Rethinking social upgrading in global value chains around worker power

Kristoffer Marslev, Cornelia Staritz, and Gale Raj-Reichert
Jun, 2024
#Trade and FDI
#Social and working conditions
The concept of ‘social upgrading’ has been instrumental in bringing the situation of workers in export sectors across the global South to the fore in research on global value chains (GVCs) and global production networks (GPNs). Responding to accusations of ‘labour-blindness’ (Taylor, 2007), scholars in the field defined social up-grading as the “process of improvement in the rights and entitlements of workers as social actors” (Barrientos et al., 2011, 324). It includes measurable standards – tangible aspects such as wage levels, contractual terms and working hours – and enabling rights, which refer to freedom of association, non-discrimination, voice and empowerment. While this was a welcome move, and has spawned many case studies, the concept has been criticized for insufficiently theorizing how and why social upgrading occurs (Selwyn, 2013). To contribute to rectifying this, we pro-pose a reconceptualization of social upgrading around worker power. We apply this approach to the case of the garment sector in Cambodia to understand how strike action, understood as an outcome of shifting state-labour relations and re-flective of intersections of worker identities, was a causal factor for social upgrad-ing. Our findings contribute to a more comprehensive understanding of the pro-cesses of social upgrading.


A reconceptualization of social upgrading around worker power

In so doing, we build on a stream of GVC/GPN research that treats workers as agents with the capacity to shape their own situation and the dynamics of globali-zation, more broadly (e.g. Coe and Jordhus-Lier, 2011; Selwyn, 2013; Newsome et al., 2015). A common starting point for unpacking such agency is the distinction be-tween associational power, which stems from the ability to mobilize collectively, and structural power, arising from positionality in the economic system and the capacity it provides for disrupting accumulation (Wright, 2000). The latter can be broken down into marketplace bargaining power, a result of tight labour markets, and workplace bargaining power, which accrues to workers who occupy strategic spots or chokepoints in production (Silver, 2003). These distinctions and their link-ages can usefully be integrated with insights from economic geography, situating agency at the intersection of a ‘vertical’ and a ‘horizontal’ dimension of GPNs (Neilson and Pritchard, 2009). While the vertical dimension represents transnational structures and relationships such as inter-firm governance, competition among suppliers or transnational civil society campaigns, the horizontal dimension describes the ways in which workers are embedded in particular locations and institutional contexts. Foregrounding workers’ struggles, and the various ways in which structural and associational power articulate along these horizontal and vertical dimensions as shown in Table 1, can offer an understanding of the causal processes of social up- and downgrading.

Yet, we call for a deeper theorization of the places in which GVCs ‘touch down’ at the local scale, emphasizing two aspects that play out largely at the horizontal di-mension: state-labour relations and the intersectionality of worker identities. First, ‘the state’ regulates worker power in fundamental ways but was for long consigned to the margins of GVC/GPN research (Werner 2021). We propose a strategic-relational approach to viewing the state as a complex social relation and a key arena of social conflict (Jessop, 1990). The nature of state-labour relations is critical for understanding how and why states regulate, mediate and intervene in capital-labour conflicts, and what opportunities and constraints workers in GVCs face in seeking concessions from suppliers, lead firms or the state itself. State strategies towards trade unions and NGOs, for instance, set the parameters for workers’ associational power; and policies on migration, social protection, education and other areas regulate labour supply, with bearings on workers’ structural power.

Second, workers’ identities are complex and multi-dimensional, as class is inter-woven with other social categories such as gender, race, age, nationality, ethnicity and community, interlinking the spheres of production and reproduction, which feminist scholars have pointed out for a long time (e.g. Mezzadri, 2020; Bair, 2010). Such an intersectionality perspective is useful to understand not only the differentiated outcomes for different types of workers in GVCs but also how worker power is confronted by a wider array of social relations beyond production. For instance, supplier firms adjust to pressures from lead firms by creating fine-grained stratifications among their workforces linked to differences (e.g. gender, ethnicity or mi-grant status), which is the basis for differential schemes of remuneration and working conditions. Yet, these positionalities can also inform shared identities and collective consciousness among workers as a basis for labour activism in GVCs (Carswell and De Neve, 2013).

