Abstract
In this Part 2, Subjective Agency, I delve into the evolving power and agency of workers as structural conditions undergo transformation. Part 1 described how the law of competition propels firms towards centralisation, gradually diminishing the monopsony power wielded by global buyers in labour-intensive sectors. While manufacturers gain more power and significance, capable of undermining workers’ power, they also become susceptible to disruption. This section introduces the concept of Degree of Spatial Inflexibility (dsi). It posits that as a manufacturer’s monopoly power increases, the monopsony power in the supply chain decreases. However, this elevated monopoly power simultaneously establishes a higher dsi, amplifying the potential for worker disruption. It is essential to recognise that structural power shifts do not guarantee automatic benefits for workers; instead, they necessitate workers to capitalise on these vulnerabilities. This establishes a dialectical relationship between the evolving conditions of capital outlined in Part 1 and the subjective agency of workers, as detailed in Part 2 below.
On 17 September 1982 David Dubinsky died, aged 90. An obituary in the New York Times described a life of commitment to the ilgwu, where he had served as president for more than 3 decades, from 1932 to 1966.1 Born in what is now Belarus, at age 13 he was already working as a unionised baker, and during the failed Russian revolution of 1905 he took inspiration from a mass rally for the Jewish Workers’ Union, or Bund. The next year, at age 14, he was elected its assistant secretary. And after several arrests for union-related activity, including attempts to organise strikes, he escaped from police custody while en route to Siberia. In 1911, he appeared in New York City. By 1932, the 5′4″ Dubinsky was running an American union: the ilgwu, which had been organised out of New York City’s garment district. Buoyed by the jobbers’ agreement strategy, which forced both retailers and suppliers to sign union agreements, union membership soared under his leadership. His detractors, however, accused the Belarusian of supporting imperialism, undermining strikes, and ‘collaborating with manufacturers to fleece the consuming public’.2 The end of Dubinsky’s presidency coincided with the end of an era, as the jobbers’ strategy, the ilgwu’s highly successful three-part strategy, concluded and globalisation swept the garment sector out of New York City and London and into the Third World. In the early twentieth century, the ilgwu was forced to confront a new, vertically disintegrated business model, one that would become commonplace across industrial capitalism, and spread across the world. This system of outsourced production made direct negotiations with low-value employers very difficult, leading to a war of attrition.
On the face of things, it is the suppliers/producers who are responsible for the wages and working conditions of shopfloor labour, since they are the ones who set wage policy and maintain the factories. But suppliers – however villainous or well-intentioned their owners may be – operate within the limits imposed on them by buyers through the value chain. And, as I described in detail in Part 1,3 when buyers have a high Degree of Monopsony Power (dmp) they can demand a lot for very little, pushing down margins and giving labour much less to bargain for.
But the ilgwu’s efforts were not entirely in vain. Concerted actions at the points of production, as well as consumption, had forced contractors to include the price of labour in source price negotiations with suppliers. These jobbers’ agreements also effectively ‘ring fence’ any additional labour costs that contractors might incur, affording garment workers more wiggle room to bargain.4 The is an important beachhead for workers. As Anner et al.5 observe, ‘the principal cause for the persistent violation of workers’ rights in the global apparel supply chain is the pricing mechanism between buyers and their suppliers.’ What workers can demand of the contractors employing them cannot therefore be separated from source prices.6
Firms deploy a wide variety of different strategies and tactics to survive and thrive under competition. From complex pricing manoeuvres to product innovations, differentiation, and marketing, all aim to create and exploit a loyal base of consumers. But in addition to drumming up business, firms also need to cut the costs of doing business (the one brings more money in while the other lowers the cost of getting it – and the difference is profit). And cutting costs is accomplished though new technologies, scientific management, outsourcing, and so on; anything and everything that does the same for less, within reasonable bounds. The third front in enterprise, of course, is the competition itself, whose own positions and activities in the market must be considered by a firm’s management. Everyday competition rewards competent firms, who best their rivals through gumption, cost-effectiveness, or wily connivance, with bigger chunks of the market and higher profits.
Likewise, labour – in its efforts to self-determine and resist subordination – is in competition with the bosses over who gets a bigger slice of the pie. It does this by combining workers into associations and trade unions and through organising campaigns for better working conditions and pay. Organised labour can, among other things, pressure political institutions to implement regulations shielding workers from the worst predations of capital.
Employment relations can be explained in part by the conflicts between capital and labour. Labour is, after all, the only living and subjective factor in production. It performs tasks that vary in time, intensity, skill, quality of outcomes, the faculties used, oversight, etc. In Marxian terms,7 labour is an employer’s ‘variable capital’. Variable, in that it is elastic; the degree of exploitation involved (i.e. its profitability) can be increased by cutting pay, extending the work day, or intensifying the work day.
Braverman8 believes labour-power has an ‘intelligent and purposive character, which gives it infinite adaptability’ and ‘infinite potential’. But the capitalist cannot exploit the infinite potentials of labour-power, since these are ‘limited to the subjective state of the workers, by their previous history, by the general social conditions under which they work, as well as the particular condition of the enterprise, and by the technical setting of their labour’.9 Instead, the capitalist desires only to obtain the most labour-power for the least money. This application of the profit motive reduces labour to an input comprised of fungible workers who, if they are to keep their livelihoods, must reach given quotas within given time frames while toiling under a production regime devised by the capitalist. Any resistance by organised labour will initiate a bargaining process between employer and employees. And how that bargaining process shakes out will largely be determined by capital’s relative oligopsony power.
In other words, the state of capitalism at any given time depends on the strength of labour. Because capital’s raison d’être is self-expansion, so long as there are different capitals, they will compete for profits, or additional fragments of the total available capital (minus costs). However, a firm can also free up any capital used in its own operations which is deemed superfluous. It is here that capital confronts labour, which is the ultimate source of revenue, but also a cost. To achieve maximal productivity (i.e. to draw maximal revenue), firms impose discipline on labour, coercing a set intensity of work. But to achieve minimal cost, firms pay as little as labour, and the available pool of labour, will bear. To achieve both a semblance of autonomy and a standard of living for workers better than subsistence (or worse), labour must assert itself through associational power, structural power, or a combination thereof. As already demonstrated in Part One, in the previous issue of this journal,10 competition drives down gvc dmp by reducing the number of suppliers to a few large firms which have greater bargaining power vis-à-vis buyers; however, these changes are also reflected in labour’s bargaining power within increasingly consolidated supply chains. Here in Part Two, workers’ subjective agency acts on and through the material conditions. This is not a linear relationship, of course, but a dialectical one. Structure and agency shape, and are shaped by, each other.
