Labour control and the labour question in global production networks: exploitation and disciplining in Senegalese export horticulture

Labour control and the labour question in global production networks: exploitation and... Abstract This article advocates the centrality of labour control to understand the constitutive role of labour and production within global production networks (GPNs). It draws from the global value chain/GPN literature, labour process theory and agrarian political economy to examine the architecture of labour control in the Senegalese–European horticultural GPN through an analysis of local labour control regimes that are constituted at different scales (global, national and local). It frames labour control through the interplay of labour exploitation and disciplining, identifying the different actors, institutions and dynamics shaping labour control and links places of production to broader spaces of labour control. The strong role of the Senegalese state indicates that labour control is closely connected to a broader ‘labour question’ that revolves around the productive/reproductive activities of households and gender subordination. The historical analysis shows the path-dependent, dialectical nature of labour control and labour resistance, and importantly, their dependency on broader production and reproduction relations. 1. Introduction The gradual tackling of a ‘labour gap’ represents a key development since the early elaboration of the global commodity chain (GCC), global value chain (GVC) and global production network (GPN) frameworks (Henderson et al., 2002; Bair, 2008; Coe et al., 2008; Taylor, 2008; Starosta, 2010; Rainnie et al., 2011; Selwyn, 2012). Whilst a number of studies highlight the role of labour organisation and class struggle within commodity chains (Selwyn, 2007; Cumbers et al., 2008), less attention has been devoted to the issue of labour control, its constitutive role within GPNs and its ability to counteract and prevent collective labour action. Building on the emerging dialogue between labour geography, GVC/GPN literature (Coe and Jordhus-Lier, 2011) and labour process theory (LPT) (Taylor et al., 2015), this article advocates the centrality of labour control for a more nuanced understanding of production within GPNs. Empirically, it shifts attention from Northern industrial formations to the countryside of the global South, with a focus on Senegalese export horticulture—a pioneer amongst horticultural GPNs and today hailed as a champion of poverty reduction (World Bank, 2007; Maertens and Swinnen, 2009). The article sheds light on evolving labour control regimes in the South that have contributed to the broader ‘horticultural revolution’ in the 1970s (Mackintosh, 1977) and the ‘supermarket revolution’ of the 1990s–2000s (Dolan and Humphrey, 2000) in the West. The prism of labour control provides a contribution to three main debates that prompt a more sophisticated understanding of the constitutive role of labour within GPNs. The first concerns the analysis of labour control regimes linked with the broader dynamics of supply chains (Anner, 2015; Mezzadri, 2016b). To grasp the complex and changing architecture of labour control underpinning GPNs, the article develops the analysis of the local labour control regime (LLCR) (Jonas, 1996), which frames social relations of production in particular places in continuity with different spaces and scales (Castree et al., 2004). Labour control regime analysis has hitherto received only erratic attention within economic geography (Hastings and MacKinnon, 2016): very few studies examine LLCRs from an historical perspective (Ruwanpura, 2016) and within agriculture (Riisgard and Hammer, 2011). To advance this analysis, labour control is framed along the interplay between labour exploitation and disciplining, as respectively (i) the production of value in excess of labour remuneration and (ii) the mechanisms of mitigation, containment and prevention of conflict inherent to production, i.e. the constant subordination of labour to the labour process. The analytical distinction between exploitation and disciplining reconciles places of production (firms) with broader spaces of labour control (here households) that are shaped at different scales (global, national and local). Second, the article responds to a long-standing call to move beyond both firm- and network-centric views of production networks that provide a partial analysis of the social relations of production (Taylor, 2007). The analytical prism of LLCRs challenges much of the ‘network ontology’ inherent to the GVC/GPN literature (Bair and Werner, 2015) by highlighting specific upstream labour control edifices that emerge from the unique interplay of the dynamic restructuring of global capital, state agendas, conflicting local mediation between capital and labour and household social reproductive spaces. This reinstates the structural role of the state and ‘development’ within GPNs (Kelly, 2013; Smith, 2015), and links labour control to a broader ‘labour question’ facing the postcolonial state, i.e. the never-ending problem of managing the working class (Cooper, 1996). In turn, this highlights the centrality of households and the gendering of production and reproductive work that broadly sustain GPNs (Kelly, 2009; Cumbers et al., 2010). Third, the article contributes to recent debates within labour geography calling for a more contingent, embedded and relational understanding of workers agency (Castree, 2007; Lier, 2007; Coe and Hess, 2013; Peck, 2013). By placing workers in relation to broader structures of labour control the article promotes a ‘more embedded conception of agency that reconnects with the continued structural domination of capital in many contexts’ (Coe, 2013, 273). The case study shows that the position of workers in GPNs needs to be understood in relation to the historical evolution of LLCRs, and ‘in relation to the formation of capital, the state, the community and the labour market in which workers are incontrovertibly yet variably embedded’ (Coe and Jordhus-Lier, 2011, 4). On the one hand this demonstrates the path-dependent and dialectical relationship between capital and labour (Henderson et al., 2002), and on the other, their dependency on broader production and reproduction relations. It echoes Cumber et al.’s (2010, 67) invitation to see ‘the agency of capital and labour as bound in a dialectical totality’, but it also confirms the relational nature of both capital and labour agencies vis-à-vis other capital and labour segments. The following Section 2 frames labour control by drawing from LPT, the GVC/GPN literature and agrarian political economy. After a brief methodological discussion in Section 3, I explore changing forms of labour control in Senegal in Sections 4 and 5, focusing on the emergence of LLCRs based on estate farming in the 1970s and its combination with contract farming as of 2000. This is followed by concluding remarks. 2. Framing labour control Labour control has been the traditional analytical domain of LPT, which focuses on ‘the dynamics of control, consent and resistance at the point of production’ (Thomson and Smith, 2009a, 915). Following Marx (1990), Braverman (1998), and Burawoy (1992), LPT conceives the labour process as necessarily 2-fold: the metabolic transformation of nature through labour and the simultaneous production of value in excess of labour remuneration, or reproduction, that is applied to other commodities to make further commodities with a greater value (Nichols, 1980). Thus, it represents a fundamental moment of inequality in the social relations of production. This intrinsic unequal exchange between employer and employee is far from natural and therefore needs to be constantly preserved inside and outside firms. As labour is essentially indeterminate (Smith, 2006), labour control is intrinsic to the labour process and requires the organisation and supervision of workers, i.e. regulation of labour activity and the conflict this engenders. LPT, therefore, suggests that, whether evident or not, conflict is structural to employment relations (Edwards, 1986), and requires a multidimensional and evolving repertoire of labour control strategies. Traditionally preoccupied with the relational nature of labour control and labour resistance within firms, LPT has recently turned its attention to the ‘wider political economy’ affecting employer–employee relations (Thomson and Smith, 2009b; McGrath-Champ et al., 2010; Thompson, 2010). A recent edited collection (Newsome et al., 2015) explores the interrelationships between LPT and GVC/GPN analyses to reintegrate the labour process as a constitutive network element. This article furthers this agenda by building on the LLCR approach, as an analytical device, that dialectically links workplaces to broader local and global dynamics thereby offering a complex, multilevel understanding of the labour process. Jonas defines an LLCR as ‘an historically contingent and territorially embedded set of mechanisms which co-ordinate the time-space reciprocities between production, work, consumption and labour reproduction within a local labour market’ (1996, 325). Thus, LLCRs represent local, relatively stable, institutional fixes to the immanency of labour control and capital accumulation, both at the firm and at the sector level. In this conceptualisation, labour control constantly evolves through the opportunities and constraints posed by various players including workers and their social networks, state officials, local and foreign firms, and in the global South, global governmental and non-governmental institutions. Whilst much LLCR analysis has been rather capital-centric (Hastings and MacKinnon, 2016) and focused on industrial formations (Jonas, 1996; Kelly, 2002; Davies et al., 2011; Neethi, 2012), this notion can be transferred to the countryside1 to examine how network pressures interact with local dynamics of agrarian change. According to Jonas, ‘a local labour control regime involves the exercise of power by local agencies and their particular strategies of control’ (1996, 329). To emphasise the agencies (of capital, labour and other institutions such as states and households) constituting LLCRs, I frame labour control as the interplay between labour exploitation and labour disciplining. Whereas labour exploitation refers to the production of value in excess of labour remuneration, labour disciplining refers to the mechanisms of mitigation, containment and prevention of conflict inherent to production, i.e. the constant subordination of labour to the labour process. Just as labour exploitation serves the production of commodities, labour disciplining serves labour exploitation: the two are necessarily linked, ambiguously and often mutually reinforcing. Although manifestly occurring within firms, exploitation and disciplining of workers is based on social distinctions other than class (e.g. gender, race, ethnicity, etc.) and are buttressed by wider state policies, as well as the interests of foreign capital. Thus, labour disciplining requires an analysis of labour process reliance upon a broader subsumption of labour within and beyond the immediate place of production. From this perspective, LLCRs represent different articulations between labour exploitation and diverse, related instances of labour disciplining within broader spaces. Castree et al. (2004, 115) argue that the notion of the LLCR ‘recognizes that the nature of worker-employer relations in a particular locality is simultaneously determined by local and extra-local dynamics at a variety of scales up to, and including, the global’. This resonates with conceptualisations of GPN as ‘multi-scalar spaces of governance’ (Dicken and Malmberg, 2001) shaped by the jurisdiction of local, national and supranational institutions concurrently structuring production. According to GPN scholars, production in different places is simultaneously affected by the local environment and relations embedded within production networks (Henderson et al., 2002; Coe et al., 2008), i.e. the relationship between places and networks are fundamentally dialectical (Coe et al., 2010). Production is crucial even for distant lead firms as these still ‘exercise control over the labour process through the constraints they place on relations between management and labour’ (Robinson and Rainbird, 2013, 97). Thus, through governance, GPNs channel labour control pressures that shape LLCRs. These, in turn, further combine with pressures from the state. Recent critical interventions have stressed that ‘GVCs/GPNs are emergent artefacts from state action: they always “touch down” somewhere, and in every “somewhere”, there is always the hand of the state’ (Neilson et al., 2014, 2). Whilst critical studies emphasise how states are constitutive of GPNs (Glassman and Choi, 2014; Smith, 2015), and of labour control regimes (Burawoy, 1985), their constitutive role remains essentially contradictory. As Hudson points out, ‘while the state does not represent the general interests of capital, its continued existence as a particular form of social relationship depends upon reproducing the social relations of capital and the successful continuation of the accumulation process’ (2001, 53). In Africa, as elsewhere in the global South, this is the familiar territory of ‘development’, which engages the postcolonial state in the difficult task of promoting ‘modernisation’ without social disruption (Cowen and Shenton, 1995). Historically, ‘development’ has caught African governments in the predicament between attracting capitalist production and dealing with the ensuing ‘labour question’, i.e. the constitution and management of the working classes that collectively organise to improve conditions of employment and reproduction (Cooper, 1996). This is most visible within the agricultural sector, where ‘development’ has led to the simultaneous, and contradictory, support of large-scale mechanised farming and small-scale, family farming (Oya, 2012), often cohabiting at the bottom of GPNs. While the former represents the modernisation option, the latter responds to those rural masses and donor agencies who, fearing the ‘death of the peasantry’, claim that its small-scale, allegedly industrious, and family-friendly nature avoids large-scale land alienation and the creation of landless labour armies. Thus, state support for the ‘peasantry’—presented as something systematically different from, and alien to, the antagonism of class-based capitalist production—offers a powerful means to counteract class struggle and demonstrates how the state’s material and discursive power vitally structures LLCRs (Coe and Kelly, 2002). This is reflected at the local level where both GPN pressures and (more or less manifest) state intervention contribute to the continuous relevance of small-scale farmers as ‘households’. At a local scale, Cumbers et al. (2010, 50) emphasise how the notion of the LLCR ‘recognises capital’s need to do more than just exercise coercive power over labour in production, but also to attend to the conditions under which labour is reproduced’, thereby extending the analytical focus to relations between firms and households and labour control operating beyond firms. It is at this scale that different instances of labour disciplining are more evident. Here, I focus on the disciplining of workers along spatial and social lines. Spatial disciplining refers to the spatial fragmentation of workforces through subcontracting and their ‘containment’ within firms (Kelly, 2002). Social disciplining refers to the social segmentation of workers primarily through the gendering of labour markets and the subordination of women within households. Both develop through the interplay between firms and households. Whereas GPN literature has emphasised the role of subcontracting in creating hidden and informal labour armies (Castree et al., 2004), less attention has been given to its counterpart in agriculture, i.e. contract farming. Contract farming represents a local form of outsourcing that spatially splinters the labour force and externalises the labour process, together with the risks and costs of production, to local growers. The latter include a diverse and differentiated group of farmers, including smallholders, which represent a particularly resilient form of production where the rural armies of hidden, informal and unpaid workers typically orbit. This is emphasised by Bernstein who frames smallholders as petty commodity producers (PCPs), i.e. small-scale capitalist farmers who combine in the same person/household the class ‘places’ of capital and labour, often unequally distributed across family members, especially along gender/age lines, and/or typically attract the cheapest labour (Bernstein, 2003; 2010). Usually defined as ‘family farms’ or ‘peasants’, they are seldom conceived of as a set of extremely heterogeneous small-scale firms typically working small patches of land, straddling self- and wage employment, and combining a portfolio of activities at the intersection between simple reproduction and accumulation. By blurring the boundaries between firms and families, PCPs typically rely on uneven accumulation and strong levels of labour exploitation/disciplining (Kautsky, 1900; Cramer et al., 2008; Journal of Agrarian Change, 1–2, 2004). Here, the smaller scale of production and tighter labour control deter labour organisation and resistance. Thus, the notion of PCP underscores how the spatial fragmentation of labour involves a particular combination of the firm and the family within the same entity. Underlying the spatial disciplining of workers through contract farming and petty commodity production, is perhaps the most pervasive source of disciplining in the global South: the subordination of women through the intersection between the spheres of production and reproduction. In the Southern countryside, the structural bond between households and firms unleashes a substantial supply of cheap female labour that fills casual and informal employment at the bottom of GPNs (Barrientos and Kritzinger, 2004; Dolan, 2004; Tallorine et al., 2005; Goger et al., 2014). The ability of firms to fill insecure employment with women (and migrants) reflects the underlying dynamics of female subordination within households that are channelled through labour markets (Elson and Pearson, 1981). According to Elson (1999), markets are essentially ‘gendered institutions’: they convey widespread social stereotypes which buttress perceptions of typical ‘men’s work’ and ‘women’s work’, and gain from women’s disproportionate assumption of household duties, i.e. their employment in the ‘reproductive economy’. Feminist agrarian political economists have stressed that households are heterogeneous (Whitehead, 1991), and ‘are not necessarily sites of sharing and equity’ (Whitehead and Kabeer, 2001, 2; see also, Agarwal, 1986; O’Laughlin, 2007). As such, ‘the ability of peasant households to produce commodities for sale (including labour) is dependent on their ability to mobilise and organise labour (mostly female) to carry out the domestic work necessary for their reproduction’ (Crehan, 1992, 99). In short, women’s cheap and free labour stems from their ‘patriarchal unfreedom’ (Mezzadri, 2016a). Crucially, this makes women, not men, ‘the optimal labour force for the capitalist’: their constant disciplining as housewives cheapens their cost as workers and has made them the driving force behind the proliferation of GPNs since the 1970s (Mies, 2014, 116). Overall, both spatial and social instances of labour disciplining reveal the structural relationships between firms and households representing a fundamental continuum for production and accumulation. This develops Kelly’s (2009) argument that households are an integral part of GPNs. Formally considered as external to GPN production sites, households (as both workers’ families and subcontracted smallholders) are very much included in the wider space that makes GPNs possible. These spaces are actively shaped by overseas firms, states and local production and reproduction dynamics. If on the one hand this shows that ‘the control of ‘space’ is central to the control of labour in capitalism’ (Jonas, 1996, 330), on the other it also highlights the hidden web of relations that makes production possible. By scrutinising the relations between firms and households and the underpinning disciplining of workers, it is possible to see uneven the relations and emphasise how some workers’ gains are structurally linked to other workers’ losses (Coe and Hess, 2013; Peck, 2013). As the case study of Senegal shows, the empirical investigation of different forms of disciplining emphasises the relationships between different segments of capital, labour and institutions: these are contingent and locally embedded in the particular history and political economy of the Cap-Vert Niayes, which is briefly introduced below. In Sections 4 and 5, I examine the constitution of LLCRs in the 1970s and 2000s in the Cap-Vert Niayes, at the global, national and local scale, through pressures for labour control emanating from: (i) Transnational corporations (TNCs) and international institutions; (ii) the Senegalese state; and (iii) at the firm level through the labour process and its interaction with other local social relations, in particular gender relations straddling production and reproduction. A comparative summary is included in the conclusions. 3. Case study and methodology Senegalese fresh fruit and vegetables (FFV) represent an important source of off-season supply to Europe. 