore CSR adoption by Chinese multinationals in host countries, including strategic or institutional factors affecting CSR initiatives (Miska et al., 2016; Zhang et al., 2020), the various forms of CSR practices across different contexts (Tashman et al., 2019), and undertaking CSR activities as an alternative means to gain legitimacy (Zhang et al., 2020). CSR activities also benefit Chinese multinationals in overcoming the liability of origin and outsidership in host countries (Miska et al., 2016; Tashman et al., 2019) and in building up a trust relationship with local stakeholders (Jiang et al., 2018).

All these arguments are based on the presumption that taking on social responsibilities overseas can benefit Chinese multinationals operating overseas. Yet, to date, the links between CSR and performance consequences are mixed and controversial (Kiessling et al., 2016). Some find a positive relationship, suggesting that multinationals involved in CSR may receive higher evaluations from their consumers worldwide, thus improving their global image and potentially boosting profitability (Husted et al., 2016; Kiessling et al., 2016). Some, however, detect a negative CSR–performance link, arguing that CSR practices may lead multinationals to bear more costs and at the same time fail to reap additional benefits of strategic differentiation and innovation in their foreign market operations (Orlitzky et al., 2003; Tang et al., 2012). Apart from research design heterogeneity, these mixed findings can be attributed to a lack of deep understanding of the learning effects of the EMNEs’ overseas CSR initiatives.

According to the knowledge-based view (KBV), performance variances among firms are determined by firms’ ability to learn new knowledge and implement that knowledge for value-added activities (Grant, 1996; Kogut and Zander, 2003). As global market latecomers, Chinese multinationals tend to engage in localized learning activities to tap into local networks and identify business opportunities, and thereby become more attuned to host-country environments (Fan et al., 2016; Li and Fleury, 2020). Since learning requires a responsive mentality and it is dependent upon individual effort (Fan et al., 2016; Peng and Beamish, 2014), a micro-foundational approach is needed to explain the learning effects of CSR initiatives. A micro-foundational view is helpful because it looks across the different levels of analysis and can thoroughly explain the individual role in learning new knowledge while pursuing MNEs’ CSR practices (Devinney, 2013; Shea and Hawn, 2019). Most importantly, a micro-foundational study has the potential to bridge macro–micro links and develop a more unified conclusion on the CSR–performance relationship (Shea and Hawn, 2019).

To address this concern, there are questions regarding the micro-foundations of overseas CSR practices (Hafenbrädl and Waeger, 2017; Shea and Hawn, 2019): How is CSR leveraged by individuals? What is the extent of the impacts of CSR on Chinese multinationals at both organizational- and individual-level? To address these research questions, it is worthwhile to examine the individual-level learning mechanisms’ effects on outcomes of overseas CSR initiatives, because the individual actors (i.e. expatriates) play important roles in international knowledge transfer (Antal and Sobczak, 2004; Chang et al., 2012; Vlajčić et al., 2019; Zhang et al., 2020). This study contends that seeking knowledge overseas is the key motivator for Chinese multinationals embedding in foreign markets through implementing CSR initiatives. That knowledge includes both general knowledge that can be reverse-transferred to headquarters (HQs) and subsidiaries, and local knowledge, such as responding more effectively to local economic and institutional conditions and collaborating efficiently with foreign stakeholders (cf. Fan et al., 2016). From this perspective, the learning paths of Chinese multinationals may offer a nuanced way of predicting the different behavioral consequences of CSR and add a knowledge-based mechanism to the CSR–performance link.

Building on the KBV, this study adopts a quantitative approach, using structural equation modeling (SEM) to test the first-hand data collected from expatriates and their supervisors of 38 Chinese multinationals. The results show that the Chinese multinationals’ engagement in overseas CSR can lead to an increase in international commitment (IC), which facilitates their expatriates gaining more learning opportunities and developing better cross-cultural competence (CCC). Ceteris paribus, the findings also reveal that IC and individuals’ CCC positively affect organizational-level reverse-knowledge transfer (RKT) and individual-level work performance. The findings contribute to the literature by highlighting the knowledge-seeking motives of EMNEs and by bringing the KBV into CSR research. The study acknowledges individual-level efforts in dealing with CSR-related complexity and provides managerial implications for EMNEs in dealing with the complexity of international assignments. In particular, managers should leverage the CSR tool kits for mitigating the liability of foreignness and outsidership through effective localized learning. Moreover, encouraging individual employees to develop CCC (Caligiuri and Tarique, 2012; Yamazaki and Kayes, 2004) can help multinationals benefit from knowledge spillover and improve learning efficiencies through CSR practices.

Theoretical background

In the literature, KBV is regarded as an outgrowth of the resource-based view, which proposes that knowledge determines performance differences because it is the most important strategic resource of a firm (Grant, 1996; Kogut and Zander, 2003). It considers firms’ capability in accessing new knowledge and developing new skills for added strategic value (Grant, 1996). This view has been adopted in the literature to discuss the knowledge-seeking and learning activities of multinationals (Kogut and Zander, 2003). From the KBV perspective, the competence of expatriates determines whether they can identify and aggregate dispersed knowledge, which enables use of a micro-foundations lens to investigate the interactions and learning paths in the process of CSR (Barney and Felin, 2013; Han et al., 2019). Because the expatriates are acting as knowledge conduits between knowledge bases of home and host countries, their capabilities and interactions with local markets are essential for dealing with CSR complexity (Wang et al., 2014). Hence, this study incorporates the micro-foundations and KBV to analyze the CSR practices by Chinese multinationals in host countries: that is, unpacking the learning process by individuals and showing the influence of individual capabilities on firms’ competitive advantages (Barney and Felin, 2013; Kogut and Zander, 2003).

To engage in localized learning, local responsiveness, even mutual trust-building, is required (Fan et al., 2016). In the literature, the assumptions of micro-foundations emphasize the individual approach, indicating that knowledge yield in individual interaction process is not simply additive but also emergent (Barney and Felin, 2013; Devinney, 2013). In this regard, CSR adoption can seamlessly connect the knowledge bases of individuals and enable expatriates to interact with local markets (Antal and Sobczak, 2004; Hafenbrädl and Waeger, 2017; Luo et al., 2017). This not only enables multinationals to access new knowledge via embedding in the host market and interacting with local stakeholders, it also helps multinationals combine their existing knowledge bases with localized learning and thus create organizational-level heterogeneity (Fan et al., 2016; Matten and Moon, 2008; Tang et al., 2012). Moreover, it requires expatriates to largely incorporate CSR into their operational activities, especially those expatriates who can transfer HQs information and corporate culture to host market business practices (Chang et al., 2012). CSR therefore provides expatriates and managers with platforms to develop CCC (e.g. new skills, knowledge, and adjustment abilities), which is important in promoting the expansion of Chinese multinationals (Gorondutse and Hilman, 2016; Shea and Hawn, 2019).

However, prior studies often regard CSR as “compulsory investment” for multinationals to get rid of liabilities of origin, outsidership, and newness (Fiaschi et al., 2017; Marano et al., 2017). The knowledge mechanisms in CSR practices have been largely ignored. Indeed, the overseas CSR investment of global latecomers is also paying “tuition fees” that enable these multinationals to access new knowledge and learning opportunities in a new market (Asmussen and Fosfuri, 2019). First, multinationals need to afford “translation or communication costs” to achieve a certain code of conduct for doing business in the local market via constantly investing in CSR in order to interact with local stakeholders (Asmussen and Fosfuri, 2019). Second, as latecomers to global markets, Chinese multinationals are engaging in CSR activities to gather, process, and act on new knowledge from the market. Such CSR-related activities can be regarded as knowledge-seeking costs (Chen and Tan, 2016; Fan et al., 2016). Third, through embedding and interacting with the host markets, expatriates can notice and learn complementary knowledge that is critical for their operative upgrading, and which can improve the accuracy of learning and reduce the risks of the knowledge redundancy or duplication that often happens in acquisition activities (Antal and Sobczak, 2004; Matten and Moon, 2008). Taking all these aspects together, this study suggests that incorporating KBV to examine the learning mechanisms in CSR practices provides a more nuanced approach to revealing the CSR–performance link.

An integrative model and research hypotheses

This study proposes an integrative model to explore CSR and performance consequences in the context of Chinese multinationals. The model embraces the theoretical framework of KBV, motivations of firms, and the micro-foundation perspective. Aligning with these theories, the basic premises of the model are built on the commonly accepted means to goals relationships through exploring the micro-foundations mechanism. In the context of EMNEs, individuals (i.e. expatriates in this research context) acting on CSR activities can improve the degree of IC in host countries, which in turn requires individuals to embrace a greater level of cross-culture competence. These “means” can eventually lead to better individual work performance, which potentially enlarges the chances to improve organizational-level strategic “goals,” such as RKT. The integrative model is depicted in Figure 1, and research hypotheses are developed.

