Beyond the enclave: towards a critical political economy of China and Africa

This article provides a political economy framework for analysing China’s engagements with Africa. It situates the rise of China in the context of the changing balance of power in the world system and particularly China’s re-entry into spheres of influence in Africa that have been the purview of the former European colonial powers for two centuries or more. It begins by arguing that current approaches to China, Africa and international relations are fragmented in particular ways which prevents the development of a more critical political economy. It then examines a pervasive theme in China–Africa relations, which assumes that the Chinese work through enclaved investments to secure the resources of low-income economies, though in this sense the Chinese are no different from other investors. Where they do differ is in their bundling of aid, trade and FDI and their use of imported labour, which has been termed ‘surgical colonialism’. The article does not dispute the existence of Chinese enclaves, but argues that we need more empirical evidence on the levels of labour importation in relation to local labour market conditions. This requires a more nuanced understanding of state–capital dynamics in those countries where the Chinese operate although the model appears to be one of elite brokerage. However, the enclaved investments and inter-elite bargaining are only part of the story and the closing sections analyse the role of independent Chinese businesses in Africa’s social and political development, which moves us beyond the enclave. First unnumbered footnote: I would like to acknowledge the support of the UK Economic and Social Research Council in funding the research project (Refs: RES-062-23-0487 and RES-075-25-0019) from which this paper derives. More details are available at www.geography.dur.ac.uk/projects/china-africa/ and http://risingpowers.open.ac.uk/. The comments of the anonymous reviewers were really helpful in helping shaping this paper.

I would like to acknowledge the support of the UK Economic and Social Research Council in funding the research project (Refs: RES-062-23-0487 and RES-075-25-0019) from which this paper derives. More details are available at www.geography.dur.ac.uk/projects/china-africa/ and http://risingpowers.open.ac.uk/. The comments of the anonymous reviewers were really helpful in helping shaping this paper.

INTRODUCTION
While the China in Africa story has been headlines for nearly a decade (e.g. Hilsum, 2006) there have been relatively few attempts to theorize what this means for the political economy of international development (Alden and Large, 2011;Ayers, 2013;Campbell, 2008;Tull, 2006). Much of what has been written is theory as hypothetical conjecture which posits the emerging China-Africa relationship in either zero-sum or positive-sum ways and is supported by limited and partial data. A typical example was the accusation that China's concessional lending is a form of 'rogue aid' (Naim, 2007) countered equally bluntly by the official Chinese argument that its interventions in Africa can only yield 'win-win' outcomes involving mutually beneficial business partnerships. A second approach to analysis is more inductive in focusing on a bilateral linkage between China and a particular country and analysing what this means for local development. Such studies have been invaluable (see Ampiah and Naidu, 2008;Corkin, 2013) in countering the first tendency of unfounded speculation, but they have failed to cohere into a sustained theorization of the broader processes at work (Henderson, 2011). Moreover these inter-state studies fall, as Ayers (2013) usefully argues, into a methodological statism that ignores wider structural forces; a critique I revisit in this article. In contrast to these ideological and/or descriptive approaches to theory building, Cox (1987: 32) argues that we can produce 'critical' theory 'concerned with how the existing order came into being'.
This article attempts to construct a critical political economy of China's engagements with Africa both deductively and inductively using data collected in China, Africa, and Europe (Tan-Mullins et al., 2010;Power et al., 2012), as well as engaging with a range of secondary sources. To this end, the article proceeds with an analysis of the difficulties of theorizing 'South-South' international development in view of the fragmentation of academic knowledge. I use these fragmentations to identify the need for inter-disciplinary approaches that keep in play the global and local. From here, I examine those approaches to China-Africa that use the ideas of the enclave and dispossession, which are rooted in Marxian political economy. While useful, I argue that they are partial and that we need to examine the dynamics of labour recruitment and labour relations, as well as how smaller Chinese firms operate since these are necessarily less enclaved.