2 Historically, Senegalese horticulture developed in the shadow of the groundnut sector, the traditional engine of the Senegalese economy. Today, FFV exports supplying European markets represent one of its most dynamic sub-sectors. These exports are concentrated in French beans, cherry tomatoes and mangoes. Production is focused in the Niayes area, a narrow belt stretching from the capital Dakar to the northern city of Saint Louis (Figure 1). The Southern part—the Cap-Vert Niayes—is a relatively small area situated between Dakar and Thiès (Figure 2), and corresponds to the local scale of this study.3 This region has known horticultural activities since the 1940s.4 Exports took off in the early 1970s through Bud Senegal, one of the earliest large-scale agribusiness schemes in Africa; they peaked in the mid-1970s and then rose again from 2000.5 These periods correspond to the establishment of tighter, vertically coordinated GPNs and LLCRs marked by estate farming, i.e. large-scale farms employing wage labour. The analysis of LLCRs emerging in both periods draws substantially from the seminal work of Maureen Mackintosh’s Gender, Class and Rural Transition, Agribusiness and the Food Crisis in Senegal (1989),6 and from the author’s fieldwork carried out in 2007–2008 in the same area. Figure 1 View largeDownload slide Senegal and the Niayes region. Figure 1 View largeDownload slide Senegal and the Niayes region. Figure 2 View largeDownload slide The Cap-Vert Niayes. Figure 2 View largeDownload slide The Cap-Vert Niayes. Mackintosh’s work drew from fieldwork carried out in the mid-1970s across Europe, Dakar and the Niayes villages supplying labour to Bud estates. Here fieldwork activities explored the complex intertwining between relations of production and consumption at the village level. Information was gathered by combining quantitative and qualitative data, and by interviewing Bud management and workers.7 Her work represents a precursor to the more nuanced GVC/GPN and labour control regime studies where local relations of production and reproduction are analysed as part of a wider process of global capitalist restructuring and post-colonial state-building that underpinned the global horticultural revolution of the 1970s. In 2007–2008, I carried out fieldwork in the Niayes, including the villages adjacent to former Bud estates. Compared to the 1970s, when Bud dominated the sector, and despite export concentrations within three export firms, in the 2000s the sector was crowded by several firms. The fieldwork aimed at collecting the life histories, accumulation trajectories and strategies of local exporters–producers who had driven the sector in the 1990s–2000s. This was done with multiple semi-structured interviews with exporters (24 out of 30 active export firms), and their managerial staff at different levels. The information gathered was triangulated through secondary sources and interviews with local subcontracting and non-subcontracting growers and around 50 other stakeholders including state officials and NGOs. This article places the agency of local horticultural capital in the GPN in historical perspective and mediates it with other crucial agencies (foreign capital, the state and the workers) to draw an historical picture of the LLCRs. Despite the absence of significant collective action by workers since Bud’s operations, workers’ agency emerged strongly from interviews with employers, managers, state officials and local growers, and from participant observation of their activities: the lack of substantial, direct interviews with workers represents itself an under-representation of their agency. To further substantiate key findings, fieldwork data is combined with simultaneous and later studies emphasising some core dimensions observed in the late 2000s, notably, a marked feminisation of the workforce, the continuous relevance of smallholders, and an extremely segmented, differentiated labour market (Maertens and Swinnen, 2012; RDS, 2012, 2013; Swinnen et al., 2013; Oya, 2015). 4. Labour control in the 1970s: from estate farming to contract farming To grasp the constitutive and path-dependant nature of labour control, the empirical analysis starts by examining the LLCR that sustained the early development of the sector in the 1970s. As emphasised above, this was a result of a combination of pressures emanating at global, national and local scales. Based on the direct control of workers on large estates (estate farming), the 1970s LLCR emerged when a subsidiary of a US company established itself in the Niayes to supply the European off-season FFV market. ‘Bud Senegal’ operated for almost a decade supplying exclusively its parent-marketing agency in Europe, House of Bud Société Anonyme (HOBSA). Within a few years the export of FFV took off but then petered out in the same dramatic way when the company went bankrupt in 1979 to the benefit of HOBSA (see below). The scheme was the first effort to organise export horticulture on estate farms with direct employment of wage labour on a large scale. At its peak a total area of 600–650 ha were cultivated, employing a daily average of 1600 workers (Mackintosh, 1989). Although short-lived, Bud’s deep transformation of the countryside represented an extremely path-dependent instance of labour control. The ‘problems’ created by an emerging rural ‘proletariat’ cast an enduring scepticism over the opportunity of employing a large number of workers ‘under the same roof’. As this section explains however, the organisation of production through estate farms and large-scale direct employment fulfilled the interests of different players: Bud managers, the Senegalese government and international donors. All three were influential in shaping the form of the labour process and the prevailing LLCR. 4.1. Global dynamics and chain governance The development of estate farming in Senegal by Bud was a typical expression of the industrialisation of agriculture emanating from USA, marked by the concentration of agro-industrial capital and the ascendancy of transnational corporations integrating farming activities horizontally, vertically and globally (Friedmann and McMichael, 1989; Heffernan, 2000). Bud Antle was a leader in Californian horticulture, a maverick in technological innovation, and a champion of anti-unionism (Friedland et al., 1981; Fredrichs, 1984). In the 1960s, the firm invested in Europe where it set up a marketing agency, HOBSA. The demand for FFVs was growing significantly and triggered new opportunities for traders, brokers and wholesalers (Mackintosh, 1977). This incentivised the sourcing and relocation of FFV production in the South by large-scale European brokers and/or giant processors like Nestle, Findus, etc. By 1980, there were around 140 investments in African countries by leading US and European food processing firms (Dinham and Hines, 1983). Among them Bud was a pioneer: an offshoot of Bud Antle, the emerging HOBSA–Bud Senegal supply chain linked Senegalese land and workers to European consumers through the supply of off-season FFVs. Like Bud, many of these investments were branded as ‘development’ because, from the 1960s, the internationalisation strategies of emerging giant corporations found support in key institutions like the World Bank and Food and Agricultural Organisation (George, 1976). The emergent ‘development discourse’ was constructed on ideas of food security, modernisation, private entrepreneurship and export orientation to gain foreign exchange. In the late 1960s, the effects of the Sahelian drought had widely captured international interest, and projects directed by agribusiness firms for savannah lands were promptly supported by aid agencies and development funds. HOBSA managers were successful in exploiting this conjuncture and funded Bud Senegal with the contributions from the World Bank, European donors and the Senegalese government (Mackintosh, 1989). The choice of large-scale production and wage-employment was shared by Bud’s management and by donors. It reflected the link between foreign firms and plantations, that in Africa stretched back to colonialism, where the high costs of plantations were offset by low running costs and the negligible land and labour costs (Dinham and Hines, 1983). After independence and the nationalisation of key cash crop economies, the link between foreign firms and plantations gradually developed in the direction of higher value crops, including cut flowers and FFVs. The economic arguments were the ‘technical superiority’ of both scientific production techniques and scientific management of workers (Gibbon, 2011). The use of large-scale irrigation, mechanisation and a stronger supervision of workers were deemed necessary to achieve a greater standardisation of produce as required by European markets. The direct control over production also replicated Bud Antle’s history and operations in USA: Bud Senegal represented the African counterpart of a highly mechanised, technologically advanced Californian horticulture achieved through large-scale farming. Fundamentally, however, Bud opted for estate farming because donors and the Senegalese government were willing to assume other major set-up costs (land and water). Once these were paid for, Bud could assume the cost of directly managing hundreds of workers. 4.2. The state Confronted with persistent drought and the crisis of the groundnut sector as a central source of revenue, for the Senegalese government Bud’s ‘modern’ farming option was alluring. It matched the state’s ambitious plan for agricultural diversification and broad reorientation of its development strategy away from rain-fed agriculture and in favour of modernisation, mechanisation and large irrigation schemes. Large-scale tracts of land in the Cap-Vert Niayes were handed over to Bud as virtually free, renewable land leases (Franke and Chasin, 1980). Other important concessions included cheap water, import-export tax exemptions and crucially the allocation of the marketing monopoly to HOBSA. As noted earlier, the emerging supply chain was an ‘artefact’ based on state action (Neilson et al., 2014, 2). More fundamentally however, and beyond the expectation of increased foreign exchange, state backing for estate farming expressed a wider ambition to intensify control mechanisms in the countryside, hitherto based on trade monopoly (through marketing boards). The crisis of the groundnut economy reflected the crisis of state resource extraction mechanisms, hitherto centred on trade monopoly and patronage relationships between producers and rural notables for the political containment of the rural population (Diop, 1992). In a countryside characterised by weakened state control of peasants, direct employment by Bud on irrigated estates provided an opportunity to reinforce the market dependence of rural populations, not as sellers of crops, but as sellers of labour. 8 Thus, estate farming represented the opportunity to establish ‘a very sharp increase in intervention, by government and private capital, to manage and transform the labour process in agriculture’ (Mackintosh, 1989, 8), i.e. to curtail relative autonomy within farming. Estate farming provided the state with an LLCR functional to agricultural modernisation and labour commodification, boosting both productivity and labour control. As argued below, the contradictory nature of this ambition became evident within the labour process, precisely because of the increased commodification and spatial aggregation of a large labour force that estate farming entailed. 4.3. The labour process How was labour control organised at the firm level? First, the higher costs and risks of directly employing labour were mitigated by a strong regime of labour exploitation and disciplining: at Bud the labour process was shaped by the adoption of mechanisation, labour informality and segmentation, especially along gender lines. Second, as these strategies failed to contain labour organisation, Bud gradually shifted to subcontracting, i.e. splintering the labour force over a multitude of small farms. In short, when the social disciplining of workers proved insufficient, management resorted to spatial discipline across multiple, spaced out sites. While mechanisation and especially widespread informal labour were pervasive from the outset, the main mechanism of disciplining at the firm level was a regime of highly gendered labour segmentation. Bud’s labour control strategies reproduced wider inequalities within Senegalese society, especially related to gender and migrants, and reflected management’s ability to capitalise on local household reproduction dynamics heavily based on female domestic work (Mackintosh, 1989). Within three years of activity the largest estate had created the category of ‘women’s work’ as a generally lower status, lower paid type of job. Managers portrayed women as an ‘unskilled and archaic’ labour mass. The discursive representation of the ‘problem of the women’ as an undisciplined crowd legitimised and reinforced informal and casual employment. In turn, informality offered women the possibility to fulfil household duties, combine and solve the conflict between wage and domestic work. In other words, the ability to push the risks and costs of insecure employment onto women depended on their social disciplining: household inequality was extended to the firm and legitimised through discursive representations. Thus, Bud’s labour regime reflected the broad ‘gendering’ of local labour markets on the basis of stereotyped characteristics ascribed to women and their subordinated role within households (Elson and Pearson, 1981; Elson, 1999). This was reinforced by presenting part-time and flexible work conditions as opportunities rather than constraints, options matching women’s need to juggle work and the household. This obscured and reinforced disproportionate female engagement in the sphere of reproduction and constrained their access to the productive economy. Hence, from the perspective of the LLCR the disciplining of women across the household-firm continuum sustained the employment of hundreds of workers, i.e. the informal, casual incorporation of women stood in relation to the sustaining of an enterprise based on the direct employment and control of workers on a large scale. Despite Bud’s reproduction of gender subordination, and the wider combination of labour disciplining and exploitation strategies, the company still faced a mass of workers whose direct employment offered the opportunity to resist and dispute different labour control mechanisms. In general, walkouts, stoppages and strikes occurred throughout the 1970s. Workers protested piece-rate employment, payment below the minimum wage and the unpredictability of employment. Some strikes required military intervention. On a small estate, female workers independently and collectively protested against the employment of non-local women. The tension culminated in September 1979, as 3000 Senegalese workers demonstrated in Dakar against the decision to close the company. As argued below, however, rather than leading to success, the growing organisation of workers foreshadowed ‘the continued structural domination of capital’ (Coe, 2013, 273) prompting the shift to an LLCR based on contract farming: workers’ agency led to more effective labour control. By 1979, the state had taken over Bud, which had registered financial losses each year. The government continued subsidising with no financial return as surpluses moved upstream to HOBSA. Estate farming had proven economically and politically expensive, especially for the government which as both the ‘employer’ and the ‘state’ was confronted with securing accumulation and development simultaneously (Cowen and Shenton, 1995). But how could the state reproduce capital accumulation without producing a class of conscious and organised workers? The answer to this ‘labour question’ laid essentially in the retrieval of the ‘peasantry’, i.e. a development model based on family farming as the way to modernity. This pivoted on local growers perceived as ‘more politically acceptable, less likely to create labour problems, and better in promoting individual work incentives’ (Mackintosh, 1989, 16). As one civil servant claimed at the time: ‘Our aim is definitely not to create waged agricultural worker, but to organize (encadrer) the peasantry’ (Mackintosh, 1989, 187). Within the official discourse therefore, contract farming came to replace estate farming as a more family-friendly inclusion in the GPN. In reality, contracting with local growers offered opportunities to transfer significant costs to small producers, sidestep minimum wage regulations and thwart mounting collective labour action on estates. The state’s recourse to the ‘peasant option’ offered the chance to discursively and spatially separate wage-workers whilst entangling them in the household. Through subcontracting, the labour process was displaced and transferred to smallholders employing family and migrant labour and enjoying the comparative advantage of avoiding ‘labour shirking’ (Rassas, 1988). In the 1980s, the government contracted with around 650 families, allocating to each small plots of land (Sall, 1993). Directly counterpoised to the estate, the family was presented as a homogenous, equitable and ‘natural’ instance of development, thus obscuring the unequal social and economic relations structuring rural households as PCPs. Overall, the shift to contract farming evidenced a broad strategy where spatial disciplining (labour displacement and dispersal within small farms) served to reinforce social disciplining (stronger to the extent that the farm approximates the household). The stability of the ensuing LLCRs based on contract farming and the incorporation of small-scale farms that continued throughout the 1980s and 1990s was broad evidence of how ‘space itself is deployed as a weapon in the contestation between capital and labour’ (Kelly, 2002, 397). Following Bud, the spatial fragmentation of labour within small family-farms continued. Throughout the 1980s, contract farming was the dominant form of upstream production involving thousands of Niayes family farmers and a dozen local entrepreneurs attracted by Bud operations (Horton, 1987). The strict vertical coordination between Bud Senegal and HOBSA, became a looser supply chain anchored to spot-market purchases rather than long-term supply agreements. Things changed, however, when European supermarkets gradually became the driving agents in the chain. Within an unfolding supermarket revolution (see below), European supermarket demands began to reshape the LLRC once more by encouraging a return to estate farming alongside small subcontractors. 5. Labour control in the 2000s: combining estate and contract farming After two decades of contract framing, the steady reappearance of estate farming was evidence of an emerging new LLCR governing Niayes export horticulture. From the 1990s, some domestic exporters began to invest in large-scale farming to comply with stricter requirements embodied in private European standards. In 2007–2008, there were roughly 30 active export firms. The majority were exporters–producers, i.e. firms internalising production, and a (shrinking) minority of sole exporters relying on contract farming. Exports were concentrated in relatively few firms: the three largest export firms accounted for two-thirds of all exports. The chain was broadly bifurcated: on the one hand the largest exporters–producers supplied European supermarkets through private certifications and marginally wholesale markets, and on the other smaller exporters exclusively supplied wholesale markets or European importers. In principle, direct production on estate farms supplied supermarkets while most contracted local growers sourced other European markets where food standards were looser. Like Bud before them, exporters–producers in the 2000s adopted articulated labour control strategies, combining labour exploitation and disciplining and profiting from the state’s simultaneous interest in boosting capital-intensive, large-scale farming without displacing small-scale farmers. Reproducing many of the 1970s labour-control practices, exporters–producers engaged in broad labour disciplining strategies to maximise exploitation and counterbalance the high costs and risks of upstream vertical integration. However, whereas in the 1970s the shift to contract farming was subsequent to the investment made to set up estates and labour collective action, from the 2000s contracting with local growers was largely previous and simultaneous to large-scale farming, i.e. a cost-saving platform supporting estates (Baglioni, 2015). The emerging LLCR therefore combined both estate and contract farming. As earlier, the pressures driving this combination originated from different levels, inside and outside the network. 5.1. Global dynamics and chain governance The re-emergence of estate farming in Senegal stemmed from the broader ‘supermarket revolution’ that had dramatically restructured Western retail markets (Humphrey, 2007). Displacing the traditional role of wholesale markets, by the 1990s European supermarkets had steadily gained market share becoming major marketing outlets for domestic and foreign suppliers. Moreover, fierce competition had prompted the adoption of new technologies and long-distance sourcing strategies. Overall, the simultaneous concentration and competition among retailers entailed a broad restructuring of African–European horticultural GPNs. In a classic study, Dolan and Humphrey (2000) show how the growing market power of supermarkets is translated into buyer-driven horticultural commodity chains characterised by intensified vertical coordination and concentration at strategic nodes, as well as increased competitive pressure on producers upstream. The chain ‘buyer-driveness’ was further intensified by supermarket competition for the supply of safe food following the food-scares from the 1990s onwards. Emanating from the escalating industrialisation of agriculture worldwide, these food crises prompted the proliferation of European public food safety and quality standards (Freidberg, 2004). Public standards were rapidly incorporated and widened by supermarkets competing to provide the best food quality and safety while protecting the environment and workers’ livelihoods (Henson and Reardon, 2005; Gibbon et al., 2010). Suppliers wishing to sell to European supermarkets were confronted with complex standard schemes and certifications such as EurepGap (now GlobalGap) that were far more demanding than public ones. Whereas in the 1970s TNCs and donor agencies branded plantations as ‘development’, in the 2000s the new mantra was ‘sustainable supply chain management’ (Müller et al., 2009) and the absorption of the developmental function within standard themselves. Supply chains represented the new development frontier, especially for African smallholders who could rely on private players engaged in creating a dynamic and prosperous agricultural sector (Amanor, 2009). The assimilation of ‘development’ by the market reflected the broad shift from state-led to market-led development characterising neoliberal globalisation. Poor working/workers’ conditions in the global South, exposed by Northern governmental and non-governmental institutions, spurred ethical initiatives targeting GPNs as the ideal terrain to improve workers lives (Hughes, 2007). However, as the labour process upstream came to be organised according to these new forms of governmentality, the burden of ‘decent production’ fell on producers. Producers, rather than supermarkets, were required to assume the costs of compliance with standards. This left broader power relations untouched: by portraying ethics as a problem of sourcing, supermarkets were able to ‘distance themselves from their own role in creating an environment for exploitation’ (Du Toit, 2001, 3).9 These broad global changes reflected downstream in contradictory pressures shaping the new Senegalese LLCR. On the one hand, supermarkets required a stricter control of labour. To obtain a private certification such as EurepGap or Tesco’s Nature Choice, producers had to adhere to rules covering all areas of production, crop handling and regulations over worker safety and health. According to all exporters, these rules could not be complied with through contract farming, where their control over farming was limited. Hence, the need to engage in direct production to control directly the labour process. On the other hand, direct control over farming and labour greatly increased the production costs, requiring a strong regime of labour exploitation and disciplining, for which contract farming represented an important tool. Overall, weak supermarket monitoring systems increased LLCR room for manoeuvre, allowing combined estate and contract farming to cut costs and prevent widespread labour organisation. 