Labour Bargaining Power within Increasingly Consolidated Supply Chains
Policy papers within the past decade have highlighted the importance of global supply chains to labour relations, emphasising – among other things – the roles of skill11 and international labour standards.12 What is neglected in the discussion, however, is worker agency, specifically its exercise through collective bargaining. The supply-chain literature is dominated by analysis that is developed into the Global Commodity Chains (gcc), Global Value Chains (gvc), and Global Production Networks (gpn) frameworks. gcc/gvc/gpn analysis dwells mostly on how lead firms figure in the power balance, portraying them as prime movers in the production process, as well as the broader chain activity, and enmeshed political structures. This framework is sometimes criticised as being too ‘firm-centric’.
The treatment of chains/networks as static power balances between buyers and suppliers has obscured the role of labour, rendering it an exogenous factor in production. Much of the literature seems to have forgotten that the underlying thrust of the gvc – and capitalism itself – consists in the relentless accumulation of capital, leading to a turbulent and uninterrupted evolution, forever changing the character of production. Here I treat labour not as an exogenous factor of production, or a variable in the cost function, but as a dynamic and decisive element in production, co-determining the value chain and its political environment. Ultimately, it is the third major force shaping the gvc, alongside buyers and suppliers. By bringing labour back into the equation, I intend to fill a glaring blind spot in a literature which is already preoccupied with questions of power and governance.
gcc/gvc/gpn approaches are also limited by how they conceive of power. They frequently regard it as a function of the technical and organisational position of each firm in the value chain, isolated from competition. For Gereffi et al.13 power is simply the ‘degree of coordination’ and ‘asymmetry’ achieved by lead firms vis-à-vis suppliers in the value chain. For the gpn, power is the influence firms, institutions, and stakeholders have over a production network. Power, by these estimations, primarily concerns decision-making and resource allocation, in which leverage comes from the firm’s position in a production network.14
Labour is, however, an active agent within production, exerting a pressure on the gvc in proportion to its organisation and activity. But, in much of the literature, it is conspicuously absent, or present only as spectator; even in the gpn, which tries to formalise labour as ‘collective power’. Indeed, as I have argued here, gvcs competition is comprised of three fronts; a horizontal contest between capitals producing similar commodities; a vertical tug-and-pull between suppliers and buyers; and an intrafirm conflict between capital and labour. Each of these battles over value distribution creates an evolving configuration, affecting linked governance structures.
An early attempt to include labour in a gvc framework was undertaken by Frenkel and Kuruvilla,15 who argued that patterns in labour relations are the combined effect of three factors: competition, industrial peace, and employment-income protection. Later, Riisgaard and Hammer16 analysed labour through the prisms of power and drive in the banana and flower value-chains, with special focus on international solidarity. Despite a few divergences with my own views, there is much to be gleaned from Riisgaard and Hammer whose work guides some of the analysis herein. In particular, how the power of suppliers vis-à-vis buyers affects labour; and how the power of labour affects the supplier–buyer relationship. Long-term contracts, the current relationship between suppliers and buyers, and the control buyers have over suppliers, and so on, are all shaped in part by the relative strength of labour. As Riisgaard and Hammer observed, the strengthening of supplier firms depends on labour.
Another forerunner, Lakhani, Kuruvilla and Avgar were among the first to marry theories of the gvc to employment relations.17 Their framework helped move firm-centric employment relations research toward a networked analysis, providing a foundation for something beyond the level of case studies. But their work was incomplete, insofar as it treated gvc evolution as a top-down, automatic affair. As I demonstrate, rather than unidirectional, this process is dialectical.
In that same year, 2013, Ben Selwyn argued against the ‘automatic’ interpretation of gvc change.18 Although Selwyn made several valuable contributions, the argument against social upgrading in the ilo’s Decent Work Agenda is most useful here. Selwyn’s thesis – that improvements in labour conditions do not simply ‘trickle down’ but are won by workers – does colour the thinking behind the analysis herein. Comparing and contrasting the bottom-up and top-down interpretations, as readers of the recent literature are liable to do, is a very instructive process, immensely enriching one’s understanding of how value is actually created and distributed.
A Human Resources special issue, Global Supply Chains and Social Relations at Work: Brokering Across Boundaries,19 spotlighted brokerage and supply-chain intermediaries. But those are bit parts; my focus is on the gvc main cast: buyers, suppliers, and labour.
Finally, we have the bjir 2018 symposium, ‘Corporate Social Responsibility and Labor Standards’,20 which also concerns itself primarily with what is secondary. As I argue in both Part 1 and 2, a corporation – whether it be supplier, buyer, or other – is animated, above all, by the profit drive. And under that compulsion, meaningful labour standards can only be established where the rubber meets the road, and material conditions (structural power) constrain worker agency (associational power). Questions of corporate social responsibility (csr) are window dressing. Mark Anner’s piece on wildcat strikes and the ‘sourcing squeeze’ in Vietnam from the same issue, however, is useful grounding for several of the theories developed here.
Structural Power in the gvc
Erik Olin Wright’s21 formulations of associational and structural power are valuable assets for analysing bargaining power in the gvc. Under Wright’s rubric, structural power is the ‘power that results simply from the location of workers within the economic system’ and labour’s ability to interrupt the production process and thereby exact concessions from employers. Associational Power, however, is the collective power which emerges from the representative institutions and organisations of workers. Dimensions of associational power include union density, collective-bargaining agreement coverage, and the participation of labour collectives in firm and non-firm decision-making.
Meanwhile, employers design and employ business strategies that extend or intensify work, reduce wages, and so on. Often, they attack the associational and structural power of labour directly, erecting obstacles to unionisation such as the dispersal – through one means or another – of work; enacting systems of control over production and labour processes; and prevailing upon politicians to dilute the regulatory frameworks that legalise union power.22
In response, labour leverages its associational and structural powers and expands; it rallies the shop floor, establishes representative institutions, and promotes pro-labour legislation, at both the national (laws) and the international (gfas) levels.23 And, when all its ducks are in a row, labour initiates economic struggles. Wright sees the relationship between capital and labour as a fluid ‘class compromise’, shifting with the balance of power. But it is more complicated than what Wright suggests, especially regarding gvcs. Although ‘polarisation’ (i.e. buyers, producers, and labour) is a useful analytic tool for exploring how specificity of place figures in capital–labour relations (encompassing a multitude of interactions, hierarchies, and antagonisms), its greatest virtue lies in its universality.
