INTRODUCTION
Contemporary internalization theory suggests that multinational enterprises (MNEs) expand internationally to deploy and exploit their firm-specific assets (FSAs) (Rugman & Verbeke, Reference Rugman and Verbeke1992). However, the extent and organization of an MNE's international operations depend on the transferability of its FSAs (Rugman & Verbeke, Reference Rugman and Verbeke2004). In particular, given that many MNEs’ FSAs are bounded by location (Rugman & Verbeke, Reference Rugman and Verbeke1992), MNEs often adopt a regional strategy to achieve global integration and local responsiveness simultaneously when deploying FSAs in the global arena (Qian, Li, Li, & Qian, Reference Qian, Li, Li and Qian2008; Rugman, Reference Rugman2005). This regional strategy may drive MNEs to establish regional headquarters (RHQ) to manage and coordinate their regional operations (Birkinshaw, Braunerhjelm, Holm, & Terjesen, Reference Birkinshaw, Braunerhjelm, Holm and Terjesen2006; Lasserre, Reference Lasserre1996; Yeung, Poon, & Perry, Reference Yeung, Poon and Perry2001).
Based on observations of MNEs in developed countries, studies overwhelmingly conclude that MNEs need to set up RHQ in distant regions only to deal with the geographical constraints of their FSAs (Enright, Reference Enright2005a, Reference Enright2005b). However, increasing evidence shows that many Chinese MNEs, which often lack transferrable FSAs at the early stage of internationalization (Luo & Tung, Reference Luo and Tung2007), also set up RHQ in their home region, Asia. For example, Bank of China established its Asian headquarters in Hong Kong in 2001, TCL set up its Asian headquarters in Singapore in 2007, and Haier formed its Asian headquarters in Japan in 2012. Indeed, studies have largely ignored the establishment of RHQ by Chinese firms in their home region. Because the FSAs within a region are associated with a smaller compound distance (Verbeke & Kano, Reference Verbeke and Kano2012), the establishment of RHQ in the home region – which is clearly in opposition to the predictions made by prior studies – calls for theoretical explanation. Therefore, we pose the following research question: What determines Chinese MNEs’ establishment of Asian RHQ (ARHQ)?
To answer this question, we draw on contemporary internalization theory and, in particular, the location- and non-location-bound FSA framework developed by Rugman and Verbeke (e.g., Rugman, Reference Rugman2005; Rugman & Verbeke, Reference Rugman and Verbeke2001, Reference Rugman and Verbeke2003, Reference Rugman and Verbeke2004). Based on this framework, we contend that ARHQ enhance Chinese MNEs’ FSAs, which are subject to market, political, and cultural risks and hence bounded by location (lack of transferability) in overseas markets. We argue that the location boundedness of Chinese MNEs’ FSAs differs across two types of regionalization strategies – that is, foreign direct investment (FDI) and export regionalization, which have different effects on ARHQ establishment. In addition, we adopt a contingency approach to investigate how Chinese MNEs’ home-city regional integration and firm-specific international management capabilities may affect the relationship between regionalization and ARHQ establishment.
Our study contributes to the regional strategy literature. First, in a departure from prior studies, it shows that, unlike traditional MNEs in developed countries, MNEs from emerging economies such as China can build RHQ in their home regions as a platform for augmenting the operational FSAs in intraregional FDI and as an indicator of their legitimacy (e.g., ARHQ in Hong Kong/Singapore) in interregional exporting. Our results support and enrich the springboard perspective (Luo & Tung, Reference Luo and Tung2007) by highlighting the importance of ARHQ as springboards for emerging MNEs’ global expansion. Second, we theoretically differentiate the location boundedness of the FSAs required for FDI and export regionalization strategies and empirically test their effects on ARHQ establishment. As such, we find that the incentives for MNEs to establish ARHQ differ, depending on the MNEs’ operating modes and the geographic scope of their international activities. Finally, our contingency approach identifies the external and internal conditions (i.e., home-city regional integration versus firm-specific international management capability) under which the effects of the two types of regionalization strategies are more salient. These dependent conditions have drawn growing theoretical interest in the management research (Boyd, Haynes, Hitt, Bergh, & Ketchen, Reference Boyd, Haynes, Hitt, Bergh and Ketchen2012; Edwards, Reference Edwards2010) and been further explored in our study.
THEORETICAL BACKGROUND AND HYPOTHESES
Based on the classic work of Buckley and Casson (Reference Buckley and Casson1976), contemporary internalization theory emphasizes the generation, exploitation, and rejuvenation of firm-specific advantages (FSAs) and matches them with the country-specific advantages (CSAs) of host countries (Verbeke & Kano, Reference Verbeke and Kano2012). These FSAs arise when an MNE has developed special knowledge, or a capability that is unavailable to others and cannot be duplicated, except in the long run and at a high cost. Such FSAs arise from (1) upstream research and development expenditures that lead to new products, new production processes, or good relationships with supply-chain members; (2) downstream FSAs in the marketing or distribution of differentiated product lines or marketing interface activities (Rugman, Verbeke, & Nguyen, Reference Rugman, Verbeke and Nguyen2011); or (3) some unique element of a company's management structure or core routines that confer an operational FSA (Rugman & McIlveen, Reference Rugman and McIlveen1985).
Focusing on FSA transferability, Rugman and Verbeke (Reference Rugman and Verbeke2004) further differentiate location- and non-location-bound FSAs. They argue that the extent and organization of an MNE's international operations depend on the location boundedness or transferability of its FSAs (Rugman, Reference Rugman2005). Non-location-bound FSAs are capabilities that can be internationally transferred at a low marginal cost, such as cutting-edge technologies, marketing prowess, strong brands, and outstanding production quality embodied in globally exported products, if highly valued by worldwide customers (Rugman & Verbeke, Reference Rugman and Verbeke1992; Verbeke, Reference Verbeke2013). An MNE can implement global integration by producing and distributing products/services of a homogeneous type and quality worldwide. Location-bound FSAs are firm-specific capabilities that can be exploited only within a limited geographic scope (Rugman, Reference Rugman2005) and hence necessitate a localization effort.
The location boundedness of FSAs may come from downstream, upstream, and operational activities. First, downstream FSAs – the knowledge strengths which are deployed in activities involving a direct interface with customers, and which are required to achieve successful market penetration – are embedded in home countries but not necessarily in other countries (Rugman, Reference Rugman2005). Second, upstream FSAs – which are deployed in activities that lack a direct interface with customers but are critical to creating an efficient internal production system (Rugman, Reference Rugman2005) – must be aligned with specific environmental elements to function (Cuervo-Cazurra, Maloney, & Manrakhan, Reference Cuervo-Cazurra, Maloney and Manrakhan2007; Kogut & Zander, Reference Kogut and Zander2003). Third, operation-related FSAs – such as organizational philosophy and institutional memory in addition to organizational routines, policies, procedures, and processes (Bartlett & Ghoshal, Reference Bartlett and Ghoshal1989) – are embedded in an MNE's corporate headquarters and home country (Lo, Mahoney, & Tan, Reference Lo, Mahoney and Tan2010).
Regionalization (Rugman, Reference Rugman2005; Rugman & Verbeke, Reference Rugman and Verbeke2001, Reference Rugman and Verbeke2002, Reference Rugman and Verbeke2003) extends the geographic scope of location-bound FSAs from within a country to within a region. For example, the triad regions such as North America, Western Europe, and Asia-Pacific have been identified (Beaverstock, Smith, & Taylor, Reference Beaverstock, Smith and Taylor1999; Poon, Reference Poon1997; Rugman, Reference Rugman2005). In a time of semiglobalization (Ghemawat, Reference Ghemawat2003), intraregional market integration and interregional market barriers coexist. As a result, the transferability of FSAs within regions increases while access to external regional markets is limited. Similarly, Rugman (Reference Rugman2005) points out that transferability of MNEs’ FSAs declines with distance (Rugman, Reference Rugman2005) due to unfamiliarity with culture and institutional systems in addition to liability to regional foreignness (e.g., Ghemawat, Reference Ghemawat2003). In response, MNEs set up RHQ in distant regions to facilitate the deployment of FSAs.
