Which Way through the Open Door? Reflections on the Internationalization of Chinese Firms



Received internationalization theory argues that firms occupy domestic space before going abroad; in other words, large, oligopolistic firms are most likely to internationalize. The experience of China, whose economy is fragmented and whose firms are small by global standards, suggests otherwise. We construct a model of small firm internationalization driven by the relative transaction costs of crossing domestic (in the case of China, provincial) and international borders. When the costs of crossing domestic borders exceed the costs of crossing international borders, firms will internationalize at a relatively early stage of development. In the case of China, local protectionism and inefficient domestic logistics increase the costs of doing business domestically; moreover, protection of property rights in the West and the advantages afforded Chinese owned firms reconstituted as foreign entities operating in China decrease the costs of ‘going out’. We coin the term ‘institutional arbitrage' to capture Chinese firms’ pursuit of efficient institutions outside of China. We argue that strategic exit from the home country rather than strategic entry into foreign markets may explain the internationalization of many Chinese firms.


In the wash of the Cultural Revolution in 1978, China initiated an ‘open door’ policy with an aim to reconnect with the world and to attract foreign investment and modern technology. At the time, both Chinese policy-makers as well as prospective foreign investors saw movement through the door flowing essentially one way – from the outside-in. Most of the policy decisions faced by the Chinese leadership at the time concerned how to handle foreign firms on Chinese soil (Wang & Chen, 1984). Now, suddenly, after 30 years of unprecedented economic growth, the traffic through the door is beginning to move in the other direction with the Chinese government actively encouraging firms to operate abroad and setting up the China Investment Corporation (CIC) to assist them in the process (Brainard & Fenby, 2007). Thus, not only has China become the world's largest recipient of inward foreign direct investment (FDI) – the inward flow of FDI into China increased from US$1 billion in 1983 to US$53 billion in 2003 and US$60 billion in 2004 and 2005, not counting investments into the financial sector, US$3 billion in 2004 and US$12 billion in 2005 – but the country is beginning to invest significant assets abroad. Indeed, according to a survey of investment promotion agencies published by the United Nations Conference on Trade and Development (UNCTAD, 2004), China is poised to become one of the world's biggest overseas investors. Garelli (2003), writing in IMD's World Competitiveness Yearbook, argued that Asian nations, led by China and India, are set to engineer a profound shift in the global economic landscape. According to Garelli, nations such as China and India will beset Western markets on an even vaster scale than Japan. Yet a 2005 report by the Economist Intelligence Unit argues that Chinese enterprises are not nearly as strong as they should be, given the domestic market opportunities they enjoy (Economist Intelligence Unit, 2005).

So, how should we be thinking of the incipient move abroad by Chinese firms? Do current theories of internationalization offer us the right kind of guidance for dealing with the challenges posed by an internationalizing China? And if not, would we be better off seeking a more ‘China specific’ theory to account for the particular facts of the Chinese case, or should we rather be aiming for a more comprehensive rethinking of our existing theories (Erdener & Shapiro, 2005)? In what follows, we provide grounds for thinking that make sense of China's initial moves abroad and, while not actually invalidating existing theories of internationalization, require a shift of perspective that would simultaneously place them in a new light and extend them in new directions.

The paper is structured as follows. First, we briefly look at early and current theories of internationalization. Most of these theories have had the internationalization of the large Western firm as their primary focus, although in recent years this has begun to change. Next, we examine how certain key features of China's economy changed following the introduction of the open door policy in 1978. We argue that administrative decentralization has led to the paradox of a hypercompetitive yet fragmented domestic economy in which transaction and other operating costs remain extraordinarily high. This has made it difficult for domestic firms to build up any competitive advantage that would remain either sustainable or exploitable outside the domestic space. Next, we discuss the Chinese case in the light of existing internationalization theory and introduce the notion of institutional arbitrage, a process that allows firms whose domestic growth is constrained by domestic transaction and operating costs to exploit the ready availability of low-cost institutional and cultural assets located outside their domestic space. In this section, we also present the implications of our analysis in a graphic form. Finally, a conclusion follows in the last section.