Structural power is an especially significant feature in gvcs, where production formations require a system of tiered production that is decentralised and re-integrated through the vicissitudes of the market, allowing labour – if it can – to interdict capital accumulation at key junctures. Labour therefore possesses great latent power in countries, regions, particular economic activities, even specific firms, which have acquired a central position in a value-chain or production network.
The ‘positioning’ of labour in the production process is also distinguished by two different aspects: marketplace structural power, which arises from the relative tightness of a labour market (in terms of skills, unemployment rate, and non-wage income); and workplace structural power, arising from the position of labour in ‘tightly integrated production processes, where a localised work stoppage in a key node can cause disruptions on a much wider scale than the stoppage itself’.24 However, the latter is more important in gvcs, where labour is systematically deskilled.
The bargaining power of labour, of course, corresponds to its structural power. And while the cynical but savvy exploitation of uneven development led to a system of world production centred around the Global North, it appears the next stage of development is already on the horizon. In 2010, for instance, the economies of the Global South accounted for half of fdi inflows – a first. This fact, coupled with the year-on-year increase in fdi outflows from those same economies – the unctad report notes that ‘Emerging economies are the new fdi powerhouses’, and that most outflows stayed within the Global South – indicate their growing significance as sites of production and consumption, and as sources of investment.25 And, as mentioned earlier, as a region, industry, or firm becomes more important and profitable, the higher are the stakes which associated labour can exploit (i.e. the higher its relative structural power). While in the 1930s and 1940s, the ilgwu innovated a strategy of applying pressure at the sites of both production and consumption, workers in the Global South are now less and less reliant on the Global North, especially as retailers turn to the Global South for new consumers. Global South garment workers could also be charting new territory for international, South–South solidarity, as they successfully target retailers and brands at the shopfront level.26
Power and Network Centrality
Insights from network theory allow for a more concrete understanding of labour’s power position, of what it depends on and how it is affected by changes in global production, like outsourcing. Networks are graphs consisting of connected nodes or points. The lines that connect the nodes are called links or edges. The literature on gpns imagines production as a network of interconnected activities executed by firm and non-firm actors – labourers, national states, intergovernmental entities, etc. Each actor in a production network is a node and each node is connected to the other nodes through links. These links represent established relationships.
Since analysis of gpns uses networking theory, one would expect a more rigorous application. But the study of network structural properties is limited. Take, for example, network centrality, a fundamental concept in network theory emphasising the importance of nodes. While Coe and Yeung admit networks have structural properties affecting the larger power balance, they dwell instead on the relational characteristics of networks and conceptualise power as being ‘as much a structural property as a contingent and contextually defined practice among interconnected actors in a network’.27 Nevertheless, the structural properties of networks remain valuable to the study of global production networks, especially regarding labour.28
Labourers employed in the ‘key nodes’ of globally integrated production systems possess greater bargaining power (workplace bargaining power) vis-à-vis capital, which can affect the entire value chain and production network. However, as Beverly Silver points out, subcontracting and ‘vertical disintegration’ were introduced to erode that power.29 Spatially and organisationally flexible production systems are therefore a tool for controlling labour costs by constraining labour’s power.
To better understand power relations from the standpoint of labour, it is necessary to study the network properties of global production, including network centrality which captures the importance of node positioning within networks. There are also now other centralities in network theory, each shedding light on ‘different aspects of the position that a node has, which can be useful when working with information flows, bargaining power, infection transmission, influence and other sorts of important behaviours on a network’.30 Network centrality is a function of the structural positioning of nodes in a network distinguishing four families of centrality: Degree, Betweenness, Closeness, and Neighbours’ Characteristics.31 Network centrality adds to our understanding of capital–labour relations as well as labour’s structural power. The greater the centrality of a firm in the production network, the greater the importance of associated labour and therefore the greater its latent bargaining power.
In the end, of course, capital–labour relationships are co-determined through subjective agency, and the confrontation of labour’s associational and marketplace power with capital’s oligopoly and oligopsony power. But firm centrality is nonetheless a useful proxy for gauging the latent workplace power of labour in value chains and production networks, since their respective strategic positions are two sides of the same coin.
The myriad organisational, spatial, and functional factors that affect the number of nodes in a gvc/gpn will, by extension, affect centrality and the relative power of firms. Whenever lead firms implement business strategies promoting flexible production systems, for example, and start outsourcing, they dilute the centrality of supply firms, whose labour forces are thereby weakened as well. Vertical integration, on the other hand, creates ‘mega suppliers’ whose workforces are strategically significant. The fewer the nodes in a network, the higher centrality of suppliers and more bargaining power is involved, for both firms and labour.
Conflict between capital and labour in each gvc segment does not occur in isolation but usually as the result of larger market forces. And whenever capital wants to undercut labour, it can move to vertically disintegrate; narrowing the topmost circle of competition while enlarging the bottom. As case studies attest, the latent power of workers lies in the relative power of suppliers vis-à-vis buyers. An increasing number of scholars are now using competition and market power to explain the dynamics of organisational structures under globalisation. Ben Selwyn, for instance, argues that, ‘Capital–labour relations are based on an inherently conflicting and ongoing process (sometimes hidden and sometimes open) where each class attempts to maximise their share of the surplus created in the process of accumulation. Various authors recognise this dynamic process, but argue that contemporary globalisation has reduced significantly labour’s power to appropriate a greater share of surplus.’32
Degree of Spatial Inflexibility
I argue that it is the combination of material conditions and subjective worker agency that produces bargaining outcomes, good or bad. To concretise this idea, let us treat the buyer-driven and producer-driven value chains as two ends of the same spectrum. Here I introduce the concept of Degree of Spatial Inflexibility (dsi) which is central to workers’ structural power. dsi is the scope of geographic possibility within which production can take place; in other words, the constraints on how global buyers can move production to optimise capital accumulation. A low Degree of Monopsony Power (dmp) results in a high dsi and thereby greater bargaining power for workers.
dsi builds on David Harvey’s ‘spatial fix’, which is capital’s use of space to temporarily resolve crises of profitability. There are two sometimes overlapping forms: regulatory and market. Regulatory DSI is the set of geographic limits imposed on capital by states, supranational bodies, and trade agreements; and was strongest during postwar embedded liberalism. Market DSI is the set of de facto geographic limitations which are baked into a given stage of capitalist development. During early capitalist development, for instance, crude technology, insufficient surpluses, and a tiny bourgeoisie constrained market growth, producing a high dsi. During advanced capitalism, however, the drives to centralise, redistribute wealth upward, erecting high entry barriers, etc., eventuate in a handcuffing of garment buyers, for instance, to a fixed few mega-suppliers.