5.2. The state As in the 1970s, state agency was a pillar of the new, tighter GPN and corresponding LLCR. Mirroring the support to Bud operations, exporters in the 2000s relied on considerable state backing, from the central government and/or local officials heading rural communities. This was mostly in the form of relentless land accumulation, necessary for estate farming. Exporters could access land through highly favourable, state-granted, long-term land concessions and/or gradually amassing small-scale (informal) land acquisitions through the intercession of state officials within rural communities. For the government, large-scale horticultural farms represented the new ‘modern’ agriculture, associated with inclusion in profitable supply chains, agribusiness and mechanised, capital-intensive farming. The sector figured as a core pillar of Senegalese agricultural development strategies in the 2000s (Fall, 2005), and was supported by the World Bank. 10 The state however, while promoting agribusiness, also continued to pursue the long-standing valorisation of small-scale family farming (Senghor, 2006). This led to various public projects aimed at strengthening smallholder inclusion within supply chains,11 possibly reinforcing the bifurcated nature of the network where large- and small-scale farming continue in parallel. Despite its incoherence (Anseeuw, 2010; Oya and Ba, 2013), state participation in the agricultural sector has long reflected the ambitious need to keep a foot in both camps: that is, to pursue modernisation via large-scale commercial agriculture and small-scale family farming (Oya, 2012). Discourses on the intrinsic value of ‘peasants’ as champions of supposedly family-friendly and human-scale farming endure. These counter-pose modern estates with industrious and diligent peasants that make the most out of very skewed resources (but systematically mask higher levels of labour exploitation and disciplining, including gendered allocation of resources and duties). In 2007, following a conversation on Bud strikes, an exporter and state official claimed: The smallholder is more productive because he is under risk. The more he produces the more he gains. Also he is more responsible because he works his own land. Wage labour is not good for agriculture…, because the worker gets the same pay whether he produces or not, whether he produces a lot or a little. Also with wage labour you have the problem of unionisation, which means less work. While the prevailing ideological construction of industrious peasants reveals the states’ discursive power in shaping the LLCR (Coe and Kelly, 2002), the preservation of small producers confirms the state’s efforts to maximise the interplay between labour exploitation and disciplining, structural to PCP. Here, the spatial fragmentation of workers, their closer surveillance and greater exploitation under the form of household duties hinders collective labour action, and represents the successful management of the ‘labour question’ by the state. Indeed, simultaneous support for large-, and small-scale farming demonstrates the inherent path-dependency of class struggle. The collective action of Bud estate workers leading to the 1970s strikes had cast a long shadow over the sector. State officials and exporters invoked those episodes to demonstrate the ‘side effects’ of estate farming. For them, subcontracting was a means to avoid Bud ‘mistakes’. 5.3. The labour process In the late 2000s, export firms were extremely diverse: the largest being around 1400 ha while smaller firms covered between 10 and 100 ha. In general, the higher the degree of vertical chain coordination, the more complex the division of labour within the firm, reflecting the multiplication and diversification of activities that compliance with European requirements entailed. As in the 1970s, the cost of employing wage labour was compensated for in different ways. The lessons learned from Bud ‘mistakes’ resonated in a broad concern with curbing labour costs and preventing collective action at the heart of the new LLCR. Mechanisation and informal employment still served to mitigate the cost of maintaining a large labour force. But due to the availability of cheap labour, especially women for harvesting, picking and packaging, the mechanisation of activities was patchy and extremely varied. Firms combined diverse forms of employment: permanent staff for management, seasonal staff for some farming and transport activities, and informal, daily and piece-work employment for the majority. In the largest firms the number of permanent employees did not exceed 100 while daily seasonal workers could peak at over 2000–2500.12 No manager could be precise about the number of workers employed within his/her operations.13 Mirroring Bud, informal employment was disproportionally female. A quantitative study of the sector estimated that women represented 90% of the workforce in mango/bean firms and 60% in tomato firms (Maertens and Swinnen, 2012). The gender division of labour was still fairly rigid: despite one woman owning and actively managing an export-producing firm, and a few secretaries, females dominated in unskilled, informal and poorly paid occupations. While picking, sorting and packaging remained essentially ‘women’s jobs’, men were more likely employed in management, cultivation and transport. According to managers, women received lower wages than men and were more often paid on a piece-rate. As before, gender segmentation was buttressed by the pervasive disciplining of women according to supposedly natural attributes. The construction of narratives describing women’s higher performance in tedious, and less skilled tasks was ubiquitous in managers’ accounts. Women’s ‘nimble fingers’ and docility emerged repeatedly in managers’ attempts to highlight what women were best at, e.g. sorting French beans or idly waiting for hours for the next truck to arrive. Ironically, this contradicted the managers’ enduring obsession with ‘the problem of the women’ as a ‘wild’ mass, unaccustomed or unsuited for skilled or continuous work therefore requiring special and flexible arrangements. These were invariably invoked to accommodate women’s needs to combine wage and domestic work. In 2008, a supervisor of one of the largest companies described female pickers as follows: They are happy to work by the piece. In this way they can fill a few buckets enough for their d.q. [dépense quotidienne] and go home. Women living further away do more buckets. The system just works perfectly for them. While the ability of some women to turn the terms of their employment to their advantage emphasised instances of micro-agency (Carswell and De Neve, 2013), the latter were generally anchored to compliance with household duties, underlining their structural subordination within the household noted earlier. Building on this micro-agency, management legitimised insecure female employment by picturing it as a natural continuum of household activities. This depiction however became complicated by the increasing competition for women that refined exporters’ recruitment strategies. The growing need for women at peak times prompted larger employers to collect women from more distant villages thereby widening the labour frontier to the outskirts of the Niayes. Exporters would frequently claim to be the best and ‘most caring’ employers, emphasising the need to strengthen the extra economic nature of employment. Alongside the social disciplining of women workers as housewives, the high costs of large-scale employment and standards compliance were crucially sustained through contract farming, i.e. through the spatial disciplining of all workers. Contracting offered exporters the possibility to push labour control to local growers, partially fragment the labour force within smaller units of production, and critically, further divide rural workers between those ‘included’ and those ‘excluded’ by standard compliance. While this distinction has been broadly observed in the literature (Evers et al., 2014; Goger et al., 2014), the structural relation between the two has received less attention (Bair and Werner, 2015). Despite the re-emergence of large estates, and the important role of large local growers, small growers were still largely included, although their involvement was generally unrecorded and unpredictable, often on the ‘spot’. Smallholders were both the sole source of supply for small exporters and a critical source of hedging for larger exporter–producers. In other words, and echoing the strong pressures constituting PCP, smallholders occupied a position in the chain that simultaneously created and restrained their accumulation (Gibbon and Neocosmos, 1995; Bernstein, 2003). Their competitive advantage vis-à-vis large farms consisted essentially in lower labour costs and income/market diversification strategies to make up for uneven and meagre accumulation (interviews; Maertens and Swinnen, 2009, 2012). Lower labour costs stemmed from cheap, seasonal, immigrant labour and a variegated and shifting pool of relatives, including women contributing to farming, trade and household reproductive duties (RDS, 2012). Women’s greater exploitation within small farms is evidenced by a large survey of the sector that highlighting how women gained more from employment on large farms than from their family contracting activities (Maertens and Swinnen, 2012). Furthermore, given the concentration of export production in just a few months, smallholders would produce on contract while growing other crops for local markets and—where possible—engage in a diversified portfolio of income-generating activities functional to sustaining subcontracting, where accumulation was insufficient and irregular. Therefore, subcontracting to non-compliant farms was possible because of the ability of small farmers to capture and exploit untapped sources of cheap labour and to engage in other sources of reproduction/accumulation. In short, the incorporation of some workers on certified farms was sustainable due to the labour of other cheap workers on non-compliant farms, and the unpaid work of family and female labour. Conversely, the ability of exporters to enter and sustain chain inclusion rested on the ability of small farmers to ‘get by’ with a more adverse incorporation (Phillips, 2011) in the network. Overall, the combination of estate and contract farming constituted a crucial element of the new LLCR. This combination, by articulating different instances of social and spatial disciplining, served to sustain the economic costs of a tighter, supermarket-driven network, based on the mutual upstream relations between different segments of capital (large and small) and labour (permanent/seasonal workers on certified farms, daily male and female workers, hidden workers on small farms, unpaid family labour including the reproductive work of women). The relations between large- and small-scale firms, encapsulating the ‘modern’ and the ‘African’ ways, respectively, also reflected the state/donors’ development agendas centred on coalescing firms and households through the constant branding of the industrious ‘African smallholder’. Ultimately, the precarious and uneven outsourcing to small growers highlighted once again the structural link between firms and households: the production/reproduction of an indistinct, uneven and hidden cheap labour frontier serving the broad sustenance of the network. As this article has highlighted, the dialectical and path-dependant nature of the LLCR has entailed a deepening of labour control in the Niayes throughout the 2000s. This was largely due to the preservation of households as the engine of both reproduction and production activities. Is this still the case today? And what are the consequences for labour? Whilst available evidence emphasises the ‘significant’ contribution of smallholders to FFV exports (RDS, 2013), the relations between estate and contract farming remains largely unexplored. Thus, the degree to which exporters rely on subcontracting to cut costs and comply with supermarket standards remains hard to measure, due to both the lack of further research and the ‘underground’ nature of much contract farming. Meanwhile, the prolonged growth of export horticulture has increased pressures on land and labour driving increasing investments north of the Niayes, especially in the Senegal river region (RDS, 2013; Van den Broeck et al., 2016). Despite being extremely dynamic and diversified, the Niayes labour markets remain largely unexamined (Oya, 2015). While it may be too early to observe radical changes in the Cap-Vert Niayes LLCR, the prolonged growth of export horticulture is likely to deepen the socio-economic differentiation among producers and workers alike, constantly changing the terms and forms of inclusion within the GPN. In particular, the growing demand for female workers at harvest may strengthen women’s bargaining power vis-à-vis employers. Given the strong unequal gender relations that still characterise this area (RDS, 2012), and based on the relational nature of worker agency illustrated in this article, perhaps it is more likely that the expansion in the female labour market will lead to further internal differentiation, improving the terms of inclusion of some at the expense of others. 5. Conclusions The historical development of the Senegalese–European horticultural production network shows the centrality of labour control at the point of production for a variegated range of actors. A comparison of the labour control regimes analysed in this article is included in Table 1. Large European firms driving supply chains, development agencies, the Senegalese state, local firms, households and workers all shape the complex architecture of LLCRs. These operate by exploiting and disciplining workers in different ways: labour-saving mechanisation, employment informality, gender segmentation and subordination, as well as the spatial separation of workers across firms and households. The interplay between labour exploitation and disciplining contributes to labour regime analysis showing how labour control operates across a broader space than the workplace, in particular by structuring relationships among firms, households, and genders. Thus, the case of Senegal illustrates how places of production cannot function without broader spaces of labour control, while at the same time the constitutive and scalar nature of labour control challenges network-essentialist literature. Table 1 Niayes LLCRs in 1970s and 2000s Scale  Actors  1970s LLCR based on estate farming  2000s LLCR based on estate and contract farming  Global scale  TNCs, Development agencies and donors  Bud Antle vertical integration (direct labour control) Large-scale, mechanised farming in Senegal as the fastest and most efficient way to supply European markets Development discourse based on ‘modern agriculture’ as large-scale, mechanised, farming and ‘technical superiority’ of ‘scientific production’ and ‘scientific management of workers’ Agency: financial support to set up Bud Senegal  Supermarkets vertical coordination (indirect labour control) Standards’ compliance drives direct production while low standard enforcement and diversified GPNs enable contract farming Development discourse based on supply chain inclusion and absorption of the development function within standards Agency: technical and infrastructural support for both large- and small-scale farms  National scale  State  Agricultural diversification and foreign exchange Increased control of the countryside by pushing small-scale farmers towards wage work  Agricultural diversification and foreign exchange Continuous control of the countryside by avoiding the constitution of large-scale workforces; discourse on ‘peasants’ as the backbone of the country      Agency: financial, infrastructural and political support for Bud Senegal  Agency: infrastructural and technical support to large-scale farming and technical support to small-scale farmers  Local scale  Labour process  Large capital agency: Labour exploitation via large-scale wage employment, mechanisation, informality and segmentation Social disciplining via gendered segmentation of tasks, jobs and labour markets: creation of an informal and casual large-scale female labour force Spatial disciplining via a shift to contract farming following collective labour action  Large capital agency: Local exporters combining estate farming (to comply with supermarket standards) and contract farming (to finance compliance) Labour exploitation via large-scale wage employment, mechanisation, labour informality and segmentation Social disciplining via gendered segmentation of tasks, jobs and labour markets: sustainment and expansion of an informal, large-scale female labour force Spatial disciplining via combining estate and contract farming to prevent collective labour action      Small capital agency: Small-scale farmers combining contract farming, different markets and economic activities Labour exploitation via cheap and unpaid (family) labour, labour informality Social and spatial disciplining via firms as households    Labour agency: Collective labour actions, strikes walk-outs Women combining housework and wage-work  Labour agency: Women combining housework and wage-work Absence of manifest collective action  Scale  Actors  1970s LLCR based on estate farming  2000s LLCR based on estate and contract farming  Global scale  TNCs, Development agencies and donors  Bud Antle vertical integration (direct labour control) Large-scale, mechanised farming in Senegal as the fastest and most efficient way to supply European markets Development discourse based on ‘modern agriculture’ as large-scale, mechanised, farming and ‘technical superiority’ of ‘scientific production’ and ‘scientific management of workers’ Agency: financial support to set up Bud Senegal  Supermarkets vertical coordination (indirect labour control) Standards’ compliance drives direct production while low standard enforcement and diversified GPNs enable contract farming Development discourse based on supply chain inclusion and absorption of the development function within standards Agency: technical and infrastructural support for both large- and small-scale farms  National scale  State  Agricultural diversification and foreign exchange Increased control of the countryside by pushing small-scale farmers towards wage work  Agricultural diversification and foreign exchange Continuous control of the countryside by avoiding the constitution of large-scale workforces; discourse on ‘peasants’ as the backbone of the country      Agency: financial, infrastructural and political support for Bud Senegal  Agency: infrastructural and technical support to large-scale farming and technical support to small-scale farmers  Local scale  Labour process  Large capital agency: Labour exploitation via large-scale wage employment, mechanisation, informality and segmentation Social disciplining via gendered segmentation of tasks, jobs and labour markets: creation of an informal and casual large-scale female labour force Spatial disciplining via a shift to contract farming following collective labour action  Large capital agency: Local exporters combining estate farming (to comply with supermarket standards) and contract farming (to finance compliance) Labour exploitation via large-scale wage employment, mechanisation, labour informality and segmentation Social disciplining via gendered segmentation of tasks, jobs and labour markets: sustainment and expansion of an informal, large-scale female labour force Spatial disciplining via combining estate and contract farming to prevent collective labour action      Small capital agency: Small-scale farmers combining contract farming, different markets and economic activities Labour exploitation via cheap and unpaid (family) labour, labour informality Social and spatial disciplining via firms as households    Labour agency: Collective labour actions, strikes walk-outs Women combining housework and wage-work  Labour agency: Women combining housework and wage-work Absence of manifest collective action  Strong state intervention recasts the role of development within the proliferation of GPNs in the global South. Ultimately, this intervenes between the fundamental requirement to control labour to produce commodities and the consequent management of the labour question, two fundamental questions within capitalism. Often, in the African countryside this has led to dual rural development policies: on the one hand buttressing highly capitalised, modern, large-scale farms, while on the other supporting smallholders as mechanisms of inclusionary development alien to class struggle. The material and discursive support for small-scale production blurs social conflict and transfers labour control to within the boundaries of the household. As inclusion within GPNs furthers the commodification of labour, and the risk of labour organisation, the protracted support for ‘peasants’ becomes a powerful answer to the labour question. From this perspective, rather than simply a ‘natural’ development of flexible, just-in-time sourcing of supermarkets, the continuous proliferation of subcontracting and the uncharted reliance on smallholders in the global South reveal heavy footprints of the state and development policies. The constitutive role of households also shows how labour control and the labour question revolve around gender relations; how the bond between households and firms features a far more structural interdependence between the productive and the reproductive economy; and how agency in one sphere is linked to agency/control in the other. In Senegal, the subordination of women across firms and households has unleashed abundant cheap labour to sustain upstream production for four decades. Thus, while the labour control perspective necessarily overlooks internal differentiation, agency and forms of female ‘empowerment’ (Deere, 2005; Selwyn, 2010), in Senegal it highlights the central contention that ‘gender hierarchies have not been erased from markets’ (Razavi, 2009, 211). This prompts a crucial question for a nuanced understanding of workers’ agency everywhere: the ability of some workers to reap the benefit of European supermarket standards compliance can hardly be disjoined from a broad feminisation of the workforce whose informality is internalised as a universal opportunity. Indeed, the case of Senegal situates worker agency in a very complex network of path-dependant, contingent and embedded relations among different segments of capital, differentiated and segmented labour forces, and the state. Shifting regimes of labour control emphasise the relationship between capital and labour ‘as bound in a dialectical totality’ (Cumbers et al., 2010, 67) where the increasing control of labour can fuel organised labour resistance, in turn thwarted by deepening labour control regimes. In Senegal, the latter somehow deepen the relations between firms and households structuring the new supermarket-driven production networks. So far, the agencies of different sections of capital, donors, the state and workers have led to ‘the structural domination of capital’ in the Niayes. How workers react, actively shape and challenge this domination remains a crucial question. Acknowledgements I would like to thank Liam Campling, Gerard Hanlon, Jonathan Pattenden, and Adrian Smith for the precious support and invaluable comments offered. A very early version of this article was presented at the Symposium on Labour and Large-Scale Agriculture in Post-Independent Africa and South-East Asia (organised by DIIS, CIRAD and PLAAS) held in Cape Town in September 2012 and benefited from the helpful comments of Benoit Daviron, Peter Gibbon, Bruno Losch, and other participants. I am also very grateful to the editor of the Journal of Economic Geography and the anonymous peer reviewers, who have provided constructive feedback on earlier versions of this article. Footnotes 1 See Pattenden (2016) for a different framing of the LLCR in agrarian formations. 