Internationalization has today become a major strategic component of most business firms' activities (Meilin, 1992). Key models of the process describe it as taking place gradually and in distinct stages (Johanson & Vahlne, 1990; Meilin, 1992; Vernon, 1966). As they gradually learn about foreign markets, firms shift progressively from exporting to higher forms of involvement such as FDI in entering foreign markets (Zhao, Luo, & Suh, 2004). Such models assume that it is more costly for a firm to operate abroad than in its domestic market, and some further assume that learning how to operate abroad is also more costly than learning how to operate in the domestic market, thus incurring what Hymer (1976) labeled a ‘liability of foreignness’. The extra costs associated with operating abroad have to be counterbalanced by some compensating advantages (Caves, 1982; Hymer, 1976; Kindleberger, 1984). Dunning (1980, 1988) identified three of these: ownership, localization and internalization (the OLI framework).

Dunning's framework looks to economic theory and, in particular, to transaction and factor costs for its main explanatory variables. Arrow (1969) defines these broadly as the costs of running the economic system (p. 48). Coase (1937) conceives them more narrowly. For him, market transaction costs are incurred in discovering who it is that one wishes to deal with, in informing the people who one wishes to deal with of the proposed terms, in conducting the negotiations leading up to a bargain, in drawing up the contract, in undertaking the inspection needed to ensure that the terms of the contract are being observed and so on. In what follows, we shall draw on Arrow's more broad-based concept of transaction costs and refer to such costs as Arrovian TCs. Where necessary, we shall refer to the more narrowly focused concept of transaction costs as Coasian TCs.

However they are defined, all types of transaction costs are sensitive to the spatial and institutional distances that separate transacting parties. Crossing national boundaries is assumed to impose a number of cultural and institutional discontinuities upon firms that can increase the problems of bounded rationality and the scope for opportunistic behaviour that they face. These discontinuities have to be bridged in ways that typically increase transaction costs. Thus, we can hypothesize a positive relationship between transaction costs and spatial distance that will vary according to the nature of the space that has to be crossed.

Most of the early work on internationalization tended to focus on the large firm. But the assumption in the internationalization literature has been that a firm first grows large in the domestic market on the back of some market based or product based competitive advantage and then goes abroad, using that advantage to compensate for the added costs of operating abroad. Yet if, as Ansoff (1965) has argued, geographical expansion into new markets is one of the most important paths for firm growth, it constitutes a particularly important growth strategy for small and medium-sized enterprises (SMEs) whose business scope has been geographically confined (Barringer & Greening, 1998). During recent years, a significant development within the broad internationalization trend has been the increasingly active role played by SMEs in international markets (Oviatt & McDougall, 1994, 1999; Wolff & Pett, 2000).

The internationalization of SMEs can be expected to gain further momentum because the world economy is becoming increasingly integrated with continued declines in government imposed barriers and continued advances in information and communication technology (ICTs) (Lu & Beamish, 2001). For this reason, the distinction between domestic and foreign operations is now eroding (Geringer, Beamish, & Costa, 1989). The newly internationalizing SMEs are nevertheless particularly vulnerable to Hymer's (1976)‘liability of foreignness’ and to Stinchcombe's (1965)‘liability of newness’. The liability of foreignness means that the global entrepreneur may incur higher costs than local host country competitors. This initial disadvantage may diminish over time as the global entrepreneur accumulates experience in host country markets. Yet while large firms may well be able to absorb the costs associated with the liability of foreignness, is this also true for SMEs? We contend that, although for SMEs the liability of foreignness is the primary concern when first entering foreign markets via FDI, global entrepreneurs may still experience lower costs outside their home market than inside it, particularly when their home markets are emerging ones. Later, we argue that this is indeed the case for Chinese entrepreneurs in small and private SMEs not protected by the state.

We summarize the picture presented by the literature on internationalization as follows. Firms grow in response to managerial and other slack (Penrose, 1959). At some point, they outgrow their possibilities in the domestic market, whether or not they have succeeded in occupying the whole of the domestic market space. They then look for opportunities to expand abroad (Ansoff, 1965). But, given the liability of foreignness, the spatial distances and the institutional barriers involved, operating abroad is costlier than operating in the domestic market (Hymer, 1976). Firms therefore need some competitive advantage to compensate for the extra costs of first moving and then operating abroad (Caves, 1982), and firms that have grown domestically on the back of some competitive advantage are now well placed to move abroad to exploit that advantage (Dunning, 1988). Thus, firms that do move abroad tend to be large and oligopolistic (Knickerbocker, 1973). That said, ICTs and globalization are now lowering the costs of moving and operating abroad, making it easier for small firms to do so (Silverman, 1999).