Regionalization Implementation Strategies and ARHQ
A regional strategy is traditionally simplified as a company's dependence on a region as a source of revenue (Rugman & Verbeke, Reference Rugman and Verbeke2004). However, MNEs may implement different strategies, such as FDI or export regionalization, to deploy FSAs in a regional market. FDI regionalization refers to a firm embedding itself in one region by engaging in FDI to exploit regional market opportunities or use regional resources. FDI refers to subsidiaries that MNEs establish in host countries via a wholly owned (greenfield or acquisition) or equity-based joint venture entry mode. Export regionalization refers to a firm embedding itself in a region by engaging in export trading activities to exploit market opportunities. Both FDI and export regionalization can lean more toward either a home region or other regions. As shown in Figure 1, regional strategies can be categorized into four types based on these dimensions: intra- and interregional FDI regionalization and intra- and interregional export regionalization.
Matrix of regional implementation strategies
In general, the types of FSAs required by expansion differ across the FDI and export regionalization strategies. In FDI, by adopting the entry modes of a wholly owned or equity-based joint venture, MNEs typically intend to maintain some level of control over their overseas operations (Bartlett, Beamish, & Ghoshal, Reference Bartlett, Beamish and Ghoshal2010; Geringer & Hebert, Reference Geringer and Hebert1989). The preference for high control is more prominent among Chinese executives due to both their national pride and their belief that direct control is the only way to coordinate economic activity (Morck, Yeung, & Zhao, Reference Morck, Yeung and Zhao2008). Hence, Chinese MNEs demand downstream, upstream, and operational FSAs as a full supporting package transferred from the parent company to subsidiaries. In terms of exporting, upstream non-location-bound FSAs – that is, technological sophistication or operational excellence – are embedded in the product through the production process in the MNE's home country or production base. Hence, little or no extra effort is required to transfer upstream FSAs in a host country. Instead, MNEs may have to adapt some downstream FSAs, such as packaging, marketing, and distribution activities, when implementing export regionalization (Verbeke, Reference Verbeke2013). Moreover, the transferability of FSAs is lower between regions than within regions as a company is less familiar with institutional systems and liable to foreignness in other regions (e.g., Ghemawat, Reference Ghemawat2003). Considering the different activities and distance involved in the four types of regionalization strategies, we argue that Chinese MNEs’ incentives to establish ARHQ vary across the strategies.
Intraregional FDI and ARHQ
Although a few big Chinese companies possess some non-location-bound FSAs (Mathews, Reference Mathews2006; Morck et al., Reference Morck, Yeung and Zhao2008; Sun, Reference Sun2009),[ Footnote 1 ] most smaller Chinese firms lack transferrable upstream and operational FSAs in cross-border subsidiaries, even within their home region (Gugler & Boie, Reference Gugler, Boie, Chaisse and Gugler2008; Hobdari et al., Reference Hobdari, Sinani, Papanastassiou and Pearce2007; Mathews, Reference Mathews2006; Yiu, Lau, & Bruton, Reference Yiu, Lau and Bruton2007). First, the upstream FSAs that Chinese MNEs deploy in their subsidiaries are location bound. Chinese firms have developed FSAs based mainly upon re-combinations with home CSAs (Rugman, Nguyen, & Wei, Reference Rugman, Nguyen and Wei2014). Chinese MNEs possess upstream FSAs, such as cheap labor, personnel and business network ties (Marinov, Child, & Marinov, Reference Marinov, Child, Marinov, Asmussen, Pedersen, Devinney and Tihanyi2011), and strong relationships with the home government, which allows them access to various institutional supports in terms of tax incentives and low-interest loans (Rugman, Nguyen, & Wei, Reference Rugman, Nguyen and Wei2014). All of these are embedded in the home country, but are not necessarily available in host countries. Indeed, the FDI activities of emerging MNEs reflect attempts to acquire or augment complementary upstream FSAs. Second, when operating their subsidiaries in host countries, MNEs often transfer their organizational routines, cultures, policies, procedures, and processes to those foreign subsidiaries. These elements are developed by each MNE's parent company and may not be applicable to new subsidiaries in other countries.
Due to the boundedness of upstream and operational FSAs in their home country, Chinese MNEs must establish ARHQ when they commit more FDI to Asia. ARHQ may facilitate the acquisition and integration of regional information and resources, thus equipping Chinese MNEs’ subsidiaries with greater capability to adapt upstream and operational FSAs to the region. Moreover, ARHQ may closely monitor the deployment of FSAs, thus decreasing the management and coordination costs in Asian subsidiaries (Enright, Reference Enright2005b). The practices of MNEs support such an argument. According to the regional executive director of the OCBC (Oversea-Chinese Banking Corporation) Bank Zhang Zhijian (Yu, Reference Yu2014), many Chinese MNEs set up Asian regional headquarters in Singapore for two reasons. First, Singapore's well-developed financing, logistics, and legal systems allow Chinese MNEs to manage their operations in nearby countries easily. Second, Singapore offers a good talent pool, which is critical for international expansion and management. The case of Shanghai Zhenhua Heavy Industry offers similar insights. The company previously manufactured its products in China and sold them in Southeast Asian countries. With the establishment of its ARHQ in Singapore, the company has improved its infrastructure and talent pool to perform order taking, manufacturing, and after-sale services locally in Southeast Asia (Yang & Cao, Reference Yang and Cao2014). As a result, Singapore serves as a springboard for Chinese MNEs to expand faster and farther within their home region.
In this study, we focused on the effect of intraregional FDI, but not interregional FDI for two reasons. First, an ARHQ may not be helpful in coping with location-bound FSAs when it is geographically distant from the subsidiaries. Hence, Chinese firms do not have much incentive to establish ARHQ to deal with interregional FDI. Second, most of the FDI effort that Chinese MNEs expend is concentrated in Asia. According to Lian and Ma (Reference Lian and Ma2011), Asia remains the main target area of Chinese FDI outflow, with an average proportion of 63.16% in 2003–2009. Therefore, we posit the following hypothesis.
Hypothesis 1: The higher the degree of a Chinese MNE's intraregional FDI is, the more likely it is to establish ARHQ.
Interregional exports and ARHQ
Compared to FDI, exporting is a simpler form of internationalization and less dependent on location-bound FSAs. First, the upstream FSAs embedded in the export product – that is, technological sophistication or operational excellence – are usually not bounded by location. Second, little or no operation-related FSAs need to be transferred to the host country, given that the products are transacted mainly in an arm's-length market. Third, downstream FSAs can often be transferred at a relatively low cost in a Chinese MNE's home region, where consumers share a similar cultural, institutional, and economic environment and hence have preferences that are similar to its domestic consumers. As a result, Chinese exporters rely predominantly on non-location-bound FSAs and very little on location-bound FSAs in Asia (Verbeke, Reference Verbeke2013), suggesting that interregional exports carried out by Chinese firms should be made as a focus. Indeed, it is more challenging to deploy downstream FSAs across different regions through exporting (Rugman & Verbeke, Reference Rugman and Verbeke2004). The transfer of downstream FSAs in distant regions faces impediments due to the formal and informal marketing barriers posed by the host governments’ deliberate policies and divergent cultural, institutional, and economic environments across the triad regions (Ghemawat, Reference Ghemawat2003).