2 In 2013, Senegal’s French bean exports to the EU ranked behind Morocco, Egypt, Kenya and Rwanda and behind Morocco, Egypt, Ethiopia, South Africa and Tunisia for tomatoes (FOASTAT). Senegal is the largest Western African exporter of tomatoes to the EU (République du Sénégal, 2013). 3 In line with Jonas (1996), the ‘local’ is defined here at the subregional scale. Other export horticulture areas, notably the Senegal river valley bordering with Mauritania and Casamance in the South, are characterised by rather different political economies and histories and most likely present different, more or less structured, LLCRs. 4 The Niayes accounts for 60% of Senegalese FFV production and 50% of FFV exports (RDS, 2013); export production has been growing rapidly in the Senegal river region since the mid-2000s. 5 FFV exports peaked in 1975–1976 to around 10,000 tons, of which Bud exported more than 8000 tonnes (Berniard et al., 1976) and then again from the mid-2000s when seasonal averages oscillated between 15,000–20,000 tons (Data provided by the Organisation Nationale des Producteurs Exportateurs de Fruits et Legumes—ONAPES, and FAOSTAT). 6 This was complemented with available secondary sources (a limited sample includes Amin, 1969; RDS, 1972, 1974a, 1974b; Berniard et al., 1976; Collins and Lappe, 1977; Batsch, 1980; Horton, 1987; Rassas, 1988). 7 See Mackintosh (1989, 24–41 and 201–204). 8 Faced with dropping groundnut prices, in the late 1960s Senegalese peasants were massively shifting to food production (Mbodji, 1992). 9 The effectiveness of ethical standards has also been questioned empirically (Barrientos and Smith, 2007). 10 See the Projet de Promotion des Exportation Agricoles (PPEA) supporting medium and large-scale horticultural producers through the provision of technical support and basic infrastructure (Fall, 2004). 11 For example, see the Programme de Développement des Marchés Agricoles du Sénégal (PDMAS): the EU Pesticide Initiative Programme (Jaud and Cadot, 2012); and the government’s initiatives like the Plan Reva, and the latest Projet de Développement Inclusif et Durable de l’Agrobusiness au Sénégal (PDIDAS) envisaging a ‘sustainable’ and ‘inclusive’ development of horticulture based on strong linkages between large and small producers. 12 Data collected in interviews. 13 Colen et al. (2012) confirm the large prevalence of informal employment. References Agarwal B. ( 1986) Women, poverty and agricultural work in India. Journal of Peasant Studies , 13: 165– 220. Google Scholar CrossRef Search ADS   Amanor K. ( 2009) Global food chains, African smallholders and World Bank governance. Journal of Agrarian Change , 9: 247– 262. Google Scholar CrossRef Search ADS   Amin S. ( 1969) Le Monde Des Affaires Sénégalais . Paris: Editions de Minuit. Anner M. ( 2015) Labour control regimes and workers resistance in global supply chains. Labour History , 56: 292– 307. Google Scholar CrossRef Search ADS   Anseeuw W. ( 2010) Agricultural policy in Africa—renewal or status quo? In Padayachee V. (ed.) The Political Economy of Africa . London: Routledge, pp. 247– 265. Baglioni E. ( 2015) Straddling contract and estate farming: accumulation strategies of Senegalese horticultural exporters. Journal of Agrarian Change , 15: 17– 42. Google Scholar CrossRef Search ADS   Bair J. ( 2008). Global commodity chains: genealogy and review. In Bair J. (ed.) Frontiers of Commodity Chains Research . Ithaca, NY: Cornell University Press, pp. 1– 34. Bair J., Werner M. ( 2015) Global production and uneven development. In Newsome K. et al.   (eds) Putting Labour in its Place . New York: Palgrave, pp. 119– 134. Google Scholar CrossRef Search ADS   Barrientos S., Kritzinger A. ( 2004) Squaring the circle: global production and the informalization of work in South Africa fruit export. Journal of International Development , 16: 81– 92. Google Scholar CrossRef Search ADS   Barrientos S., Smith S. ( 2007) Do workers benefit from ethical trade? Third World Quarterly , 28: 713– 729. Google Scholar CrossRef Search ADS   Batsch ( 1980) Les mésaventures d’une entreprise de maraîchage au Sénégal. Le Monde Diplomatique . Berniard S., Chatain E., Lavenant H. ( 1976) Le Sénégal en chiffres . Dakar: Société Africaine d’Edition. Bernstein H. ( 2003) Farewells to the peasantry. Transformation: Critical Perspectives on Southern Africa , 52: 1– 19. Google Scholar CrossRef Search ADS   Bernstein H. ( 2010) Class Dynamics of Agrarian Change . Halifax and Winnipeg: Fernwood Publishing. Braverman H. ( 1998) Labor and Monopoly Capital . New York: Monthly Review Press. Burawoy M. ( 1985) The Politics of Production . London: Verso. Burawoy M. ( 1992) Manufacturing Consent . Chicago: The University of Chicago Press. Carswell G., De Neve Geert ( 2013) Labouring for global markets. Geoforum , 44: 62– 70. Google Scholar CrossRef Search ADS   Castree N. ( 2007) Labour geography: a work in progress. International Journal of Urban and Regional Research , 31, 853– 862. Google Scholar CrossRef Search ADS   Castree N., Coe N. M., Ward K., Samers M. ( 2004) Spaces of Work . London: Sage. Coe N. M. ( 2013) Geographies of production III. Progress in Human Geography , 37: 271– 284. Google Scholar CrossRef Search ADS   Coe N. M., Dicken P., Hess M. ( 2008) Global production networks: realizing the potential. Journal of Economic Geography , 8: 271– 295. Google Scholar CrossRef Search ADS   Coe N. M., Dicken P., Hess M., Yeung H. ( 2010) Making connections: global production networks and world city networks. Global Networks , 10: 138– 149. Google Scholar CrossRef Search ADS   Coe N. M., Hess M. ( 2013) Global production networks, labour and development. Geoforum , 44: 4– 9. Google Scholar CrossRef Search ADS   Coe N. M., Jordhus-Lier D. C. ( 2011) Constrained agency? Re-evaluating the geographies of labour. Progress in Human Geography , 35: 211– 233. Google Scholar CrossRef Search ADS   Coe N, Kelly P. F. ( 2002) Languages of labour: representational strategies in Singapore’s labour control regime. Political Geography , 21: 341– 371. Google Scholar CrossRef Search ADS   Colen L., Maertens M., Swinnen J. ( 2012) Private standards, trade and poverty. The World Economy , 35: 1073– 1088. Google Scholar CrossRef Search ADS   Collins J., Lappe F. M. ( 1977) Multinational agribusiness and the global supermarket. Social Scientist , 6: 3– 12. Google Scholar CrossRef Search ADS   Cooper F. ( 1996) Decolonization and African Society . Cambridge: Cambridge University Press. Google Scholar CrossRef Search ADS   Cowen M., Shenton R. ( 1995) The invention of development. In Crush J. (ed.) Power of Development . London: Routledge, pp. 27– 43. Cramer C., Oya C., Sender J. ( 2008) Lifting the blinkers: a new view of power, diversity and poverty. Journal of Modern African Studies , 46: 361– 392. Google Scholar CrossRef Search ADS   Crehan K. ( 1992) Rural households. In Bernstein H., Crow B., Johnson H. (eds) Rural Livelihoods. Crises and Responses . Oxford: Oxford University Press, pp. 87– 112. Cumbers A., Helms G., Swanson K. ( 2010) Class, agency and resistance in the old industrial city. Antipode , 42: 46– 73. Google Scholar CrossRef Search ADS   Cumbers A., Nativel C., Routledge P. ( 2008) Labour agency and union positionalities in global production network. Journal of Economic Geography , 8: 369– 387. Google Scholar CrossRef Search ADS   Davies S., Hammer N., Willimas G., Raman R., Ruppert S. C., Volynets L. ( 2011) Labour standards and capacity in global subcontracting chains. Industrial Relations Journal , 42: 124– 138. Google Scholar CrossRef Search ADS   Deere C. D. ( 2005) The Feminisation of Agriculture? Economic Restructuring in Rural Latin America . UNRISD: Geneva. Dicken P., Malmberg A. ( 2001) Firms in territories. Economic Geography , 77: 345– 363. Google Scholar CrossRef Search ADS   Dinham B., Hines C. ( 1983) Agribusiness in Africa . London: Erath Resources Research Publication. Diop M. C. (ed.) ( 1992) Sénégal. Trajectoires d’un État . Dakar: CODESRIA. Dolan C. ( 2004) On farm and packhouse: employment at the bottom of a global commodity chain, Rural Sociology , 69: 99– 126. Google Scholar CrossRef Search ADS   Dolan C., Humphrey J. ( 2000) Governance and trade in fresh vegetables. Journal of Development Studies , 37: 147– 176. Google Scholar CrossRef Search ADS   Du Toit A. ( 2001) Ethical Trading, A Force of Improvement, or Corporate Whitewash?, ODI Natural Resource perspectives, N. 71, PLAAS Edwards P. ( 1986) Conflict at Work . Oxford: Blackwell. Elson D. ( 1999) Labour markets as gendered institutions. World Development , 27: 611– 627. Google Scholar CrossRef Search ADS   Elson D., Pearson R. ( 1981) Nimble fingers make cheap workers. Feminist Review , 7: 87– 107. Google Scholar CrossRef Search ADS   Evers B., Maggie O., Barrientos S., Krishnan A., Amoding F., Ndlovu L. ( 2014) Global and Regional Supermarkets . Manchester: Capturing the Gains Working Paper. Fall A. A. ( 2004) Evaluation De La Rentabilité Économique Et Financière Du Projet De Promotion Des Exportations Agricole (PPEA) . Sénégal: Projet de Promotion des Exportations Agricole, Mimistère de l’Agriculture, de l’Elevage et de l’Hidraulique. Fall A. A. ( 2005) Evaluation Economique Du Programme De Développement Des Marchés Agricoles Du Sénégal (Pdmas) . Sénégal: Mimistère de l’Agriculture et de l’Hidraulique. Franke R. W., Chasin B. H. ( 1980) Seeds of Famine . Montclair: Allanheld, Osmund & Co. Fredrichs A. ( 1984) Technological change and the growth of agribusiness: a case study of california lettuce production. In Berardi G. M., Geisler C. C. (eds) The Social Consequences and Challenges of New Agricultural Technologies . Boulder, CO: Westview Press. Freidberg S. ( 2004) French Beans and Food Scares . New York: Oxford University Press. Friedland W., Barton A. E., Thomas R. ( 1981) Manufacturing Green Gold . Cambridge: Cambridge University Press. Friedmann H., McMichael P. ( 1989) Agriculture and the state system. Sociologica Ruralis , XXIX: 93– 117. Google Scholar CrossRef Search ADS   George S. ( 1976) How the Other Half Dies . London: Penguin Books. Gibbon P. ( 2011) Experiences of Plantation and Large-Scale Farming in the 20th Century Africa. Danish Institute for International Studies (DIIS) Working Paper. Gibbon P., Lazaro E., Ponte S. ( 2010) Global Agro-Food Trade and Standards: Challenges for Africa . New York: Palgrave Macmillan. Google Scholar CrossRef Search ADS   Gibbon P., Neocosmos M. ( 1985) Some problems in the political economy of "African socialism". In Bernstein and H., Campbell B. K. (eds) Contradictions of Accumulation in Africa . Beverly Hills, CA: Sage, pp. 153– 206. Glassman J., Choi Y. J. ( 2014) the Chaebol and the US military-industrial complex. Environment and Planning A , 46: 1160– 1180. Google Scholar CrossRef Search ADS   Goger A., Hull A., Barrientos S., Gereffi G., Godfrey S. ( 2014) Capturing the Gains in Africa: Making the Most of Global Value Chain Participation . Durham, North Carolina (USA): Duke University, Centre on Globalization, Governance and Competitiveness. Hastings T., MacKinnon D. ( 2016) Re-imbedding agency at the workspace scale: workers and labour control in Glasgow call centres. Environment and Planning A , 49: 104– 120. Google Scholar CrossRef Search ADS   Heffernan W. D. ( 2000) Concentration of ownership and control in agriculture. In Magdoff F., Bellamy Foster J., Buttel F. (eds) Hungry for Profit . New York: Monthly Review Press. Henderson J., Dicken P., Hess M., Coe N., Wai-Chung Yeung H. ( 2002) Global production networks and the analysis of economic development. Review of International Political Economy , 9: 436– 464. Google Scholar CrossRef Search ADS   Henson S., Reardon T. ( 2005) Private agri-food standards: implications for food policy and the agri-food system. Food Policy , 30: 241– 253. Google Scholar CrossRef Search ADS   Horton J. ( 1987) Characteristics of the Horticultural Export Enterprises Utilizing Contract Farming Schemes in Senegal. IDA Working Paper No. 84, Clark University, Worcester, MA (USA). Hudson R. ( 2001) Producing Places . New York: The Guildford Press. Hughes A. ( 2007) Multi-stakeholder approaches to ethical trade. Journal of Economic Geography , 7: 421– 437. Google Scholar CrossRef Search ADS   Humphrey J. ( 2007) The Supermarket revolution in developing countries. Journal of Economic Geography , 7: 433– 450. Google Scholar CrossRef Search ADS   Jaud M., Cadot O. ( 2012) A second look at the pesticide initiative program: evidence from Senegal. World Trade Review , 11: 490– 506. Google Scholar CrossRef Search ADS   Jonas E. G. ( 1996) Local labour control regimes: uneven development and the social regulation of production. Regional Studies , 30: 323– 338. Google Scholar CrossRef Search ADS   Kautsky K. ( 1900) La Question Agraire . Paris: Giard & Brière. Kelly P. F., ( 2002) Spaces of labour control: comparative perspectives from Southeast Asia. Transactions of the Institute of British Geographers , 27: 395– 411. Google Scholar CrossRef Search ADS   Kelly P. F. ( 2009) From global production networks to global reproduction networks: households, migration, and regional development in Cavite, the Philippines. Regional Studies , 43: 449– 461. Google Scholar CrossRef Search ADS   Kelly P. F. ( 2013) Production networks, place and development: thinking through global production networks in Cavite, Philippines. Geoforum , 44: 82– 92. Google Scholar CrossRef Search ADS   Lier D. C. ( 2007) Places of Work, scales of organising: a review of labour geography. Geography Compass , 1: 814– 833. Google Scholar CrossRef Search ADS   Mackintosh M. ( 1977) Fruit and vegetables as an international commodity. Food Policy , 2: 277– 292. Google Scholar CrossRef Search ADS   Mackintosh M. ( 1989) Gender, Class and Rural Transition, Agribusiness and the Food Crisis in Senegal . London: Zed Books. Maertens M., Swinnen J. M. ( 2009) Trade, standards, and poverty: evidence from Senegal. World Development , 37: 161– 178. Google Scholar CrossRef Search ADS   Maertens M., Swinnen J. M. ( 2012) Gender and modern supply chains in developing countries. Journal of Development Studies , 48: 1412– 1430. Google Scholar CrossRef Search ADS   Marx K. ( 1990), Capital Volume I . London: Penguin. Mbodji M. ( 1992) L’Etat de l’économie arachidière: une crise trentenaire. In Diop M. C. (ed.) Sénégal Trajectoire d’un État . Dakar: CODESRIA, pp. 95– 135. McGrath-Champ P., Herod A., Rainnie A. ( 2010) Working Space . Cheltenham: Edward Elgar. Mezzadri A. ( 2016a) Class, gender and the sweatshop. Third World Quarterly , 37: 1877– 1900. Google Scholar CrossRef Search ADS   Mezzadri A. ( 2016b) The Sweatshop Regime . Cambridge: University Press. Google Scholar CrossRef Search ADS   Mies M. ( 2014) Patriarchy and Accumulation on a World Scale . London: Zed Books. Müller C., Vermeulen J. V., Glasbergen P. ( 2009) Perceptions on the demand side and realities on the supply side. Sustainable Development , 17: 295– 310. Google Scholar CrossRef Search ADS   Neethi P. ( 2012) Globalization lived locally: investigating Kerala’s local labour control regimes. Development and Change , 43: 1239– 1263. Google Scholar CrossRef Search ADS   Neilson J., Pritchard B., Yeung H. ( 2014) Global value chains and global production networks in the changing international political economy. Review of International political Economy , 21: 1– 8. Google Scholar CrossRef Search ADS   Newsome K., Taylor P., Bair J., Rainnie A. ( 2015) Putting Labour in Its Place . New York: Palgrave Macmillan, pp. 1– 26. Google Scholar CrossRef Search ADS   Nichols T. ( 1980) Marx on the labour process. In Nichols T. (ed.) Capital and Labour . Glasgow: Fontana Paperbacks. O’Laughlin B. ( 2007) A bigger piece of a very small pie: intrahousehold resource allocation and poverty reduction in Africa. Development and Change , 38: 21– 44. Google Scholar CrossRef Search ADS   Oya C. ( 2012) Contract farming in Sub-Saharan Africa: a survey of approaches, debates and issues. Journal of Agrarian Change , 12: 1– 33. Google Scholar CrossRef Search ADS   Oya C. ( 2015) Rural labour markets and agricultural wage employment in semi-arid Africa. In Oya C., Pontara N. (eds) Rural Wage Employment in Developing Countries . London: Routledge, pp. 37– 68. Oya C., Ba C. O. ( 2013) Les politiques agricoles 2000–2012. In Diop M. C. (ed.) Sénégal 2000-2012 . Paris: Karthala, pp. 149– 178. Pattenden J. ( 2016) Working at the margins of global production networks. Third World Quarterly , 37: 1809– 1833. Google Scholar CrossRef Search ADS   Peck J. ( 2013) Making space for labour. In Featherstone D., Painter J. (eds) Spatial Politics . Oxford: Wiley-Blackwell. Phillips N. ( 2011) Informality, global production networks and the dynamics of ‘adverse incorporation’. Global Networks , 3, 380– 397. Google Scholar CrossRef Search ADS   Rainnie A., Herod A., McGrath-Champ S. ( 2011) Review and positions: global production networks and labour. Competition and Change , 15: 155– 169. Google Scholar CrossRef Search ADS   Rassas B. ( 1988) Contract Farming in the Horticultural Sector in Senegal. IDA Working Papers No. 85, Clark University, Worcester, MA (USA). Razavi S. ( 2009) Engendering the political economy of agrarian change. Journal of Peasant Studies , 36: 197– 226. Google Scholar CrossRef Search ADS   RDS – République du Sénégal ( 1972) Rapport De Présentation De La Convention D’établissement Bud Sénégal. Dakar: Direction des Archives Du Sénégal. RDS – République du Sénégal ( 1974a) Bud Sénégal Société D’économie Mixte Avenant N.1 A La Convention D’établissement Du 4 Février 1972. Dakar: Direction des Archives Du Sénégal. RDS – République du Sénégal ( 1974b) Projet D’aménagement De Deux Périmètres Maraichages Irrigues De 200 Ha Chacun Pour Y Installer 400 Familles Des Paysans. Dakar: Direction des Archives Du Sénégal. RDS – République du Sénégal ( 2012) Étude Diagnostique Sur Légalité Entre Les Femmes Et Les Hommes Dans Les Niayes, PADEN. Dakar, Senegal. RDS – République du Sénégal ( 2013) Études De Filières Horticoles, PADEN. Dakar, Senegal. Riisgard L., Hammer N. ( 2011) Prospects for labour in global value chains. British Journal of Industrial Relations , 49: 168– 190. Google Scholar CrossRef Search ADS   Robinson P. K., Rainbird H. ( 2013) International supply chains and the labour process. Competition and Change , 17: 91– 107. Google Scholar CrossRef Search ADS   Ruwanpura K. N. ( 2016) Garments without guilt? Uneven labour geographies and ethical trading - Sri Lankan labour perspectives. Journal of Economic Geography , 16: 423– 446. Google Scholar CrossRef Search ADS   Sall B. ( 1993) De La Modernité Paysanne En Afrique Noire. Le Sénégal . Paris: L’Harmattan. Selwyn B. ( 2007) Labour processes and workers’ bargaining power in export grape production, North East Brazil. Journal of Agrarian Change , 7: 526– 553. Google Scholar CrossRef Search ADS   Selwyn B. ( 2010) Gender wage work and development in North East Brazil. Bulletin of Latin American Research , 29: 51– 70. Google Scholar CrossRef Search ADS   Selwyn B. ( 2012) Beyond frim-centrism: re-integrating labour and capitalism into global commodity chain analysis. Journal of Economic Geography , 12: 205– 226. Google Scholar CrossRef Search ADS   Senghor F. ( 2006) Plan Reva. Note Introductive. Dakar: Ministère de l’Agriculture, de l’Hydraulique Rurale et de la Sécurité Alimentaire, Sénégal. Smith A. ( 2015) The state, institutional frameworks and the dynamics of capital in global production networks. Progress in Human Geography , 39: 290– 315. Google Scholar CrossRef Search ADS   Smith C. ( 2006) The double indeterminacy of labour power. Work Employment and Society , 20: 389– 402. Google Scholar CrossRef Search ADS   Starosta G. ( 2010) Outsourcing of manufacturing and the rise of the giant global contractors. New Political Economy , 15: 543– 563. Google Scholar CrossRef Search ADS   Swinnen J. F. M., Colen L., Maertens M. ( 2013), Constraints to smallholder participation in high-value agriculture in West Africa. In Elbehri A. (ed.) Rebuilding West Africa’s Food Potential . FAO/IFAD. Available online at: http://www.fao.org/docrep/018/i3222e/i3222e09.pdf. Tallorine A., Dolan C., Smith S., Barrientos S. ( 2005) Reaching the marginalised? gender value chains and ethical trade in African horticulture. Development in Practice , 15: 559– 571. Google Scholar CrossRef Search ADS   Taylor M. ( 2007) Rethinking the global production of uneven development. Globalizations , 4: 529– 542. Google Scholar CrossRef Search ADS   Taylor M. ( 2008) Power, Conflict and the Production of the Global Economy. In Taylor M. (ed.) Global Economy Contested . London: Routledge. Google Scholar CrossRef Search ADS   Taylor P., Newsome K., Bair J., Rainnie A l. ( 2015) Putting Labour in its Place . New York: Palgrave Macmillan. Google Scholar CrossRef Search ADS   Thompson P. ( 2010) The capitalist labour process: concepts and connections. Capital and Class , 34: 7– 14. Google Scholar CrossRef Search ADS   Thomson P., Smith C. ( 2009a) Labour power and labour process: contesting the marginality of the sociology of work. Sociology , 43: 913– 930. Google Scholar CrossRef Search ADS   Thomson P., Smith C. (eds) ( 2009b) Working Life . Basingstoke: Palgrave. Van den Broeck G., Van Hoyweghen K., Maertens M. ( 2016) Employment conditions in the Senegalese horticultural export industry: a worker perspective. Development Policy Review , 34: 301– 319. Google Scholar CrossRef Search ADS   Whitehead A. ( 1991) Women in food production in Sub-Saharan Africa. In Sen A., Dreze J. (eds) The Political Economy of Hunger . Oxford: Clarendon Press, pp. 425– 473. Whitehead A., Kabeer N. ( 2001) Living with Uncertainty: Gender, Livelihoods and Pro-Poor Growth in Rural Sub-Saharan Africa, IDS Working Paper 134, Institute of Development Studies, Brighton, Sussex, England. World Bank ( 2007) World Development Report 2008 . New York: Oxford University Press. © The Author (2017). Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oup.com http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Journal of Economic Geography Oxford University Press

Labour control and the labour question in global production networks: exploitation and disciplining in Senegalese export horticulture

Loading next page...