Figure 1.

Theoretical framework.

CSR practices and IC

When conducting business abroad, multinationals are required to reconcile and assess the composite interests and feasibility of all the internal and external stakeholders (Li and Hambrick, 2005). IC involves the extent to which a multinational makes irreversible investments in a foreign country (Johanson and Vahlne, 1977, 2009). With a rationale similar to that of employees making an organizational commitment, increasing IC expresses a Chinese multinationals’ emotional attachments to the foreign market (Zhang et al., 2014). In the literature, scholars suggest that, since a lack of local knowledge will lead to an increase of psychic distance and liability of foreignness, it is necessary for multinationals to gradually enhance their commitment to their new markets (Johanson and Vahlne, 1977, 2009). In particular, increasing commitment to foreign markets enables multinationals, especially global market latecomers, to have wider interactions and communications with foreign stakeholders, and consequently build up trust relationships in foreign markets (Li and Fleury, 2020; Li and Hambrick, 2005). By embracing KBV, it is important for Chinese multinationals to embed in their overseas markets in order to seek more learning opportunities and acquire local knowledge that is essential for their production upgrading (Johanson and Vahlne, 1977, 2009; Nadkarni and Perez, 2007).

Engaging in CSR will result in an increase in IC of Chinese multinationals, given the “beyond regulatory requirements” of CSR practices (Husted et al., 2016). The practices often involve extra capital expense or investment, which is adopted voluntarily by multinationals in order to ease the liability of foreignness in foreign markets (Marano et al., 2017). This is especially true for Chinese multinationals that have endured significant criticism for their absence of advanced and sustainable business models when they are faced with increasing tensions for their overseas operations (Chen and Tan, 2016). As a result, these multinationals tend to engage in CSR practices, such as donations to support local communities and investments in the improvement of domestic infrastructure (Husted et al., 2016). Through these CSR activities, Chinese multinationals can embed more in local markets for the purpose of accessing broader learning opportunities (Chen and Tan, 2016). Hypothesis 1 is thus proposed as follows:

Hypothesis 1: Overseas CSR practices are positively related to the IC of Chinese multinationals.

CSR practices and CCC

As Chinese multinationals’ CSR activities have expanded globally in recent decades, researchers have examined various aspects of how market latecomers cope with challenges in the process of localization in foreign markets (Chang et al., 2012; Fan et al., 2016; Tashman et al., 2019). At an individual level, previous studies imply two cross-conditions that dictate the efficiencies of employees’ local market engagement, particularly targeted at expatriates who are able to bridge the connections and knowledge transfer between the host and home countries (Vlajčić et al., 2019). One part of the study underlines the holistic process of expatriates’ adaptation to foreign markets (Yamazaki, 2010). The other highlights expatriates’ learning and local engagement that largely shape the individual’s development of specific skills, especially the development of CCC that is viewed as a key specialized situational ability for expatriates for dealing with environmental complexity (Hafenbrädl and Waeger, 2017; Peng and Beamish, 2014).

Defined as the expatriates’ ability to function effectively in an environment with a different cultural background, CCC denotes an all-encompassing term that elucidates the individual characteristics for coping with ambiguity, managing stress, building relationships with others, and leading to the success of job tasks (Wang et al., 2014; Yamazaki, 2010). Building on the KBV, CSR engagement facilitates the development of expatriates’ CCC to a degree (Hafenbrädl and Waeger, 2017; Johnson et al., 2006; Yamazaki and Kayes, 2004). Since CSR activities require individuals to interact with foreign market stakeholders, expatriates are forced to adapt to the new environment and develop CCC to cope with daily struggles to achieve work–life balance (Caligiuri and Tarique, 2012; Wang et al., 2014). In addition, as CSR activities are often involved in long-term projects, challenges generated by CSR tasks provide more incentives and pressures for expatriates to develop CCC (Yamazaki and Kayes, 2004). Given the arguments above, the following hypothesis is proposed:

Hypothesis 2: Overseas CSR practices are positively related to the CCC of expatriates from Chinese multinationals.

IC and CCC

As the KBV underlines the importance of knowledge heterogeneity and capabilities in sustaining competitive advantages of a firm, this view has been widely adopted in emerging market research due to the intense knowledge-seeking motivations of global market latecomers (Chen and Tan, 2016; Fan et al., 2016). Scholars have contended that, for Chinese multinationals, embedding in host countries via CSR practices can help them gain wider local knowledge, which is essential for enabling multinationals to overcome the liability of foreignness and of country of origin (Li and Hambrick, 2005; Yin and Zhang, 2012). Some have also argued that CSR entails conformance with local laws and firms embedded in social relations, enhancing the firms’ business legitimacy in follow-up commitment and expansion (Gorondutse and Hilman, 2016; Husted et al., 2016). While most studies have emphasized the organizational-level influences of increasing IC in foreign countries, the individual-level influences have been much less studied (Shea and Hawn, 2019).

This study advocates that the individual expatriates can better develop CCC to deal with cross-cultural tasks related to increased market commitment. First, the enhanced local market cooperation and available resources allow expatriates to have greater bargaining power with local market stakeholders in negotiations and communications (Wang et al., 2014; Yamazaki and Kayes, 2004). Second, increasing their market commitment allows expatriates to reduce costs for learning local knowledge and lessens the liability of outsidership that might hinder expatriates from seeking complementary knowledge (Li and Hambrick, 2005). Third, expatriates can develop more difficult-to-codify skills that are essential for cross-cultural management, because increasing local stakeholders’ recognition facilitates them in accessing more implicit knowledge (Wang et al., 2014). Taken all together, if the Chinese multinationals increase their commitment to a foreign country, expatriates are able to improve CCC due to the increased local market interactions and learning benefits. Hence:

Hypothesis 3: IC is positively related to the development of expatriates’ CCC.

IC and achieving organizational and individual goals

While Chinese multinationals engaging themselves in CSR activities can better access local knowledge and learning opportunities, how to utilize their commitment to achieve both organizational and individual goals remains in doubt. It is important to note that overwhelmingly the literature in the CSR field tends to use financial indicators to measure the outcomes of multinationals’ local market engagement, such as profitability, productivity, or accounting gains (Young and Makhija, 2014). However, short-term financial indicators might not be able to fully capture the outcomes of long-term and loosely coupled CSR activities (Orlitzky et al., 2003; Tang et al., 2012). Hence, this study adopts a long-term view and cross-level perspectives to measure commitment-related outcomes. Generally, committing to foreign markets via CSR demands knowledge transfer between HQs and subsidiaries (Yang et al., 2008). In particular, RKT as an organizational-level outcome can proxy the efficiency of a subsidiary utilizing commitment benefits to improve negotiation power and strategic position at HQs (Torugsa et al., 2013).

In the literature, RKT is defined as implicit and explicit knowledge transfer from foreign subsidiaries to the HQs. It can be measured by the variations of knowledge reserves of the recipient unit, that is, the HQs (Chang et al., 2012). Based on the KBV, committed firms and individuals in foreign markets is important for promoting subsequent RKT. Through increasing IC, Chinese firms can gain access to wider knowledge bases in host countries, which facilitate market latecomers in acquiring additional new knowledge (Nadkarni and Perez, 2007). Embedding in local markets can also assist firms to involve themselves in more interactions and learning opportunities with local stakeholders, especially when the subsidiaries are located in those countries framed as “pockets of innovation” (Kiessling et al., 2016). This provides greater knowledge sources for expatriates in order to proceed with RKT. Thus, the following hypothesis is proposed:

Hypothesis 4a: IC is positively related to the RKT of Chinese multinationals.

It is also necessary to consider the impacts of commitment variations on expatriates’ performance (EPs), since, in the literature, organizational achievements in various business strategies are argued to be formed by the actions of individuals (Vlajčić et al., 2019; Zhu et al., 2018). In general, increasing market commitment allows expatriates to obtain more recognition from local stakeholders, and consequently reduce their psychological pressures in foreign market adaptations (Yamazaki and Kayes, 2004). Along with increasing commitment, expatriates are provided with greater support to deal with cross-cultural conflicts and language barriers (Kraimer et al., 2001). In addition, since individual performances are often considered to be related to their surrounding environments, the CSR-related interactions generated by market embeddedness procedures empower expatriates to learn new knowledge and skills to cope with their daily struggles, such as work–life balance (Yamazaki, 2010; Yamazaki and Kayes, 2004). In this regard, expatriates who can adapt efficiently to foreign environments in the CSR implementation process can perform better in fitting into organizational demands and environmental requirements. Accordingly, the following hypothesis is proposed:

Hypothesis 4b: IC is positively associated with the EPs.