Most countries implement some form of trade barriers (Poon, Reference Poon1997). Many Chinese products are subject to import prohibitions or restrictions for failing to meet specific standards, many of which are implemented by countries in the triads other than the Asia-Pacific. For example, from 1995 to 2005, 29 of 31 complaints against China came from North America and the European Union. Therefore, traders in other triad countries are usually more cautious when doing business with Chinese MNEs because of the legal risks involved.
Consumers who are unfamiliar with products made in China tend to use an extrinsic cue – that is, the ‘country of origin’ stereotype – to assess them (Cuervo-Cazurra et al., Reference Cuervo-Cazurra, Maloney and Manrakhan2007; Peterson & Jolibert, Reference Peterson and Jolibert1995). Although China is increasingly respected for its high-quality products[ Footnote 2 ] and reputable brands,[ Footnote 3 ] products in most product categories that have low brand awareness but are tagged as ‘made in China’ are thought to be low in quality, unsafe, and unreliable. Hence, their associated purchase intention is low, particularly in developed countries (Blecken, Reference Blecken2007; Chinen & Sun, Reference Chinen and Sun2011; Kabadayi & Lerman, Reference Kabadayi and Lerman2011; Maheswaran, Reference Maheswaran2013). For example, according to the World Advertising Research Centre, 81% of Americans prefer a Japanese product to a Chinese product. Few German consumers are able to name a Chinese brand without prompting a negative response, and 70% have no preference for brands from emerging markets. Such informal market barriers devalue Chinese MNEs’ downstream FSAs, making the transfer of downstream FSAs across regions more challenging (Bauche, Reference Bauche2014).
Hence, Chinese MNEs with intensive interregional export activities are motivated to set up ARHQ in their home region for their symbolic value (Birkinshaw et al., Reference Birkinshaw, Braunerhjelm, Holm and Terjesen2006). By setting up an ARHQ, a Chinese MNE demonstrates internally to the product division and externally to the general public, financial community, and international trading partners that the company attaches strategic importance and commits itself to the greater Asia region. In this way, it signals that it is no longer constrained by the norms, standards, and expectations of its home country (Zaheer, Reference Zaheer1995). Instead, it adopts an international standard usually applied in its ARHQ to ensure the quality, pricing, safety, and environmental protection of its trading practices. The well-developed legal institution of ARHQ decreases exporters’ propensity to engage in opportunistic behavior and enables importers from other triad regions to rely on the legal system to resolve contract enforcement disputes (Li, Vertinsky, & Zhang, Reference Li, Vertinsky and Zhang2013). For example, Ningbo Shenzhou International Group Holdings set up its ARHQ for the purpose of exporting to the United States. Thus, it enjoys preferential treatment under the U-Cambodia Trade and Investment Framework Agreement and is able to circumvent the rule-of-origin restriction and trading barriers. Hong Weiqiang, the chargé d'affaires at the Singapore embassy in China, said that Chinese MNEs could enjoy easier market access to other regions because Singapore has a more comprehensive free trade arrangement with various countries and regions than China (Zhu, Reference Zhu2014).
Establishing an ARHQ also helps to dampen the negative ‘country of origin’ stereotype in two ways. First, Chinese MNEs may establish ARHQ in a neighboring country with a favorable reputation for assembly or design (Chao, Reference Chao1993; Han & Terpstra, Reference Han and Terpstra1988) and thus benefit from offering a hybrid product (Chao, Reference Chao1993). For example, Haier has established its ARHQ in Japan to signal that it has adopted Japanese design and technology in its production process. Second, ARHQ in global cities enhances brand ‘globalness’, which is usually labeled as high quality and prestige (Holt, John, & Taylor, Reference Holt, John and Taylor2004; Johansson & Ronkainen, Reference Johansson and Ronkainen2005) and thereby increases purchase likelihood (Steenkamp, Batra, & Alden, Reference Steenkamp, Batra and Alden2003). It is easier for Chinese MNEs to bypass formal and informal barriers when exporting to the other two triads. Therefore, we propose the following hypothesis.
Hypothesis 2: The higher the degree of a Chinese MNE's interregional exports are, the more likely it is to establish ARHQ.
The Contingent Effects of External and Internal Conditions
Given the foregoing baseline arguments, we examined whether and how Chinese MNEs set up ARHQ following regionalization are contingent on external and internal conditions. We emphasized a Chinese MNE's home base, given its influence on the location boundedness of the MNE's FSAs and its international management capability as a special type of FSA.
External condition: Home-city regional integration
An MNE's home city is the city in which the MNE has grown up and the top management center is located. A home city provides a home-specific advantage and serves as the base of the MNE's location-specific FSAs (Rugman & Verbeke, Reference Rugman and Verbeke2004). The dynamism of the home-city environment depends on not only its own resource endowment but also its level of integration in the region. In this study, we focused on economic integration because it is the result and ultimate manifestation of the historical cultural, geographic, business, and political linkages of countries and cities (Anderson & Marcouiller, Reference Anderson and Marcouiller2002; Ghemawat, Reference Ghemawat2001; Portes & Rey, Reference Portes and Rey2005; Rauch, Reference Rauch2001). Moreover, although China as a whole has been involved in series of regional trade agreements with other economies in Asia, not every city in China is embedded in Asia versus other regions of the world to the same extent because of their unique historical links with trading partners, industry compositions, and economic structures. For example, most of the Chinese immigrants in Southeast Asia came from cities in Fujian or Guangdong province. Therefore, businesses from these cities may face lower barriers when expanding to Southeast Asia than those from other cities in China because of their traditional ethnic connections. Cities that have made significant contributions to the automobile industry may target more of their trading at South America, which has led China's auto export market in recent years (Xinhua, 2014). Therefore, Chinese MNEs based in different cities are subject to different economic dynamics.
A higher level of regional integration in a home city increases the mobility of the individuals, goods, and services within the region (Viner, Reference Viner1950) and increases the region's production- and market-related knowledge (Matusik & Hill, Reference Matusik and Hill1998). Therefore, an MNE's HQ has more chances to access business information and screen potential opportunities to deploy their FSAs (Romanelli & Khessina, Reference Romanelli and Khessina2005). In addition, frequent interactions encourage exchange participants to form strong ties and have common experiences, leading to lower bounded rationality costs and greater cooperation (Lester & Cannella, Reference Lester and Cannella2006; Romanelli & Khessina, Reference Romanelli and Khessina2005; Rugman & Verbeke, Reference Rugman and Verbeke2005). In this way, the confidence and purchase intentions of traders, suppliers, consumers, and investors are positively influenced (Agarwal, Malhotra, & Wu, Reference Agarwal, Malhotra and Wu2002; Rugman & Verbeke, Reference Rugman and Verbeke2004). When an MNE strengthens its trust and reputation, it can more feasibly deploy its upstream and downstream FSAs in an extended area. The decreased location boundedness of upstream and downstream FSAs within Asia caused by the economic integration of home cities has different implications for the regional strategies and ARHQ decisions of MNEs.
When an MNE leverages the market opportunities introduced by its home city through intraregional FDI, its subsidiaries are in greater need of processing a larger amount of orders and enhancing interfaces with local customers, suppliers, investors, workers, governments, and other stakeholders. The challenges involved in adapting operational FSAs to local management complexities become more salient. Although the economic integration of a home city can alleviate the location boundedness of an MNE's upstream and downstream FSAs, it does not help with the transfer of operational FSAs because of a lack of close contact. Therefore, Chinese MNEs have more of an incentive to establish ARHQ and thereby to augment their operational FSAs when coordinating their FDI operations. Hence, we posit the following hypothesis.