The case of the garment industry in Cambodia

Our reconceptualization of social upgrading can be illustrated by the case of the garment industry in Cambodia. After a long phase of social downgrading – with falling real wages, a boom in temporary contracts and a wave of mass faintings – workers saw their nominal minimum wage (measurable standards) double be-tween 2012 and 2015; and by 2019, their real minimum income was twice as high as in 2012 (Marslev, 2019). In terms of enabling rights, however, the situation deteriorated. While Cambodia had a progressive labour law and a unique social clause in its past trade agreement with the US, which led to the comprehensive Better Factories Cambodia programme, it made it into the ITUC’s top 10 of the “world’s worst countries for workers” in 2016 (ITUC, 2016).

The steep wage hikes were triggered by an unprecedented strike wave in the garment sector as shown in Figure 1 that culminated in a sector-wide strike in December 2013. This exercise of associational power had important ‘horizontal’ and ‘vertical’ dimensions. Horizontally, strikes were characterized by an unusual unity in an otherwise fragmented union movement and supported by a broad coalition of occupations, social movements and the political opposition (Arnold 2017). Vertically, images of workers killed during a police crackdown of the strike forced global brands and retailers to intervene, a ‘boomerang’ process that heightened pressure on employers and the government (Merk, 2009). These events did not just lead to immediate wage gains but also institutionalized subsequent concessions through a new, annualized wage-fixing mechanism. In parallel, however, employers and the government took measures to curb workers’ associational power, including a controversial trade union law in 2016, legal harassment and anti-union tactics that led to a decline in strike activity (Marslev, 2019).

The measurable gains in the 2012-2014 strikes were also secured against the backdrop of shifts in workers’ structural bargaining power. Due to a number of processes, including a reduction of a post-Khmer Rouges baby boom, deteriorating garment wages vis-à-vis alternative employment and an inflow of investors escaping rising wages in China, factories experienced the first labour shortages, leading to a rise in workers’ marketplace bargaining power. Unions – in collaboration with transnational activist networks –targeted factories, where workers enjoyed higher workplace bargaining power, and sought concessions from suppliers with reputation-sensitive buyers. This helped drive a wedge into the united employers’ front against worker concessions (Marslev, 2019).

Our framework, however, also brings attention to the ways in which the outcomes for garment workers were shaped by state-labour relations, and thus deeply en-twined in domestic politics. The strike wave fed into the greatest challenge to Prime Minister Hun Sen and his Cambodian People’s Party (CPP)’s strong power-base in rural areas from which many garment workers originate. After two decades, the garment workforce had become a decisive bloc – representing almost a tenth of voters – and promises by the political opposition of higher minimum wag-es during the 2013 election led to a near-defeat of the ruling party. Although the CPP stayed in power, the threat became acute in December 2013, when opposition protests and garment strikes converged into a mass movement. In this light, the material concessions since 2013, and Hun Sen’s charm offensive in the 2018 election, which was won after dissolving the opposition, emerged as bids to secure the ‘garment vote’; and the crackdown as an attempt to break the alliance between independent unions and opposition forces (Marslev, 2019).

The power of Cambodian garment workers, moreover, is conditioned by intersections of class, gender, family and rural-urban relations. For many families, sending female members to work in urban areas is part of “trans-local” livelihood strategies. Women are not just expected to shoulder the bulk of household work, but to take economic responsibility by remitting back much of their incomes; a dual bur-den that generates a deep fear of unemployment and dissuades many from engaging in activism (Salmivaara, 2020). In the 2012-14 strikes, however, these social structures also became instrumental in exercising associational power: workers’ action was encouraged by rural families who – in the context of pressures on family farming, including falling rice prices, rising costs of fertilizer and heavy floods – had material stakes in higher garment minimum wages, and mobilization spread through extended family networks. Garment workers “protest[ed] not only for themselves but as the representatives of a wider household structure” (Lawreniuk and Parsons, 2018, 33).