How useful a guide will the above prove to be to the internationalization of Chinese firms? In what follows, we argue first that the economic reforms have further contributed to the fragmentation of an already fragmented domestic economy and that this has raised domestic Arrovian TCs above those that prevail outside the country (Child & Rodrigues, 2005). Consequently, we argue secondly that the internationalization of SMEs in China is best framed as the pursuit of an exit strategy from the domestic market rather than as the pursuit of an entry strategy into foreign ones.


China's open door policy was forged in the context of administrative decentralization. Today, there are three prongs to this decentralization. The first is delegation of responsibility for economic performance to the provinces and larger municipalities, whereby performance targets for local government officials replace direct administrative control (Liu & Tao, 2004). The second prong is fiscal decentralization (Dougherty & McGuckin, 2008). Beginning in 1988, a revenue contracting system was implemented under which, much like the early European practice of tax farming, provincial governments each negotiated a fixed tax quota with the central government, with collections above that level being retained at the local level. The third prong of administrative decentralization is the delegation of control of most large state-owned enterprises (SOEs) to provincial and municipal governments and of smaller ones to city and county governments. This move instantly created several hundred thousand firms, almost all miniscule by global standards (Meyer, Lu, Lan, & Lu, 2002). Today, in consequence, managers of all but the largest state-owned firms have to contend with the pervasive influence of local government on their business activities. This enormously complicates the task of building firms with substantial national, much less local global, market share (Meyer & Lu, 2005).

One must distinguish between administrative decentralization to local authorities and economic decentralization to firms (Boisot & Child, 1988; Huang, 2003). The former is a presumption in favour of socialism that aims to keep the associated information costs and incentive distortions manageable. The latter is a presumption in favour of capitalism. Yet administrative decentralization has led to a feudalization of China's industrial structure and an economic fragmentation of the national economic space (Boisot & Child, 1988). The majority of SOEs came under the control of local territorial units, each pursuing a local economic agenda and protecting ‘its’ firms, whether state-owned, collective, or even private, through various anti-competitive measures. Wu (2005, p. 56) has labelled an administratively decentralized China a ‘vassal economy’ dominated by local protectionism.

The Chinese central government's response to economic fragmentation has been limited. Yet market fragmentation and local protectionism inflict substantial Arrovian TCs on firms even though the policy of administrative decentralization has reduced the state's administrative costs. We identify four broad classes of costs.

Capacity Costs

Fragmentation of the Chinese economy has led to the miniaturization of Chinese firms, even in those industries where size constitutes a major source of competitive advantage. Relentless capacity growth in China, mainly due to local governments'penchant for fixed asset investment, has led to economy wide overcapacity and, hence, a severe profit squeeze.

Operating Costs

According to the China Federation of Logistics and Purchasing total logistics costs in 2004 were 21.6 percent of China's GDP (China Economic Review, 2007). The December 2005 recalculation of China's 2004 GDP would reduce this figure somewhat to 18 percent (Holz, 2006). This percentage is quite high in comparison with developed countries where logistics costs are at around 10 percent of GDP. Overland transport is especially costly in China. A rule of thumb in China is that it costs more to transport goods from Chengdu, the capital of Sichuan Province, to Shanghai than from Shanghai to New York.[1] From most parts of China, then, access to foreign markets is easier and cheaper than access to most of the country's domestic markets.

Coasean TCs

These are the costs of consummating transactions but are usually construed as the costs of writing and executing contracts. The decision to internalize transactions rather than having them pass through markets only explains the existence of large firms when it is based on economic rather than administrative criteria. The large firms that were created in China were brought into existence primarily for the administrative convenience of central planners and, thus, rarely achieved the efficiency savings associated with transaction cost arguments (Boisot & Child, 1988). Compounding this problem was the fact that the economic fragmentation of the country made transactions among small, geographically scattered producers and customers very costly and impeded the possibilities of an economically rather than an administratively driven organizational integration. In the context of this paper and of internationalization theory, transaction costs also refer to the costs associated with doing business across boundaries – national, provincial, industrial, administrative, spatial, etc. To the extent that doing business across boundaries serves as a proxy measure for doing business at a distance, the received proposition, outlined above, becomes as follows: the greater the ‘distance’ created by boundaries, the greater the costs of doing business.