Hypothesis 3a: Home-city regional integration in Asia strengthens the positive relationship between the degree of a Chinese MNE's intraregional FDI and the establishment of its ARHQ.
For Chinese MNEs that adopt an export regionalization strategy, it should be noted that home-city regional integration enhances the transferability of downstream FSAs only within the region but not across regions. An MNE located in a home city that is highly integrated in Asia can easily obtain intraregional trade information and contracts from its home city, lowering the significance of establishing an ARHQ. However, a home city that is highly embedded in its own region decreases the degree of business contact between a Chinese MNE and its interregional partners. The MNE's products/services are oriented more toward intraregional consumers and deviate from the preferences of other regions. Hence, home-city economic integration creates more barriers for Chinese MNEs to transfer downstream FSAs to other triad regions. These MNEs must gain legitimacy and change the ‘country of origin’ stereotype if they want to exploit more interregional opportunities. They are more likely to establish an ARHQ as a symbol of legitimacy to springboard their global expansion. Therefore, we posit the following hypothesis.
Hypothesis 3b: Home-city regional integration in Asia strengthens the positive relationship between the degree of a Chinese MNE's interregional exports and the establishment of its ARHQ.
Internal condition: MNEs’ international management capabilities
To implement a regional strategy, firms must have higher-order FSAs – that is, international management capabilities–to adapt and adjust their operations and convert stand-alone FSAs into value (Teece, Pisano, & Shuen, Reference Teece, Pisano and Shuen1997). Following prior studies (Griffith & Harvey, Reference Griffith and Harvey2001; Lu et al., Reference Lu, Zhou, Bruton and Li2010; Oktemgil & Greenley, Reference Oktemgil and Greenley1997; Roth, Schweiger, & Morrison, Reference Roth, Schweiger and Morrison1991), we define international management capabilities as the static capabilities required to assemble, manage, and coordinate bundles of resources transferred through foreign operations and the dynamic capabilities required to acquire and absorb strategic resources from overseas markets (i.e., advanced technology, talent, and market information) and integrate them into international operations.
We contend that an MNE's international management capability moderates its decision to establish ARHQ for different types of regionalization strategies in different ways. Chinese MNEs with greater international management capability are able to perceive more expansion opportunities. When they explore these opportunities via FDI regionalization within Asia (i.e., setting up more subsidiaries), management becomes more complex because they have to coordinate a larger number of subsidiaries. Additionally, they need to closely interact with more business partners in the host counties. Hence, it is less effective to use their corporate headquarters in the home country to manage day-to-day subsidiary operations. We thus suggest that a Chinese MNE is more likely to build an ARHQ to implement its intraregional FDI strategy when it is equipped with better international management capability. Therefore, we posit the following hypothesis.
Hypothesis 4a: A Chinese MNE's international management capability strengthens the positive relationship between the degree of its intraregional FDI and the establishment of its ARHQ.
For Chinese MNEs that adopt an export regionalization strategy, the main challenge in transferring downstream FSAs involves overcoming formal and informal marketing barriers and making some minor adaptations for the host country. When a Chinese MNE's headquarters has a high international management capability, it can adapt to the foreign environment, establish a positive reputation among its trading partners, and overcome barriers at a low cost. The importance of an ARHQ is deemphasized even when the Chinese MNE targets other regions. Therefore, we posit the following hypothesis.
Hypothesis 4b: A Chinese MNE's international management capability weakens the positive relationship between the degree of its interregional exports and the establishment of its ARHQ.
Figure 2 summarizes the four hypotheses.
Conceptual model
METHOD
Sampling and Data Collection
The data used in our study came from a survey conducted in 2011. We collected data related to the motivation, challenges, decision criteria, aggregate volume, and distribution of Chinese firms’ outward foreign direct investment and export activities and ARHQ strategies.
To select the sample firms, we first chose the 16 provinces and municipalities that had the greatest outflow of FDI in 2011 (Beijing, Shanghai, Tianjin, Zhejiang, Liaoning, Guangdong, Jiangsu, Shandong, Henan, Hebei, Shaanxi, Fujian, Hunan, Hubei, Hainan, and Guangxi). We then randomly chose sample firms within those provinces and municipalities from a list of Chinese firms obtained from China's Ministry of Commerce. Our sampling procedure had high representativeness and broad scope, as it included firms from eastern, central, and western China and involved 16 of the 34 provincial-level areas and 69 of the 333 prefectural-level cities as of 2010. We included 9 industries and 46 subindustries in the sample to reflect the realities of ARHQ establishment.
Before we collected the data, we prepared an English-language version of the survey, which was translated into Chinese by two of the researchers on this project who had sufficient fluency in both languages. To ensure conceptual equivalence, two independent translators were employed to translate the Chinese version back into English. The researchers and translators discussed any conflicts until they reached agreement (Hoskisson et al., Reference Hoskisson, Eden, Lau and Wright2000). We conducted a pilot study in Chinese that involved 15 senior managers who were currently or had previously been in charge of international business. The pilot results supported our attempt to continue a larger-scale survey. Finally, we sent the surveys to CEOs and senior managers who were directly involved in internationalization projects at 350 Chinese firms. We cooperated with local government institutions and researchers to conduct on-site surveys, of which we collected 333. After surveys with missing values were dropped, 227 usable surveys were left, indicating a valid response rate of 64.86%. Of these, we omitted one survey, because the contributing firm had not engaged in any foreign expansion effort. The remaining 226 Chinese firms surveyed had made some efforts toward foreign expansion, through exporting, setting up foreign subsidiaries, or both.[ Footnote 4 ]
We collected information from different sources to overcome possible common method bias. We obtained the FDI and export regionalization, ARHQ, international management capability, domestic performance, international experience, technological capability, and brand awareness measures from the survey. (The relevant questions are listed in Appendix I.) The regional integration levels of an MNE's home city using secondary data from the China Statistical Yearbook 2011 were then calculated. The control variable data (i.e., a firm's ownership type age, size, and industry) came from archival data provided in the business directory. We conducted Harman's one-factor test with all the measurement items in a factor analysis without rotation and achieved a solution that accounted for 66.30% of the total variance, with the first factor accounting for 16.62%. Because a single dominant factor did not emerge, we found common method bias to be an unlikely concern in this study (Podsakoff & Organ, Reference Podsakoff and Organ1986).
We assessed potential respondent bias by comparing key demographic variables, including the number of employees, total assets, and annual profits of our sample firms, with those of the population of all the Chinese firms in the same industry, the means of which were obtained from the China Statistical Yearbook 2011. Insignificant mean differences demonstrated the representativeness of the sample.
On average, our sample firms had operated in China for 17.6 years (ranging from 2 to 82 years). In 2010, their annual sales averaged RMB 764.57 million, with 44.32% of the total from export sales. The firms had 9.99 subsidiaries in China and 2.84 foreign subsidiaries (ranging from 0 to 79).
Variables and Measurement
Dependent variable
ARHQ
We followed the measurements of previous RHQ studies to measure ARHQ (Enright, Reference Enright2005a, Reference Enright2005b; Ma & Delios, Reference Ma and Delios2010). We asked respondents whether their firms had built RHQ in Asia. The value of ARHQ was 1 if yes and 0 otherwise. However, because the Chinese MNEs were at the early stage of internationalization, few firms had ARHQ. Hence, we asked the firms without ARHQ whether they would build an ARHQ within three years. The value of ARHQ was changed to 1 if yes and 0 otherwise. We combined two types of answers – that is, ‘We have built an ARHQ’ and ‘We intend to build an ARHQ’ – into one variable. Of the 226 firms, 21 firms indicated that they had already set up an ARHQ (9.29%) and 60 firms indicated that they would set up ARHQ within three years (26.55%).