As identified in Part 1, there are two ends of the supply-chain spectrum. At one end are producer-driven chains, high-technology sectors such as automobiles and aeronautics where power lies in the hands of producers. At the other end we find buyer-driven chains, low-technology sectors such as garments and footwear where the buyers have the power. The higher bargaining power of workers in producer-driven chains is a direct result of the factors of production. Standardisation led to mechanisation which led to the automation, integration, and monopolisation of production firms in these sectors. This increased investment in technology, further thinning out the herd by raising barriers to entry, thus limiting the possibility of alternative sites of production (or a lowering of the Degree of Monopsony Power (dmp)). Workplace structural bargaining power, at its most fundamental, boils down to the ability of workers to disrupt, or threaten to disrupt, the production process. This is directly related to the Degree of Spatial Inflexibility (dsi). Producer-driven chains have a high dsi, because of low dmp, which means workers have a higher bargaining power.33 Contrary to some of the labour-process theory literature,34 that technology-led deskilling over the longer term can lead to greater bargaining power, not less.
Take the history of the garment sector, a buyer-driven sector, where state-level regulation began in the nineteenth century. Before that period, any structural power possessed by labour was owed to the localised nature of capital and of work, aka Market DSI. Tilly observes of this period that ‘although people exercised some individual rights as members of communities, churches, households, and other organizations, workers’ rights generally took a categorical form, applying to individual workers only in so far as they qualified as bona fide members of local trades.’35
By the 1850s, however, the legal grey area in which early labour activism operated gave way to more-formal collaboration with employers and the state. Through a form of ‘collective bargaining by riot’,36 workers won, for example, the legal right to strike, to associate, a set of unemployment benefits, and state support for barring foreign labour; aka Regulatory DSI. Of the burgeoning regulatory apparatus, Tilly writes that, ‘controls [were instituted] over the stocks and flows of persons, diseases, other biota, pollutants, weapons, drugs, money, other capital, technology, information, commodities, political practices and cultural forms within well-delimited territories’.37 Within this context, ‘states that could conscript, tax, and police could also regulate working conditions, organise schools, and build highways. At the same time, concentration and nationalization of capital gave workers connections and central objects of claims they had not previously known’.38
By the 1920s and again in the early 1970s, a high degree of regulatory DSI had given labour a longer lever with which to move the economy, a newfound power manifest through the trade-union movement and collective bargaining agreements. Before the crisis of the 1960s and 1970s, Western garment workers, particularly in the US and UK, had been shielded by protectionism. This capped the monopsonistic power of buyers and confined the ilgwu’s need to ‘chase the work’ within the US and parts of Canada. Even then, however, the ilgwu relied on a degree of a market DSI. They distinguished between what they called the ‘backbone shops’, that were essential to the jobbers’ business, and the ‘overflow shops’, that were used in seasonal peaks. As the former ilgwu Director of Organizing Jeffery Hermanson39 told me, the backbone shops were powerful during an era of the domestic garment value chain, stating that ‘brands depended on these [backbone] shops for capacity and specific production ability; and if we organised them and could stop their operation, we stood a good chance of winning a confrontation with the jobber.’ He contrasts the strength of the backbone with the weakness of the ‘overflow shops’; ‘we would concentrate on organising strong majorities in the backbone shops, while simply sending one or two workers to work in an overflow shop, to be in position to know what was going on inside, as we could then agree with the employer to put aside the jobber’s work during the strike’.
The crisis, however, inaugurated a new era of trade liberalisation – i.e. a lower degree of regulatory spatial inflexibility, as capital spilled over into the Third World, significantly isolating the ilgwu.
But Global North trade unions mitigated garment capital outflow with 1974’s mfa, which had been preceded by the Short-Term Cotton Agreement (1961) and the Long-Term Cotton Agreement. The mfa partially constrained major import markets (US, Canada, Europe) by establishing import quotas. The result, however, was a production apparatus which was spread more thinly, and whose regulatory dsi was too low to prevent spatial fixes should workers in any one location flex their associational muscle. The end of the mfa in 2005 was also the end of the industry’s regulatory dsi. Global buyers then steered production into a handful of cheap but labour-rich countries. Deregulation led simultaneously to higher dmp (and therefore greater value capture for buyers) and lower dsi (and therefore less structural power for workers). As manufacturers began consolidating in response to the intense competition for contracts in those labour-rich countries, buyer dmp gradually contracted, giving rise to what I call market DSI. By this point, suppliers had become mature firms, guarding their market positions with high entry barriers (via technology, etc.), and exercising increasing heft within the gvc.
Before this latest stage set in, labour had been launching campaigns organised around a rights-based framework (codes, audits, etc.), but did not have enough footing to fight a globalised, vertically disintegrated industry. But now centralisation and market DSI have rendered the relevant workforces much more important, substantially increasing their relative structural power. They are no longer so disposable.
Regulatory and market DSI each have their own drivers, underlying logics, and developmental contexts. But both – especially the former – constrain production, and therefore dmp, value distribution, and – critically – the power of buyers in the gvc.
For Harvey, globalisation is simply the modern expression of the spatial fix, a geographic mechanism for capital expansion, whose origins he traces back to 1492. He teases the concept out of Marx’s observation that capital is the ‘annihilation of space through time’. Late Capitalism’s spatial fix used deregulation to open new frontiers (i.e. to negate regulatory dsi). Market DSI is developed out of Marx’s theories of capital concentration and centralisation, and describes a situation in which the maelstrom of competition leaves only a few large and increasingly interdependent firms.40 Here inflexibility is the culmination of untrammelled market forces. Regulatory and market dsi sometimes overlap and are often complementary phenomena. But they can also be inversely related. Using Harvey as a foundation, dsi puts the production, reproduction, and reconfiguration of space front and centre in analysing the contours of the gvc and the political economy of capitalism itself.