Analyzing social upgrading as the outcome of the exercise of – and shifting sources of –worker power, as illustrated by the case of Cambodia, not only provides a more realistic assessment of how and why social upgrading occurs; it also brings attention to the complex power dynamics that underpin – and sometimes contravene – institutional attempts to ameliorate poor working conditions, offering valuable nuances to governance-based approaches to social upgrading. While the case presented above involved a major strike wave and visible capital-labour conflict, we argue that our reconceptualization can also be applied to contexts where worker struggles manifest themselves in more covert ways or where capital-labour conflict is channeled into more negotiated and institutionalized forms. The case, further, shows that social up- and downgrading can go hand in hand, as workers achieved minimum wage increases and new wage-setting frameworks, but at the cost of rising work intensity and a backlash on enabling rights. This is related to the counter strategies of firms and the state that need to be conceptualized along worker power in GVCs. Our conceptual approach contributes more broadly to GVC research on understanding power more comprehensively through interactions between firms, workers and states at the vertical and horizontal dimensions.


This blog post builds on Marslev et al. (2022), where more extensive discussions of the theoretical framework and empirical case can be found.
Arnold, D. (2017). Civil society, political society and politics of disorder in Cambo-dia. Political Geography, 60, 23-33.

Bair, J. (2010). On difference and capital: Gender and the globalization of produc-tion. Signs, 36(1), 203–226.

Barrientos, S., Gereffi, G., & Rossi, A. (2011). Economic and social upgrading in global production networks: A new paradigm for a changing world. International Labour Review, 150(3–4), 319–340.

Carswell, G., & De Neve, G. (2013). Labouring for global markets: Conceptualising labour agency in global production networks. Geoforum, 44, 62-70.

Coe, N.M., & Jordhus-Lier, D. C. (2011). Constrained agency? Re-evaluating the geographies of labour. Progress in Human Geography, 35(2), 211-233.

ITUC (2016). ITUC global rights index: The world’s worst countries for workers 2016. Brussels: International Trade Union Confederation.

Jessop, B. (1990). State theory: Putting the capitalist state in its place. Polity Press.

Lawreniuk, S., & Parsons, L. (2018). For a few dollars more: Towards a translocal mobilities of labour activism in Cambodia. Geoforum, 92, 26-35.

Marslev, K. (2019). The political economy of social upgrading. Unpublished PhD Thesis.

Marslev, K., Staritz, C., & Raj-Reichert, G. (2022). Rethinking social upgrading in global value chains: Worker power, state‒labour relations and intersectionality. Development and Change.

Merk, J. (2009). Jumping scale and bridging space in the era of corporate social responsibility: Cross-border labour struggles in the global garment industry. Third World Quarterly, 30(3), 599–615.

Mezzadri, A. (2020). The sweatshop regime: Labouring bodies, exploitation and garments made in India. Cambridge: Cambridge University Press.

Neilson, J. & Pritchard, B. (2009). Value chain struggles. Wiley-Blackwell.

Newsome, K., Taylor, P., Bair, J., & Rainnie, A. (2015). Putting labour in its place: labour process analysis and global value chains. Palgrave.

Salmivaara, A. (2020). ‘What if they don’t renew my contract?’ Cambodian garment workers, social reproduction and the gendered dull compulsion of economic relations. In A. Fishwick & A. Hammer (Eds.), Political economy of work in the Global South. Red Globe Press.

Selwyn, B. (2013). Social upgrading and labour in global production networks: A critique and an alternative conception. Competition & Change, 17(1), 75–90.

Silver, B.J. (2003). Forces of labor. Cambridge: Cambridge University Press.

Taylor, M. (2007). Rethinking the global production of uneven development. Globalizations, 4(4), 529–542.

Werner, M. (2021). Geographies of production II: Thinking through the state. Progress in Human Geography, 45(1), 178-189.

Wright, E. O. (2000). Working-class power, capitalist-class interests, and class compromise. American Journal of Sociology, 105(4), 957–1002.


Cornelia Staritz

University of Vienna

Gale Raj-Reichert

Bard College Berlin

Scroll to Top