Administrative Costs

Fragmentation of the Chinese economy has imposed substantial administrative costs on firms operating across provincial boundaries. The administration of the corporate earned income tax has been especially burdensome. In principle, all private enterprises founded after 2001 make earned income tax payments to the State Administration of Taxation or SAT. In fact, provincial and municipal tax bureaus have continued to collect earned income tax payments from private enterprises. A further administrative cost affecting firms operating across provincial boundaries arises from inconsistent implementation of intellectual property laws. While trademarks and trade names are registered with the central government, their protection is delegated to local administrations of industry and commerce (AICs) rather than the judiciary (State Council Information Office, 2007) – a further stimulus to local protectionism.

In sum, China's policy of administrative decentralization, which was aimed at unclogging blockages in the central bureaucracy, had the unintended consequence of clogging the arteries of the economy. Administrative decentralization fragmented the economy by pitting provinces and municipalities against each other in competition to achieve economic targets. This led to redundant capacity and both formal and informal barriers to domestic trade.

Foreign firms operating in China face the same capacity, operating, transaction and administrative cost problems as do Chinese firms. Yet, because so much of their value chain is located outside the country –Dunning (1988) would label this their localization advantage – they enjoy a competitive advantage that remains unavailable to domestic firms as long as the latter remain purely domestic. Furthermore, the size of foreign firms, to a far greater extent than that of their Chinese counterparts, reflects economically driven rather than administratively driven internalization decisions. For them, therefore, the cumulative internalization decisions – i.e., to conduct transactions through internal hierarchies rather than markets (Coase, 1937) – that account for their larger size express an underlying competitive strength rather than, as has often proved to be the case for large Chinese firms, an underlying competitive liability. Finally, given the continuing ambiguity that attaches to property rights in China, foreign firms enjoy an ownership advantage for the coordination of their operations over a far greater part of their value chain, namely, that part located outside China, than is the case for their domestic Chinese competitors. What we see, then, is that foreign firms entering the Chinese market enjoy all three of the competitive advantages – ownership, localization and internalization – identified by Dunning (1980, 1988) in his OLI framework. We hypothesize that this situation provides powerful incentives for Chinese firms to move some of their operations abroad, if only to neutralize the advantage enjoyed by foreign firms in the Chinese domestic space.

All of the above is consistent with received internationalization theory. In what follows, however, we offer a somewhat different interpretation of the challenges confronting Chinese firms.


Given the higher Arrovian TCs that Chinese firms confront at home compared with abroad and given the intensity of foreign competition that they encounter in their domestic market, it is plausible to argue that Chinese firms will internationalize at an earlier stage of their development than anticipated by received theories. This raises a question of framing: is such a forced internationalization process best viewed as an entry strategy into foreign markets or as an exit strategy from the domestic market? Clearly there is no full exit. Chinese firms will keep a domestic operation going, if only for the option value that it presents. Now, if all other countries were like China, then Chinese firms would realize no advantage by going abroad; indeed, all of the disadvantages of doing so assumed in the received theory of internationalization would apply. However, as we have seen, China's economy is unusually fragmented, even in comparison with countries at a similar level of development, and this fragmentation, largely an outcome of the policy of administrative decentralization, has made it costly for businesses to operate between provinces. China's logistics costs, for example, are higher than India's. Though estimates of logistics costs in proportion to India's GDP vary from 10 to 15 percent,[2] all of these estimates are substantially below the 18 to 21 percent reported consistently in China. China's airports, railways, roads, bridges and tunnels, then, do not appear to be the problem. In the coastal and central regions of the country, these are more modern than India's. Rather, it is the way business is done in China – in short, Chinese institutions.

If, in fact, the source of blockage inside China is institutional rather than physical, then, arguably, Chinese firms going abroad are doing so in pursuit of more efficient institutions. In other words, in an effort to neutralize the localization advantages of foreign firms in China, they are engaging in what we might label ‘institutional arbitrage’, interpreted here as an exploitation of the differences between different institutional arrangements operating in different jurisdictions.

The notion of institutional arbitrage is sharply at odds with the standard view of internationalization. The standard view portrays firms based in emerging economies as seeking ever lower operating costs and, hence, locating their subsidiaries in ever poorer countries (Wells, 1983). Institutional arbitrage, by contrast, portrays firms from emerging economies as seeking out institutions capable of supporting international operations and, hence, incorporating, listing and establishing selected parts of their operations – often involving headquarters related activities – in developed economies. Such arbitrage is already in evidence among Chinese firms. ‘Round tripping’, the practice of Chinese domestic firms shifting assets out of the country and coming back in, posing as foreign investors, provides one example of an institutional arbitrage operation (Huang, 2003). The opportunities and need for Chinese firms to practice institutional arbitrage, we believe, make China an important exception to received theories of internationalization but not, as we shall argue presently, the only exception.