FRAGMENTED STATES OF THEORY
The ability of academic research to fashion a political economy of China-Africa relations has been hindered by three sets of mutually reinforcing fragmentationsregional, disciplinary and representational. One fragmentation is between experts on China's development, and those studying the impacts of China on various parts of the global South. The former focus on the country's internal dynamics and how China projects itself internationally, particularly in relation to existing powers (e.g. Ash et al., 2007;Shirk, 2007), but include little analysis of the effects on developing countries. Some of these studies characterize Chinese actors as relatively state-dominated and guided by a centralized logic producing the 'Beijing Consensus' on development (Ferchen, 2013;Halper, 2010). The second group tends to be scholars of other countries or regions who assess the difference China makes to the development prospects of the context they study, but who know little about the internal dynamics that drive China's behaviour. Thus we have regionally specific studies (e.g. Gonzalez-Vicente, 2012;Jenkins et al., 2008;Rotberg, 2008) that, as noted, fail to cohere into a systematic framework. Alongside western 'stereotypes', the Chinese literature on China's international development interventions is largely ideological in portraying China's relations with other Southern countries in uncritical 'win-win' terms (e.g. Li, 2007), with limited empirical research.
Hence, there is a need to link disparate pockets of regionally-specific knowledge, and to expand beyond our limited (and often value-laden) perspective, in order to analyse how China's domestic and overseas development processes relate and with what effects at a series of scales.
The second fragmentation is disciplinary. China's development footprint has been conceptualized (Kaplinsky, 2008) as comprising a series of inter-linked modes of interaction: trade links, investment flows, aid, institutions of global governance, flows of people, and environmental implications. Not only were these modes of interaction theorized in relative ignorance of China, but most research to date has focused on the economic relationships (Jenkins et al., 2008;Kaplinsky and Morris, 2008), and on those areas where data are available, notably trade and Foreign Direct Investment (FDI) (UN, 2007). As a result, analysts in the development field have tended to treat these linkages in isolation and to underplay the politics of these international relations. By contrast, scholars of international relations have tended to focus on Sino-US relations (Ikenberry, 2008;Wang and Pauly, 2013) or other major power relationships (Hurrell, 2006) whereas political scientists of China have often focused on the elite politics of the Party-state leadership (Kallio, 2011). Few have engaged with the development studies community or those concerned with core issues of development, such as well-being and equity and environmental sustainability. We thus need an inter-disciplinary approach to understanding China's development role and impacts.
Playing into these first two fragmentations is a third, and one that predates studies of China in Africa, namely that African perspectives have largely been absent in the debates (Mohan, 2008). Using Spivak's (1988) framing of re-presentation, Africa is spoken of or spoken for in all these accounts. At best, African political elites appear in analyses of bilateral relations (e.g. Large, 2008), or in the vein of Junger (2007), western journalists select some horror show regarding 'poor' Africans and extrapolate to the whole continent.
Even the 'Asian Drivers' agenda discussed above (e.g. Kaplinsky, 2008), is premised on a conflation of all Asian countries as sharing some essential characteristics and that they do the 'driving', which denies African agency in this relationship. In turn this leads to (or arguably stems from) 'Africa', 'China' or 'Asia' being treated as singular and homogenous (Chan, 2008). Hence, we need to re-insert African agency into the picture (Corkin, 2013;Mohan and Lampert, 2013). Better ethnographic work around Chinese in Africa and Africans in China (e.g. Bredeloup, 2012) is emerging which disaggregates ideas of a singular 'Africa' or 'China' and teases out more complex and contradictory relationships across space and culture. I return to such approaches below.
In seeking to develop a more critical political economy, there are three possible scenarios.
The first is that China represents nothing particularly novel for Africa so that while its revived and enhanced dealings with Africa are new, we do not need additions to our conceptual toolkit, based largely on neo-Marxist readings of imperialism (see Ayers, 2013;Tull, 2006). A second scenario is that China is a different type of actor and presages an era of 'South-South' cooperation that will radically alter the geography of international development and so requires some new theoretical insights. Such logic has led to the 'Beijing Consensus' thesis; that China offers an alternative to established development ideologies. In this vein, some have called for a 'post-colonial' reading of China which posits it as different by being both powerful and sharing with Africa a history of being colonized by western powers (see Six, 2009). Chinese exceptionalism has also been treated in Weberian fashion, which sees China's capitalism as culturally embedded and informed by a deep-rooted Confucian logic (Dirlik, 1997;Greenhalgh, 2004), which might engender different kinds of relationships with African economies. A third scenario is more complex and argues, like Ferguson (2006), that not only has the global political economy shifted as well as Africa's incorporation into it, but that China is a new type of actor which we poorly understand (Jacques, 2009).