CCC and achieving organizational and individual goals

Several studies have analyzed a broad range of managerial influences of CCC, such as global leadership, learning, and environmental adaptation (Caligiuri and Tarique, 2012; Yamazaki and Kayes, 2004), inter alia. However, the effects of CCC in achieving RKT have not been properly measured (Vlajčić et al., 2019). Since scholars have identified cross-border knowledge transfer as one of the key sources for latecomers for enhancing competitive advantages (Vlajčić et al., 2019), it is worth discussing how expatriate capability affects the efficiency of RKT. As suggested by Pedersen et al. (2003), the function of RKT from host country to home country is the key to distinguishing domestic firms and multinationals. In particular, transferring tacit knowledge requires the efforts of expatriates, as tacit understandings are uncodified and can only be obtained through individuals’ direct interactions in host environments (Pedersen et al., 2003): that is, the individuals’ CCC largely determines the efficiency of diffusion, creations, and transfer of knowledge of Chinese multinationals (Chang et al., 2012).

In the literature on knowledge transfer, codification has been recognized as the key aspect for promoting RKT (Vlajčić et al., 2019). It is part of a communication model and requires individuals to transmit local messages to the receiver (i.e. HQs) (Cowan et al., 2000). However, despite the uses of codebooks, expatriates should understand the underlying meanings and procedures of applying the acquired new knowledge from overseas (Cowan et al., 2000). In fact, CCC empowers expatriates to interpret local knowledge and enables them to effectively reverse-transfer the context-dependent knowledge back to their home countries (Johnson et al., 2006; Vlajčić et al., 2019). Moreover, expatriates with CCC can not only distinguish complementary knowledge from the local market, they are also able to transform overlapping knowledge into unique ideas by combining Western and Eastern cultures, thereby magnifying the learning benefits that result in RKT being better achieved (Li and Fleury, 2020; Yamazaki, 2010). Hence, a hypothesis is proposed as follows:

Hypothesis 5a: Expatriates’ CCC is positively related to RKT of Chinese multinationals.

Directly related to the individuals’ goal, CCC positively affects expatriates’ work performance for the following reasons. First, prior studies show that cross-cultural intelligence allows expatriates to acclimatize to cross-cultural settings and unfamiliar environments, and thereby improve their communication skills for interacting with foreign stakeholders (Caligiuri and Tarique, 2012; Johnson et al., 2006; Pedersen et al., 2003). Second, new stimuli from host countries can be utilized by individuals in generating higher productivity at work (Vlajčić et al., 2019). Chinese expatriates with higher levels of CCC may manage any ambiguity they may encounter in the process of CSR activities, and therefore they may respond more efficiently to the novel, different, and unpredictable problems that might happen in CSR activities (Caligiuri and Tarique, 2012; Yamazaki, 2010). Third, with better adaptability, expatriates can better deal with stress imposed by CSR requirements, maintain work–life balance, and consequently perform efficiently in completing their international assignments (Caligiuri and Tarique, 2012; Wang et al., 2014). Hence, the following hypothesis is proposed:

Hypothesis 5b: Expatriates’ CCC is positively associated with their performance.

Methodology

Sample and data collection

A survey-oriented quantitative method is adopted in this study to test the proposed integrative model and the hypotheses associated with it. The matched survey data come from both the expatriates and HQs managers of Chinese multinationals. Chinese firms are targeted in this study because these firms often face great challenges in establishing trust relationships and obtaining legitimacy in foreign countries (Tashman et al., 2019; Zhu et al., 2018). Moreover, there is a growing global awareness of the social and environmental impacts of Chinese firms’ international assignments around the world (Marano et al., 2017; Wang et al., 2014). The surveys were distributed and collected between 2015 and 2016, with 150 valid responses (a 37.6% response rate).

While there is no public list of all Chinese expatriates and managers (Wang et al., 2014), this study randomly selected 495 expatriates and 103 HQs managers from Chinese-affiliated firms and from multiple industries. Two sets of surveys (targeted at expatriates and their managers) were designed and sent to participants after contacting the HR departments in the firms’ overseas divisions. The major parts of the survey questions were designed using a five-point Likert-type scale, ranging from “1” (strongly disagree) to “5” (strongly agree). Some questions were designed using only two options: “Yes” (coded 1) and “No” (coded 0).

The survey questions were translated into Chinese and back-translated to English, in order to prevent language-related misunderstandings. Of the 314 returned questionnaires, 225 were complete, with an overall response rate of 37.6%. After eliminating invalid responses, the final data set in this study contained 150 valid responses from 38 Chinese multinationals. Of these, most firms were in manufacturing (86.8%). Fifteen firms were state-owned and 26 were of relatively large size, with average net sales above 500 million RMB. The characteristics of firms and respondents are shown in Appendix 1, Table 1A.

The data sources of the main constructs are from different sets of the surveys (responses from expatriates and from their managers) in order to avoid common method variance (CMV) (Podsakoff et al., 2003). For example, expatriates were asked to evaluate their CCC, and let their managers explain the level of IC. In particular, while the data of the two outcome constructs (RKT and EP) were both collected from the managers’ responses, evaluating EPs by their HQs managers can avoid CMV caused by respondents’ self-reported bias (Podsakoff et al., 2003).

Measures

Dependent variables

Following the measurement proposed by Chang et al. (2012), this study uses seven items to construct RKT as the firm-level outcome. In particular, the HQ managers were asked to measure the extent of new knowledge that they received from expatriation activities; such knowledge includes that related to new markets, new managerial and production techniques, and quality management procedures and manuals (Chang et al., 2012; Yang et al., 2008). The Cronbach’s α value of RKT is 0.882, indicating a high level of reliability of the construct (Lu et al., 2010).

To better understand the consequences of CSR adoption, EP is measured as the individual-level outcome variable. Following the measurement proposed by Bader et al. (2015), this study uses five items to construct EP. The managers were asked to evaluate EPs based on goal achievements, quality of work results, teamwork performance, and time management efficiencies (Bader et al., 2015; Kraimer et al., 2001). The mean value of the construct is 3.943, with reliability at a satisfactory level (Cronbach’s α = 0.822).

Explanatory variables

Since CSR practices can be represented by different dimensions of sustainable development (economic, social, and environmental), this study follows the scales proposed by Torugsa et al. (2013) to measure different dimensions of CSR as a latent construct. The economic dimension of CSR measures to what degree a firm can integrate with local customers, suppliers, and stakeholders and act on economic responsibility concerns to solve profit-maximizing issues (Torugsa et al., 2013; Young and Makhija, 2014). These dimensions can reflect the extent to which the Chinese firms can achieve long-term economic performance and contribute to the improvement of local production efficiencies (Torugsa et al., 2013). The economic dimension is measured by eight items (Cronbach’s α = 0.736; see Appendix 1, Table 1B). To avoid variations of the impact of each CSR dimensions on performance outcomes, the three dimensions are aggregated into the “CSR practices” variable for the SEM analysis.

Prior studies have suggested that Chinese multinationals experience difficulties in generating social-related CSR (Marano et al., 2017; Torugsa et al., 2013). In particular, these firms have to incur additional costs and human resources in improving employees’ satisfaction and enhancing the local community’s benefits (Torugsa et al., 2013). In this study, the measurements proposed by Torugsa et al. (2013) are adopted in order to identify whether the firm can act as a good citizen in the host country. The construct consists of eight items on a five-point scale (Cronbach’s α = 0.827; see Appendix 1, Table 1B).

The environmental dimension of CSR adoption is used to examine whether the environmental impacts of the firm’s production are effectively monitored and systematically managed (Fiaschi et al., 2017; Henriques and Sadorsky, 1999). As Chinese firms are expanding increasingly in cross-border activities, their environmental performance has come under close scrutiny by local stakeholders (Fiaschi et al., 2017; Tashman et al., 2019). Following the existing measurement scales (see Appendix 1, Table 1B), the participants were asked to answer 10 questions regarding their environmental and ecological awareness (Fiaschi et al., 2017; Torugsa et al., 2013). The reliability of this construct reports as high (Cronbach’s α = 0.929).

IC is measured using the scales provided by Gregersen and Black (1992). The expatriates’ managers were asked to answer five questions related to the degree of investment in the project, the budget plans, and whether they will continue to discover more business opportunities in the local market, and so on. The construct of IC consists of five items on a five-point scale (mean value = 3.872, Cronbach’s α = 0.704). Likewise, CCC is calculated based on the measurement proposed by Leslie et al. (2002). The expatriates were asked to evaluate their capability developments in the foreign markets. Five items were used to construct CCC, which yields a high level of reliability (Cronbach’s α = 0.878). The details of each constructs are shown in Appendix 1, Table 1B.