Independent variables
Intra-region FDI
Following Rugman and Verbeke (Reference Rugman and Verbeke2004), we measured intraregional FDI as the ratio of an MNE's number of overseas subsidiaries in Asia to its total number of overseas subsidiaries. The greater the intraregional FDI ratio, the more the MNE was embedded in its home region in terms of FDI.
Inter-region exporting
We measured interregional exporting based on an MNE's export intensity – that is, the ratio of its exports from Asia to its total exports (Banalieva, Santoro, & Jiang, Reference Banalieva, Santoro and Jiang2012; Delios & Beamish, Reference Delios and Beamish2005; Rugman & Verbeke, Reference Rugman and Verbeke2008). The greater the interregional export, the less the MNE was embedded in its home region in terms of exporting.
Home-city regional integration
Intraregional trade share has been widely used to capture a country's economy integration into the region (Iapadre, Reference Iapadre2004; Lombaerde, Reference Lombaerde2006). Prior studies have instrumented such a measure by (1) identifying the most advanced type of trade agreement in which a country is involved and (2) obtaining the trade share index – the ratio of sales made within that trade agreement to total sales (Barrera & Hass, Reference Barrera and Haas1969; Kegley & Howell, Reference Kegley and Howell1975). The goal of this conceptualization and measure is to evaluate the effect of a trade agreement.
We modified the measure to capture the home city's regional integration by using the ratio of a home city's import and export trading with Asia to its total world trade. A value closer to 1 indicated that the home city had more economic interaction with Asian countries than with countries in other regions and vice versa.
Such a modified measure differed from the widely used measure in two ways. First, we examined the regional integration of a home city into Asia as a whole instead of focusing on only one trade agreement. Our study focused on whether the home city of an MNE interacts frequently with every other Asian country and not merely countries within specific trade agreements. Numerous trade arrangements involve China within the Asian region, including those signed on the Asia-Pacific Economic Cooperation (APEC) forums, the China-ASEAN (Association of Southeast Asian Nations) Free Trade Agreement (CAFTA), and Trilateral Cooperation among the People's Republic of China, Japan, and the Republic of Korea (Trilateral Agreement). All these agreements allow China to interact with other Asian countries frequently and with low barriers. Second, rather than developing one index for China as a whole, we differentiated the regional integration indices for different cities within the country. As mentioned previously, although China as a whole has been involved in a series of regional trade agreements with other Asian economies, not every city in China is embedded in Asia versus other regions of the world to the same extent because of their unique historical links with trading partners, industrial composition, and economic structure. Our data suggests that, of the 61 sample cities, 23 had more interregional exports than intraregional trading.
International management capability
Although many studies have acknowledged the importance of an MNE's international management capability (e.g., Lu et al., Reference Lu, Zhou, Bruton and Li2010; Oktemgil & Greenley, Reference Oktemgil and Greenley1997), no standard measure of it has been developed. We developed an international management capability construct that combined three items to measure an MNE's static and dynamic capability. We measured an MNE's static capability according to the following statement: ‘We have the capability to manage overseas operations’ (Roth et al., Reference Roth, Schweiger and Morrison1991). Furthermore, we instrumented an MNE's dynamic capability according to two statements: ‘We have the capability to adapt to a foreign culture and distance during overseas operations’ and ‘We have the capability to learn about and absorb foreign advanced technology during overseas operations’. We asked all the sample firms to rate the preceding statements using a seven-point Likert scale (‘1’ = totally disagree, ‘7’ = totally agree). Because these were perceptual scales, we tested their construct reliability and validity. We conducted exploratory factor analysis (EFA) to test the validity. The factor loadings of the three items were 0.94, 0.94, and 0.93, contributing 88% of the variance. Cronbach's alpha was 0.93, confirming the reliability of the scale.
Control variables
We included several control variables in our model that might have altered the results. Firm size was measured using the natural log of the total sales of a firm. Firm age was measured as the natural log of the number of years since the establishment of a firm. We also controlled for ownership type using a dummy variable equal to 1 if a firm was state owned and 0 if it was privately owned. We controlled for industry effect using a dummy variable to indicate whether a firm was in the manufacturing industry. Because Chinese littoral and nonlittoral provinces and cities exhibited differences, we controlled for a firm's home location, with a dummy variable equal to 1 if the firm was in a littoral location and 0 otherwise. The littoral regions included Guangdong, Shandong, Jiangsu, Zhejiang, Liaoning, Hebei, Fujian, Hainan, and Guangxi provinces; and the littoral cities included Shanghai, Tianjin, and Beijing.
We also controlled for management overseas background using a dummy variable to reflect whether a firm's top management team (TMT) had at least one overseas returnee (‘1’ = yes, ‘0’ = no). To eliminate the effect of domestic performance variance, we controlled for a firm's performance, which we measured using a combined variable with six items – ‘Please rate your level of satisfaction with your firm's sales, sales growth rate, market share, market share growth, margins, and margin growth compared with those of your domestic competitors (‘1’ = totally dissatisfied, ‘7’ = totally satisfied)’. Based on prior studies, international experience, technological capability, and brand awareness may affect regional headquarters establishment. We controlled for international experience, technological capability, and brand awareness by asking the firms to assess whether they had international experience, whether their products and technology were competitive, and whether foreign consumers well understood the company and brand using seven-point scales (‘1’ = totally disagree, ‘7’ = totally agree). In addition, we controlled for total export intensity and total foreign overseas subsidiaries to eliminate the effect of differences in the level of internationalization.
Analytic Techniques
Table 1 lists the means, standard deviations, and correlations of the variables included in our analysis. The variance inflation factors for the regression models did not exceed 10, indicating no serious multicollinearity problems (Neter, Wasserman, & Kutner, Reference Neter, Wasserman and Kutner1990). Because our dependent variable was a dummy variable with a value of 1 or 0, it was feasible to use the probit model to test our model. In addition, the data might have been correlated within an industry cluster due to the presence of a common, unobserved, cluster-specific term. The MNEs in some industries might have been more likely to establish ARHQ because of their high reliance on a local supply chain. Therefore, we estimated the probit regressions clustered by industry.
Means, standard deviations, and correlations (N = 226)
* Significant at the 0.1 level (two-tailed test) when Pearson correlations > 0.13 or < -0.13.
RESULTS
In our model, to mitigate the possibility of multicollinearity, we mean-centered all the independent variables before creating interaction terms. We used five models to test our hypotheses (Table 2). We built the models incrementally from the baseline (model 1), to which we added the independent variables and moderators. The changes in the model χ2 values were mostly significant (p < 0.1). The addition of the respective variables in each model improved the overall model fit.
Probit model for headquarters in Asia (N = 226)
a Notes: Standard errors are in parentheses. + p < 0.10; * p < 0.05; ** p < 0.01; *** p < 0.001 (two-tailed).
Model 2, with the main effects of intraregional FDI and interregional exports, was used to test Hypothesis 1 and Hypothesis 2. Hypothesis 1 predicted a positive effect of intraregional FDI on ARHQ establishment. As shown in model 2, an MNE's intraregional FDI was positively related to ARHQ establishment (β = 0.801, p < 0.01). Therefore, Hypothesis 1 was supported. Hypothesis 2 proposed a positive effect of interregional exports on ARHQ establishment. MNE's interregional exports were positively related to ARHQ establishment (β = 0.411, p < 0.1). Therefore, Hypothesis 2 was marginally supported.