This article seeks to develop the third of these scenarios but draws inspiration from some of the critical insights of neo-Marxist theory from scenario one, as well as the idea of 'decentring' the west from explanations of globalization from scenario two. As such, I am keen to take the more material concerns of political economy and put them into productive engagement with the culturally-centred approaches of post-colonialism (see Glassman, 2011). Post-colonial theory has argued for recognizing the complex intermingling of the subjectivities of colonizer and colonized so that we do not reduce the colonial relationship to one of singular, external domination but one where 'subaltern' agency may play a part. The rest of the article develops the analysis, first, by situating China within an analysis of global restructuring. From there I examine ideas around enclaves before complicating that analysis using my own and others' data.

A POSTCOLONIAL (GEO)POLITICAL ECONOMY OF CHINA AND AFRICA
As Harvey (1993) observed, political economy is necessary for making connections between seemingly unconnected events or processes. I start by situating the rise of China within a wider process of capitalist restructuring before examining how more localized processes of accumulation play out, particularly notions of how Chinese firms operate in enclaved ways.

Globalization, neoliberalism and the rise of China
Some African economies are rising after two decades (arguably much longer) on the back burner (McKinsey, 2010). Without reducing this tentative success to any single factor, the increased demand for African resources from the growing economies of Asia has been important for both increasing demand and seeing the appreciation of commodity prices (Kaplinsky and Farooki, 2011). Hence, one starting point in examining the political economy of China in Africa is to understand the trajectory of China's development (Arrighi, 2007;Henderson, 2011). Ayers (2013) sets up a useful hypothesis in terms of whether there is something unique to the BRIC economies around their economic policies or whether their rise is part of a wider restructuring of global capitalism. I concur with her that it is the second issueglobal restructuringthat is most significant. As such, we necessarily decentre China from some of our explanations since other developed and rising powers are implicated in these changes.
The general context for this is identifying and explaining capitalist variation. At its heart is a debate around whether capitalism is composed of transhistorical and universal components that are found everywhere or whether meaningful national or regional differences are evident in the organisation of capitalism. The 'varieties of capitalism' debate (Peck and Theodore, 2007) refers to essentially Weberian approaches seeking to understand the differences that institutions make to the organization and trajectory of different capitalisms, understood to be nationally-centred. Rather than explore this debate in detail we follow Peck and Theodore's (ibid.) attempt to signal a 'variegated capitalism' approach that looks at capitalism as inherently productive of combined and uneven development (Rosenberg, 2006). This is not to argue that China's development is reducible to some essential model of capitalism and so lacks specificities, but rather that capitalism is a global system and so it is unhelpful to view capitalist processes as nationally-bounded and endlessly varied.
China's development is part of this wider restructuring involving what Harvey (2003) terms the revitalization of primitive accumulation. Far from this 'accumulation by dispossession' being a temporary feature of capitalism, it is an incomplete and recurring phenomenon, given new legitimacy under neo-liberalism in which displacement of people and a violent proletarianization are the norm for many in the developing world (Glassman, 2006). This moves us away from considering China as a bounded actor and, for our purposes, situates the China-Africa relationship as composed of a series of interconnected transformations, not least the movement of global production to Southeast Asia from the late 1970s and the opening up of African economies during the Structural Adjustment Programmes of the 1980s. However, by situating China in 'the context of a significant spatial reorganization of global capitalism' (Ayers, 2013: 238), we cannot simply read off Chinese activities in Africa from an abstract 'reorganization' thesis but need to account for how specific actors and discourses on the Chinese side, and more widely, enable these specific activities to come into being. That said, the Chinese firms central to its internationalization are capitalist, thus we have to acknowledge the accumulation imperative in any relationship with African actors.