Control variables

Several control variables are included in this study to avoid biased results. At country level, Target Country (dummy variable) and Market Size are used to control for host-country variances. Specifically, 35 dummy variables were used to control for the differences of target countries, and market size of the target country was calculated by the natural logarithm of gross domestic product. At industry level, the Industry dummies for 19 industries were included in the data set to control for industry impacts on Chinese firms’ CSR adoption and the outcome variables (Peng and Beamish, 2014).

At individual level, the age of expatriates (Mellahi and Guermat, 2004), education background (Jokinen et al., 2008; Suutari and Brewster, 2000), marital status (Stahl et al., 2002), and the expatriates’ international experience (Inkson et al., 1997; Selmer, 2002) were controlled. The age of expatriates was a control variable because young expatriates often face more pressure than seniors do in adjusting to fit into the foreign country (Mellahi and Guermat, 2004). The educational background of expatriates could also affect their learning capabilities (Jokinen et al., 2008; Suutari and Brewster, 2000). The survey participants were asked to answer whether they had educational experience in a foreign country or not (“yes” coded “1,” “no” coded “0”).

In the demographic characteristics of participants, their marital status was controlled. In expatriate research, spouses and families play predominant roles in affecting the expatriate’s decision-making and the level of job satisfaction and work performance (Stahl et al., 2002). Hence, expatriates’ marital status was controlled. The participants were asked to indicate whether they had a spouse or not (“1” if the answer was yes, “0” otherwise). Expatriates’ overseas experience can also be used to support the development of their global mindset and job performance (Selmer, 2002). In order to complete international assignments, expatriates are not only required to have rich knowledge of their HQ’s strategy and goals but also need to have the ability to work and live in a foreign country (Inkson et al., 1997). With international experience, individuals can adjust themselves more efficiently and adapt to the new work environment (Inkson et al., 1997; Selmer, 2002).

CMV and validity tests

Because data for the main constructs were collected from the two different sets of surveys (one for expatriates and one for HQs managers), and some of the constructs were not self-reported (RKT and EP), the empirical tests in this study are not susceptible to CMV issues (Chang et al., 2012; Podsakoff et al., 2003). Moreover, potential response bias was checked by conducting t-tests to compare the differences between early and late responses (Podsakoff et al., 2003). The results show that there is no statistically significant difference between the two groups. Moreover, in order to assess item validity and reliability of the constructs, further tests were conducted using different reflective indicators, including convergent validity, discriminant validity, and item reliabilities (Hulland, 1999; Podsakoff et al., 2003). First, the results show that all factor loadings of the constructs are greater than 0.70. Hence, evidence for convergent validity is presented by observing the Cronbach’s α and the coefficients of construct reliability (Podsakoff et al., 2003). Second, the values of average variance extracted (AVE) are significantly greater than 0.50. The correlations between each of the constructs are presented in Table 1 and are used to compare with the square root of AVE (Hulland, 1999). Discriminant validity is confirmed, because the square root of AVE is larger than the correlations between each pair of constructs (Hulland, 1999).

Results

Table 1 presents the descriptive statistics between the variables. It shows that the correlations between different dimensions of CSR and RKT are 0.254 (Economic dimension), 0.195 (Social dimension), and 0.248 (Environmental dimension). The correlations between all the dimensions of CSR and EP are also less than 0.50. Similarly, IC and CCC show different degrees of correlation with the predictors and outcome variables, and most of them are less than 0.50. Hence, the multicollinearity issues of the variables may not be a significant issue. The variance inflation factor (VIF) method is used to test for multicollinearity between the variables (Hair et al., 1998), and the result indicates that all the VIF values are less than 4 (significantly below the critical value of 10). Therefore, multicollinearity is not a serious concern in this study.

Table 1.

Descriptive analysis and correlations.

	Mean	SD	VIF	1	2	3	4	5	6	7	8	9	10	11	12	13	14
1. Reverse knowledge transfer	3.393	0.666	1.260	1
2. Expatriate’s performance	3.943	0.435	1.480	0.238**	1
3. CSR (economic)	3.781	0.351	3.510	0.254**	0.456**	1
4. CSR (social)	4.019	0.426	1.980	0.195*	0.402**	0.612**	1
5. CSR (environmental)	4.039	0.518	2.890	0.248**	0.298**	0.732**	0.623**	1
6. International commitment	3.872	0.473	1.610	0.148	0.125	0.558**	0.311**	0.443**	1
7. Cross-cultural competence	3.887	0.589	1.230	0.390**	0.294**	0.272**	0.264**	0.317**	0.187*	1
8. Target country	9.093	7.581	1.890	−0.027	−0.069	0.122	−0.096	−0.093	0.223**	−0.023	1
9. Market size	6.685	2.481	1.800	−0.024	0.022	−0.091	0.110	0.066	−0.185*	−0.003	−0.631**	1
10. Industry	6.393	5.381	1.180	0.158	0.148	0.179*	0.201*	0.035	−0.008	0.065	0.067	0.110	1
11. Expatriate age	32.696	6.504	1.180	0.034	0.145	0.057	0.049	0.114	0.006	0.191*	−0.049	0.001	0.084	1
12. Expatriate education	0.213	0.411	1.070	−0.119	0.039	−0.063	−0.019	−0.043	0.032	−0.060	−0.039	0.083	−0.072	0.017	1
13. Marital status	0.147	0.355	1.200	0.006	0.020	−0.097	−0.108	−0.148	−0.068	0.112	−0.005	−0.059	0.071	0.310**	0.060	1
14. International experience	0.260	0.440	1.120	0.090	0.057	0.170*	0.058	0.131	0.123	0.094	0.218**	−0.188*	0.067	0.035	0.099	−0.074	1

Note: SD: standard deviation; VIF: variance inflation factor; CSR: corporate social responsibility.

* Correlation is significant at the 0.05 level (two-tailed).

** Correlation is significant at the 0.01 level (two-tailed).

To test the proposed hypotheses, the SEM method is adopted because it can reveal the causal relationships between the variables and at the same time control for measurement error (Anderson and Gerbing, 1988; Lu et al., 2010). The original model in this study was set to test the full mediation effects, with no direct paths from CSR to the outcome variables (Lu et al., 2010). Model 1 is also a full mediation test, but the direct paths from CSR to RKT and EP are added. From model 2 to model 5, the original model is revised in order to rule out of some of the alternative explanations. For example, the mediation effect of IC is tested in model 2, and the sequential mediation effect of CCC is tested in model 3. The full mediation effects of IC and CCC on the relationship between CSR and RKT are observed in model 4, and the mediation effects between CSR and EP are examined in model 5.

Model 6 presents the two-stage mediation effects, with the paths of “CSR → IC → CCC → RKT” and “CSR → IC → CCC → EP.” The rationale of this model reflects the idea that, through enhancing local commitment, expatriates can better develop their capability to deal with cross-cultural environments, and this will eventually influence their work performance and degree of RKT. In addition, model 7 only controls for the nonindividual-level variables, and model 8 only considers the individual-level controls.

Hypothesis testing

Figure 2 portrays the path estimates of the original model, with the control variables included. Hypotheses 1 and 2 state that Chinese multinationals engagement in overseas CSR practices will lead to an increase of IC, as well as to an improvement in their expatriates’ CCC. From the parameter estimates, the two paths, “CSR → IC” (β = 0.524, p < 0.01) and “CSR → CCC” (β = 0.236, p < 0.01), are statistically significant. These show that hypotheses 1 and 2 are accepted. Hypothesis 3 states that, through the increase of commitment in foreign markets, CSR adoption can enhance expatriate CCC development. This hypothesis is accepted based on the parameter estimates in Figure 2 (β = 0.255, p < 0.1).

Hypotheses 4a and 4b argue that the degree of IC of Chinese multinationals will positively mediate the relationships of “CSR → RKT” and “CSR → EP.” Figure 2 shows that hypothesis 4a can be accepted with a relatively weak significance level (β = 0.174, p = 0.086), whereas hypothesis 4b can be strongly accepted (β = 0.212, p < 0.01), indicating that an increase of commitment to the foreign markets can significantly improve EPs. Hypotheses 5a and 5b propose that CCC can positively mediate the relationship between CSR-RKT and CSR-EP. The SEM results shown in Figure 2 indicate that, through the enhancement of expatriates’ CCC, individuals’ work performance and the efficiency of RKT will be increased. Therefore, both hypotheses 5a (β = 0.332, p < 0.01) and 5b (β = 0.300, p < 0.01) are accepted.

Figure 2.