Model 5 tested Hypothesis 3 and Hypothesis 4. Hypothesis 3a and Hypothesis 3b respectively predicted that home-city regional integration strengthens the positive effect of intraregional FDI and interregional exports on ARHQ establishment. In model 5, the main effect of home-city regional integration showed no significant effect on an MNE's establishment of ARHQ (β = 0.946, p > 0.1). The interaction between home-city regional integration in Asia and intraregional FDI showed no significant effect on ARHQ establishment (β = 1.558, p > 0.1). The interaction between home-city regional integration and interregional exports had a positively significant effect on ARHQ establishment (β = 8.504, p < 0.001). Thus, Hypothesis 3b was supported, but Hypothesis 3a was not supported.
Hypothesis 4a and Hypothesis 4b respectively proposed that an MNE's international management capability strengthens the positive relationship between intraregional FDI and ARHQ establishment and weakens the positive relationship between interregional exports and ARHQ establishment. In model 5, international management capability was positively significant (β = 0.211, p < 0.1). The interaction between international management capability and intraregional FDI had a significantly positive effect on ARHQ establishment (β = 0.285, p < 0.1). Thus, Hypothesis 4a was marginally supported. However, the interaction between international management capability and interregional exports showed no significant effect on ARHQ establishment (β = -0.283, p > 0.1). Thus, Hypothesis 4b was not supported.
Figures 3 and 4 plot the results to further probe the findings related to significant moderating effects. In Figure 3, all the variables in model 5 of Table 2 except interregional export and home-city regional integration are constrained to their mean values. The values for interregional exports range from three standard deviations below or above the mean. We also set a low level of home-city regional integration as one standardized deviation below its mean score and the high level as one standard deviation above the mean. In Figure 4, all the variables in model 5 of Table 2 except intraregional FDI and international management capability are constrained to their mean values. Intraregional FDI takes values ranging from three standard deviations below or above the means. We also set the low level of international management capability as one standardized deviation below its mean score and the high level as one standard deviation above the mean. Figures 3 and 4 depict the effect of interregional exports on ARHQ establishment for low and high levels of home-city regional integration and the effect of intraregional FDI on ARHQ establishment for low and high levels of international management capability, respectively.
Moderating effect of interregional exports on home-city regional integration (HCRI) – ARHQ establishment relationship
Moderating effect of intraregional foreign direct investment (FDI) on international management capability (IMC) – ARHQ establishment relationship
Robustness Check and Further Evidence
Robustness tests
Additionally, we did several more robustness tests using alternative measures for our main variables. In our main test, we combined into one variable the two types of answers ‘We have built ARHQ’ and ‘We intend to build ARHQ’. In the first robustness test, we used the actual establishment of an ARHQ and intention to establish an ARHQ as dependent variables, respectively, to test the full model and obtained consistent results. Second, we used an alternative measure for focal firm's interregional exports – that is, the ratio of sales to the rest of the world to total sales–to test the full model. The results were also consistent. Third, we used an alternative measure for the home-city's regional integration level – the symmetric index of intraregional trade intensity (Dalum, Laursen, & Villumsen, Reference Dalum, Laursen and Villumsen1998) – to test the full model, and similar results were yielded. Lastly, we employed instrumental variable regression to correct for potential endogeneity issues (Reeb, Sakakibara, & Mahmood, Reference Reeb, Sakakibara and Mahmood2012). Specifically, we chose instrumental variables[ Footnote 5 ] to respectively instrument intraregional FDI and its interaction with home-city regional integration and international management capability, as well as interregional exports, and its interaction with home-city regional integration and international management capability. Results were also consistent with the main tests.
Further evidence
Although the statistical analyses provide support for most of our hypotheses, some of the important underlying mechanisms related to FSAs were not directly tested in the main analyses. Ideally, we would use the relevant FSA data before the MNEs established their ARHQ to test the argument. However, our survey comprised only cross-sectional data from 2010, preventing us from retracing the decision scenario at the time of an MNE's regional expansion and ARHQ establishment. In this case, we used the available information from our full survey to provide some evidence in support of those mechanisms.
Lack of transferrable FSAs in FDI regionalization
In Hypothesis 1, we argued that FDI regionalization is associated with non-transferrable FSAs and hence demands ARHQ. To verify the argument, we need to reveal the nature of non-transferrable FSAs in the Chinese MNEs. We examined the related question in the survey. We asked respondents to rate the difficulties they faced when setting up FDI operations in host countries on a seven-point Likert scale (‘1’ = very low difficulty, ‘7’ = very high difficulty). As shown in Figure 5, the sample firms rate connecting with supply chain partners at the highest level of difficulty (4.24), followed by managing cultural differences and adapting staff to local operations (4.21 and 3.70) and low brand awareness among local consumers (4.16). The results confirm the low transferability of the upstream (difficulty in creating relational assets), downstream (difficulty in creating brand awareness), and operational FSAs (difficulty in managing cultural differences and adapting staff to local operations) in the FDI regionalization of Chinese MNEs. In terms of FDI regionalization, ARHQ are established mainly to build local supply chain relationships, recruit and manage local workers, and to adapt the product or service.
Difficulties faced by Chinese multinational enterprises (MNEs) when setting up foreign direct investment (FDI)
ARHQ as a Symbol of Legitimacy for Export Regionalization
Hypothesis 2 argues that ARHQ help Chinese MNEs overcome the negative ‘country of origin’ stereotype in other regions via association with a favorable image of the ARHQ location when the MNEs are involved in interregional exports. We checked the locations of the ARHQs to find evidence to support this explanation. Not surprisingly, among the 22 Chinese MNEs that set up ARHQ, 13 did so in Hong Kong or Singapore. Of the 60 Chinese MNEs that planned to set up ARHQ, 44 planned to do so in Hong Kong or Singapore. We argue that setting up an ARHQ in Hong Kong or Singapore may decrease the concerns of interregional business partners and hence facilitate more interregional exports, given the well-established legal infrastructure and business transparency in those two cities. Such an argument was well supported by the finding that Chinese MNEs with ARHQ in Hong Kong or Singapore had a significantly higher interregional export ratio and similar FDI in Asia than did MNEs with ARHQ in other Asian countries (diff = 35.33%; p < 0.01). Interregional exports could be a particularly important motivation for setting up an ARHQ in Hong Kong or Singapore.
To further verify that ARHQ help Chinese MNEs alleviate the negative ‘country of origin’ stereotype, we conducted a probit regression using a dummy variable indicating ARHQ in Hong Kong or Singapore as the dependent variable. We tested Hypotheses 1 to Hypothesis 4 using the models in Table 3. The results for model 7 are similar to those for our main model. Hypothesis 1, Hypothesis 2, and Hypothesis 3b were well supported. The interaction between intraregional FDI and international management capability was not significant. Thus, Hypothesis 4a was not supported. The result may imply that Chinese MNEs tend not to use ARHQ in Hong Kong or Singapore to coordinate their Asian FDI. The interaction between interregional export and international management capability in model 7 was negatively significant (β = -0.488; p < 0.1), supporting Hypothesis 4b. Such a result suggests that Chinese MNEs are less likely to establish ARHQ in Hong Kong or Singapore when they have better international management capabilities and more interregional exports. This is reasonable, given that Chinese MNEs do not need to establish legitimacy through ARHQ in Hong Kong or Singapore when they have enough international management capability to leverage their FSAs across regions.
Probit regression model for headquarters in Hong Kong/Singapore (N = 226)
Notes: Standard errors are in parentheses. + p < 0.10; * p < 0.05; ** p < 0.01; *** p < 0.001 (two-tailed).
To confirm the legitimacy role played by Hong Kong and Singapore, we added the interaction term between brand awareness and interregional export. The results for model 8 showed that the interaction term was negatively significant (β = -0.969; p < 0.01), suggesting that better brand awareness and more interregional exports discourage an MNE from setting up ARHQ in Hong Kong or Singapore. These consistent results confirm that an ARHQ in Hong Kong or Singapore is used as an indicator of legitimacy that can help Chinese MNEs overcome the ‘country of origin’ stereotype. Figure 6 plots the results.