China's 'state-orchestrated market capitalism' (Ampiah and Naidu, 2008) has produced consistently high growth for two decades, but energy security is required to sustain this growth. China has therefore been forced to look beyond its borders for sources of energy and other natural resources. This resulted in China's 'Going Out' strategy whereby China sought to achieve its energy needs by encouraging outward investment and subsidizing investment by Chinese companies in extracting natural resources overseas (Rutherford et al., 2008). Commercial engagements have thus come to play a central role in China's strategy to strengthen its economy. In addition to capital formation, market-seeking exports are another impetus, with de Haan (2009) noting that while imports from Africa are primarily in oil and gas, exports include machinery, vehicles, textiles and manufactured products.
Although the Chinese state has had an important role to play in propelling this investment, as we shall see, the role of private Chinese firms is increasingly significant and is driven by 'African market opportunities, competition within China and the presence of a strong entrepreneurial spirit' (Gu, 2009: 570). China's SOEs are major stakeholders, especially in the ring-fenced investment projects that we examine presently, but around 85 per cent of Chinese companies investing in Africa are privately owned (ibid.). While a few large privately owned companies like Huawei Technologies, Holley Group and Zhongxing ZTE Corporation have operations in Africa, the majority of these private firms are small and medium enterprises (SMEs). Most of these companies were motivated to invest overseas because of access to larger markets in comparison to highly competitive domestic markets. The Chinese government is also encouraging private overseas investment under the auspices of both the 'Going Out' strategy and its Africa Policy, which includes financial support, information dissemination, tax incentives, credit and loans.
If China's development is part of a wider transformation of the global economy and it is essentially hybrid and conditioned by translocal channels, we also need to analyse the nature of the political economies in Africa. For African societies, globalization is a longstanding reality and, as Ferguson (2006) argues, has re-emphasized Africa's position in the global economy as a supplier of raw materials. It is also important to emphasize from the foregoing summary of different types of Chinese firms that capital is fragmented so that state activity and politics cannot be homogenized by reducing it to the needs of a 'unitary' capital (Glassman, 1999). Similarly, international capital is never completely 'external' since it combines with fragments of local capital. As Ferguson (2006) notes, the internationalization of capital makes the relationships between capital and the state more complex, and breaks away from a rigid territorialization of the political and economic, which assumes capital has a nationality. This is also a process that entrenches the power of elites who can control access to these resources (Carmody, 2011). Hence, there are different modes and scales over which globalization occurs which informs the discussion below on Chinese SMEs.

Enclaves and Concessions
Situating China's role in Africa within this broader restructuring process enables us to examine particular aspects of the relationship in ways that do not ex ante treat the Chinese as 'worse' than other actors vying for Africa's resources, the effect of which is to exonerate western actors (Yan and Sautman, 2013). It not only urges empirical investigation of what Chinese firms actually do, but situates this comparatively with respect to other actors. In this section, I examine the processes of investment and labour use through the idea of the enclave. The enclave discourse certainly fits with a reading of neo-liberal Africa in a global war for resources, particularly oil (e.g. Ferguson, 2006). But commentators across the political spectrum argue that 'the Chinese' do not integrate, which the more conservative (e.g. Hitchens, 2008) and liberal internationalist (e.g. Human Rights Watch, 2011) take as evidence of their exploitative intentions.
Echoing Harvey's accumulation by dispossession thesis, Bergesen (2008) refers to Chinese engagement in Africa as a form of 'surgical colonialism' that 'involves a minimum of local disruption' (ibid.: 4). This return to the mercantilist enclave then opens up familiar neo-Marxist critiques around a lack of local multipliers, repatriation of surplus and, ultimately, limited sustainability. But where China differs, according to Bergesen, is that it has state backing in the form of aid to effectively subsidize these operations and 'buy off' local elites. However, the discourse of respecting sovereignty distinguishes the Chinese from western powers insofar as the latter have insisted on conditionality which delimits autonomy while formally respecting sovereignty. How far is this enclave discourse accurate?