 
/lp/ou_press/labour-control-and-the-labour-question-in-global-production-networks-DlIao1d8j6
Publisher
Oxford University Press
Copyright
© The Author (2017). Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oup.com
ISSN
1468-2702
eISSN
1468-2710
D.O.I.
10.1093/jeg/lbx013
Publisher site
See Article on Publisher Site

Abstract

Abstract This article advocates the centrality of labour control to understand the constitutive role of labour and production within global production networks (GPNs). It draws from the global value chain/GPN literature, labour process theory and agrarian political economy to examine the architecture of labour control in the Senegalese–European horticultural GPN through an analysis of local labour control regimes that are constituted at different scales (global, national and local). It frames labour control through the interplay of labour exploitation and disciplining, identifying the different actors, institutions and dynamics shaping labour control and links places of production to broader spaces of labour control. The strong role of the Senegalese state indicates that labour control is closely connected to a broader ‘labour question’ that revolves around the productive/reproductive activities of households and gender subordination. The historical analysis shows the path-dependent, dialectical nature of labour control and labour resistance, and importantly, their dependency on broader production and reproduction relations. 1. Introduction The gradual tackling of a ‘labour gap’ represents a key development since the early elaboration of the global commodity chain (GCC), global value chain (GVC) and global production network (GPN) frameworks (Henderson et al., 2002; Bair, 2008; Coe et al., 2008; Taylor, 2008; Starosta, 2010; Rainnie et al., 2011; Selwyn, 2012). Whilst a number of studies highlight the role of labour organisation and class struggle within commodity chains (Selwyn, 2007; Cumbers et al., 2008), less attention has been devoted to the issue of labour control, its constitutive role within GPNs and its ability to counteract and prevent collective labour action. Building on the emerging dialogue between labour geography, GVC/GPN literature (Coe and Jordhus-Lier, 2011) and labour process theory (LPT) (Taylor et al., 2015), this article advocates the centrality of labour control for a more nuanced understanding of production within GPNs. Empirically, it shifts attention from Northern industrial formations to the countryside of the global South, with a focus on Senegalese export horticulture—a pioneer amongst horticultural GPNs and today hailed as a champion of poverty reduction (World Bank, 2007; Maertens and Swinnen, 2009). The article sheds light on evolving labour control regimes in the South that have contributed to the broader ‘horticultural revolution’ in the 1970s (Mackintosh, 1977) and the ‘supermarket revolution’ of the 1990s–2000s (Dolan and Humphrey, 2000) in the West. The prism of labour control provides a contribution to three main debates that prompt a more sophisticated understanding of the constitutive role of labour within GPNs. The first concerns the analysis of labour control regimes linked with the broader dynamics of supply chains (Anner, 2015; Mezzadri, 2016b). To grasp the complex and changing architecture of labour control underpinning GPNs, the article develops the analysis of the local labour control regime (LLCR) (Jonas, 1996), which frames social relations of production in particular places in continuity with different spaces and scales (Castree et al., 2004). Labour control regime analysis has hitherto received only erratic attention within economic geography (Hastings and MacKinnon, 2016): very few studies examine LLCRs from an historical perspective (Ruwanpura, 2016) and within agriculture (Riisgard and Hammer, 2011). To advance this analysis, labour control is framed along the interplay between labour exploitation and disciplining, as respectively (i) the production of value in excess of labour remuneration and (ii) the mechanisms of mitigation, containment and prevention of conflict inherent to production, i.e. the constant subordination of labour to the labour process. The analytical distinction between exploitation and disciplining reconciles places of production (firms) with broader spaces of labour control (here households) that are shaped at different scales (global, national and local). Second, the article responds to a long-standing call to move beyond both firm- and network-centric views of production networks that provide a partial analysis of the social relations of production (Taylor, 2007). The analytical prism of LLCRs challenges much of the ‘network ontology’ inherent to the GVC/GPN literature (Bair and Werner, 2015) by highlighting specific upstream labour control edifices that emerge from the unique interplay of the dynamic restructuring of global capital, state agendas, conflicting local mediation between capital and labour and household social reproductive spaces. This reinstates the structural role of the state and ‘development’ within GPNs (Kelly, 2013; Smith, 2015), and links labour control to a broader ‘labour question’ facing the postcolonial state, i.e. the never-ending problem of managing the working class (Cooper, 1996). In turn, this highlights the centrality of households and the gendering of production and reproductive work that broadly sustain GPNs (Kelly, 2009; Cumbers et al., 2010). Third, the article contributes to recent debates within labour geography calling for a more contingent, embedded and relational understanding of workers agency (Castree, 2007; Lier, 2007; Coe and Hess, 2013; Peck, 2013). By placing workers in relation to broader structures of labour control the article promotes a ‘more embedded conception of agency that reconnects with the continued structural domination of capital in many contexts’ (Coe, 2013, 273). The case study shows that the position of workers in GPNs needs to be understood in relation to the historical evolution of LLCRs, and ‘in relation to the formation of capital, the state, the community and the labour market in which workers are incontrovertibly yet variably embedded’ (Coe and Jordhus-Lier, 2011, 4). On the one hand this demonstrates the path-dependent and dialectical relationship between capital and labour (Henderson et al., 2002), and on the other, their dependency on broader production and reproduction relations. It echoes Cumber et al.’s (2010, 67) invitation to see ‘the agency of capital and labour as bound in a dialectical totality’, but it also confirms the relational nature of both capital and labour agencies vis-à-vis other capital and labour segments. The following Section 2 frames labour control by drawing from LPT, the GVC/GPN literature and agrarian political economy. After a brief methodological discussion in Section 3, I explore changing forms of labour control in Senegal in Sections 4 and 5, focusing on the emergence of LLCRs based on estate farming in the 1970s and its combination with contract farming as of 2000. This is followed by concluding remarks. 2. Framing labour control Labour control has been the traditional analytical domain of LPT, which focuses on ‘the dynamics of control, consent and resistance at the point of production’ (Thomson and Smith, 2009a, 915). Following Marx (1990), Braverman (1998), and Burawoy (1992), LPT conceives the labour process as necessarily 2-fold: the metabolic transformation of nature through labour and the simultaneous production of value in excess of labour remuneration, or reproduction, that is applied to other commodities to make further commodities with a greater value (Nichols, 1980). Thus, it represents a fundamental moment of inequality in the social relations of production. This intrinsic unequal exchange between employer and employee is far from natural and therefore needs to be constantly preserved inside and outside firms. As labour is essentially indeterminate (Smith, 2006), labour control is intrinsic to the labour process and requires the organisation and supervision of workers, i.e. regulation of labour activity and the conflict this engenders. LPT, therefore, suggests that, whether evident or not, conflict is structural to employment relations (Edwards, 1986), and requires a multidimensional and evolving repertoire of labour control strategies. Traditionally preoccupied with the relational nature of labour control and labour resistance within firms, LPT has recently turned its attention to the ‘wider political economy’ affecting employer–employee relations (Thomson and Smith, 2009b; McGrath-Champ et al., 2010; Thompson, 2010). A recent edited collection (Newsome et al., 2015) explores the interrelationships between LPT and GVC/GPN analyses to reintegrate the labour process as a constitutive network element. This article furthers this agenda by building on the LLCR approach, as an analytical device, that dialectically links workplaces to broader local and global dynamics thereby offering a complex, multilevel understanding of the labour process. Jonas defines an LLCR as ‘an historically contingent and territorially embedded set of mechanisms which co-ordinate the time-space reciprocities between production, work, consumption and labour reproduction within a local labour market’ (1996, 325). Thus, LLCRs represent local, relatively stable, institutional fixes to the immanency of labour control and capital accumulation, both at the firm and at the sector level. In this conceptualisation, labour control constantly evolves through the opportunities and constraints posed by various players including workers and their social networks, state officials, local and foreign firms, and in the global South, global governmental and non-governmental institutions. Whilst much LLCR analysis has been rather capital-centric (Hastings and MacKinnon, 2016) and focused on industrial formations (Jonas, 1996; Kelly, 2002; Davies et al., 2011; Neethi, 2012), this notion can be transferred to the countryside1 to examine how network pressures interact with local dynamics of agrarian change. According to Jonas, ‘a local labour control regime involves the exercise of power by local agencies and their particular strategies of control’ (1996, 329). To emphasise the agencies (of capital, labour and other institutions such as states and households) constituting LLCRs, I frame labour control as the interplay between labour exploitation and labour disciplining. Whereas labour exploitation refers to the production of value in excess of labour remuneration, labour disciplining refers to the mechanisms of mitigation, containment and prevention of conflict inherent to production, i.e. the constant subordination of labour to the labour process. Just as labour exploitation serves the production of commodities, labour disciplining serves labour exploitation: the two are necessarily linked, ambiguously and often mutually reinforcing. Although manifestly occurring within firms, exploitation and disciplining of workers is based on social distinctions other than class (e.g. gender, race, ethnicity, etc.) and are buttressed by wider state policies, as well as the interests of foreign capital. Thus, labour disciplining requires an analysis of labour process reliance upon a broader subsumption of labour within and beyond the immediate place of production. From this perspective, LLCRs represent different articulations between labour exploitation and diverse, related instances of labour disciplining within broader spaces. Castree et al. (2004, 115) argue that the notion of the LLCR ‘recognizes that the nature of worker-employer relations in a particular locality is simultaneously determined by local and extra-local dynamics at a variety of scales up to, and including, the global’. This resonates with conceptualisations of GPN as ‘multi-scalar spaces of governance’ (Dicken and Malmberg, 2001) shaped by the jurisdiction of local, national and supranational institutions concurrently structuring production. According to GPN scholars, production in different places is simultaneously affected by the local environment and relations embedded within production networks (Henderson et al., 2002; Coe et al., 2008), i.e. the relationship between places and networks are fundamentally dialectical (Coe et al., 2010). Production is crucial even for distant lead firms as these still ‘exercise control over the labour process through the constraints they place on relations between management and labour’ (Robinson and Rainbird, 2013, 97). Thus, through governance, GPNs channel labour control pressures that shape LLCRs. These, in turn, further combine with pressures from the state. Recent critical interventions have stressed that ‘GVCs/GPNs are emergent artefacts from state action: they always “touch down” somewhere, and in every “somewhere”, there is always the hand of the state’ (Neilson et al., 2014, 2). Whilst critical studies emphasise how states are constitutive of GPNs (Glassman and Choi, 2014; Smith, 2015), and of labour control regimes (Burawoy, 1985), their constitutive role remains essentially contradictory. As Hudson points out, ‘while the state does not represent the general interests of capital, its continued existence as a particular form of social relationship depends upon reproducing the social relations of capital and the successful continuation of the accumulation process’ (2001, 53). In Africa, as elsewhere in the global South, this is the familiar territory of ‘development’, which engages the postcolonial state in the difficult task of promoting ‘modernisation’ without social disruption (Cowen and Shenton, 1995). Historically, ‘development’ has caught African governments in the predicament between attracting capitalist production and dealing with the ensuing ‘labour question’, i.e. the constitution and management of the working classes that collectively organise to improve conditions of employment and reproduction (Cooper, 1996). This is most visible within the agricultural sector, where ‘development’ has led to the simultaneous, and contradictory, support of large-scale mechanised farming and small-scale, family farming (Oya, 2012), often cohabiting at the bottom of GPNs. While the former represents the modernisation option, the latter responds to those rural masses and donor agencies who, fearing the ‘death of the peasantry’, claim that its small-scale, allegedly industrious, and family-friendly nature avoids large-scale land alienation and the creation of landless labour armies. Thus, state support for the ‘peasantry’—presented as something systematically different from, and alien to, the antagonism of class-based capitalist production—offers a powerful means to counteract class struggle and demonstrates how the state’s material and discursive power vitally structures LLCRs (Coe and Kelly, 2002). This is reflected at the local level where both GPN pressures and (more or less manifest) state intervention contribute to the continuous relevance of small-scale farmers as ‘households’. At a local scale, Cumbers et al. (2010, 50) emphasise how the notion of the LLCR ‘recognises capital’s need to do more than just exercise coercive power over labour in production, but also to attend to the conditions under which labour is reproduced’, thereby extending the analytical focus to relations between firms and households and labour control operating beyond firms. It is at this scale that different instances of labour disciplining are more evident. Here, I focus on the disciplining of workers along spatial and social lines. Spatial disciplining refers to the spatial fragmentation of workforces through subcontracting and their ‘containment’ within firms (Kelly, 2002). Social disciplining refers to the social segmentation of workers primarily through the gendering of labour markets and the subordination of women within households. Both develop through the interplay between firms and households. Whereas GPN literature has emphasised the role of subcontracting in creating hidden and informal labour armies (Castree et al., 2004), less attention has been given to its counterpart in agriculture, i.e. contract farming. Contract farming represents a local form of outsourcing that spatially splinters the labour force and externalises the labour process, together with the risks and costs of production, to local growers. The latter include a diverse and differentiated group of farmers, including smallholders, which represent a particularly resilient form of production where the rural armies of hidden, informal and unpaid workers typically orbit. This is emphasised by Bernstein who frames smallholders as petty commodity producers (PCPs), i.e. small-scale capitalist farmers who combine in the same person/household the class ‘places’ of capital and labour, often unequally distributed across family members, especially along gender/age lines, and/or typically attract the cheapest labour (Bernstein, 2003; 2010). Usually defined as ‘family farms’ or ‘peasants’, they are seldom conceived of as a set of extremely heterogeneous small-scale firms typically working small patches of land, straddling self- and wage employment, and combining a portfolio of activities at the intersection between simple reproduction and accumulation. By blurring the boundaries between firms and families, PCPs typically rely on uneven accumulation and strong levels of labour exploitation/disciplining (Kautsky, 1900; Cramer et al., 2008; Journal of Agrarian Change, 1–2, 2004). Here, the smaller scale of production and tighter labour control deter labour organisation and resistance. Thus, the notion of PCP underscores how the spatial fragmentation of labour involves a particular combination of the firm and the family within the same entity. Underlying the spatial disciplining of workers through contract farming and petty commodity production, is perhaps the most pervasive source of disciplining in the global South: the subordination of women through the intersection between the spheres of production and reproduction. In the Southern countryside, the structural bond between households and firms unleashes a substantial supply of cheap female labour that fills casual and informal employment at the bottom of GPNs (Barrientos and Kritzinger, 2004; Dolan, 2004; Tallorine et al., 2005; Goger et al., 2014). The ability of firms to fill insecure employment with women (and migrants) reflects the underlying dynamics of female subordination within households that are channelled through labour markets (Elson and Pearson, 1981). According to Elson (1999), markets are essentially ‘gendered institutions’: they convey widespread social stereotypes which buttress perceptions of typical ‘men’s work’ and ‘women’s work’, and gain from women’s disproportionate assumption of household duties, i.e. their employment in the ‘reproductive economy’. Feminist agrarian political economists have stressed that households are heterogeneous (Whitehead, 1991), and ‘are not necessarily sites of sharing and equity’ (Whitehead and Kabeer, 2001, 2; see also, Agarwal, 1986; O’Laughlin, 2007). As such, ‘the ability of peasant households to produce commodities for sale (including labour) is dependent on their ability to mobilise and organise labour (mostly female) to carry out the domestic work necessary for their reproduction’ (Crehan, 1992, 99). In short, women’s cheap and free labour stems from their ‘patriarchal unfreedom’ (Mezzadri, 2016a). Crucially, this makes women, not men, ‘the optimal labour force for the capitalist’: their constant disciplining as housewives cheapens their cost as workers and has made them the driving force behind the proliferation of GPNs since the 1970s (Mies, 2014, 116). Overall, both spatial and social instances of labour disciplining reveal the structural relationships between firms and households representing a fundamental continuum for production and accumulation. This develops Kelly’s (2009) argument that households are an integral part of GPNs. Formally considered as external to GPN production sites, households (as both workers’ families and subcontracted smallholders) are very much included in the wider space that makes GPNs possible. These spaces are actively shaped by overseas firms, states and local production and reproduction dynamics. If on the one hand this shows that ‘the control of ‘space’ is central to the control of labour in capitalism’ (Jonas, 1996, 330), on the other it also highlights the hidden web of relations that makes production possible. By scrutinising the relations between firms and households and the underpinning disciplining of workers, it is possible to see uneven the relations and emphasise how some workers’ gains are structurally linked to other workers’ losses (Coe and Hess, 2013; Peck, 2013). As the case study of Senegal shows, the empirical investigation of different forms of disciplining emphasises the relationships between different segments of capital, labour and institutions: these are contingent and locally embedded in the particular history and political economy of the Cap-Vert Niayes, which is briefly introduced below. In Sections 4 and 5, I examine the constitution of LLCRs in the 1970s and 2000s in the Cap-Vert Niayes, at the global, national and local scale, through pressures for labour control emanating from: (i) Transnational corporations (TNCs) and international institutions; (ii) the Senegalese state; and (iii) at the firm level through the labour process and its interaction with other local social relations, in particular gender relations straddling production and reproduction. A comparative summary is included in the conclusions. 3. Case study and methodology Senegalese fresh fruit and vegetables (FFV) represent an important source of off-season supply to Europe. 