SEM results. SEM: structural equation modeling.

Alternative models and robustness check

To exclude other possible explanations and check for the robustness of the findings, this study compares different alternative models and follows the bootstrapping procedure to create a new data set (Lu et al., 2010). Specifically, a variety of alternative models are constructed in Table 3 and to retest the hypotheses in Table 4. First, the direct paths from CSR to RKT and EP are included in model 1, and the χ ² difference tests presented in Table 3 show that the original model fits the data better. Second, although IC can mediate the relationship between the predictor and outcomes, CCC might not have direct influence on the relationships. To exclude this possibility, model 2 is constructed, and the χ ² difference test shows that the original model, (Root mean square error of approximation, RMSEA = 0.064) fits the data better than model 2 (RMSEA = 0.068, $Δ χ_{(b, m 2)}^{2}$ = 31.597, p < 0.01). Similarly, model 3 is designed not to include the mediation effect of IC in that the paths of “CSR → IC → CCC,” “CSR → IC → RKT,” and “CSR → IC → EP” are constrained to zero. The χ ² test shows that the original model fits better than model 3 (RMSEA = 0.065, $Δ χ_{(b, m 3)}^{2}$ = 12.674, p < 0.01). Therefore, this possibility has been ruled out.

Table 3 also demonstrates the models with single outcome variables only, that is, either paths to RKT (model 4) or to EP (model 5), are constrained to zero. As is given in Table 3, the model fit of model 4 (RMSEA = 0.067) and model 5 (RMSEA = 0.075) is decreased and the χ ² difference tests also indicate that the original model fits better than model 4 ( $Δ χ_{(b, m 4)}^{2}$ = 226.003, p < 0.01) and model 5 ( $Δ χ_{(b, m 5)}^{2}$ = 97.740, p < 0.01). Therefore, these potential alternative models are excluded (see Table 2).

Table 2.

Results of alternative models.

Model	χ ²	$Δ χ^{2}$	RMSEA	RMSEA (90% CI)	AIC	CFI	SRMR
Original model	652.100	—	0.064	0.055 (0.073)	961.100	0.874	0.076
Model 1	651.646	$Δ χ_{_{(b, m 1)}}^{2}$ = 0.454***	0.064	0.055 (0.073)	965.646	0.873	0.076
Model 2	683.697	$Δ χ_{_{(b, m 2)}}^{2}$ = 31.597***	0.068	0.059 (0.076)	989.697	0.859	0.101
Model 3	664.774	$Δ χ_{_{(b, m 3)}}^{2}$ = 12.674***	0.065	0.056 (0.074)	968.774	0.869	0.084
Model 4	464.873	$Δ χ_{_{(b, m 4)}}^{2}$ = 226.003***	0.067	0.056 (0.077)	714.873	0.890	0.075
Model 5	426.097	$Δ χ_{_{(b, m 5)}}^{2}$ = 97.740***	0.075	0.063 (0.086)	660.097	0.858	0.077
Model 6	665.149	$Δ χ_{_{(b, m 6)}}^{2}$ = 13.094***	0.065	0.056 (0.074)	969.149	0.869	0.086
Model 7	511.197	$Δ χ_{_{(b, m 7)}}^{2}$ = 140.903***	0.065	0.054 (0.075)	749.197	0.896	0.756
Model 8	548.278	$Δ χ_{_{(b, m 8)}}^{2}$ = 103.822***	0.064	0.054 (0.074)	789.278	0.885	0.757

Note: CSR: corporate social responsibility; RKT: reverse-knowledge transfer; EP: expatriate performance; CCC: cross-cultural competence; IC: international commitment. Original model: full mediation (no direct paths from CSR to RKT and EP); RMSEA: root mean square error of approximation; CFI: comparative fit index; AIC: information criterion; SRMR: standardized root mean squared error. Model 1: full mediation (with direct paths from CSR to RKT and EP). Model 2: the mediation effects of CCC were constrained to zero, including the paths of “CSR → CCC → RKT” and “CSR → CCC → EP.” Model 3: the mediation effects of IC were constrained to zero, including “CSR → IC → CCC,” “CSR → IC → RKT,” and “CSR → IC → EP.” Model 4: only observe the paths of predictors to RKT. Model 5: only observe the paths of predictors to EP. Model 6: two-stage mediation model, only observe the paths of “CSR → IC → CCC → RKT” and “CSR → IC → CCC → EP,” other mediation effects were constrained to zero. Model 7: only control for non-expatriate-level variables, including “Target Country,” “Market Size,” and “Industry.” Model 8: only control for the expatriate-level variables, including “Expatriate Age,” “Expatriate Education,” “Marital Status,” and “International Experience.”

*p < 0.05; **p < 0.01; ***p < 0.001.

Table 3.

Alternative SEM models for hypothesis testing.

Hypothesis testing		Original model	Model 1	Model 2	Model 3	Model 4	Model 5	Model 6	Model 7	Model 8
H1	CSR → IC	0.524***	0.515***	0.531***	0.582***	0.527***	0.525***	0.511***	0.525***	0.479***
H2	CSR → CCC	0.236**	0.226*	0.418***	0.443***	0.241**	0.235**	Constrained	0.248**	0.209**
H3	CSR → IC → CCC	0.256*	0.261**	0.286**	Constrained	0.254*	0.253*	0.438***	0.244*	0.226*
H4a	CSR → IC → RKT	0.171*	0.136	0.313**	Constrained	0.158*	Constrained	Constrained	0.165	0.146
H4b	CSR → IC → EP	0.200***	0.198**	0.392***	Constrained	Constrained	0.197**	Constrained	0.185**	0.202***
H5a	CSR → CCC → RKT	0.333***	0.326***	Constrained	0.374***	0.328***	Constrained	Constrained	0.351***	0.369***
H5b	CSR → CCC → EP	0.299***	0.298***	Constrained	0.339***	Constrained	0.295***	Constrained	0.283***	0.301***
	CSR → IC → CCC → RKT							0.375***
	CSR → IC → CCC → EP							0.340***
df		405	403	407	408	280	233	408	315	339
Δdf			2	2	3	125	172	3	90	66
p-Value		0	0	0	0	0	0	0	0	0

Note: SEM: structural equation modeling; CSR: corporate social responsibility; IC: international commitment; CCC: cross-cultural competence; RKT: reverse-knowledge transfer; EP: expatriate performance.

*p < 0.05; **p < 0.01; ***p < 0.001.

Table 4.

Hypothesis testing: Bootstrap result (n = 1000).

Hypotheses	Paths	Bootstrapping				95% CI
Hypotheses	Paths	Significance level	p-Value	CR	SE	Lower	Upper
H1	CSR → IC	***	<0.001	5.606	0.093	0.509	0.539
H2	CSR → CCC	**	0.054	1.921	0.123	0.216	0.256
H3	CSR → IC → CCC	*	0.062	1.866	0.137	0.233	0.277
H4a	CSR → IC → RKT	*	0.086	1.716	0.101	0.158	0.190
H4b	CSR → IC → EP	***	0.007	2.674	0.079	0.199	0.225
H5a	CSR → CCC → RKT	***	<0.001	3.569	0.093	0.317	0.347
H5b	CSR → CCC → EP	***	<0.001	4.268	0.070	0.289	0.311

Note: SE: standard error; CSR: corporate social responsibility; IC: international commitment; CCC: cross-cultural competence; CR: critical ratio; RKT: reverse-knowledge transfer; EP: expatriate performance.

*p < 0.05; **p < 0.01; ***p < 0.001.

Model 6 is constructed to check for the robustness of hypothesis 5, in which all the paths excluded, “CSR → IC → CCC,” “CSR → IC → CCC → RKT,” and “CSR → IC → CCC → EP,” are constrained to zero. The results presented in Table 2 show that the model fit is acceptable (RMSEA = 0.065). In Table 3, the path of “CSR → IC → CCC → RKT” (β = 0.375, p < 0.001) and the path of “CSR → IC → CCC → EP” (β = 0.340, p < 0.001) show a high significance level to reject the null hypothesis. The results presented in Figure 3 show that, through the increase of IC, expatriates can better develop their cross-cultural skills. This is consistent with the arguments in the literature suggesting that, via interacting and communicating with the local markets, expatriates can adjust themselves more efficiently and maintain a good work–life balance (Chang et al., 2012).

Figure 3.

Alternative two-stage mediation results.

Model 7 only includes the nonindividual-level control variables: Target Country dummies, Market Size, and Industry dummies. Similarly, Model 8 only considers the individual-level control variables to test the hypotheses. The model fit results (RMSEA = 0.065) are similar to the original model. The results in Table 3 show that the direct effects of CSR on the two outcome variables remain insignificant. In addition, the impacts of IC and CCC remain unchanged. Therefore, the robustness of the empirical findings in this study is guaranteed.