Driving down an ungraded road north of Techiman in central Ghana you would not be aware of the major development nearby. As you drop down a hill you come to a security gate manned by two Ghanaian guards, but even then nothing looks much different on the other side. Descending further you pass blue-roofed, prefabricated buildings, which look like barracks before coming to an impromptu food market and bus stop. To the right are more substantial buildings in orderly rows intermingled with large tanks containing fly ash for concrete production. This is the Sinohydro work camp for the Bui Dam, a hydroelectric project being undertaken for the Government of Ghana and supported to a significant degree by concessional loans from the ExIm Bank of China. The heavy trucks and equipment are all Chinese, so is the cooking oil as well as the cigarettes that the Chinese workers smoke. There's even a small farm growing Chinese vegetables for the Chinese workers. Such enclaves are now common across much of Africa and other parts of the developing world where bilateral deals are struck between China and the recipient government and then Chinese transnational firms tender for the contracts and set up these relatively self-contained projects (Brautigam, 2009).
Speaking of resource extraction in Africa, Ferguson (2005: 378) argues 'this economic investment has been concentrated in secured enclaves, often with little or no economic benefit to the wider society…..see how different the political-economic logic of the privately secured enclave is from the universalizing grid of the modernist state'. These 'secured enclaves' tend to be in the minerals sector with their gated compounds and limited linkages with the local economy, but in many ways the Chinese state-owned enterprises (SOEs) involved in major infrastructure projects, such as Sinohydro, operate in similar ways. Ferguson contrasts the enclave model with a post-independence 'social' model where mining houses often constructed company towns around a sense of paternalism and thereby had deeper roots in African societies allied to a developmental state agenda.
Associated with the newer enclaved mode of insertion into Africa is, Ferguson argues, a bifurcated governance model in which the increasingly unviable formal state structures are 'hollowed out' fiscally and in terms of authority and personnel, while the viable enclaves are governed efficiently as private entities in a similar vein to pre-colonial mercantilist entrepôts. What is interesting is how the viable and unviable are linked since the Chinese use concessional financing to African states to ease their insertion into these lucrative resource and infrastructure markets (Brautigam, 2009). In this way, a mineral concession is secured through a financial concession. These settlements are brokered at the elite level in ways that bypass channels of debate and accountability within African states and polities (Carmody et al., 2012;Mohan and Lampert, 2013), suggesting that only parts of the African state are unviable (or at least rendered irrelevant).
In a similar vein, the Chinese have recently initiated a series of Special Economic Zones (SEZs) across Africa (Brautigam and Tang, 2011). Although these zones are mainly invested in and built by Chinese SOEs, their main occupants are small and medium-sized enterprises (SMEs), amounting to 85 per cent of the businesses (Tang and Zhang, 2011).
Although most of these SEZs are very recent or under construction it is likely that they will operate like other Chinese SEZs where there is a 're-territorialization of the nation state in order to accommodate and attract the flow of capital, trade in goods and labour' (Arnold and Pickles, 2011: 21). Like the enclaved projects, these zones operate as spaces of neo-liberal 'exception' which actually function to normalize a specific model of capital accumulation, underwritten in particular ways by state authority.
Hence the outcomes of China's involvement in Africa will primarily be shaped by these spatially complex state-capital dynamics. We then have to examine different fractions of capitalsome of which may be enclavedand what role states play in enabling these to succeed or how capital itself exploits (unintended) differences in state policies. As noted, Chinese capital is, at minimum, split between the large, state-influenced TNCs, like Sinohydro, and a myriad of smaller, independent entrepreneurs. Whereas in the past Chinese firms and the state were coincidental, now there is relative autonomy of Chinese firms from state agendas, although the ties between the Chinese Communist Party and the large Chinese multinationals remain quite strong. This is a profoundly political process as different classes seek to transform both the Chinese state and its African counterpart in pursuit of their class interests. Here Hagmann and Peclard's (2010: 544) work on the African state is instructive in seeking to 'understand how local, national and transnational actors forge and remake the state through processes of negotiation, contestation and bricolage'. This is also important as Chinese policy responds to local political conditions (Eisenmann, 2008) despite an avowed intention of non-interference.

Dispossession and Labour
Returning to Bergesen's account of China's 'surgical colonialism', an argument complicating a straightforward process of accumulation by dispossession is that Chinese firms import cheap labour. Bergesen argues 'traditional' neo-colonialism relies on absolute exploitation of African land and labour (e.g. apartheid) whereas the Chinese practice 'national self-exploitation' by importing their labour. Murray Li (2010: 69) has also argued that while Ferguson's enclaves enable super-exploitation with minimal contact, 'There is another dynamic, however, that is potentially more lethal: one in which places (or their resources) are useful, but the people are not, so that dispossession is detached from any prospect of labour absorption'. So does China presage a new form of imperialism, which connects poverty in China with a surgical colonialism in impoverished Africa?