In the received theory of internationalization, the capacity, operational, transaction and administrative cost increases associated with distance generated by domestic boundaries are low. As with first class postage within a country, many of these costs remain the same regardless of distance. At the national border, however, there is a discontinuity in the costs caused by a bunching of various boundaries there: linguistic, legal, monetary, psychological, etc. Beyond the national border, costs are assumed to increase noticeably faster. Thus it costs more to do business abroad than at home and the more so as distance increases beyond the national border. Received internationalization theory asks what kinds of firms can bear these costs. Almost always the answer is large, dominant firms that have built up some competitive advantage in their domestic market.

Figure 1, however, suggests a different, somewhat more complicated and – at present, at least – China specific story. It is this: within the home province, Arrovian TCs can be taken as low and constant. However, as indicated in the figure, these costs rise sharply with distance beyond the provincial border but drop in a step function at the national border so that it costs far less to do business with the nearest country than with the furthest province. Once beyond the national border, Arrovian TCs continue to increase with distance, but the slope or rate of increase with distance is much lower internationally than domestically. Several implications of the figure should be noticed. First, the step function cost reduction at the national border is the driver of institutional arbitrage: doing business abroad is cheaper because the costs of blockages at home exceed the costs of crossing national borders, or, more fully, the costs of crossing national borders less the benefits of foreign institutions. Second, firms from point A on the distance from home base scale to the national boundary will leapfrog domestic space at an early stage rather than occupying it more fully prior to going abroad because it is cheaper to do so. Moreover, firms will leapfrog as far as point B on the distance scale depending on where they started. One firm currently leapfrogging domestic space is Gome (pronounced go-me) Appliances, China's largest retail electronics dealer. Gome's founder, and by some accounts the wealthiest entrepreneur in China, Huang Guangyu, has stated his intention to internationalize his business because costs are more favourable abroad than at home.[3] A firm ignoring domestic space altogether is Donlim,[4] located in Shunde Township, Foshan City, Guangdong Province, which designs and manufactures small electrical appliances – tea kettles, coffee makers, hand mixers and the like – for distribution by multinationals. Donlim claims 70 percent of the European and 40 percent of the US market in these categories but has little or no domestic distribution. The third implication is that the contest between Chinese firms and foreign multinationals will be largely decided outside of China. Given the magnitude of domestic Arrovian TCs, Chinese companies retain little if any home country advantage and are disadvantaged by their small scale compared with the multinationals. However, other things being equal, Chinese companies that are able to leapfrog from their home province straight into global markets and expand the scale of their operations outside their home country will enjoy the same labour costs and Arrovian TCs as foreign multinationals manufacturing in China for export. ‘Other things’, of course, include corporate governance, management, marketing, and other critical business skills, most of which are not readily available in China.

Figure 1.

Revised model of internationalization
Firms between A and national border internationalize before occupying domestic space.

The expansion of small and medium-sized Chinese firms in the domestic market exposes them to the liability of newness. Their expansion abroad exposes them to the liability of foreignness. To grow successfully, they need to find a balance between these two types of liability. Our hypothesis is that, given the current Arrovian TCs they face as small firms in China, expanding in the domestic market could actually prove to be riskier for them than moving abroad. A more provocative way of putting this is that a small or medium sized Chongqing based firm attempting to invest in Wuhan could well encounter a greater ‘liability of foreignness’ in crossing provincial borders than if it crossed national borders and invested, say, in Singapore or even in London. If such a hypothesis has any validity, then we can infer that Chinese firms will move abroad at a smaller average size than their Western and Japanese predecessors did. They will be helped in this by the availability of new information and communications technologies (ICTs) that, by further reducing the costs of transacting across international borders, also further lower the size threshold at which it becomes viable for a firm to contemplate international operations. Figure 2 shows the impact of ICTs on the decision to internationalize. The dotted line in the diagram depicts transaction costs prior to the introduction of new ICTs whereas the solid line depicts transaction costs subsequent to the introduction of new ICTs. In effect, Figure 2 offers us a new variant of the late development effect. In explaining the rapid development of modern Japan, Dore argued that countries that industrialize late can build on institutions and technologies that were not available to those that industrialize early (Dore, 1973). And given that we are dealing with a path-dependent process, the trajectory followed by the early developers provides little basis for predicting that followed by the late developers. We believe that this argument will apply to the internationalization of Chinese firms.