Although dsi best describes the vertical relationship of buyers and producers, it can also be seen in other, more producer-driven sectors, like the automotive industry. The allocation of market power that guaranteed the stable, oligopolistic conditions of the American auto industry unravelled, not only because of Japanese competition, but also because of the relative strength of its unions.
The US automobile industry’s high levels of unionisation were the effect of United Automotive Workers (uaw) organising from the 1940s until the end of the 1970s, when international competition from Asia and Germany threatened America’s Big Three automakers (Chrysler, Ford, and gm).41 The powerful uaw represented Canadian auto workers as well, until 1985, when the Canadian Auto Workers (caw) union was established.
Here we find that autoworkers had high associational power due to both regulatory and market dsi. And for that reason, US automobile firms responded to international competition by changing how production was organised. First, they outsourced it, to low-cost regions in Europe, Asia, and Mexico; and to the independent American auto parts sector, which had much lower union density. At the same time, they implemented labour policies carefully skirting the collective bargaining agreements made with the uaw. Similarly, Mahutga demonstrates that ‘entry barriers’ relate directly to the bargaining power of suppliers in the garment and automobile sector – a phenomenon which functions much like dsi.42
Labour and Wage Distribution
There is still much debate in the academy over wage distribution. The neoclassical school understands international wage differentials as a harmonious expression of marginal productivity (generally linked to education level and human capital). Perfect, or near perfect, markets are an a priori assumption for their growth theory, and wages are indexed to marginal productivity. On this view, a worker can obtain raises simply by being more productive. Bargaining, collective or individual, never enters the picture, since workers receive what the market determines, and surplus value is a fiction. Worker agency is negligible.
According to the standard neoclassical growth model, production has two inputs: capital and labour. Labour’s share derives from the production function, so wages should correlate with profits. And yet the standard production function cannot explain a profit share that increases as wage share decreases – namely because it assumes they are linked. That may have been reasonable in the postwar era, but the late 1970s proved to be the beginning of a new, clarifying period. Post-Keynesian Neo-Kaleckians, such as Onaran and Galanis, have improved on the profit-led growth model, showing that the global decline in labour share since the late 1970s/early 1980s has contributed significantly to the overall decline in economic growth.43
Stockhammer assesses the macroeconomics of income distribution from the standpoint of political economy. He analyses 43 developing countries and 28 advanced countries over a period of 37 years (1970–2007) to understand why wages have fallen. He uses a more broad-minded approach, taking account of welfare-state retrenchment, financialisation, globalisation, and technology when considering the tug-and-pull between labour and capital, and the effect on income distribution over time. Like Onaran and Galanis, Stockhammer concludes that ‘wage shares’ are not linked to productivity.44
Despite the ‘mainstream’ popularity of the neoclassical growth model, it has – understandably – little purchase in gcc/gvc/gpn literature. Here I draw on the tradition of post-Keynesian/Kaleckian economics, for which income distribution depends on the degree of monopoly; and the tradition of Goodwin, for whom income distribution depends on unemployment levels. The latter takes a more neoclassical redistributive approach, formulating an inverse relationship between labour share and growth (profit-led growth); while the former describes how more equal income distribution contributes to growth (wage-led growth). Both approaches, however, have their fans, and have been influential in non-neoclassical macroeconomics for the last half century.
I maintain that the relationship between growth and income distribution can be explained with reference to Kalecki as well as to Goodwin. And I am not alone in such cross-pollination. There is in fact a wide array of scholarship informed by both traditions (for example Stephen Marglin and Amit Badouri, whose research has spawned innumerable studies). Unemployment, of course, affects the bargaining power of workers. And Kaleckians maintain that a high degree of monopoly increases a firm’s bargaining power. These ideas have made significant inroads outside the neoclassical school.45
Malcolm Sawyer, inquiring about the role of trade unions in influencing real wages distribution, bases his work on Kalecki,46 who argued that the pressure placed by trade unions (or similar workplace action) could impact real wages. Sawyer states trade unions may make greater demands on employers where profit margins are high. In addition Sawyer states that ‘the effect of a money wage push by unions depends on the firms’ ability to pass on the wage increase as a price increase.’ Since more competitors decrease the ability to ‘pass on’ wage increases, the increased monopoly power of the firm increases the possibilities for workers’ bargaining. Sawyer concludes, ‘Hence, the structure of wage determination (e.g. decentralised or centralised) may be relevant to the determination of money wages and their impact on prices and real wages. […] The degree of monopoly is seen as modified by activities in the labour market, and hence the real wage is influenced by the labour market.’47
Outside of neoclassical economics, wage rates are in fact affected by bargaining. In Neo-Marxian Economics, wages (or ‘markup’), for example, often depend on the unemployment rate (the reserve army of the unemployed); and in Post-Keynesian–Kaleckian Economics wages depend on degrees of monopoly power or the relative growth rate (analytically, the same as the rate of exploitation). In the Goodwin Class Struggle Model, Richard Goodwin uses a Marxian-Keynesian approach to explain the relationship between class conflict, employment as a proportion of the total labour force, and the wage/labour share as a measure of national income. Economic booms produce a rising employment–population ratio, which drives wages and wage share higher. This creates a profit squeeze and often a reaction. After another boom, the cycle is repeated. The bargaining process (or ‘class conflict’) is triggered by fluctuations in the unemployment rate.48 Expanding on Goodwin’s work, however, German Marxist Stephan Kruger identifies labour scarcity as but one part of the larger struggle by working people to defend wages. Nevertheless, trends in profit and wage share generally follow the vagaries of class struggle.