The labour discourse argues that the Chinese simply 'export the capitalism they know best', which is a low wage model with minimum workers' rights or consideration of health and safety. Yan and Sautman (2013) forcefully argue, using the much-analysed case of Zambian mining (Fraser and Lungu, 2007;Haglund, 2009;Lee, 2009), that the data presented in an influential report by Human Rights Watch (2011) reflects a biased analysis of Chinese firms which not only seeks to present them in a poor light but overlooks the equally unsavoury practices of non-Chinese firms. The limited evidence that exists suggests that labour importation varies according to the nature of the project, the Chinese firm involved, productivity levels and the labour market conditions in Africa (Chen and Orr, 2009). Yan and Sautman (2013) show that wages in Chinese firms are not necessarily lower and that unionization levels in Chinese firms are not so different from those in other sectors. Likewise health and safety standards are no worse than in other firms. Importantly, in terms of local development, Chinese firms are resilient when many other firms, either national or international, are not. They also show that many of the negative perceptions of the Chinese in Zambia were based on a particular incident and that since then, and in the face of organized civil society action, these firms have greatly improved their practices.
In our research we found evidence of industrial negligence among Chinese firms. In Lagos, killing at least two workers and injuring others. It was claimed that the Chinese manager would not allow local workers to use his car to transport one of their injured colleagues to hospital.
Such cases undoubtedly point to poor conditions and mismanagement in some Chinese factories in Nigeria, and there is much frustration that the Nigerian state has done so little to prosecute, compensate and regulate. But as Yan and Sautman (2013) note, conditions in non-Chinese owned firms in Zambia are no better, which points to double standards on the part of foreign investors and a weakness in the regulatory capacity of African states (Haglund, 2009). However, beyond the horror of these instances, there are ways in which Chinese bosses are seen by Nigerians to have promoted local welfare and progress. At the most basic level, there is gratitude for the jobs that Chinese companies have created in a country blighted by unemployment. In Kano, for example, local trade union and industry figures highlighted the Hong Kong Chinese-owned Li Group of factories, which was established in Nigeria in the late 1950s, and is the second biggest employer in the city after the government, employing some 7,000 locals and over 20,000 Nigerians across the country. While they bemoan the poor wages and conditions offered by the Li Group, they are quick to point out that local workers have few other employment options and that 'hard bread is better than none'. An official of the Kano branch of the Manufacturers Association of Nigeria stressed that the Chinese-owned factories have been the most resilient, being largely responsible for maintaining the last vestiges of the city's, and Nigeria's, manufacturing sector.
In terms of labour recruitment from China at the Bui Dam, for example, most labourers and semi-skilled workers are Ghanaian. The contract with Sinohydro capped the upper limit of Chinese labour on the project at 600. Interestingly, the Chinese have brought in sixty Pakistanis to operate the heavy equipment who are treated as 'Chinese' for 'imported labour' quota but are even cheaper than Chinese workers. Interviews with Chinese workers revealed that they took jobs in Africa for economic reasons. Most mentioned a wage differential of 300-400 per cent compared with China. Chinese corporations in general do not encourage trade unions, and originally did not allow it at Bui. But a deputation from the Ghana TUC argued that it was enshrined both in Ghanaian law and the contract, and the union was eventually recognized and allowed to operate. By contrast in Angola, which is still recovering from civil war and where many projects are in specialist sectors, the level of labour importation is higher (Corkin, 2012). We have also seen in Nigeria Chinese labour being imported by Nigerian firms and it was this issueas opposed to Chinese firms employing Chinese workersthat led to the most vociferous trade union activity.