Moderating effect of interregional exports and international management capability (IMC)
DISCUSSION
Although many studies have focused on the phenomenon of the regionalization and emergence of RHQ, most have looked at these subjects from the perspective of a developed economy. Moreover, studies have highlighted that MNEs set up RHQ because of the difficulty of managing subsidiaries from a distance. The reason why Chinese MNEs establish RHQ in their home region has yet been adequately explored. In this study, drawing on contemporary internalization theory, we investigated this phenomenon and proposed that the motivation of Chinese MNEs in setting up an ARHQ derives from their particular form of regional strategies. This study is among the first to examine the underlying reasons for which MNEs from emerging markets establish a home-region RHQ, which has important implications for research, policy makers, and managers.
Implications for International Business Theory
As the first to address the establishment of home RHQ at the early stage of international expansion, our study has two interesting results regarding those MNEs from emerging economies, revealing a different RHQ arrangement to support emerging MNEs’ leapfrogging expansion paths. First, although the geographic and cultural distance is supposed to be short, we find that Chinese MNEs establish home RHQ rather than use their corporate headquarters in China to coordinate intraregional FDI operations. Young international players supported by the home government and pushed by the global competitive climate tend to expand quickly; however, they lack experience in international operations and transferrable upstream and operational FSAs, even in their home regions. Hence, they require RHQ that are closer to their target countries to obtain local talent, adapt their products to the local market, and build and maintain good relationships with local supply chain members.
Another interesting result is that Chinese MNEs that engage in more interregional export activities tend to establish more ARHQ. Many MNEs from developed countries with intensive interregional exports establish headquarters in other regions to stay close to their target market for more effective management and coordination, in which an ARHQ far from the interregional market would not help much. Therefore, establishment of ARHQ by Chinese MNEs for interregional exports may be better explained by legitimacy concerns, rather than operational concerns. Chinese MNEs lack legitimacy because of an unfavorable ‘country of origin’ stereotype and low brand awareness in other regions. Hence, MNEs must establish headquarters in their home regions in some global cities to increase their legitimacy. Moreover, the interregional trade and non-trade barriers created by governments are more prominent in other regions than they are within a triad region. Hence, Chinese MNEs have a greater incentive to establish RHQ in some home-region countries with better trading terms and institutional infrastructure to circumvent the barriers posed in other triad regions.
These results extend and enrich the emerging ‘springboard’ perspective (Luo & Tung, Reference Luo and Tung2007; Mathews, Reference Mathews2006; Ramamurti, Reference Ramamurti2012). The springboard perspective proposes that emerging MNEs lack transferrable FSAs due to the disadvantage of latecomer and that they take international expansion itself as a springboard to acquire various resources (Marinov et al., Reference Marinov, Child, Marinov, Asmussen, Pedersen, Devinney and Tihanyi2011; Pananond & Zeithaml, Reference Pananond and Zeithaml1998). For example, FSAs quickly expand to several foreign markets and proactively seek mergers and acquisitions to acquire technology and brands, to escape from domestic constraints, and to circumvent trade barriers (Luo & Tung, Reference Luo and Tung2007). We expand on this perspective by giving evidence of emerging MNEs’ structural arrangements to support their springboard intention. We argue that, when the emerging MNEs engage in a series of aggressive and risky international expansions, it is difficult for them to depend only on their corporate headquarters in their home countries to coordinate foreign operations. We find that ARHQ serve to supplement the FSAs of these latecomer MNEs and hence help with their expansion as a springboard in two ways. First, ARHQ can augment the operational FSAs in the intraregional FDI and hence allow Chinese MNEs to conduct a quick regional expansion. Second, ARHQ help Chinese MNEs establish their legitimacy in other regions and thus speed up their expansion from the regional to the international level. Through these functions, ARHQ as springboard nodes can advance the expansion of a Chinese MNE within and outside its home region. Our study thus enriches the studies of latecomers’ international strategies by highlighting the importance of home RHQ during international expansion.
To take a broader view, our study adds value to international theory for unconventional MNEs, including not only latecomer MNEs from emerging markets but also international entrepreneurs. Latecomer MNEs from emerging markets are typically subject to a negative ‘country of origin’ stereotype and are inexperienced in international management. International entrepreneurs or those ‘born globally’ are young and small firms going international at a very early stage (Turnbull, Reference Turnbull, Rosson and Reid1987), motivated by the globalization of markets, technological development in production, and the growth of small niche markets (Andersson, Reference Andersson2011; Madsen & Servais, Reference Madsen and Servais1997). They hence lack some important transferrable FSAs due to their newness and small scale, such as brand awareness, financial and human resources, and market knowledge (Freeman, Edwards, & Schroder, Reference Freeman, Edwards and Schroder2006; Gabrielsson, Reference Gabrielsson2005; Johne & Rowntree, Reference Johne and Rowntree1991; Oakey, Reference Oakey1991). Prior studies highlight the importance of management's personal networks to connect with the right resources and supplement these FSAs (Freeman et al., Reference Freeman, Edwards and Schroder2006; Harris & Wheeler, Reference Harris and Wheeler2005). Our study suggests that establishing an RHQ at an early stage in the home region might be an alternative strategy to access resources and overcome these latecomers’ and international entrepreneurs’ constraints.
Implications for Regionalization Research
Regionalization has been an increasingly important research area in the international business field. Building on the FSA framework developed by Rugman and Verbeke, our study confirms that types of regionalization (i.e., intraregional FDI and interregional exports) are not merely continuum solutions for coping with MNEs’ location-bound FSAs (Rugman, Reference Rugman1980; Siripaisalpipat & Hoshino, Reference Siripaisalpipat and Hoshino2000). It requires FSAs with different levels of location boundedness and, accordingly, different coordination efforts when being implemented. Hence, regionalization type is an important factor in an MNE's RHQ decision. Our findings help to address the three criticisms on the work of Rugman and Verbeke in exploring and explaining regional strategy data (Delios & Beamish, Reference Delios and Beamish2005; Steven & Bird, Reference Stevens and Bird2004; Westney, Reference Westney2006). These criticisms indicate that the theory simplifies the regional strategy as intraregional trading, is based solely on Global Fortune 500 firms, and neglects the specific regional management forms following regional strategies. Our results therefore advance Rugman and Verbeke's FSA framework in explaining regional strategy.
Our contingency hypotheses indicate which types of ARHQ functions can be substituted by external and internal resource endowments. We find that Chinese MNEs’ propensity to build ARHQ for intraregional FDI is not influenced by MNEs’ home-city regional integration but is enhanced by international management capability. These results imply that MNE's decision to establish ARHQ is driven mainly by the operational complexity of its intraregional FDI, which cannot be alleviated through information sharing and networking at corporate headquarters. Moreover, although better international management capability in corporate headquarters nurtures an MNE's ambition and speeds up its pace of expansion, it cannot substitute for the efforts of ARHQ to coordinate and address the complex management of subsidiaries. Indeed, a home-city or corporate headquarters cannot substitute for ARHQ in complementing operational FSAs.
However, if ARHQ are established for symbolic purposes, as in the case of export regionalization, they could be substituted by some internal and external resources. Because home-city regional integration limits an MNE's exposure to other regions, it becomes increasingly necessary for an MNE to set up ARHQ to increase its legitimacy. Hence, a home city highly embedded in the home region cannot effectively substitute for ARHQ in legitimizing a Chinese MNE for the purposes of interregional exports. MNEs’ international management capability seems to substitute for ARHQ in Hong Kong or Singapore driven by interregional exports, as shown in our robustness test. This is reasonable, given that export regionalization is an early and simple form of internationalization that does not require much international management effort. If corporate headquarters had the competence to manage trading outside its home region, MNEs would not commit extra resources to ARHQ simply for symbolic purposes (e.g., Rugman, Reference Rugman2005).