There is a rich body of literature in gvc/gpn/gcc arguing that the bargaining process is shaped in part by the power relationships within gvcs, and by upgrading within the factory.49 Riisgaard and Hammer, in scrutinising the cut flower and banana gvcs, conclude that, ‘analyses of labour in the global economy need to take account of how gvcs shape the terrain for labour to build international networks, strategies, and campaigns for labour rights’.50 Furthermore, analyses of gvc restructuring and governance itself need to integrate the role of labour in shaping global value production as well as in ‘actively produc[ing] economic spaces and scales in particular ways’.51
Economic Geography and International Political Economy assume a greater degree of market imperfection than is usually tolerated in neoclassical growth theory. Indeed, the question of power is central to gvc analysis especially as smaller firms become large firms. Indeed, there are an endless number of studies that prove or attempt to explain why large firms pay higher wages and offer greater benefits than small ones.52 The conclusions are varied, including larger employers seeking high-skilled employees, greater capital intensity, greater efficiency, a mechanism to forestall unionisation, less able to monitor workers and so on. Whilst these tell part of the story, what we find in the case of increasingly consolidated garment and footwear suppliers are increasing forms of workers’ organisation and mobilisation combined with employers finally capable of both absorbing greater wage increases and the power to remain in the gvc.
And the relationship between firm growth and the power of workers in the value chain is developed in detail by the ‘economic upgrading / social upgrading’ debates. Tied to the ilo’s ‘Capturing the Gains’ research programme, which launched in 2009, these debates help answer a critical question: how does upgrading and firm consolidation affect workers? In the literature spurred by the ilo’s initiative, upgrading is divided into capital upgrading (the use of new machinery/technology) and labour upgrading (making workers more productive as workers). Social upgrading, on the other hand, refers to improvements in working conditions, and worker protections and rights. Barrientos et al. note that social upgrading usually results from complex bargaining processes.53 The question is whether economic upgrading necessarily translates into social upgrading. As Gereffi and Fernandez-Stark point out, ‘within manufacturing, if we compare industries that can be classified as relatively low-tech (apparel) [… a] key task for the gvc analysis is to explain the conditions under which the economic upgrading of firms and the social upgrading of workers can be mutually reinforcing.’54
Many of the case studies in gvc/gpn research highlight instances of just such a mutually reinforcing relationship; of how workers and capital co-constitute while shaping the chain/network. Posthuma and Nathan observe that upgrading among firms in India is uneven and sector-dependent, and that garment suppliers largely remain ‘locked-in’ at low value-added and low-wage tiers.55 However, Tewari shows that even where Indian garment firms are upgrading, social upgrading is far from inevitable. Labour-market ‘intermediaries’ – ‘new’ unions, community groups, and buying agents – are another set of variables that must be considered.56 Caswell and De Neve take this farther in a study of the Tirupur garment cluster, bringing local social relations, the regional economy, and the cultural environment into the equation.57
Important here is Selwyn’s ‘labour-led’ social upgrading. For Selwyn, the struggles of workers in horticulture to ‘transform their structural power into associational power in order to extract concessions from capital constitutes a core determinant of the relations between economic and social upgrading. Put differently, if workers are able to organise in the face of capitalist management systems designed to raise the rate of exploitation, then they raise, significantly, the possibilities of achieving some form of social upgrading.’58 If workers can organise despite capital’s machinations – to intensify labour, to atomise workers – then that is half the battle.
Sigmann, Merk and Knorringa apply Selwyn’s ‘labour-led’ social-upgrading model to Wright’s concepts of associational and structural power while analysing the Freedom of Association protocol in Indonesia’s footwear industry. The Indonesian workers’ struggle benefited from a strong transnational solidarity network, which helped it implement the FoA protocol (meanwhile, there was a clear conflict between suppliers and buyers over value capture). The protocol was partially successful in applying the gains from labour struggles across the country’s entire sector to eliminate competitive advantages. Sigmann, Merk and Knorringa identify this as an example of labour-led social upgrading, since ‘it allows acts in the athletic footwear industry in Indonesia to move from a situation of confrontation to one that has the potential to catalyse cooperation.’ They draw a direct link between the power of suppliers vis-à-vis buyers to that of workers vis-à-vis suppliers: ‘In gvcs negotiations between unions and athletic sportswear producers are enmeshed in the negotiations with brands.’59
As I have argued, higher value capture at the point of production makes the relevant workforces more important, and therefore potentially more effective agitators. The relationship between economic upgrading and social upgrading is not automatic, of course – it is established by workers. The results are varied.
Critically, wage distribution in the garment sector needs to take account of gender. Systematic feminisation of the sector has been a function of both skill and power. About 95% of those employed in the garment industry are concentrated in the production segments of the value chain. These are the lowest ‘skill’ and those most gendered; as the firm upgrades and expands vertically, the skill levels required grow with them.60 In the Cut–Make–Trim phase of the production process, the skills required of workers revolve around operating machinery which sews, cuts, or presses clothing. And the bargaining power of workers is further complicated by issues of race, caste, and gender.
At first glance, it makes little economic sense for a factory owner to exclusively seek out women labourers. In many cases, that would necessitate the additional costs of providing a legally-mandated crèche (as in India, per the Factories Act 1948) and/or maternity leave (also in India, per the Maternity Benefit Act 1961), which can disincentivise the hiring of women.61 In the low-value garment sector, such costs might price a firm out of the market. Yet 80% of the garment industry workforce is comprised of women,62 now as it was in turn-of-the-century New York and London. Women-only hiring practices produce a clear, gendered division of labour within the factory. Highly skilled tailors, security guards, and managers are positions filled by men; and seamstresses, cutters, pressers, and helpers – the lion’s share of work – is given to women, since it is characterised as ‘women’s work’ (in essence, ‘unskilled’). Despite the historical continuity, the acute feminisation under globalisation compounds preexisting gender norms. Women sew, cut, press, and clean what men design; women operate machines that men service; women work on the factory floor while men stand guard; women toil while men manage and so forth. ‘Women’s work’ invariably results in less pay than in what is defined as ‘men’s work’. In addition to questions of skilled and unskilled labour, workers are hindered by gendered ideas of self-organisation and power.63
Workers’ Bargaining in Consolidated Garment Firms
Network centrality is a form of ‘structural power’ whilst size is a form of a ‘contingent power’ (it is more relative than relational). The different measures of power outlined above (centrality and size) are not binary, are often overlapping, but are almost always complimentary in one direction or the other. Indeed, size can lead to centrality, just as centrality can lead to size. Centrality is typically found in a vertical relationship whilst size is a horizontal one.