Figure 2.

Model of internationalization with new ICTs
Firms between C and national border internationalize before occupying domestic space.

None of the above requires Chinese firms actually to abandon China – far from it. But they can reduce their footprint in the country and, hence, their exposure to all the sources of high Arrovian TCs that we have identified while maintaining enough of a presence there to built on in the future should the opportunity to do so present itself. From a foreign base, they will then enjoy an advantage in their domestic market that they were never able to exploit when they were purely domestic, namely, a lower psychic distance than their foreign counterparts (Johanson & Vahlne, 1990). We thus reach the paradoxical conclusion that in order to achieve competitiveness in their domestic market, small and medium-sized Chinese firms will first have to move abroad.

Finally, what about China's largest companies? What future trajectory will their internationalization take? Overseas mergers and acquisitions (M&A), at least in the USA and Europe, are challenging. The experience to date of large Chinese firms, keeping in mind that these firms are of modest size by global standards, in Western M&A markets is not promising. A five-member consortium of China Minmetals, Baoshan Iron & Steel Co., China International Trust & Investment Corp., Jiangxi Copper Co. and Taiyuan Iron & Steel Group Co. was frustrated in its attempt to acquire Noranda, Canada's largest metals mining company, due to political opposition and rising global commodity prices. CNOOC (China National Offshore Oil Co.) was similarly frustrated in its effort to acquire Unocal, which was ultimately acquired by Chevron. Haier's US$1.3 billion offer for Maytag was more than doubled by Whirlpool, leaving Whirlpool with a nearly two-thirds share of the US appliance market (http://ApplianceMagazine.com, 2005). Lenovo acquired the IBM Personal Systems Division in 2005 but had to cede control of Lenovo management to a consortium of US private equity firms to effect the acquisition and has only recently been able to reverse losses in the USA. Recently, CHALCO (Aluminum Corporation of China) and Alcoa jointly acquired a 12 percent stake in Rio Tinto, and rumours abound that Chinese investment in Rio Tinto's archrival BHP Billiton is imminent. Arguably, internationalization in these instances has been in pursuit of natural resources (Minmetals, CNOOC, CIC's bid for Rio Tinto) and prestigious global brands (Haier, Lenovo) rather than exploitation abroad of competitive advantages developed at home. Moreover, in the case of Lenovo, strategic exit and institutional arbitrage are evident: Lenovo was far below the global scale prior to the acquisition of IBMPSD, the acquisition would not have gone through without intervention by US investors, Lenovo headquarters has been relocated from Hong Kong to North Carolina and Lenovo's top management today is largely Western or Western educated.

Only fundamental institutional changes would allow Chinese firms to grow to global scale by first developing competitive advantages at home and then exploiting these advantages abroad. This would make their internationalization process less of an exit from China and more of a conventional entry into overseas markets. One such set of institutional changes would lower domestic trade barriers by eliminating discriminatory tolls and local taxes and substituting uniform national regulations for inconsistent provincial regulations governing commerce; another set of changes would encourage industry consolidation by transforming all enterprises into shareholding firms, eliminating the fiction of ownership by ‘the whole people’ and encouraging listing of group corporations while discouraging listing of their subsidiaries. China's recent history suggests that changes of this magnitude could be accomplished speedily provided there was willingness to reform, which in this instance would mean curbing local privilege. This is one of the aims of the 2007 Anti-Monopoly Law. However, the Anti-Monopoly Law has not yet been implemented, and it remains to be seen whether the Law will have its intended effect (Fox, 2008). Given that Chinese enterprises have been legally separated from the state only since 1988 and Chinese company law has existed only since 1993, it is not surprising that a local rather than a national mindset still prevails and that fragmentation rather than excessive societal coordination of the economy (Witt & Lewin, 2007) causes Chinese firms to pursue exit options.