2 Historically, Senegalese horticulture developed in the shadow of the groundnut sector, the traditional engine of the Senegalese economy. Today, FFV exports supplying European markets represent one of its most dynamic sub-sectors. These exports are concentrated in French beans, cherry tomatoes and mangoes. Production is focused in the Niayes area, a narrow belt stretching from the capital Dakar to the northern city of Saint Louis (Figure 1). The Southern part—the Cap-Vert Niayes—is a relatively small area situated between Dakar and Thiès (Figure 2), and corresponds to the local scale of this study.3 This region has known horticultural activities since the 1940s.4 Exports took off in the early 1970s through Bud Senegal, one of the earliest large-scale agribusiness schemes in Africa; they peaked in the mid-1970s and then rose again from 2000.5 These periods correspond to the establishment of tighter, vertically coordinated GPNs and LLCRs marked by estate farming, i.e. large-scale farms employing wage labour. The analysis of LLCRs emerging in both periods draws substantially from the seminal work of Maureen Mackintosh’s Gender, Class and Rural Transition, Agribusiness and the Food Crisis in Senegal (1989),6 and from the author’s fieldwork carried out in 2007–2008 in the same area. Figure 1 View largeDownload slide Senegal and the Niayes region. Figure 1 View largeDownload slide Senegal and the Niayes region. Figure 2 View largeDownload slide The Cap-Vert Niayes. Figure 2 View largeDownload slide The Cap-Vert Niayes. Mackintosh’s work drew from fieldwork carried out in the mid-1970s across Europe, Dakar and the Niayes villages supplying labour to Bud estates. Here fieldwork activities explored the complex intertwining between relations of production and consumption at the village level. Information was gathered by combining quantitative and qualitative data, and by interviewing Bud management and workers.7 Her work represents a precursor to the more nuanced GVC/GPN and labour control regime studies where local relations of production and reproduction are analysed as part of a wider process of global capitalist restructuring and post-colonial state-building that underpinned the global horticultural revolution of the 1970s. In 2007–2008, I carried out fieldwork in the Niayes, including the villages adjacent to former Bud estates. Compared to the 1970s, when Bud dominated the sector, and despite export concentrations within three export firms, in the 2000s the sector was crowded by several firms. The fieldwork aimed at collecting the life histories, accumulation trajectories and strategies of local exporters–producers who had driven the sector in the 1990s–2000s. This was done with multiple semi-structured interviews with exporters (24 out of 30 active export firms), and their managerial staff at different levels. The information gathered was triangulated through secondary sources and interviews with local subcontracting and non-subcontracting growers and around 50 other stakeholders including state officials and NGOs. This article places the agency of local horticultural capital in the GPN in historical perspective and mediates it with other crucial agencies (foreign capital, the state and the workers) to draw an historical picture of the LLCRs. Despite the absence of significant collective action by workers since Bud’s operations, workers’ agency emerged strongly from interviews with employers, managers, state officials and local growers, and from participant observation of their activities: the lack of substantial, direct interviews with workers represents itself an under-representation of their agency. To further substantiate key findings, fieldwork data is combined with simultaneous and later studies emphasising some core dimensions observed in the late 2000s, notably, a marked feminisation of the workforce, the continuous relevance of smallholders, and an extremely segmented, differentiated labour market (Maertens and Swinnen, 2012; RDS, 2012, 2013; Swinnen et al., 2013; Oya, 2015). 4. Labour control in the 1970s: from estate farming to contract farming To grasp the constitutive and path-dependant nature of labour control, the empirical analysis starts by examining the LLCR that sustained the early development of the sector in the 1970s. As emphasised above, this was a result of a combination of pressures emanating at global, national and local scales. Based on the direct control of workers on large estates (estate farming), the 1970s LLCR emerged when a subsidiary of a US company established itself in the Niayes to supply the European off-season FFV market. ‘Bud Senegal’ operated for almost a decade supplying exclusively its parent-marketing agency in Europe, House of Bud Société Anonyme (HOBSA). Within a few years the export of FFV took off but then petered out in the same dramatic way when the company went bankrupt in 1979 to the benefit of HOBSA (see below). The scheme was the first effort to organise export horticulture on estate farms with direct employment of wage labour on a large scale. At its peak a total area of 600–650 ha were cultivated, employing a daily average of 1600 workers (Mackintosh, 1989). Although short-lived, Bud’s deep transformation of the countryside represented an extremely path-dependent instance of labour control. The ‘problems’ created by an emerging rural ‘proletariat’ cast an enduring scepticism over the opportunity of employing a large number of workers ‘under the same roof’. As this section explains however, the organisation of production through estate farms and large-scale direct employment fulfilled the interests of different players: Bud managers, the Senegalese government and international donors. All three were influential in shaping the form of the labour process and the prevailing LLCR. 4.1. Global dynamics and chain governance The development of estate farming in Senegal by Bud was a typical expression of the industrialisation of agriculture emanating from USA, marked by the concentration of agro-industrial capital and the ascendancy of transnational corporations integrating farming activities horizontally, vertically and globally (Friedmann and McMichael, 1989; Heffernan, 2000). Bud Antle was a leader in Californian horticulture, a maverick in technological innovation, and a champion of anti-unionism (Friedland et al., 1981; Fredrichs, 1984). In the 1960s, the firm invested in Europe where it set up a marketing agency, HOBSA. The demand for FFVs was growing significantly and triggered new opportunities for traders, brokers and wholesalers (Mackintosh, 1977). This incentivised the sourcing and relocation of FFV production in the South by large-scale European brokers and/or giant processors like Nestle, Findus, etc. By 1980, there were around 140 investments in African countries by leading US and European food processing firms (Dinham and Hines, 1983). Among them Bud was a pioneer: an offshoot of Bud Antle, the emerging HOBSA–Bud Senegal supply chain linked Senegalese land and workers to European consumers through the supply of off-season FFVs. Like Bud, many of these investments were branded as ‘development’ because, from the 1960s, the internationalisation strategies of emerging giant corporations found support in key institutions like the World Bank and Food and Agricultural Organisation (George, 1976). The emergent ‘development discourse’ was constructed on ideas of food security, modernisation, private entrepreneurship and export orientation to gain foreign exchange. In the late 1960s, the effects of the Sahelian drought had widely captured international interest, and projects directed by agribusiness firms for savannah lands were promptly supported by aid agencies and development funds. HOBSA managers were successful in exploiting this conjuncture and funded Bud Senegal with the contributions from the World Bank, European donors and the Senegalese government (Mackintosh, 1989). The choice of large-scale production and wage-employment was shared by Bud’s management and by donors. It reflected the link between foreign firms and plantations, that in Africa stretched back to colonialism, where the high costs of plantations were offset by low running costs and the negligible land and labour costs (Dinham and Hines, 1983). After independence and the nationalisation of key cash crop economies, the link between foreign firms and plantations gradually developed in the direction of higher value crops, including cut flowers and FFVs. The economic arguments were the ‘technical superiority’ of both scientific production techniques and scientific management of workers (Gibbon, 2011). The use of large-scale irrigation, mechanisation and a stronger supervision of workers were deemed necessary to achieve a greater standardisation of produce as required by European markets. The direct control over production also replicated Bud Antle’s history and operations in USA: Bud Senegal represented the African counterpart of a highly mechanised, technologically advanced Californian horticulture achieved through large-scale farming. Fundamentally, however, Bud opted for estate farming because donors and the Senegalese government were willing to assume other major set-up costs (land and water). Once these were paid for, Bud could assume the cost of directly managing hundreds of workers. 4.2. The state Confronted with persistent drought and the crisis of the groundnut sector as a central source of revenue, for the Senegalese government Bud’s ‘modern’ farming option was alluring. It matched the state’s ambitious plan for agricultural diversification and broad reorientation of its development strategy away from rain-fed agriculture and in favour of modernisation, mechanisation and large irrigation schemes. Large-scale tracts of land in the Cap-Vert Niayes were handed over to Bud as virtually free, renewable land leases (Franke and Chasin, 1980). Other important concessions included cheap water, import-export tax exemptions and crucially the allocation of the marketing monopoly to HOBSA. As noted earlier, the emerging supply chain was an ‘artefact’ based on state action (Neilson et al., 2014, 2). More fundamentally however, and beyond the expectation of increased foreign exchange, state backing for estate farming expressed a wider ambition to intensify control mechanisms in the countryside, hitherto based on trade monopoly (through marketing boards). The crisis of the groundnut economy reflected the crisis of state resource extraction mechanisms, hitherto centred on trade monopoly and patronage relationships between producers and rural notables for the political containment of the rural population (Diop, 1992). In a countryside characterised by weakened state control of peasants, direct employment by Bud on irrigated estates provided an opportunity to reinforce the market dependence of rural populations, not as sellers of crops, but as sellers of labour. 8 Thus, estate farming represented the opportunity to establish ‘a very sharp increase in intervention, by government and private capital, to manage and transform the labour process in agriculture’ (Mackintosh, 1989, 8), i.e. to curtail relative autonomy within farming. Estate farming provided the state with an LLCR functional to agricultural modernisation and labour commodification, boosting both productivity and labour control. As argued below, the contradictory nature of this ambition became evident within the labour process, precisely because of the increased commodification and spatial aggregation of a large labour force that estate farming entailed. 4.3. The labour process How was labour control organised at the firm level? First, the higher costs and risks of directly employing labour were mitigated by a strong regime of labour exploitation and disciplining: at Bud the labour process was shaped by the adoption of mechanisation, labour informality and segmentation, especially along gender lines. Second, as these strategies failed to contain labour organisation, Bud gradually shifted to subcontracting, i.e. splintering the labour force over a multitude of small farms. In short, when the social disciplining of workers proved insufficient, management resorted to spatial discipline across multiple, spaced out sites. While mechanisation and especially widespread informal labour were pervasive from the outset, the main mechanism of disciplining at the firm level was a regime of highly gendered labour segmentation. Bud’s labour control strategies reproduced wider inequalities within Senegalese society, especially related to gender and migrants, and reflected management’s ability to capitalise on local household reproduction dynamics heavily based on female domestic work (Mackintosh, 1989). Within three years of activity the largest estate had created the category of ‘women’s work’ as a generally lower status, lower paid type of job. Managers portrayed women as an ‘unskilled and archaic’ labour mass. The discursive representation of the ‘problem of the women’ as an undisciplined crowd legitimised and reinforced informal and casual employment. In turn, informality offered women the possibility to fulfil household duties, combine and solve the conflict between wage and domestic work. In other words, the ability to push the risks and costs of insecure employment onto women depended on their social disciplining: household inequality was extended to the firm and legitimised through discursive representations. Thus, Bud’s labour regime reflected the broad ‘gendering’ of local labour markets on the basis of stereotyped characteristics ascribed to women and their subordinated role within households (Elson and Pearson, 1981; Elson, 1999). This was reinforced by presenting part-time and flexible work conditions as opportunities rather than constraints, options matching women’s need to juggle work and the household. This obscured and reinforced disproportionate female engagement in the sphere of reproduction and constrained their access to the productive economy. Hence, from the perspective of the LLCR the disciplining of women across the household-firm continuum sustained the employment of hundreds of workers, i.e. the informal, casual incorporation of women stood in relation to the sustaining of an enterprise based on the direct employment and control of workers on a large scale. Despite Bud’s reproduction of gender subordination, and the wider combination of labour disciplining and exploitation strategies, the company still faced a mass of workers whose direct employment offered the opportunity to resist and dispute different labour control mechanisms. In general, walkouts, stoppages and strikes occurred throughout the 1970s. Workers protested piece-rate employment, payment below the minimum wage and the unpredictability of employment. Some strikes required military intervention. On a small estate, female workers independently and collectively protested against the employment of non-local women. The tension culminated in September 1979, as 3000 Senegalese workers demonstrated in Dakar against the decision to close the company. As argued below, however, rather than leading to success, the growing organisation of workers foreshadowed ‘the continued structural domination of capital’ (Coe, 2013, 273) prompting the shift to an LLCR based on contract farming: workers’ agency led to more effective labour control. By 1979, the state had taken over Bud, which had registered financial losses each year. The government continued subsidising with no financial return as surpluses moved upstream to HOBSA. Estate farming had proven economically and politically expensive, especially for the government which as both the ‘employer’ and the ‘state’ was confronted with securing accumulation and development simultaneously (Cowen and Shenton, 1995). But how could the state reproduce capital accumulation without producing a class of conscious and organised workers? The answer to this ‘labour question’ laid essentially in the retrieval of the ‘peasantry’, i.e. a development model based on family farming as the way to modernity. This pivoted on local growers perceived as ‘more politically acceptable, less likely to create labour problems, and better in promoting individual work incentives’ (Mackintosh, 1989, 16). As one civil servant claimed at the time: ‘Our aim is definitely not to create waged agricultural worker, but to organize (encadrer) the peasantry’ (Mackintosh, 1989, 187). Within the official discourse therefore, contract farming came to replace estate farming as a more family-friendly inclusion in the GPN. In reality, contracting with local growers offered opportunities to transfer significant costs to small producers, sidestep minimum wage regulations and thwart mounting collective labour action on estates. The state’s recourse to the ‘peasant option’ offered the chance to discursively and spatially separate wage-workers whilst entangling them in the household. Through subcontracting, the labour process was displaced and transferred to smallholders employing family and migrant labour and enjoying the comparative advantage of avoiding ‘labour shirking’ (Rassas, 1988). In the 1980s, the government contracted with around 650 families, allocating to each small plots of land (Sall, 1993). Directly counterpoised to the estate, the family was presented as a homogenous, equitable and ‘natural’ instance of development, thus obscuring the unequal social and economic relations structuring rural households as PCPs. Overall, the shift to contract farming evidenced a broad strategy where spatial disciplining (labour displacement and dispersal within small farms) served to reinforce social disciplining (stronger to the extent that the farm approximates the household). The stability of the ensuing LLCRs based on contract farming and the incorporation of small-scale farms that continued throughout the 1980s and 1990s was broad evidence of how ‘space itself is deployed as a weapon in the contestation between capital and labour’ (Kelly, 2002, 397). Following Bud, the spatial fragmentation of labour within small family-farms continued. Throughout the 1980s, contract farming was the dominant form of upstream production involving thousands of Niayes family farmers and a dozen local entrepreneurs attracted by Bud operations (Horton, 1987). The strict vertical coordination between Bud Senegal and HOBSA, became a looser supply chain anchored to spot-market purchases rather than long-term supply agreements. Things changed, however, when European supermarkets gradually became the driving agents in the chain. Within an unfolding supermarket revolution (see below), European supermarket demands began to reshape the LLRC once more by encouraging a return to estate farming alongside small subcontractors. 5. Labour control in the 2000s: combining estate and contract farming After two decades of contract framing, the steady reappearance of estate farming was evidence of an emerging new LLCR governing Niayes export horticulture. From the 1990s, some domestic exporters began to invest in large-scale farming to comply with stricter requirements embodied in private European standards. In 2007–2008, there were roughly 30 active export firms. The majority were exporters–producers, i.e. firms internalising production, and a (shrinking) minority of sole exporters relying on contract farming. Exports were concentrated in relatively few firms: the three largest export firms accounted for two-thirds of all exports. The chain was broadly bifurcated: on the one hand the largest exporters–producers supplied European supermarkets through private certifications and marginally wholesale markets, and on the other smaller exporters exclusively supplied wholesale markets or European importers. In principle, direct production on estate farms supplied supermarkets while most contracted local growers sourced other European markets where food standards were looser. Like Bud before them, exporters–producers in the 2000s adopted articulated labour control strategies, combining labour exploitation and disciplining and profiting from the state’s simultaneous interest in boosting capital-intensive, large-scale farming without displacing small-scale farmers. Reproducing many of the 1970s labour-control practices, exporters–producers engaged in broad labour disciplining strategies to maximise exploitation and counterbalance the high costs and risks of upstream vertical integration. However, whereas in the 1970s the shift to contract farming was subsequent to the investment made to set up estates and labour collective action, from the 2000s contracting with local growers was largely previous and simultaneous to large-scale farming, i.e. a cost-saving platform supporting estates (Baglioni, 2015). The emerging LLCR therefore combined both estate and contract farming. As earlier, the pressures driving this combination originated from different levels, inside and outside the network. 5.1. Global dynamics and chain governance The re-emergence of estate farming in Senegal stemmed from the broader ‘supermarket revolution’ that had dramatically restructured Western retail markets (Humphrey, 2007). Displacing the traditional role of wholesale markets, by the 1990s European supermarkets had steadily gained market share becoming major marketing outlets for domestic and foreign suppliers. Moreover, fierce competition had prompted the adoption of new technologies and long-distance sourcing strategies. Overall, the simultaneous concentration and competition among retailers entailed a broad restructuring of African–European horticultural GPNs. In a classic study, Dolan and Humphrey (2000) show how the growing market power of supermarkets is translated into buyer-driven horticultural commodity chains characterised by intensified vertical coordination and concentration at strategic nodes, as well as increased competitive pressure on producers upstream. The chain ‘buyer-driveness’ was further intensified by supermarket competition for the supply of safe food following the food-scares from the 1990s onwards. Emanating from the escalating industrialisation of agriculture worldwide, these food crises prompted the proliferation of European public food safety and quality standards (Freidberg, 2004). Public standards were rapidly incorporated and widened by supermarkets competing to provide the best food quality and safety while protecting the environment and workers’ livelihoods (Henson and Reardon, 2005; Gibbon et al., 2010). Suppliers wishing to sell to European supermarkets were confronted with complex standard schemes and certifications such as EurepGap (now GlobalGap) that were far more demanding than public ones. Whereas in the 1970s TNCs and donor agencies branded plantations as ‘development’, in the 2000s the new mantra was ‘sustainable supply chain management’ (Müller et al., 2009) and the absorption of the developmental function within standard themselves. Supply chains represented the new development frontier, especially for African smallholders who could rely on private players engaged in creating a dynamic and prosperous agricultural sector (Amanor, 2009). The assimilation of ‘development’ by the market reflected the broad shift from state-led to market-led development characterising neoliberal globalisation. Poor working/workers’ conditions in the global South, exposed by Northern governmental and non-governmental institutions, spurred ethical initiatives targeting GPNs as the ideal terrain to improve workers lives (Hughes, 2007). However, as the labour process upstream came to be organised according to these new forms of governmentality, the burden of ‘decent production’ fell on producers. Producers, rather than supermarkets, were required to assume the costs of compliance with standards. This left broader power relations untouched: by portraying ethics as a problem of sourcing, supermarkets were able to ‘distance themselves from their own role in creating an environment for exploitation’ (Du Toit, 2001, 3).9 These broad global changes reflected downstream in contradictory pressures shaping the new Senegalese LLCR. On the one hand, supermarkets required a stricter control of labour. To obtain a private certification such as EurepGap or Tesco’s Nature Choice, producers had to adhere to rules covering all areas of production, crop handling and regulations over worker safety and health. According to all exporters, these rules could not be complied with through contract farming, where their control over farming was limited. Hence, the need to engage in direct production to control directly the labour process. On the other hand, direct control over farming and labour greatly increased the production costs, requiring a strong regime of labour exploitation and disciplining, for which contract farming represented an important tool. Overall, weak supermarket monitoring systems increased LLCR room for manoeuvre, allowing combined estate and contract farming to cut costs and prevent widespread labour organisation. 