This study also resamples the data set and follows the bootstrapping procedure (1000 times) to check the path coefficient, using the maximum likelihood method (MacKinnon et al., 2004). It is suggested that, if the empirical results are not robust, using the bootstrapping technique may change the results, because it allows the non-normality part of the data to be estimated (MacKinnon et al., 2004). Table 4 presents the bootstrapping results, which show that all the empirical findings in Figure 2 remain the same. Based on the path estimates of the best-fitting model, it can be confirmed that the mediating effects of IC and CCC in the relationship of CSR-RKT and CSR-EP are significant. In summary, all the hypotheses proposed in this study can be accepted.

Discussion

This study investigates theoretically the CSR practices of Chinese multinationals and empirically depicts the learning paths of expatriates through a KBV. Overall, this study finds that the learning impact of overseas CSR adoption on Chinese firms manifests itself in two ways. First, the Chinese firms’ engagement in overseas CSR activities is positively associated with their IC, such that expatriates can achieve more learning opportunities through knowledge interactions and engage themselves in building a trust relationship with local market stakeholders. Second, expatriates’ CCC can be improved with an increase in the degree of host country embeddedness, which further leads to better knowledge absorptive outcomes. Ceteris paribus, the sequential mediation of firm-level IC and individual-level CCC exerts positive impact on the subsidiary’s RKT and the EPs. Overall, three theoretical contributions emerge from the findings in this study.

Implications for theories

The first contribution is in incorporating a KBV in CSR research. Although the KBV is of great influence in IB research (Kogut and Zander, 2003; Nadkarni and Perez, 2007), it remains underexplored in the CSR research field. Since overwhelmingly CSR studies tend to focus on institutional theory (Jiang et al., 2018; Marano et al., 2017) and stakeholder perspectives (Henriques and Sadorsky, 1999) to examine the surrounding influences of operating CSR projects abroad, this study argues that the knowledge-seeking motivation of EMNEs in their foreign investment projects should also be highlighted. The knowledge conduits between home-and host-country knowledge resources (i.e. expatriates) are important in facilitating overseas knowledge seeking and transfer (Chang et al., 2012). This study adopts the KBV to investigate the expatriates’ knowledge interactions during the CSR process and reveal the knowledge-seeking motivations of EMNEs in CSR research (Johnson et al., 2006; Wang et al., 2014). The initial attempt of this study paves the way to examine how knowledge-related mechanisms matter for successful CSR outcomes.

In the second contribution, this study explores the micro-foundations of CSR activities and their consequences (e.g. RKTs and EP). Findings show that the mediation effects of expatriates’ CCC on the subsidiary’s RKT and individuals’ performance can be enhanced when the firms engage themselves in local market interaction via an increase of commitment. Catering to the recent call for micro-foundations analysis in the CSR field (Han et al., 2019; Shea and Hawn, 2019), the present study investigates individual-level efforts in dealing with CSR-related complexity. Results of this study highlight the importance of expatriates in identifying appropriate knowledge sources in host countries and of their CCC in aggregating dispersed information during the overseas CSR process. More importantly, the expatriate interactions with local markets in CSR practices not only imply knowledge acquisition but spur more knowledge that leads to the individual- and firm-level outcomes. Hence, future studies are encouraged to explore deeply the micro-foundations of CSR, to examine the complex boundary conditions, such as the individual characteristics or preferences affecting the CSR–performance link, and to answer questions such as how employees, founders, stakeholders’ competence, judgments, and preferences influence the outcomes of CSR.

The third contribution goes beyond the existing CSR literature to uncover the learning paths of Chinese multinationals in the local market embedding process. It is important to note that, despite well-documented economic benefits of global latecomers conducting CSR overseas, few studies have investigated whether the firms can benefit from the knowledge-spillover and learning advantages in the implementation of environmentally and socially oriented business practices (Han et al., 2019). This study reveals that the way individuals interact and engage in CSR practices may lead to added learning outcomes that are hard to predict based on organizational-level behavior. In fact, the individual-level differences in knowledge seeking and learning are likely to convert into perceptible CSR differentials in an organizational-level analysis (Han et al., 2019). Findings, therefore, contribute to the limited but growing literature by pointing to the underlying learning mechanisms of CSR. This study also suggests that the nascent work on CSR-related learning mechanisms points to the potential for further research. Therefore, future studies in this field are encouraged to explore more individual-level learning indicators of CSR practices and how these indicators affect cross-level outcomes of CSR.

Implications for practice

Findings in this study also provide some practical implications for improving the CSR practices of Chinese multinationals. Specifically, Chinese multinationals should pay more attention to their degree of commitment in foreign markets. While prior studies often suggest that firms should be more cautious in considering an increase of commitment in a foreign market, especially for those subsidiaries located in distant markets (Gregersen and Black, 1992), this study shows that commitment can be an efficient trust-enhancing tool for firms to build up better relationships with local market stakeholders. In this regard, Chinese multinationals need to investigate discovering the “appropriate” level of commitment for penetrating a new business environment and yielding better CSR outcomes.

This study demonstrates that individuals’ CCC is essential for improving learning efficiencies during overseas CSR. Given the essential role of individuals’ learning effort for generating innovation and better performance outcomes, multinationals and their managers should encourage employees to adjust themselves and interact more with foreign market stakeholders (Caligiuri and Tarique, 2012; Yamazaki and Kayes, 2004). MNE managers should develop new training systems for expatriates to progress CCC, both in the predeparture period and in the post-departure period (Zhu et al., 2018). Employees need to engage themselves in communicating with local communities and participate more in host-country cultural activities in order to learn more about the new business system and improve their psychological adjustment.

Limitations and future research directions

There are several limitations in this study that point to avenues for future research. First, it neglects to pay attention to the target country of international assignments. However, understanding the differences of host countries may generate more insights into the foreign operations and performance differences of multinationals (Makino et al., 2004; Witt et al., 2018). In particular, the worldwide variety of business systems can lead to adaptation difficulties of multinationals and an organization–environment misfit during international assignments (Witt et al., 2018). In addition, the host country’s institutional diversity and business climate can also affect the firm’s operational efficiencies. The formal regulations and informal rules in host countries can directly influence multinationals’ participation in CSR and determine their access to local knowledge and resources (Makino et al., 2004). Hence, future studies are encouraged to consider the divergence of business systems and institutional characteristics of the target countries.

Second, CSR practices are measured as a second-order construct by considering economic, social, and environmental aspects of practices. Although this measure is shared with some prior studies (e.g. Torugsa et al., 2013; Young and Makhija, 2014), the design and measures contain a potential limitation that ignores expatriates’ various paths of international experience. That is, expatriates’ experience can be path-dependent, which means their foreign experiences are affected by the global business system of the host country in which these expatriates are accustomed to work and live. Future studies can further discuss the possibility by disaggregating specific CSR practices and testing how these more narrowly defined CSR practices affect RKT and EP.

Third, the duration of international assignments has been ignored when designing the survey questions. However, assignment duration is an important component in expatriate research. For example, expatriates’ willingness to accept new assignments will change if their perceived duration of stay in a foreign country is changed. Those multinationals that have international assignments with relatively short duration may also have less incentive to increase local commitment and may engage less in social interaction with their external stakeholders (Gregersen and Black, 1992), resulting in variations of CCC cultivation and learning outcomes. Future studies in this field should include duration variable in empirical tests and compare the impact of long- and short-term assignments on CSR outcomes.

Fourth, there are no significant time lags in data collection in this study, which means that the temporal ordering or reverse effects of the predictors and outcome variables are not allowed to be tested (Torugsa et al., 2013). Although this study uses the SEM technique to reduce the risks of generating a reverse causality problem, the sequential mediation effects should be validated in future studies by using a longitudinal research design. In particular, setting time intervals to collect data can help us better observe the performance consequences of a specific strategy implemented for a period of time, especially for CSR adoption that may take a longer time to become effective (Torugsa et al., 2013). Hence, future studies can seek to use different methods (e.g. Heckman models) to test the hypotheses in this study in order to mitigate the omitted variable bias (Certo et al., 2016).

Last, this study is limited to a Chinese context so that the results may not be applicable to multinationals from other emerging countries. This is not only because of the heterogeneity among the EMNEs but also owing to the various home-country policies that affect the firms’ international behavior (Luo et al., 2017; Ramamurti, 2012). For example, Chinese firms are motivated by the BRI policies to engage in global network building and improve their global image (Zhu et al., 2020). Subjected to the influence of contextual effects, managers and expatriates from various countries may have different preferences in performing international assignments. Hence, future research can extend the present research framework with larger sample sizes and more comprehensive variables or adopt a comparative research design to include samples from different emerging countries and investigate more context-specific CSR practices of emerging market multinationals.