While African civil society has been relatively quiet regarding Chinese activities, a number of Africa trade unions and business associations closely critiqued China's role in national economies (Baah and Jauch, 2009). Lee's (2009) comparative study of Zambia's Chambishi plant and Tanzania's Urafiki textile plant illuminates certain aspects of Chinese work practices vis-à-vis organized labour's power. Lee focuses on the casualization of labour attendant on neo-liberalism and argues that the Chambishi explosion, among other things, illustrates that the official trade unions were corrupt organizations that colluded with the Zambian government to benefit the Chinese investors. Despite this cynicism, the union-backed negotiations achieved better terms and conditions with the Chinese firm, a trend corroborated by Baah and Jauch's (2009) study across a sample of African countries. This contrasts with Tanzania, where the union was much less confrontational. Tanzania's Urafiki factory experienced successive strikes that caused morale to deteriorate, but wrestled very few concessions from the Chinese parent firm. Again, there was a perception that the Tanzanian government favoured the Chinese investors over internal actors. As one worker declared, 'The Prime Minister is backing the Chinese so they dare to ignore us because they know the government is supporting them…No party dares to declare itself anti-Chinese because they are big investors' (Lee, 2009: 662). These contrasting cases suggest that where political institutions are stronger the local benefits from Chinese, and arguably other, investments is greater. Moreover, it supports Yan and Sautman's (2013) claim that the practices of Chinese firms are not static and that they evolve in response to local and international political action.

DIVERSE CAPITAL, CONVIVIALITY AND EXCHANGE
Much of the discussion of Chinese enclaves is based on the activities of SOEs that have some connections to the Chinese state. However, we noted the diversity of Chinese firms and that a sizeable majority are privately owned and often quite small. To what extent does the enclave model apply to these enterprises? The extension of an enclave logic suggests that ethnic economies are internally coherent and fail to integrate into 'local' society. Such assumptions are based on the Weber-inspired literature on Chinese firms that tend to posit cultural closure and the essentializing of 'groups' (Greenhlagh, 1994;Meagher, 2012). Meagher (2012) usefully argues that when Chinese networks 'touch ground' in Africa they become more like African networks and there is a brokerage role between enterprises. Here she uses Brautigam's (2003) work on 'flying geese industrialisation' where the Chinese presence can kick-start local economic development.
By contrast, when Meagher usesHart's (1996) work on South Africa and Taiwan, she notes that Hart's analysis is closer to Lee's (2009) political economy of casualization and under-cutting, which is anything but developmental. This usefully points to a contextual understanding of the relationship between class, cultural capital and ethnicity (Ellis, 2011) and their connection to development impacts.
On the more enclavic end of the spectrum, our research found a tendency to socialize with other Chinese, and to downplay some of the negative aspects of failure to integrate with local society. For example, a number of Chinese migrants in Angola and Nigeria echoed the sentiment that 'The Chinese are hated'. Chinese respondents took the fact that locals stole from them or acted rudely as signs of their unpopularity. These perceived barriers, in turn, seemingly justified their own failure to engage or socialize with the locals. But within these 'Chinese' groups in Ghana we found that Chinese migrants tended to organize around age and place of origin. More specifically, a younger Shanghai crowd socialized together and apparently differentiated themselves from the older Hong Kongers. The younger crowd, which also used the Internet more, tended to organize through a Ghana-centred Chinese language website and through QQ discussion groups which is an online communication platform for 'chatting' and social networking.
These tensions are echoed in Giese and Thiel's (2012) work, which examine crosscultural employment relations in the particular context of Chinese trading companies in Ghana, and so necessarily focus on the Chinese employer-African employee relationship, which has a particular class dynamic built into it. Their analysis argues that expectations on both sides are not always met and so tensions arise, though these are rarely explosive.
Similar empirical arguments appear in Haugen and Carling (2005) about lack of trust (see also Dobler, 2008). Giese and Thiel (2012) also suggest that as relations unfold mutual learning of 'the other' will not deepen since these perceptions are so ingrained. What these studies show is that despite a framing of protagonists in national and cultural terms it is the interplay of class and other identities that shapes the tensions. As such their work cautions against static readings of inter-cultural relations and urges us to move beyond the hierarchical employee-employer relationship but also to examine joint ventures and executive roles in Chinese firms where more convivial, cross-class alliances are found.