My research centres on the Garment and Textile Workers’ Union (gatwu) in Bangalore, specifically its intensive unionisation efforts between 2016 and 2018 at the Avery Dennison Factory in the Peenya Industrial Area. To date, this marked gatwus only successful union recognition and contract in its 20-year history. Despite numerous successful campaigns and a membership exceeding 6,000, gatwu achieved recognition through a robust campaign involving mass membership drives, factory-gate rallies, hunger strikes, court cases, and strikes.
gatwu focused on organising the 420 blue-collar contract workers, out of a total of 1,200 workers at the plant. By 2018, gatwu secured recognition, automatic union-dues check off, and joint negotiations with the citu union which represented most of the permanent blue-collar workers. They were able to secure the conversion of 111 contract workers and settlement for the remainder, and the end of the practice of contracting workers. This resulted in two 3-year collective bargaining agreements, and each included significant wage increases, with the average salary rising from 20,000 to 45,000 rupees per month.
Avery Dennison, a vertically-integrated rfid, tag, inlays, and label manufacturer, holds substantial monopoly power, supplying nearly every major apparel brand worldwide, roughly 130, with factories in 52 locations worldwide. The firm’s consolidation and automation strategies yield considerable influence over the supply chain. The high levels of standardisation led to mechanisation and automation, raising barriers to entry, and lowering competition and the Degree of Monopsony Power (dmp). This, in turn, increased the power of Avery Dennison via buyers, but its attendant higher Degree of Spatial Inflexibility (dsi) also resulted in a greater disruptive power for workers. The success of the unionisation efforts transformed average salaries at 45,000 rupees, this being four times the average monthly salary of a garment worker in the area, who earns 11,000 rupees per month. The union also won employer-provided life insurance, marriage allowance, bereavement allowance, and a provision that if a worker dies every worker is given an extra day’s salary, cash equivalent to 1 million rupees.64 So how did gatwu succeed?
Avery Dennison’s unique position as a ‘second-tier supplier’ in the apparel industry, not directly supplying brands but influencing their suppliers, makes it a formidable yet strategic target for unionisation. Its size, vertical integration, capital-intensity, and large capital holdings make it ideal for the union. They design, manufacture, and deliver their products. Simply put, it is gargantuan in size but is not central in the network. In terms of distance, it is far from the centre. However, its relative size makes it an inimitable partner to suppliers (through technological investment) as well as to brands (through its global reach). It could be read as a three-node exchange. A (brands/retailers) has power over (cmt manufacturer) and (Avery Dennison), and whilst all three are in contact, the exchange relationship is only unidirectional. However, the suppliers, even large and powerful ones, are compelled to contract exclusively with Avery Dennison since there are few alternatives, and Avery Dennison retains a non-exchange relationship with the brands. As Jayram,65 a gatwu organiser, tells me:
[Avery Dennison] are designing and printing the stickers. They’re designing the tapes and printing the tapes. The supplier has no choice but to contract with Avery Dennison. For example, Avery Dennison was afraid of the worker action and so sent their workers and installed a machine inside of Shahi Exports and they produce the labels and tags for the brands produced by Shahi.
Indeed, more often than not, the brand continues to dictate who the supplier must purchase the fabric, thread, labels and other essential accessories from. Thus, this distinction between size and centrality is relevant to the question of power for actors within the value chains.
It is for this reason that the management brought in a yellow union in two other plants in India (in Gurgaon and Pune) as well as their factory in Sri Lanka.66 In Bangalore, the company has recognised citu, the union affiliated to the Communist Party of India (Marxists), for its full-time employees. Though feared historically by managers, employers now favour the Karnataka-state citu for their record of signing sweetheart deals with management and undermining strike activity. Avery Dennison is also well known for having paid-off the officials in the labour ministry. Thus, a large capital-holding firm, whilst capable of integrating, investing, and largely staying-put, still has the power to withstand independent labour-organising drives.67
Beyond these factors, the union’s inability to put pressure on global buyers, since they are central in the network but technically far from Avery Dennison, and the futility of putting pressure on suppliers who are still dependent on buyers despite their direct economic relationship, made it hard for the workers and their union to assert formal associational power despite labour-market power ensuring significantly higher wages. This is despite high union-density in the factory (more than 90%) and a robust international solidarity campaign. Here we see a firm with a high organic composition and therefore at greater risk from labour-unrest and associated sunk costs. Nonetheless both its distance from the centre (via buyers) and high barriers to entry (via suppliers) make it difficult for workers to use secondary pressures to force the company’s hand at the negotiating table.
gvcs thus evolve with industry antagonisms, reflecting the ebbs and flows of the class struggle. A clear example of where size and network centrality work in tandem is global logistics. Kim Moody’s On New Terrain analyses how consolidation changes the balance of power in the logistics industry. He argues that conglomerates are better placed to resist isolated strikes since they can rely on other sites of production. Take, as an example, Arvind Mills, in which labour unrest resulted in the reorganisation of production from a core activity (pressing warehouse) to a peripheral activity (returns warehouse). This reorganisation – made possible through consolidation – was used to undermine the bargaining power of militant workers at any single production site.68 But despite such pitfalls along the way, consolidation eventually creates the conditions in which
larger firms compete, the combined workforce of more and more firms is relatively larger, and the new production methods and links [are] more vulnerable. In the long run, this is a situation that makes the industry more susceptible to unionisation, as was the case in the 1930s after the 1916–29 merger wave that produced corporate giants such as General Motors, John Deere, and Union Carbide.69
As I have argued in Part 1, the concentration of capitals contributes to R&D reinvestment, a portion of which is allocated to labour-saving technology. Moody explains that, ‘like the formation of larger firms along definite industrial lines, greater capital intensity offers expanded opportunities for successful direct action and increased power in collective bargaining’.70 This insight dovetails with that of Lund-Thomsen and Lindgreen,71 who write of gvcs in labour-intensive sectors where, ‘if suppliers earn lower unit rates over time, workers also tend to receive lower wages. If suppliers must reduce lead times, workers will have to engage in overtime work. Thus, the optimum point for suppliers and workers is inherently linked.’ And also with that Anwar Shaikh, who writes that, ‘capital-intensive industries will also tend to have high levels of fixed costs which will make them more susceptible to the effects of slowdowns and strikes. At the same time, because labour costs are likely to be a smaller portion of their total costs, such industries are able to tolerate wage increases.’72
But the new potentialities that confront the workforces of ascendant businesses, who are more dependent on them, remain just that: potentialities. Indeed, many have observed that it is in those sectors most effective at exploiting workers, and therefore the most competitive, that one finds labour’s worst adversaries. The more capital accumulates, the more difficult it is to extract additional surplus value – a fact that can dampen worker power. As the ratio of capital to labour (in terms of firm outlays) grows, the power of workers as a class may fall. In the words of Marx,73 ‘it follows therefore that in proportion as capital accumulates, the situation of the workers, be his payment high or low, must grow worse.’ This is of course a generalisation,74 but the case of the giant footwear supplier Yue Yuen, for example, offers a concrete example of how larger capital-holding firms can undermine workers’ attempts to exploit the ‘sunk costs’ of fixed capital. In part, by buying-off provincial officials, the police, and courts – a degree of government capture beyond the capabilities of smaller firms.75
In the current context, concentration and consolidation have increased exploitation but also made firms more vulnerable to agitation. Selwyn applies the so-called ‘bullwhip effect’ – in which small disruptions in a supply-chain link lead to larger disruptions elsewhere – to labour unrest and global competition.76 He cites changes to the global distribution sector, where the introduction of new technologies and radical reductions in delivery time have become the main theatres of competition. Technological innovations – namely bar codes, high-speed conveyers with advanced routing and switch controls, reliable laser scanning of incoming containers, and increased computing capacities – have facilitated ‘time–space compression’, making the gvc more interdependent, and allowing for lean production, lead distribution, and just-in-time deliveries.