5.2. The state As in the 1970s, state agency was a pillar of the new, tighter GPN and corresponding LLCR. Mirroring the support to Bud operations, exporters in the 2000s relied on considerable state backing, from the central government and/or local officials heading rural communities. This was mostly in the form of relentless land accumulation, necessary for estate farming. Exporters could access land through highly favourable, state-granted, long-term land concessions and/or gradually amassing small-scale (informal) land acquisitions through the intercession of state officials within rural communities. For the government, large-scale horticultural farms represented the new ‘modern’ agriculture, associated with inclusion in profitable supply chains, agribusiness and mechanised, capital-intensive farming. The sector figured as a core pillar of Senegalese agricultural development strategies in the 2000s (Fall, 2005), and was supported by the World Bank. 10 The state however, while promoting agribusiness, also continued to pursue the long-standing valorisation of small-scale family farming (Senghor, 2006). This led to various public projects aimed at strengthening smallholder inclusion within supply chains,11 possibly reinforcing the bifurcated nature of the network where large- and small-scale farming continue in parallel. Despite its incoherence (Anseeuw, 2010; Oya and Ba, 2013), state participation in the agricultural sector has long reflected the ambitious need to keep a foot in both camps: that is, to pursue modernisation via large-scale commercial agriculture and small-scale family farming (Oya, 2012). Discourses on the intrinsic value of ‘peasants’ as champions of supposedly family-friendly and human-scale farming endure. These counter-pose modern estates with industrious and diligent peasants that make the most out of very skewed resources (but systematically mask higher levels of labour exploitation and disciplining, including gendered allocation of resources and duties). In 2007, following a conversation on Bud strikes, an exporter and state official claimed: The smallholder is more productive because he is under risk. The more he produces the more he gains. Also he is more responsible because he works his own land. Wage labour is not good for agriculture…, because the worker gets the same pay whether he produces or not, whether he produces a lot or a little. Also with wage labour you have the problem of unionisation, which means less work. While the prevailing ideological construction of industrious peasants reveals the states’ discursive power in shaping the LLCR (Coe and Kelly, 2002), the preservation of small producers confirms the state’s efforts to maximise the interplay between labour exploitation and disciplining, structural to PCP. Here, the spatial fragmentation of workers, their closer surveillance and greater exploitation under the form of household duties hinders collective labour action, and represents the successful management of the ‘labour question’ by the state. Indeed, simultaneous support for large-, and small-scale farming demonstrates the inherent path-dependency of class struggle. The collective action of Bud estate workers leading to the 1970s strikes had cast a long shadow over the sector. State officials and exporters invoked those episodes to demonstrate the ‘side effects’ of estate farming. For them, subcontracting was a means to avoid Bud ‘mistakes’. 5.3. The labour process In the late 2000s, export firms were extremely diverse: the largest being around 1400 ha while smaller firms covered between 10 and 100 ha. In general, the higher the degree of vertical chain coordination, the more complex the division of labour within the firm, reflecting the multiplication and diversification of activities that compliance with European requirements entailed. As in the 1970s, the cost of employing wage labour was compensated for in different ways. The lessons learned from Bud ‘mistakes’ resonated in a broad concern with curbing labour costs and preventing collective action at the heart of the new LLCR. Mechanisation and informal employment still served to mitigate the cost of maintaining a large labour force. But due to the availability of cheap labour, especially women for harvesting, picking and packaging, the mechanisation of activities was patchy and extremely varied. Firms combined diverse forms of employment: permanent staff for management, seasonal staff for some farming and transport activities, and informal, daily and piece-work employment for the majority. In the largest firms the number of permanent employees did not exceed 100 while daily seasonal workers could peak at over 2000–2500.12 No manager could be precise about the number of workers employed within his/her operations.13 Mirroring Bud, informal employment was disproportionally female. A quantitative study of the sector estimated that women represented 90% of the workforce in mango/bean firms and 60% in tomato firms (Maertens and Swinnen, 2012). The gender division of labour was still fairly rigid: despite one woman owning and actively managing an export-producing firm, and a few secretaries, females dominated in unskilled, informal and poorly paid occupations. While picking, sorting and packaging remained essentially ‘women’s jobs’, men were more likely employed in management, cultivation and transport. According to managers, women received lower wages than men and were more often paid on a piece-rate. As before, gender segmentation was buttressed by the pervasive disciplining of women according to supposedly natural attributes. The construction of narratives describing women’s higher performance in tedious, and less skilled tasks was ubiquitous in managers’ accounts. Women’s ‘nimble fingers’ and docility emerged repeatedly in managers’ attempts to highlight what women were best at, e.g. sorting French beans or idly waiting for hours for the next truck to arrive. Ironically, this contradicted the managers’ enduring obsession with ‘the problem of the women’ as a ‘wild’ mass, unaccustomed or unsuited for skilled or continuous work therefore requiring special and flexible arrangements. These were invariably invoked to accommodate women’s needs to combine wage and domestic work. In 2008, a supervisor of one of the largest companies described female pickers as follows: They are happy to work by the piece. In this way they can fill a few buckets enough for their d.q. [dépense quotidienne] and go home. Women living further away do more buckets. The system just works perfectly for them. While the ability of some women to turn the terms of their employment to their advantage emphasised instances of micro-agency (Carswell and De Neve, 2013), the latter were generally anchored to compliance with household duties, underlining their structural subordination within the household noted earlier. Building on this micro-agency, management legitimised insecure female employment by picturing it as a natural continuum of household activities. This depiction however became complicated by the increasing competition for women that refined exporters’ recruitment strategies. The growing need for women at peak times prompted larger employers to collect women from more distant villages thereby widening the labour frontier to the outskirts of the Niayes. Exporters would frequently claim to be the best and ‘most caring’ employers, emphasising the need to strengthen the extra economic nature of employment. Alongside the social disciplining of women workers as housewives, the high costs of large-scale employment and standards compliance were crucially sustained through contract farming, i.e. through the spatial disciplining of all workers. Contracting offered exporters the possibility to push labour control to local growers, partially fragment the labour force within smaller units of production, and critically, further divide rural workers between those ‘included’ and those ‘excluded’ by standard compliance. While this distinction has been broadly observed in the literature (Evers et al., 2014; Goger et al., 2014), the structural relation between the two has received less attention (Bair and Werner, 2015). Despite the re-emergence of large estates, and the important role of large local growers, small growers were still largely included, although their involvement was generally unrecorded and unpredictable, often on the ‘spot’. Smallholders were both the sole source of supply for small exporters and a critical source of hedging for larger exporter–producers. In other words, and echoing the strong pressures constituting PCP, smallholders occupied a position in the chain that simultaneously created and restrained their accumulation (Gibbon and Neocosmos, 1995; Bernstein, 2003). Their competitive advantage vis-à-vis large farms consisted essentially in lower labour costs and income/market diversification strategies to make up for uneven and meagre accumulation (interviews; Maertens and Swinnen, 2009, 2012). Lower labour costs stemmed from cheap, seasonal, immigrant labour and a variegated and shifting pool of relatives, including women contributing to farming, trade and household reproductive duties (RDS, 2012). Women’s greater exploitation within small farms is evidenced by a large survey of the sector that highlighting how women gained more from employment on large farms than from their family contracting activities (Maertens and Swinnen, 2012). Furthermore, given the concentration of export production in just a few months, smallholders would produce on contract while growing other crops for local markets and—where possible—engage in a diversified portfolio of income-generating activities functional to sustaining subcontracting, where accumulation was insufficient and irregular. Therefore, subcontracting to non-compliant farms was possible because of the ability of small farmers to capture and exploit untapped sources of cheap labour and to engage in other sources of reproduction/accumulation. In short, the incorporation of some workers on certified farms was sustainable due to the labour of other cheap workers on non-compliant farms, and the unpaid work of family and female labour. Conversely, the ability of exporters to enter and sustain chain inclusion rested on the ability of small farmers to ‘get by’ with a more adverse incorporation (Phillips, 2011) in the network. Overall, the combination of estate and contract farming constituted a crucial element of the new LLCR. This combination, by articulating different instances of social and spatial disciplining, served to sustain the economic costs of a tighter, supermarket-driven network, based on the mutual upstream relations between different segments of capital (large and small) and labour (permanent/seasonal workers on certified farms, daily male and female workers, hidden workers on small farms, unpaid family labour including the reproductive work of women). The relations between large- and small-scale firms, encapsulating the ‘modern’ and the ‘African’ ways, respectively, also reflected the state/donors’ development agendas centred on coalescing firms and households through the constant branding of the industrious ‘African smallholder’. Ultimately, the precarious and uneven outsourcing to small growers highlighted once again the structural link between firms and households: the production/reproduction of an indistinct, uneven and hidden cheap labour frontier serving the broad sustenance of the network. As this article has highlighted, the dialectical and path-dependant nature of the LLCR has entailed a deepening of labour control in the Niayes throughout the 2000s. This was largely due to the preservation of households as the engine of both reproduction and production activities. Is this still the case today? And what are the consequences for labour? Whilst available evidence emphasises the ‘significant’ contribution of smallholders to FFV exports (RDS, 2013), the relations between estate and contract farming remains largely unexplored. Thus, the degree to which exporters rely on subcontracting to cut costs and comply with supermarket standards remains hard to measure, due to both the lack of further research and the ‘underground’ nature of much contract farming. Meanwhile, the prolonged growth of export horticulture has increased pressures on land and labour driving increasing investments north of the Niayes, especially in the Senegal river region (RDS, 2013; Van den Broeck et al., 2016). Despite being extremely dynamic and diversified, the Niayes labour markets remain largely unexamined (Oya, 2015). While it may be too early to observe radical changes in the Cap-Vert Niayes LLCR, the prolonged growth of export horticulture is likely to deepen the socio-economic differentiation among producers and workers alike, constantly changing the terms and forms of inclusion within the GPN. In particular, the growing demand for female workers at harvest may strengthen women’s bargaining power vis-à-vis employers. Given the strong unequal gender relations that still characterise this area (RDS, 2012), and based on the relational nature of worker agency illustrated in this article, perhaps it is more likely that the expansion in the female labour market will lead to further internal differentiation, improving the terms of inclusion of some at the expense of others. 5. Conclusions The historical development of the Senegalese–European horticultural production network shows the centrality of labour control at the point of production for a variegated range of actors. A comparison of the labour control regimes analysed in this article is included in Table 1. Large European firms driving supply chains, development agencies, the Senegalese state, local firms, households and workers all shape the complex architecture of LLCRs. These operate by exploiting and disciplining workers in different ways: labour-saving mechanisation, employment informality, gender segmentation and subordination, as well as the spatial separation of workers across firms and households. The interplay between labour exploitation and disciplining contributes to labour regime analysis showing how labour control operates across a broader space than the workplace, in particular by structuring relationships among firms, households, and genders. Thus, the case of Senegal illustrates how places of production cannot function without broader spaces of labour control, while at the same time the constitutive and scalar nature of labour control challenges network-essentialist literature. Table 1 Niayes LLCRs in 1970s and 2000s Scale  Actors  1970s LLCR based on estate farming  2000s LLCR based on estate and contract farming  Global scale  TNCs, Development agencies and donors  Bud Antle vertical integration (direct labour control) Large-scale, mechanised farming in Senegal as the fastest and most efficient way to supply European markets Development discourse based on ‘modern agriculture’ as large-scale, mechanised, farming and ‘technical superiority’ of ‘scientific production’ and ‘scientific management of workers’ Agency: financial support to set up Bud Senegal  Supermarkets vertical coordination (indirect labour control) Standards’ compliance drives direct production while low standard enforcement and diversified GPNs enable contract farming Development discourse based on supply chain inclusion and absorption of the development function within standards Agency: technical and infrastructural support for both large- and small-scale farms  National scale  State  Agricultural diversification and foreign exchange Increased control of the countryside by pushing small-scale farmers towards wage work  Agricultural diversification and foreign exchange Continuous control of the countryside by avoiding the constitution of large-scale workforces; discourse on ‘peasants’ as the backbone of the country      Agency: financial, infrastructural and political support for Bud Senegal  Agency: infrastructural and technical support to large-scale farming and technical support to small-scale farmers  Local scale  Labour process  Large capital agency: Labour exploitation via large-scale wage employment, mechanisation, informality and segmentation Social disciplining via gendered segmentation of tasks, jobs and labour markets: creation of an informal and casual large-scale female labour force Spatial disciplining via a shift to contract farming following collective labour action  Large capital agency: Local exporters combining estate farming (to comply with supermarket standards) and contract farming (to finance compliance) Labour exploitation via large-scale wage employment, mechanisation, labour informality and segmentation Social disciplining via gendered segmentation of tasks, jobs and labour markets: sustainment and expansion of an informal, large-scale female labour force Spatial disciplining via combining estate and contract farming to prevent collective labour action      Small capital agency: Small-scale farmers combining contract farming, different markets and economic activities Labour exploitation via cheap and unpaid (family) labour, labour informality Social and spatial disciplining via firms as households    Labour agency: Collective labour actions, strikes walk-outs Women combining housework and wage-work  Labour agency: Women combining housework and wage-work Absence of manifest collective action  Scale  Actors  1970s LLCR based on estate farming  2000s LLCR based on estate and contract farming  Global scale  TNCs, Development agencies and donors  Bud Antle vertical integration (direct labour control) Large-scale, mechanised farming in Senegal as the fastest and most efficient way to supply European markets Development discourse based on ‘modern agriculture’ as large-scale, mechanised, farming and ‘technical superiority’ of ‘scientific production’ and ‘scientific management of workers’ Agency: financial support to set up Bud Senegal  Supermarkets vertical coordination (indirect labour control) Standards’ compliance drives direct production while low standard enforcement and diversified GPNs enable contract farming Development discourse based on supply chain inclusion and absorption of the development function within standards Agency: technical and infrastructural support for both large- and small-scale farms  National scale  State  Agricultural diversification and foreign exchange Increased control of the countryside by pushing small-scale farmers towards wage work  Agricultural diversification and foreign exchange Continuous control of the countryside by avoiding the constitution of large-scale workforces; discourse on ‘peasants’ as the backbone of the country      Agency: financial, infrastructural and political support for Bud Senegal  Agency: infrastructural and technical support to large-scale farming and technical support to small-scale farmers  Local scale  Labour process  Large capital agency: Labour exploitation via large-scale wage employment, mechanisation, informality and segmentation Social disciplining via gendered segmentation of tasks, jobs and labour markets: creation of an informal and casual large-scale female labour force Spatial disciplining via a shift to contract farming following collective labour action  Large capital agency: Local exporters combining estate farming (to comply with supermarket standards) and contract farming (to finance compliance) Labour exploitation via large-scale wage employment, mechanisation, labour informality and segmentation Social disciplining via gendered segmentation of tasks, jobs and labour markets: sustainment and expansion of an informal, large-scale female labour force Spatial disciplining via combining estate and contract farming to prevent collective labour action      Small capital agency: Small-scale farmers combining contract farming, different markets and economic activities Labour exploitation via cheap and unpaid (family) labour, labour informality Social and spatial disciplining via firms as households    Labour agency: Collective labour actions, strikes walk-outs Women combining housework and wage-work  Labour agency: Women combining housework and wage-work Absence of manifest collective action  Strong state intervention recasts the role of development within the proliferation of GPNs in the global South. Ultimately, this intervenes between the fundamental requirement to control labour to produce commodities and the consequent management of the labour question, two fundamental questions within capitalism. Often, in the African countryside this has led to dual rural development policies: on the one hand buttressing highly capitalised, modern, large-scale farms, while on the other supporting smallholders as mechanisms of inclusionary development alien to class struggle. The material and discursive support for small-scale production blurs social conflict and transfers labour control to within the boundaries of the household. As inclusion within GPNs furthers the commodification of labour, and the risk of labour organisation, the protracted support for ‘peasants’ becomes a powerful answer to the labour question. From this perspective, rather than simply a ‘natural’ development of flexible, just-in-time sourcing of supermarkets, the continuous proliferation of subcontracting and the uncharted reliance on smallholders in the global South reveal heavy footprints of the state and development policies. The constitutive role of households also shows how labour control and the labour question revolve around gender relations; how the bond between households and firms features a far more structural interdependence between the productive and the reproductive economy; and how agency in one sphere is linked to agency/control in the other. In Senegal, the subordination of women across firms and households has unleashed abundant cheap labour to sustain upstream production for four decades. Thus, while the labour control perspective necessarily overlooks internal differentiation, agency and forms of female ‘empowerment’ (Deere, 2005; Selwyn, 2010), in Senegal it highlights the central contention that ‘gender hierarchies have not been erased from markets’ (Razavi, 2009, 211). This prompts a crucial question for a nuanced understanding of workers’ agency everywhere: the ability of some workers to reap the benefit of European supermarket standards compliance can hardly be disjoined from a broad feminisation of the workforce whose informality is internalised as a universal opportunity. Indeed, the case of Senegal situates worker agency in a very complex network of path-dependant, contingent and embedded relations among different segments of capital, differentiated and segmented labour forces, and the state. Shifting regimes of labour control emphasise the relationship between capital and labour ‘as bound in a dialectical totality’ (Cumbers et al., 2010, 67) where the increasing control of labour can fuel organised labour resistance, in turn thwarted by deepening labour control regimes. In Senegal, the latter somehow deepen the relations between firms and households structuring the new supermarket-driven production networks. So far, the agencies of different sections of capital, donors, the state and workers have led to ‘the structural domination of capital’ in the Niayes. How workers react, actively shape and challenge this domination remains a crucial question. Acknowledgements I would like to thank Liam Campling, Gerard Hanlon, Jonathan Pattenden, and Adrian Smith for the precious support and invaluable comments offered. A very early version of this article was presented at the Symposium on Labour and Large-Scale Agriculture in Post-Independent Africa and South-East Asia (organised by DIIS, CIRAD and PLAAS) held in Cape Town in September 2012 and benefited from the helpful comments of Benoit Daviron, Peter Gibbon, Bruno Losch, and other participants. I am also very grateful to the editor of the Journal of Economic Geography and the anonymous peer reviewers, who have provided constructive feedback on earlier versions of this article. Footnotes 1 See Pattenden (2016) for a different framing of the LLCR in agrarian formations. 