Conclusion

This study examines the micro-foundations of CSR research through a KBV. By focusing on the expatriates’ role in facilitating CSR activities, this study empirically demonstrates the individual learning paths in CSR practices and the outcomes of CSR at both the organizational level (i.e. RKT) and individual level (i.e. EP). The results show that increasing commitment to foreign markets via CSR involvement helps expatriates improve their CCC, which can lead to better cross-level outcomes. Overall, this research contributes to the limited but growing literature that goes beyond organizational-level analysis in the CSR field and enriches the theoretical lens of CSR research by incorporating the KBV and depicting the learning paths of emerging market multinationals.

Footnotes

Authors’ note

We would like to thank editor, Professor Teck Yong Eng and two anonymous reviewers for their insightful suggestions.

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) disclosed receipt of the following financial support for the research, authorship and/or publication of this article: The second author thanks financial support from National Natural Science Foundation of China (Grant ID: 71972148) and Ministry of Education, PRC (Grant ID: 18YJA630097).

ORCID iDs

Sihong Wu

Di Fan

Appendix 1

Table 1B.

Construct measurement.

Constructs	Factor loadings	t-Value	α	AVE
Reverse Knowledge Transfer (Chang et al., 2012; Yang et al., 2008)			0.882	0.804
HQs can gain new knowledge about the new markets.	0.704	12.089
New managerial techniques.	0.676	12.975
New production and technical knowledge.	0.704	11.677
Skills for product development.	0.824	7.887
International human resource management knowledge.	0.772	9.538
Quality management procedural and technical manuals.	0.845	8.550
Knowledge of safety management practices.	0.819	9.139
Expatriate’s Performance (Bader et al., 2015)			0.822	0.769
Expatriates can help to achieve the foreign investment goals.	0.903	11.883
Expatriates have high quality of work results.	0.882	10.777
Expatriates have good time management and high productivity.	0.752	10.741
Expatriates can provide work assistance to their colleagues.	0.800	10.546
Expatriates are overall good at protecting the interests of their HQs.	0.731	10.909
CSR Practices (Economic) (Jenkins, 2004; Torugsa et al., 2013)			0.736	0.799
The company works with the government officials to protect its interests.	0.833	16.385
Adopts long-term perspective to sustain cash flow and shareholder benefits.	0.672	17.169
The company try to minimize production costs without affecting output.	0.812	17.127
Highlights the CSP for distinguishing products/process from others.	0.714	16.804
Sells waste to generate revenue.	0.527	16.614
Obtained product quality certification such as ISO 9000.	0.417	16.885
Works on building an industrial chain for social responsibility management.	0.826	17.002
Emphasizes the creation of spin-off technologies for by-products.	0.727	16.916
CSR Practices (Social) (Jenkins, 2004; Torugsa et al., 2013)			0.827	0.806
Employees to participate in the company’s decision-making process.	0.536	16.780
The company values work–life balance and family harmony.	0.796	16.869
Company invest in staff training and development.	0.611	16.882
Provides employees with equal opportunities in promoting senior positions.	0.905	16.839
Emphasizes the improvement of health and safety of its employees.	0.71	16.911
Actively engage in philanthropy.	0.765	16.978
Sponsors local community development.	0.658	16.907
Values the interests of all stakeholders when making investment decisions.	0.728	17.063
CSR Practices (Environmental) (Jenkins, 2004; Torugsa et al., 2013)			0.929	0.854
Adopts environmental auditing to evaluate production impacts.	0.646	16.895
Emphasizes ecological environment in procurement process.	0.812	16.858
Emphasizes the employees’ environmental awareness.	0.787	16.833
Emphasizes the controls of waste emissions.	0.825	16.922
Emphasizes the efficiencies of water recycles.	0.852	16.932
Emphasizes the efficiencies of energy utilization.	0.720	16.958
Minimizes the use of hazardous chemicals and finds less harmful alternatives.	0.690	16.981
Can separate and effectively treat hazardous waste.	0.702	17.020
Has obtained environmental certification such as ISO 14000.	0.795	16.416
Minimizes the environmental impacts during the entire product life cycle.	0.792	16.930
International Commitment (Gregersen and Black, 1992)			0.704	0.725
The multinational has invested significant resources in its overseas projects.	0.770	19.974
The multinational plans to operate in foreign markets for a long time.	0.710	20.442
The parent firm gives priority to foreign projects when making budget.	0.615	13.822
HQs plan to increase its investment in the foreign market.	0.817	15.576
HQs actively seek more business opportunities in the foreign market.	0.702	18.886
Cross-Cultural Competence (Leslie et al., 2002)			0.879	0.770
Trainings to work in a multinational environment.	0.868	19.262
Ability to motivate multicultural team members to work more efficiently.	0.879	18.364
Ability to work with team members with different cultural background.	0.823	16.301
Ability to exchange information with foreign employees.	0.747	16.073
Can change or update management techniques to fit in host country.	0.783	26.101	ã€€	ã€€

Note: CSR: corporate social responsibility; HQs: headquarters; AVE: average variance extracted.

References

Anderson

Gerbing

(1988) Structural equation modeling in practice: a review and recommended two-step approach. Psychological Bulletin 103(3): 411–423.

Antal

Sobczak

(2004) Beyond CSR: organisational learning for global responsibility. Journal of General Management 30(2): 77–98.

Asmussen

Fosfuri

(2019) Orchestrating corporate social responsibility in the multinational enterprise. Strategic Management Journal 40(6): 894–916.

Bader

Berg

Holtbrügge

(2015) Expatriate performance in terrorism-endangered countries: the role of family and organizational support. International Business Review 24(5): 849–860.

Barney

JAY

Felin

(2013) What are microfoundations? Academy of Management Perspectives 27(2): 138–155.

Caligiuri

Tarique

(2012) Dynamic cross-cultural competencies and global leadership effectiveness. Journal of World Business 47(4): 612–622.

Certo

Busenbark

Woo

, et al. (2016) Sample selection bias and Heckman models in strategic management research. Strategic Management Journal 37(13): 2639–2657.

Chang

Gong

Peng

(2012) Expatriate knowledge transfer, subsidiary absorptive capacity, and subsidiary performance. Academy of Management Journal 55(4): 927–948.

Chen

Tan

(2016) Foreign knowledge acquisition through inter-firm collaboration and recruitment: implications for domestic growth of emerging market firms. International Business Review 25(1): 221–232.

10.

Cowan

David

Foray

(2000) The explicit economics of knowledge codification and tacitness. Industrial and Corporate Change 9(2): 211–253.

11.

Devinney

(2013) Is microfoundational thinking critical to management thought and practice? Academy of Management Perspectives 27(2): 81–84.

12.

Fan

Cui

, et al. (2016) Localized learning by emerging multinational enterprises in developed host countries: a fuzzy-set analysis of Chinese foreign direct investment in Australia. International Business Review 25(1): 187–203.

13.

Fiaschi

Giuliani

Nieri

(2017) Overcoming the liability of origin by doing no-harm: emerging country firms’ social irresponsibility as they go global. Journal of World Business 52(4): 546–563.

14.

Forbes (2020) Jack Ma, Alibaba donate more masks as high-profile China philanthropic push continues. Available at: www.forbes.com/sites/russellflannery/2020/03/21/jack-ma-alibaba-donate-more-masks-as-high-profile-china-philanthropic-push-continues/ (accessed 21 March 2020).

15.

Gorondutse

Hilman

(2016) The moderating effect of organisational culture on the commitment to corporate social responsibility (CSR) and the performance of SMEs in Nigeria. Journal of General Management 42(1): 65–77.

16.

Grant

(1996) Toward a knowledge-based theory of the firm. Strategic Management Journal 17(S2): 109–122.

17.

Gregersen

Black

(1992) Antecedents to commitment to a parent company and a foreign operation. Academy of Management Journal 35(1): 65–90.

18.

Hafenbrädl

Waeger

(2017) Ideology and the micro-foundations of CSR: Why executives believe in the business case for CSR and how this affects their CSR engagements. Academy of Management Journal 60(4): 1582–1606.

19.

Hair

Anderson

Tatham

, et al. (1998) Multivariate Data Analysis. Upper Saddle River, NJ: Prentice Hall.

20.

Han

Jennings

Liu

, et al. (2019) Going home and helping out? Returnees as propagators of CSR in an emerging economy. Journal of International Business Studies 50(6): 857–872.

21.

Henriques

Sadorsky

(1999) The relationship between environmental commitment and managerial perceptions of stakeholder importance. Academy of Management Journal 42(1): 87–99.