Our evidence showed hard capitalist logics determining business behaviour as opposed to a favouring of co-ethnics. There are complex migrant trajectories such that there was no inherent trust within a 'Chinese' community (Ho, 2008) and extensive relationships with African businesses. That is not to say that these business relationships are random, but that ethnicity, race and nationality are over-determined in the literature on Chinese migrant businesses (Yan and Sautman, 2013), which in turn diverts attention from the potential benefits to African economies. For example, recent research in Zambia (Fessehaie and Morris, 2011) and Angola (Corkin, 2013) suggests that the Chinese are not as enclaved as earlier reports suggested and local firms have secured a growing number of contracts.
Furthermore, while skilled and managerial positions in Chinese enterprises tend to be dominated by Chinese staff, we found that in many cases locals have attained some of these positions. Chinese bosses emphasize that this makes good business sensenot only does it reduce the cost of recruiting increasingly expensive staff from China but local staff are particularly helpful in understanding the local market, managing local workers and negotiating the demands of local officials. In this way, many Chinese bosses have come to rely on senior local staff and enduring Sino-African relationships have formed. For example, in a small Hong Kong-Chinese manufacturing group in Kano, the Chinese manager spoke admiringly of the three Nigerian technicians who have worked with the company for over a decade since being recruited straight after graduating from a local university. The manager was proud that the technicians had gradually taken over the responsibilities of their older Chinese counterparts and risen to senior positions. Similarly, in one of the Chinese factories in Tema near Accra, the manager is a Ghanaian who has become the owner's closest and most trusted deputy having worked for him since the company was founded some thirty years ago. He oversees the fifty-eight local staff and three Chinese technicians at the factory and regards the Chinese owner as 'a brother' who has helped him greatly over the years by enabling him to educate his children and sending him to China for medical treatment.
The forging of connections to China by African traders also appears to have encouraged Chinese entrepreneurs to come to the continent. In a twist to the enclave narrative around Chinese firms importing their own labour we found evidence that African companies across a range of sectors increasingly view China as a source of skilled and/or hardworking labour and are actively bringing over Chinese workers in an attempt to increase productivity and provide higher quality goods and services (Esteban, 2010). This trend is demonstrated by Mr Daniel, a Nigerian entrepreneur based in Lagos, who employs Chinese staff in a number of his businesses. Mr Daniel's furniture business had been based on importing finished furniture from China. In 2004 when the Nigerian government banned the importation of furniture he realized his only option was to manufacture locally. Feeling that local labour would not be skilled or reliable enough he gleaned from his network of contacts in southern China that Chinese labour was 'cheap' and able to bear basic living conditions. He therefore recruited a team of four furnituremakers from one of his former suppliers in Guangdong Province. On the basis of this positive experience Mr Daniel made recruiting labour from China an explicit strategy in the expansion of his businesses. Notably, he hired fourteen workers from China for a construction company he has established with a Chinese partner.
This analysis of labour relations in SMEs shows a far less enclaved relationship and arguably one where a simplistic model of Chinese-led exploitation is harder to sustain. This is not to deny the capitalist imperative in these relationships but to demonstrate that 'lower end' globalization is occurring alongside the more obvious world of large TNCs and the re-territorialization of inter-state relations. And in this there are active elements of 'African' capital which are spreading some benefits more locally although the intercultural alliances remain within a solid petite-bourgeois middle-class ambit (Lawson et al., 2012) CONCLUSION I have argued that in order to develop a critical political economy of China-Africa relations we need to combine disciplines and theoretical perspectives, particularly insights from culturally-sensitive post-colonialism. By doing this we do not treat China's liberalization as something uniquely 'eastern' nor should we simply invert our theoretical lens and replace the Washington consensus with the Beijing variety. Rather than seeing China as an exception we need to attend to the ways in which China's economic vision has affinities with and connections to other development orthodoxies, particularly with capitalism as a global process. In terms of the spatiality of these interconnections I have tried to show that ethnographic study can illuminate far more complex, surprising and contradictory patterns than a simple reading of an orientalist, neoliberal globalization. Of course we have to anchor our explanations in an understanding of capitalist transformation, but in doing so we must avoid static readings of space, namely the enclave, and of social difference, which over-determines race and ethnicity as the key markers. Rather, new transnational class formations are in evidence that are likely to mark the next few decades of an emerging 'South-South' globalization.