2 In 2013, Senegal’s French bean exports to the EU ranked behind Morocco, Egypt, Kenya and Rwanda and behind Morocco, Egypt, Ethiopia, South Africa and Tunisia for tomatoes (FOASTAT). Senegal is the largest Western African exporter of tomatoes to the EU (République du Sénégal, 2013). 3 In line with Jonas (1996), the ‘local’ is defined here at the subregional scale. Other export horticulture areas, notably the Senegal river valley bordering with Mauritania and Casamance in the South, are characterised by rather different political economies and histories and most likely present different, more or less structured, LLCRs. 4 The Niayes accounts for 60% of Senegalese FFV production and 50% of FFV exports (RDS, 2013); export production has been growing rapidly in the Senegal river region since the mid-2000s. 5 FFV exports peaked in 1975–1976 to around 10,000 tons, of which Bud exported more than 8000 tonnes (Berniard et al., 1976) and then again from the mid-2000s when seasonal averages oscillated between 15,000–20,000 tons (Data provided by the Organisation Nationale des Producteurs Exportateurs de Fruits et Legumes—ONAPES, and FAOSTAT). 6 This was complemented with available secondary sources (a limited sample includes Amin, 1969; RDS, 1972, 1974a, 1974b; Berniard et al., 1976; Collins and Lappe, 1977; Batsch, 1980; Horton, 1987; Rassas, 1988). 7 See Mackintosh (1989, 24–41 and 201–204). 8 Faced with dropping groundnut prices, in the late 1960s Senegalese peasants were massively shifting to food production (Mbodji, 1992). 9 The effectiveness of ethical standards has also been questioned empirically (Barrientos and Smith, 2007). 10 See the Projet de Promotion des Exportation Agricoles (PPEA) supporting medium and large-scale horticultural producers through the provision of technical support and basic infrastructure (Fall, 2004). 11 For example, see the Programme de Développement des Marchés Agricoles du Sénégal (PDMAS): the EU Pesticide Initiative Programme (Jaud and Cadot, 2012); and the government’s initiatives like the Plan Reva, and the latest Projet de Développement Inclusif et Durable de l’Agrobusiness au Sénégal (PDIDAS) envisaging a ‘sustainable’ and ‘inclusive’ development of horticulture based on strong linkages between large and small producers. 12 Data collected in interviews. 13 Colen et al. (2012) confirm the large prevalence of informal employment. References Agarwal B. ( 1986) Women, poverty and agricultural work in India. Journal of Peasant Studies , 13: 165– 220. Google Scholar CrossRef Search ADS   Amanor K. ( 2009) Global food chains, African smallholders and World Bank governance. Journal of Agrarian Change , 9: 247– 262. Google Scholar CrossRef Search ADS   Amin S. ( 1969) Le Monde Des Affaires Sénégalais . Paris: Editions de Minuit. Anner M. ( 2015) Labour control regimes and workers resistance in global supply chains. Labour History , 56: 292– 307. Google Scholar CrossRef Search ADS   Anseeuw W. ( 2010) Agricultural policy in Africa—renewal or status quo? In Padayachee V. (ed.) The Political Economy of Africa . London: Routledge, pp. 247– 265. Baglioni E. ( 2015) Straddling contract and estate farming: accumulation strategies of Senegalese horticultural exporters. Journal of Agrarian Change , 15: 17– 42. Google Scholar CrossRef Search ADS   Bair J. ( 2008). Global commodity chains: genealogy and review. In Bair J. (ed.) Frontiers of Commodity Chains Research . Ithaca, NY: Cornell University Press, pp. 1– 34. Bair J., Werner M. ( 2015) Global production and uneven development. In Newsome K. et al.   (eds) Putting Labour in its Place . New York: Palgrave, pp. 119– 134. Google Scholar CrossRef Search ADS   Barrientos S., Kritzinger A. ( 2004) Squaring the circle: global production and the informalization of work in South Africa fruit export. Journal of International Development , 16: 81– 92. Google Scholar CrossRef Search ADS   Barrientos S., Smith S. ( 2007) Do workers benefit from ethical trade? Third World Quarterly , 28: 713– 729. Google Scholar CrossRef Search ADS   Batsch ( 1980) Les mésaventures d’une entreprise de maraîchage au Sénégal. Le Monde Diplomatique . Berniard S., Chatain E., Lavenant H. ( 1976) Le Sénégal en chiffres . Dakar: Société Africaine d’Edition. Bernstein H. ( 2003) Farewells to the peasantry. Transformation: Critical Perspectives on Southern Africa , 52: 1– 19. Google Scholar CrossRef Search ADS   Bernstein H. ( 2010) Class Dynamics of Agrarian Change . Halifax and Winnipeg: Fernwood Publishing. Braverman H. ( 1998) Labor and Monopoly Capital . New York: Monthly Review Press. Burawoy M. ( 1985) The Politics of Production . London: Verso. Burawoy M. ( 1992) Manufacturing Consent . Chicago: The University of Chicago Press. Carswell G., De Neve Geert ( 2013) Labouring for global markets. Geoforum , 44: 62– 70. Google Scholar CrossRef Search ADS   Castree N. ( 2007) Labour geography: a work in progress. International Journal of Urban and Regional Research , 31, 853– 862. Google Scholar CrossRef Search ADS   Castree N., Coe N. M., Ward K., Samers M. ( 2004) Spaces of Work . London: Sage. Coe N. M. ( 2013) Geographies of production III. Progress in Human Geography , 37: 271– 284. Google Scholar CrossRef Search ADS   Coe N. M., Dicken P., Hess M. ( 2008) Global production networks: realizing the potential. Journal of Economic Geography , 8: 271– 295. Google Scholar CrossRef Search ADS   Coe N. M., Dicken P., Hess M., Yeung H. ( 2010) Making connections: global production networks and world city networks. Global Networks , 10: 138– 149. Google Scholar CrossRef Search ADS   Coe N. M., Hess M. ( 2013) Global production networks, labour and development. Geoforum , 44: 4– 9. Google Scholar CrossRef Search ADS   Coe N. M., Jordhus-Lier D. C. ( 2011) Constrained agency? Re-evaluating the geographies of labour. Progress in Human Geography , 35: 211– 233. Google Scholar CrossRef Search ADS   Coe N, Kelly P. F. ( 2002) Languages of labour: representational strategies in Singapore’s labour control regime. Political Geography , 21: 341– 371. Google Scholar CrossRef Search ADS   Colen L., Maertens M., Swinnen J. ( 2012) Private standards, trade and poverty. The World Economy , 35: 1073– 1088. Google Scholar CrossRef Search ADS   Collins J., Lappe F. M. ( 1977) Multinational agribusiness and the global supermarket. Social Scientist , 6: 3– 12. Google Scholar CrossRef Search ADS   Cooper F. ( 1996) Decolonization and African Society . Cambridge: Cambridge University Press. Google Scholar CrossRef Search ADS   Cowen M., Shenton R. ( 1995) The invention of development. In Crush J. (ed.) Power of Development . London: Routledge, pp. 27– 43. Cramer C., Oya C., Sender J. ( 2008) Lifting the blinkers: a new view of power, diversity and poverty. Journal of Modern African Studies , 46: 361– 392. Google Scholar CrossRef Search ADS   Crehan K. ( 1992) Rural households. In Bernstein H., Crow B., Johnson H. (eds) Rural Livelihoods. Crises and Responses . Oxford: Oxford University Press, pp. 87– 112. Cumbers A., Helms G., Swanson K. ( 2010) Class, agency and resistance in the old industrial city. Antipode , 42: 46– 73. Google Scholar CrossRef Search ADS   Cumbers A., Nativel C., Routledge P. ( 2008) Labour agency and union positionalities in global production network. Journal of Economic Geography , 8: 369– 387. Google Scholar CrossRef Search ADS   Davies S., Hammer N., Willimas G., Raman R., Ruppert S. C., Volynets L. ( 2011) Labour standards and capacity in global subcontracting chains. Industrial Relations Journal , 42: 124– 138. Google Scholar CrossRef Search ADS   Deere C. D. ( 2005) The Feminisation of Agriculture? Economic Restructuring in Rural Latin America . UNRISD: Geneva. Dicken P., Malmberg A. ( 2001) Firms in territories. Economic Geography , 77: 345– 363. Google Scholar CrossRef Search ADS   Dinham B., Hines C. ( 1983) Agribusiness in Africa . London: Erath Resources Research Publication. Diop M. C. (ed.) ( 1992) Sénégal. Trajectoires d’un État . Dakar: CODESRIA. Dolan C. ( 2004) On farm and packhouse: employment at the bottom of a global commodity chain, Rural Sociology , 69: 99– 126. Google Scholar CrossRef Search ADS   Dolan C., Humphrey J. ( 2000) Governance and trade in fresh vegetables. Journal of Development Studies , 37: 147– 176. Google Scholar CrossRef Search ADS   Du Toit A. ( 2001) Ethical Trading, A Force of Improvement, or Corporate Whitewash?, ODI Natural Resource perspectives, N. 71, PLAAS Edwards P. ( 1986) Conflict at Work . Oxford: Blackwell. Elson D. ( 1999) Labour markets as gendered institutions. World Development , 27: 611– 627. Google Scholar CrossRef Search ADS   Elson D., Pearson R. ( 1981) Nimble fingers make cheap workers. Feminist Review , 7: 87– 107. Google Scholar CrossRef Search ADS   Evers B., Maggie O., Barrientos S., Krishnan A., Amoding F., Ndlovu L. ( 2014) Global and Regional Supermarkets . Manchester: Capturing the Gains Working Paper. Fall A. A. ( 2004) Evaluation De La Rentabilité Économique Et Financière Du Projet De Promotion Des Exportations Agricole (PPEA) . Sénégal: Projet de Promotion des Exportations Agricole, Mimistère de l’Agriculture, de l’Elevage et de l’Hidraulique. Fall A. A. ( 2005) Evaluation Economique Du Programme De Développement Des Marchés Agricoles Du Sénégal (Pdmas) . Sénégal: Mimistère de l’Agriculture et de l’Hidraulique. Franke R. W., Chasin B. H. ( 1980) Seeds of Famine . Montclair: Allanheld, Osmund & Co. Fredrichs A. ( 1984) Technological change and the growth of agribusiness: a case study of california lettuce production. In Berardi G. M., Geisler C. C. (eds) The Social Consequences and Challenges of New Agricultural Technologies . Boulder, CO: Westview Press. Freidberg S. ( 2004) French Beans and Food Scares . New York: Oxford University Press. Friedland W., Barton A. E., Thomas R. ( 1981) Manufacturing Green Gold . Cambridge: Cambridge University Press. Friedmann H., McMichael P. ( 1989) Agriculture and the state system. Sociologica Ruralis , XXIX: 93– 117. Google Scholar CrossRef Search ADS   George S. ( 1976) How the Other Half Dies . London: Penguin Books. Gibbon P. ( 2011) Experiences of Plantation and Large-Scale Farming in the 20th Century Africa. Danish Institute for International Studies (DIIS) Working Paper. Gibbon P., Lazaro E., Ponte S. ( 2010) Global Agro-Food Trade and Standards: Challenges for Africa . New York: Palgrave Macmillan. Google Scholar CrossRef Search ADS   Gibbon P., Neocosmos M. ( 1985) Some problems in the political economy of "African socialism". In Bernstein and H., Campbell B. K. (eds) Contradictions of Accumulation in Africa . Beverly Hills, CA: Sage, pp. 153– 206. Glassman J., Choi Y. J. ( 2014) the Chaebol and the US military-industrial complex. Environment and Planning A , 46: 1160– 1180. Google Scholar CrossRef Search ADS   Goger A., Hull A., Barrientos S., Gereffi G., Godfrey S. ( 2014) Capturing the Gains in Africa: Making the Most of Global Value Chain Participation . Durham, North Carolina (USA): Duke University, Centre on Globalization, Governance and Competitiveness. Hastings T., MacKinnon D. ( 2016) Re-imbedding agency at the workspace scale: workers and labour control in Glasgow call centres. Environment and Planning A , 49: 104– 120. Google Scholar CrossRef Search ADS   Heffernan W. D. ( 2000) Concentration of ownership and control in agriculture. In Magdoff F., Bellamy Foster J., Buttel F. (eds) Hungry for Profit . New York: Monthly Review Press. Henderson J., Dicken P., Hess M., Coe N., Wai-Chung Yeung H. ( 2002) Global production networks and the analysis of economic development. Review of International Political Economy , 9: 436– 464. Google Scholar CrossRef Search ADS   Henson S., Reardon T. ( 2005) Private agri-food standards: implications for food policy and the agri-food system. Food Policy , 30: 241– 253. Google Scholar CrossRef Search ADS   Horton J. ( 1987) Characteristics of the Horticultural Export Enterprises Utilizing Contract Farming Schemes in Senegal. IDA Working Paper No. 84, Clark University, Worcester, MA (USA). Hudson R. ( 2001) Producing Places . New York: The Guildford Press. Hughes A. ( 2007) Multi-stakeholder approaches to ethical trade. Journal of Economic Geography , 7: 421– 437. Google Scholar CrossRef Search ADS   Humphrey J. ( 2007) The Supermarket revolution in developing countries. Journal of Economic Geography , 7: 433– 450. Google Scholar CrossRef Search ADS   Jaud M., Cadot O. ( 2012) A second look at the pesticide initiative program: evidence from Senegal. World Trade Review , 11: 490– 506. Google Scholar CrossRef Search ADS   Jonas E. G. ( 1996) Local labour control regimes: uneven development and the social regulation of production. Regional Studies , 30: 323– 338. Google Scholar CrossRef Search ADS   Kautsky K. ( 1900) La Question Agraire . Paris: Giard & Brière. Kelly P. F., ( 2002) Spaces of labour control: comparative perspectives from Southeast Asia. Transactions of the Institute of British Geographers , 27: 395– 411. Google Scholar CrossRef Search ADS   Kelly P. F. ( 2009) From global production networks to global reproduction networks: households, migration, and regional development in Cavite, the Philippines. Regional Studies , 43: 449– 461. Google Scholar CrossRef Search ADS   Kelly P. F. ( 2013) Production networks, place and development: thinking through global production networks in Cavite, Philippines. Geoforum , 44: 82– 92. Google Scholar CrossRef Search ADS   Lier D. C. ( 2007) Places of Work, scales of organising: a review of labour geography. Geography Compass , 1: 814– 833. Google Scholar CrossRef Search ADS   Mackintosh M. ( 1977) Fruit and vegetables as an international commodity. Food Policy , 2: 277– 292. Google Scholar CrossRef Search ADS   Mackintosh M. ( 1989) Gender, Class and Rural Transition, Agribusiness and the Food Crisis in Senegal . London: Zed Books. Maertens M., Swinnen J. M. ( 2009) Trade, standards, and poverty: evidence from Senegal. World Development , 37: 161– 178. Google Scholar CrossRef Search ADS   Maertens M., Swinnen J. M. ( 2012) Gender and modern supply chains in developing countries. Journal of Development Studies , 48: 1412– 1430. Google Scholar CrossRef Search ADS   Marx K. ( 1990), Capital Volume I . London: Penguin. Mbodji M. ( 1992) L’Etat de l’économie arachidière: une crise trentenaire. In Diop M. C. (ed.) Sénégal Trajectoire d’un État . Dakar: CODESRIA, pp. 95– 135. McGrath-Champ P., Herod A., Rainnie A. ( 2010) Working Space . Cheltenham: Edward Elgar. Mezzadri A. ( 2016a) Class, gender and the sweatshop. Third World Quarterly , 37: 1877– 1900. Google Scholar CrossRef Search ADS   Mezzadri A. ( 2016b) The Sweatshop Regime . Cambridge: University Press. Google Scholar CrossRef Search ADS   Mies M. ( 2014) Patriarchy and Accumulation on a World Scale . London: Zed Books. Müller C., Vermeulen J. V., Glasbergen P. ( 2009) Perceptions on the demand side and realities on the supply side. Sustainable Development , 17: 295– 310. Google Scholar CrossRef Search ADS   Neethi P. ( 2012) Globalization lived locally: investigating Kerala’s local labour control regimes. Development and Change , 43: 1239– 1263. Google Scholar CrossRef Search ADS   Neilson J., Pritchard B., Yeung H. ( 2014) Global value chains and global production networks in the changing international political economy. Review of International political Economy , 21: 1– 8. Google Scholar CrossRef Search ADS   Newsome K., Taylor P., Bair J., Rainnie A. ( 2015) Putting Labour in Its Place . New York: Palgrave Macmillan, pp. 1– 26. Google Scholar CrossRef Search ADS   Nichols T. ( 1980) Marx on the labour process. In Nichols T. (ed.) Capital and Labour . Glasgow: Fontana Paperbacks. O’Laughlin B. ( 2007) A bigger piece of a very small pie: intrahousehold resource allocation and poverty reduction in Africa. Development and Change , 38: 21– 44. Google Scholar CrossRef Search ADS   Oya C. ( 2012) Contract farming in Sub-Saharan Africa: a survey of approaches, debates and issues. Journal of Agrarian Change , 12: 1– 33. Google Scholar CrossRef Search ADS   Oya C. ( 2015) Rural labour markets and agricultural wage employment in semi-arid Africa. In Oya C., Pontara N. (eds) Rural Wage Employment in Developing Countries . London: Routledge, pp. 37– 68. Oya C., Ba C. O. ( 2013) Les politiques agricoles 2000–2012. In Diop M. C. (ed.) Sénégal 2000-2012 . Paris: Karthala, pp. 149– 178. Pattenden J. ( 2016) Working at the margins of global production networks. Third World Quarterly , 37: 1809– 1833. Google Scholar CrossRef Search ADS   Peck J. ( 2013) Making space for labour. In Featherstone D., Painter J. (eds) Spatial Politics . Oxford: Wiley-Blackwell. Phillips N. ( 2011) Informality, global production networks and the dynamics of ‘adverse incorporation’. Global Networks , 3, 380– 397. Google Scholar CrossRef Search ADS   Rainnie A., Herod A., McGrath-Champ S. ( 2011) Review and positions: global production networks and labour. Competition and Change , 15: 155– 169. Google Scholar CrossRef Search ADS   Rassas B. ( 1988) Contract Farming in the Horticultural Sector in Senegal. IDA Working Papers No. 85, Clark University, Worcester, MA (USA). Razavi S. ( 2009) Engendering the political economy of agrarian change. Journal of Peasant Studies , 36: 197– 226. Google Scholar CrossRef Search ADS   RDS – République du Sénégal ( 1972) Rapport De Présentation De La Convention D’établissement Bud Sénégal. Dakar: Direction des Archives Du Sénégal. RDS – République du Sénégal ( 1974a) Bud Sénégal Société D’économie Mixte Avenant N.1 A La Convention D’établissement Du 4 Février 1972. Dakar: Direction des Archives Du Sénégal. RDS – République du Sénégal ( 1974b) Projet D’aménagement De Deux Périmètres Maraichages Irrigues De 200 Ha Chacun Pour Y Installer 400 Familles Des Paysans. Dakar: Direction des Archives Du Sénégal. RDS – République du Sénégal ( 2012) Étude Diagnostique Sur Légalité Entre Les Femmes Et Les Hommes Dans Les Niayes, PADEN. Dakar, Senegal. RDS – République du Sénégal ( 2013) Études De Filières Horticoles, PADEN. Dakar, Senegal. Riisgard L., Hammer N. ( 2011) Prospects for labour in global value chains. British Journal of Industrial Relations , 49: 168– 190. Google Scholar CrossRef Search ADS   Robinson P. K., Rainbird H. ( 2013) International supply chains and the labour process. Competition and Change , 17: 91– 107. Google Scholar CrossRef Search ADS   Ruwanpura K. N. ( 2016) Garments without guilt? Uneven labour geographies and ethical trading - Sri Lankan labour perspectives. Journal of Economic Geography , 16: 423– 446. Google Scholar CrossRef Search ADS   Sall B. ( 1993) De La Modernité Paysanne En Afrique Noire. Le Sénégal . Paris: L’Harmattan. Selwyn B. ( 2007) Labour processes and workers’ bargaining power in export grape production, North East Brazil. Journal of Agrarian Change , 7: 526– 553. Google Scholar CrossRef Search ADS   Selwyn B. ( 2010) Gender wage work and development in North East Brazil. Bulletin of Latin American Research , 29: 51– 70. Google Scholar CrossRef Search ADS   Selwyn B. ( 2012) Beyond frim-centrism: re-integrating labour and capitalism into global commodity chain analysis. Journal of Economic Geography , 12: 205– 226. Google Scholar CrossRef Search ADS   Senghor F. ( 2006) Plan Reva. Note Introductive. Dakar: Ministère de l’Agriculture, de l’Hydraulique Rurale et de la Sécurité Alimentaire, Sénégal. Smith A. ( 2015) The state, institutional frameworks and the dynamics of capital in global production networks. Progress in Human Geography , 39: 290– 315. Google Scholar CrossRef Search ADS   Smith C. ( 2006) The double indeterminacy of labour power. Work Employment and Society , 20: 389– 402. Google Scholar CrossRef Search ADS   Starosta G. ( 2010) Outsourcing of manufacturing and the rise of the giant global contractors. New Political Economy , 15: 543– 563. Google Scholar CrossRef Search ADS   Swinnen J. F. M., Colen L., Maertens M. ( 2013), Constraints to smallholder participation in high-value agriculture in West Africa. In Elbehri A. (ed.) Rebuilding West Africa’s Food Potential . FAO/IFAD. Available online at: http://www.fao.org/docrep/018/i3222e/i3222e09.pdf. Tallorine A., Dolan C., Smith S., Barrientos S. ( 2005) Reaching the marginalised? gender value chains and ethical trade in African horticulture. Development in Practice , 15: 559– 571. Google Scholar CrossRef Search ADS   Taylor M. ( 2007) Rethinking the global production of uneven development. Globalizations , 4: 529– 542. Google Scholar CrossRef Search ADS   Taylor M. ( 2008) Power, Conflict and the Production of the Global Economy. In Taylor M. (ed.) Global Economy Contested . London: Routledge. Google Scholar CrossRef Search ADS   Taylor P., Newsome K., Bair J., Rainnie A l. ( 2015) Putting Labour in its Place . New York: Palgrave Macmillan. Google Scholar CrossRef Search ADS   Thompson P. ( 2010) The capitalist labour process: concepts and connections. Capital and Class , 34: 7– 14. Google Scholar CrossRef Search ADS   Thomson P., Smith C. ( 2009a) Labour power and labour process: contesting the marginality of the sociology of work. Sociology , 43: 913– 930. Google Scholar CrossRef Search ADS   Thomson P., Smith C. (eds) ( 2009b) Working Life . Basingstoke: Palgrave. Van den Broeck G., Van Hoyweghen K., Maertens M. ( 2016) Employment conditions in the Senegalese horticultural export industry: a worker perspective. Development Policy Review , 34: 301– 319. Google Scholar CrossRef Search ADS   Whitehead A. ( 1991) Women in food production in Sub-Saharan Africa. In Sen A., Dreze J. (eds) The Political Economy of Hunger . Oxford: Clarendon Press, pp. 425– 473. Whitehead A., Kabeer N. ( 2001) Living with Uncertainty: Gender, Livelihoods and Pro-Poor Growth in Rural Sub-Saharan Africa, IDS Working Paper 134, Institute of Development Studies, Brighton, Sussex, England. World Bank ( 2007) World Development Report 2008 . New York: Oxford University Press. © The Author (2017). Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oup.com

Journal

Journal of Economic GeographyOxford University Press

Published: Jan 1, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off