22.

Hulland

(1999) Use of partial least squares (PLS) in strategic management research: a review of four recent studies. Strategic Management Journal 20(2): 195–204.

23.

Husted

Montiel

Christmann

(2016) Effects of local legitimacy on certification decisions to global and national CSR standards by multinational subsidiaries and domestic firms. Journal of International Business Studies 47(3): 382–397.

24.

Inkson

Arthur

Pringle

, et al. (1997) Expatriate assignment versus overseas experience: contrasting models of international human resource development. Journal of World Business 32(4): 351–368.

25.

Jenkins

(2004) A critique of conventional CSR theory: an SME perspective. Journal of General Management 29(4): 37–57.

26.

Jiang

Zalan

Herman

, et al. (2018) Mapping the relationship among political ideology, CSR mindset, and CSR strategy: a contingency perspective applied to Chinese managers. Journal of Business Ethics 147(2): 419–444.

27.

Johanson

Vahlne

(1977) The internationalization process of the firm—a model of knowledge development and increasing foreign market commitments. Journal of International Business Studies 8(1): 23–32.

28.

Johanson

Vahlne

(2009) The Uppsala internationalization process model revisited: from liability of foreignness to liability of outsidership. Journal of International Business Studies 40(9): 1411–1431.

29.

Johnson

Lenartowicz

Apud

(2006) Cross-cultural competence in international business: toward a definition and a model. Journal of International Business Studies 37(4): 525–543.

30.

Jokinen

Brewster

Suutari

(2008) Career capital during international work experiences: contrasting self-initiated expatriate experiences and assigned expatriation. The International Journal of Human Resource Management 19(6): 979–998.

31.

Kiessling

Isaksson

Yasar

(2016) Market orientation and CSR: performance implications. Journal of Business Ethics 137(2): 269–284.

32.

Kogut

Zander

(2003) Knowledge of the firm and the evolutionary theory of the multinational corporation. Journal of International Business Studies 34(6): 516–529.

33.

Kraimer

Wayne

Jaworski

RAA

(2001) Sources of support and expatriate performance: the mediating role of expatriate adjustment. Personnel Psychology 54(1): 71–99.

34.

Leslie

Dalton

Ernst

, et al. (2002) Managerial Effectiveness in a Global Context. Greensboro, NC: Center for Creative Leadership.

35.

Fleury

(2020) Overcoming the liability of outsidership for emerging market MNEs: a capability-building perspective. Journal of International Business Studies 51(1): 23–37.

36.

Hambrick

(2005) Factional groups: a new vantage on demographic faultlines, conflict, and disintegration in work teams. Academy of Management Journal 48(5): 794–813.

37.

Zhou

Bruton

, et al. (2010) Capabilities as a mediator linking resources and the international performance of entrepreneurial firms in an emerging economy. Journal of International Business Studies 41(3): 419–436.

38.

Luo

Wang

Zhang

(2017) Whose call to answer: institutional complexity and firms’ CSR reporting. Academy of Management Journal 60(1): 321–344.

39.

MacKinnon

Lockwood

Williams

(2004) Confidence limits for the indirect effect: distribution of the product and resampling methods. Multivariate Behavioral Research 39(1): 99–128.

40.

Makino

Isobe

Chan

(2004) Does country matter? Strategic Management Journal 25(10): 1027–1043.

41.

Marano

Tashman

Kostova

(2017) Escaping the iron cage: liabilities of origin and CSR reporting of emerging market multinational enterprises. Journal of International Business Studies 48(3): 386–408.

42.

Matten

Moon

(2008) “Implicit” and “explicit” CSR: a conceptual framework for a comparative understanding of corporate social responsibility. Academy of Management Review 33(2): 404–424.

43.

Mellahi

Guermat

(2004) Does age matter? An empirical examination of the effect of age on managerial values and practices in India. Journal of World Business 39(2): 199–215.

44.

Miska

Witt

Stahl

(2016) Drivers of global CSR integration and local CSR responsiveness: evidence from Chinese MNEs. Business Ethics Quarterly 26(3): 317–345.

45.

Nadkarni

Perez

(2007) Prior conditions and early international commitment: the mediating role of domestic mindset. Journal of International Business Studies 38(1): 160–176.

46.

Orlitzky

Schmidt

Rynes

(2003) Corporate social and financial performance: a meta-analysis. Organization Studies 24(3): 403–441.

47.

Pedersen

Petersen

Sharma

(2003) Knowledge transfer performance of multinational companies. Management International Review 43(3): 69–90.

48.

Peng

Beamish

(2014) MNC subsidiary size and expatriate control: resource-dependence and learning perspectives. Journal of World Business 49(1): 51–62.

49.

Podsakoff

MacKenzie

Lee

, et al. (2003) Common method biases in behavioral research: a critical review of the literature and recommended remedies. Journal of Applied Psychology 88(5): 879–903.

50.

Ramamurti

(2012) What is really different about emerging market multinationals? Global Strategy Journal 2(1): 41–47.

51.

Reuters (2018) Chinese firm to invest $800 million on Sri Lanka Port City underground road. Available at: http://www.reuters.com/article/idUSKBN1HJ170 (accessed 02 April 2008).

52.

Selmer

(2002) Practice makes perfect? International experience and expatriate adjustment. Management International Review 42(1): 71–87.

53.

Shea

Hawn

(2019) Microfoundations of corporate social responsibility and irresponsibility. Academy of Management Journal 62(5): 1609–1642.

54.

Stahl

Miller

Tung

(2002) Toward the boundaryless career: a closer look at the expatriate career concept and the perceived implications of an international assignment. Journal of World Business 37(3): 216–227.

55.

Suutari

Brewster

(2000) Making their own way: international experience through self-initiated foreign assignments. Journal of World Business 35(4): 417–436.

56.

Tang

Hull

Rothenberg

(2012) How corporate social responsibility engagement strategy moderates the CSR–financial performance relationship. Journal of Management Studies 49(7): 1274–1303.

57.

Tashman

Marano

Kostova

(2019) Walking the walk or talking the talk? Corporate social responsibility decoupling in emerging market multinationals. Journal of International Business Studies 50(2): 153–171.

58.

Torugsa

O’Donohue

Hecker

(2013) Proactive CSR: an empirical analysis of the role of its economic, social and environmental dimensions on the association between capabilities and performance. Journal of Business Ethics 115(2): 383–402.

59.

Vlajčić

Caputo

Marzi

, et al. (2019) Expatriates managers’ cultural intelligence as promoter of knowledge transfer in multinational companies. Journal of Business Research 94: 367–377.

60.

Wang

Luo

, et al. (2014) Autonomy delegation to foreign subsidiaries: an enabling mechanism for emerging-market multinationals. Journal of International Business Studies 45(2): 111–130.

61.

Witt

Kabbach de Castro

Amaeshi

, et al. (2018) Mapping the business systems of 61 major economies: a taxonomy and implications for varieties of capitalism and business systems research. Socio-Economic Review 16(1): 5–38.

62.

Yamazaki

(2010) Expatriate adaptation. Management International Review 50(1): 81–108.

63.

Yamazaki

Kayes

(2004) An experiential approach to cross-cultural learning: a review and integration of competencies for successful expatriate adaptation. Academy of Management Learning and Education 3(4): 362–379.

64.

Yang

Mudambi

Meyer

(2008) Conventional and reverse knowledge flows in multinational corporations. Journal of Management 34(5): 882–902.

65.

Yin

Zhang

(2012) Institutional dynamics and corporate social responsibility (CSR) in an emerging country context: evidence from China. Journal of Business Ethics 111(2): 301–316.

66.

Young

Makhija

(2014) Firms’ corporate social responsibility behavior: an integration of institutional and profit maximization approaches. Journal of International Business Studies 45(6): 670–698.

67.

Zhang

Deephouse

van Gorp

, et al. (2020) Individuals’ perceptions of the legitimacy of emerging market multinationals: ethical foundations and construct validation. Journal of Business Ethics. Epub ahead of print 16 August 2020. DOI: 10.1007/s10551-020-04599-x.

68.

Zhang

Fan

Zhu

(2014) High-performance work systems, corporate social performance and employee outcomes: exploring the missing links. Journal of Business Ethics 120(3): 423–435.

69.

Zhu

De Cieri

Fan

, et al. (2018) Expatriate management in emerging market multinational enterprises (EMNEs): reflection and future research agenda. The International Journal of Human Resource Management 29(11): 1787–1798.

70.

Zhu

Warner

Sardana

(2020) Internationalization and destination selection of emerging market SMEs: issues and challenges in a conceptual framework. Journal of General Management 45(4): 206–216.