With globalization and increasing connectivity, organizational factors of production are becoming more mobile. The phenomenon of headquarters relocation, for example, is on the rise and is becoming increasingly important. To illustrate: Tetra Pak moved its headquarters from Sweden to Lausanne following the owner's tax dispute with the Swedish government; the relocation of SAB-Miller and Anglo American to London followed the uncertainties associated with the arrival of ANC rule in South Africa; Indian pharmaceutical company Aurigene's relocation of its headquarters to Boston was driven in large part by the more stringent patent regulations for drug development in the USA. There are also cases of multinational corporations from Mexico, China, Brazil and India moving their headquarters to major financial and commercial centres. A 2003 report from UNCTAD confirmed the emergence of a world market for corporate headquarters (UNCTAD, 2003; see also Birkinshaw & Hood, 1998).

The implications of such trends are clear. Countries in the future will be competing to retain their own firms in their domestic space as much as they will be competing to attract foreign ones. We believe that China will illustrate the functioning of this new dynamic in a particularly stark form, but its case will hardly be unique. We see the behaviour of Chinese firms with respect to their home country as conveying a wider message. Given the choice of exit, voice and loyalty (Hirschman, 1970), Chinese firms concerned with their future survival will have stronger reasons to choose exit than the other two options. Yet in a global world, does exit really mean exit? After all, as the practice of round tripping has shown us, moving abroad first can increase a firm's bargaining power when returning home. In effect, what our analysis highlights is the growing irrelevance of the distinction between home and host country under conditions of increasing factor mobility.

This paper has put forward a hypothesis that is empirically testable: on account of high domestic Arrovian TCs, Chinese firms will internationalize at a smaller size than their Western and Japanese counterparts. They will do so in order to escape the competitive disadvantages that they confront in the domestic market and that outweigh the competitive advantages of a large market size. Specifically, they engage in what we call institutional arbitrage to capture advantages of the same legal and economic protections outside China enjoyed by foreign firms operating within China. Huang (forthcoming, p. 11) makes the same point if more dramatically: ‘China's success has less to do with creating efficient institutions and more to do with permitting access to efficient institutions outside of China.’ Given the lower Arrovian TCs associated with the move abroad, Chinese firms will be less in need of a competitive advantage in doing so. Most of the literature on foreign operations has framed these in terms of strategic market entry. Our analysis points to strategic market exit as being of equal interest even if, in the case of Chinese firms, the exit is likely to prove temporary.


  • [1] 

    See, for example, http://www.drewry.co.uk/get_file.php?id=999, which states that ‘Inland transport costs to the coast from provinces like Sichuan, for export to overseas markets, are often higher than the maritime transport cost from China to the destination port . . .’ (Drewry Shipping Consultants Limited, 2007).

  • [2] 

    For example, in 2004, Mr. N. Kumar, Chairman of Confederation of Indian Industry's National Council of Logistics, pegged Indian logistics costs at 13 percent of GDP and logistics costs in developed countries at 6–7 percent (Hindu Business Line, 2004). Similar figures have been reported as recently as October 2007 (Hindu Business Line, 2007).

  • [3] 

    Presentation at Boao Forum CEO Summit, Shenzhen, November 20, 2005.

  • [4] 

    Donlim's legal name is Guangdong Xinbao Electrical Appliances Holdings Co. See http://www.donlim.com. The data on Donlim's distribution are from an unpublished interview with Donlim management on January 21, 2008, conducted by Marshall W. Meyer.

Max Boisot (boisot@attglobal.net) is Professor of Strategic Management at the Birmingham Business School, the University of Birmingham; Senior Research Fellow at the Snider Center for Entrepreneurial Research, The Wharton School, the University of Pennsylvania; and Associate Fellow at the Said Business School, Oxford University. He holds a Ph.D. in technology transfer from the Imperial College London University. His current research interests focus on the nature of knowledge flows within and between organizations and cultures.

Marshall W. Meyer (meyer@wharton.upenn.edu) is the Richard A. Sapp Professor of Management in The Wharton School as well as Professor of Sociology and Associate Member of the Center for East Asian Studies at the University of Pennsylvania. He holds a Ph.D. in sociology from the University of Chicago. His publications in MOR include ‘Managing Indefinite Boundaries: The Strategy and Structure of a Chinese Business Firm’ with Xiaohui Lu (MOR 1.1, pp. 57–86), ‘China's Second Economic Transition: Building National Markets’ (MOR 4.1, pp. 3–15) and ‘No Free Lunch: Dilemmas of Product Quality in China’ (MOR 4.2, pp. 157–165).