Implications for Policy Makers and Managers
Many cities around the world are trying to build themselves into global or headquarter cities (Dicken & Kirkpatrick, Reference Dicken and Kirkpatrick1991; Singapore EDB Annual Report, 2005/6). To support such efforts, scholars have conducted research to investigate the expected features of RHQ established in a distant region by MNEs from developed countries (Goerzen, Asmussen, & Nielsen, Reference Goerzen, Asmussen and Nielsen2013; Nachum & Wymbs, Reference Nachum and Wymbs2005). Our study provides evidence that an MNE from an emerging economy can establish separate headquarters in its home region (other than just relying on its corporate headquarters in its home country) to augment its FSAs. Hence, global cities might be attractive to those emerging foreign investors.
The preceding findings provide some important managerial implications. First, we offer insights into the strategic importance of establishing an ARHQ. A Chinese MNE that has a high level of intraregional FDI or interregional exports may consider establishing an ARHQ as a springboard. Second, MNEs should evaluate the effects of home-city regional integration. An MNE pursuing interregional markets may not benefit from a home city that is deeply embedded in its home region. Instead, it may consider relocating its headquarters to a more interregionally oriented location. MNEs should also evaluate the limitations of international management capability at their corporate headquarters. At an early stage of expansion, it seems helpful for corporate headquarters to possess international management capability to achieve simple adaptation. However, this is not the case for coordinating complex operations. Third, this study offers some insights for the public policymakers in Asian cities. To seek new sources of economic growth, Asian cities may consider upgrading their institutional and economic infrastructure to meet international standards. In doing so, they may become a ‘second home city’ for MNEs seeking to achieve further globalization.
Limitations and Future Research Directions
This study has some limitations, and our findings should be interpreted with caution. First, this study considers the Asian region as the home region for Chinese MNEs without distinguishing among Asian countries in detail. Chinese MNEs may have different strategies for choosing Hong Kong or Jakarta as locations for their ARHQ. We explored locations in Hong Kong and Singapore versus other cities in our robustness test. Future research may explore the strategic motivations behind ARHQ establishment in other sub-regions, countries, or cities.
Second, although we argue that ARHQ may serve as springboard, we cannot systematically test this argument. Our cross-sectional data limited us from testing the causal inferences and dynamic evolution of ARHQ establishment and regionalization strategies. Future studies may consider using longitudinal data to explore these areas.
Third, we also note that the reasons why some Chinese firms establish regional headquarters include not only value-creating activities but also some special purposes, such as cash flow and tax benefits (Meyer et al., Reference Meyer, Ding, Li and Zhang2014; Morck et al., Reference Morck, Yeung and Zhao2008; Sutherland & Anderson, Reference Sutherland and Anderson2015). Although our interviews with senior managers of some Chinese firms suggest that the major reasons for the establishment of their ARHQ related to value-creating activities, which are consistent with our hypotheses, future studies can further examine the weight of the value-creating activities so that we could have a more complete picture of the functionality of ARHQ.
Fourth, because Chinese MNEs are at their early stage of overseas expansion, few MNEs have established ARHQ, although an increasing number of them show the intention to do so. Our small sample size also dictated that our results should be interpreted with caution. Future research may test our hypotheses using a larger sample as Chinese MNEs expand their international operations to a larger scale.
Fifth, we focus on economic integration when examining home-city regional integration. We acknowledge that regional integration has economic, political, and cultural aspects. However, we believe that economic integration is a result and a manifestation of geographic proximity and historical political, business, and cultural links (Anderson & Marcouiller, Reference Anderson and Marcouiller2002; Ghemawat, Reference Ghemawat2001; Portes & Rey, Reference Portes and Rey2005; Rauch, Reference Rauch2001). Economic integration is primarily a result of formal political efforts embodied in the free trade agreements made between China and other Asian economies, such as those negotiated at APEC summits, CAFTA, and the Trilateral Agreement. Additionally, business and social networks that have historically operated across national borders have helped to decrease various informal trade barriers, such as weak enforcement of international contracts (Anderson & Marcouiller, Reference Anderson and Marcouiller2002) and inadequate information about international trading opportunities (Portes & Rey, Reference Portes and Rey2005). In this way, they have promoted economic integration and intraregional trade (Rauch, Reference Rauch2001).
Sixth, FDI and trade are not mutually exclusive and represent a continuum rather than discrete alternatives in many cases. An MNE may engage in both intraregional FDI and interregional trade simultaneously. In further consideration of this possibility, we attempted to insert the interaction between intraregional FDI and interregional trade in our regression model and predicted its positive relation to ARHQ establishment. However, we found no significant results for such an interactive effect and therefore did not report the results.
Finally, although our study focused on China, we believe that our findings may be applied more generally to other large emerging economies, such as India or Brazil, which exhibit regional differences in terms of the quality of their legal and other institutions. Studies conducted in those countries would help to determine the generalizability of our results.
CONCLUSION
Building on contemporary internalization theory, we developed a model to examine the determinants of Chinese MNEs’ establishment of ARHQ. Because Chinese MNEs lack international operational experience, they must establish ARHQ as a platform to augment the upstream and operational FSAs in their intraregional FDI and to indicate their legitimacy as well as complement the downstream FSAs in their interregional export activities. Home-city regional integration moderates the relationship between interregional exports and ARHQ establishment, and an MNE's international management capability moderates the relationship between its intraregional FDI and ARHQ establishment.
APPENDIX I Survey questions for some key variables
Part 1. Establishment of Asian regional headquarters (ARHQ)
-
1. Has your firm built ARHQ?
-
a. Yes; b. No
-
-
2. Where are your ARHQ?
-
a. Hong Kong; b. Singapore; c. Other Asian country
-
-
3. If your firm has not built ARHQ, do you plan to establish them within the next three years?
-
a. Yes; b. No
-
-
4. Where do you plan to establish your ARHQ?
-
a. Hong Kong; b. Singapore; c. Other Asian country
-
Part 2. FDI vs. trading regionalization
-
1. In 2010, what was the proportion of your exports to Asia to your total exports?
-
2. How many subsidiaries did you have in Asia at the end of 2010?
-
3. How many subsidiaries did you have in Europe at the end of 2010?
-
4. How many subsidiaries did you have in North America at the end of 2010?
-
5. How many subsidiaries did you have in Latin America at the end of 2010?
-
6. How many subsidiaries did you have in Africa at the end of 2010?
-
7. How many subsidiaries did you have in Australia at the end of 2010?
Part 3. International management capability
-
To what extent do you agree or disagree with the following statements (1 = strongly disagree; 7 = strongly agree)?
-
a. We are capable of managing overseas operations.
-
b. We are capable of adapting to the foreign cultural distance during overseas operations.
-
c. We are capable of learning and absorbing foreign advanced technology during overseas operations.
-
Part 4. Control variables
-
1. Performance
Please rate your level of satisfaction with your firm's sales, sales growth rate, market share, growth of market share, margins and margin growth compared with those of your domestic competitors (1 = totally dissatisfied; 7 = totally satisfied).
-
2. International experience and technological capability (reverse coded)
-
To what extent do you agree or disagree with the following statements (1 = strongly disagree; 7 = strongly agree)?
-
a. My firm lacks international experience.
-
b. My firm's products and technology lack competitiveness.
-
c. Customers in the host countries do not know much about my company and product.
-
-
3. Management overseas background
Is your firm's CEO an overseas returnee?
-
a. Yes; b. No
-