The recent history of the United Parcels Service (ups) provides an object lesson in how an emphasis on time pressures labour, increasing work intensity and provoking labour unrest. Selwyn remarks that, ‘whilst these innovations, particularly those in the labour regime, took place with the acquiescence of labour unions and reflected the latter’s weak associational power, it altered but did not diminish workers’ structural power.’77 In 1997, the 185,000 members of the ups Teamsters Union struck, bringing ups’s global supply chain to a grinding halt. Within two weeks, only 10% of ups deliveries were at normal capacity and the company was losing $50m a day. In due course, the workers won a 25–35% five-year pay rise and the union retained control of the employee pension fund. As Selwyn concludes, ‘For all its global and labour regime restructuring … ups’s Achilles Heel was the capacity of its workforce to withdraw its labour and disrupt the entire ups supply chain. The globalization of the distribution industry supply chain and the decreased lead-times only accentuate ups’s vulnerability.’78 Selwyn also adduces several automobile factory strikes in the 1990s, which were successful for similar reasons, observing that ‘workers’ ability to disrupt production is in some ways intensified, precisely because of firms’ attempts at time–space compression.’79
It is the gvc’s changing composition (reflecting larger processes in global capitalism) combined with surpluses at the point of production and network centrality that render the system vulnerable to labour pressure, widening the aperture for strategic uses of associational power to win a higher value share for workers. Florian Butollo comes to similar conclusions, drawing parallels between early twentieth-century Fordism and recent labour history in China. In both periods, he points out, the benefits to labour were not ‘passive modification of the labour process as a consequence of technological change. Quite the contrary […] the decisive moment that lead to a transition of the mode of regulation towards an acceptance of trade unions, the implementations of collective bargaining, and the construction of the welfare state was a series of militant labour conflicts during the years of the Great Depression.’ ‘Workers’ agency’, he continues, ‘needs to play a prominent role in aligning economic development in a way that allows it.’80
Finally, to bring our discussion back to the garment sector, the cases here reveal several distinct phenomena in the garment gvc and industrial relations. First, an increase in the scale and market diversification of specialised Southern suppliers (i.e. the emergence of the mega supplier) shifts the power balance between them and the Northern buyers, weakening the latter’s bargaining power. These changes bring both obstacles and opportunities for workers. The various codes of conduct or auditing regimes were essential to assist the workers’ campaign – but that is the limit of their utility. However, the implications of this dynamic for workers’ rights depend, in part, on where the Northern buyer stood on the issue of labour rights. On the one hand, buyers relentlessly search for firms with labour costs that are low and undergird the global race to the bottom. On the other hand, Northern brands are highly scrutinised by ngos, consumer groups, and anti-sweatshop activists in their own countries, and as a consequence can be relatively more responsive (compared to Southern suppliers) to demands for labour protections. Diminishing Northern buyers’ bargaining power could negatively impact workers’ rights, at least in the short term.81 Thus, on the one hand, the strategy deployed for two decades that relies on the dual pressure on Northern ngos and anti-sweatshop activists becomes less effective. On the other hand, the mega-supplier itself can now be more actively scrutinised through a ‘direct’ spotlight from the inside – from the shop floor – with workers shifting strategy by more directly targeting their employers and involving local and international media and allies for the purpose of applying secondary pressure.
References
Adolph, Christopher, Vanessa Quince and Aseem Prakash 2017, ‘The Shanghai Effect: Do Exports to China Affect Labor Practices in Africa?’, World Development, 89: 1–18.
Anner, Mark 2011, Solidarity Transformed: Labor Responses to Globalization and Crisis in Latin America, Ithaca, NY: Cornell University Press.
Anner, Mark, Jennifer Bair and Jeremy Blasi 2012, ‘Buyer Power, Pricing Practices, and Labor Outcomes in Global Supply Chains’, Institutions Program Working Paper Series: inst2012-11, available at: <https://wsr-network.org/wp-content/uploads/2018/02/inst2012-0011.pdf>.
Barrientos, Stephanie, Gary Gereffi and Arianna Rossi 2011, ‘Economic and Social Upgrading in Global Production Networks’, International Labor Review, 150, 3–4: 319–340.
Braverman, Harry 1974, Labor and Monopoly Capital, New York: Monthly Review Press.
Brown, Charles and James L. Madoff 1989, ‘The Employer-Size Wage Effect’, Journal of Political Economy, 97, 5: 1027–1059.
Burdett, Kenneth and Dale T. Mortensen 1998, ‘Wage Differentials, Employer Size, and Unemployment’, International Economic Review, 39, 2: 257–273.
Butollo, Florian 2014, The End of Cheap Labour?, Chicago: The University of Chicago Press.
Carpe, Matthieu, Carl Chiarrella, Peter Flaschel and Willi Semmler 2011, Financial Assets, Debt and Liquidity Crises, Cambridge: Cambridge University Press.
Carswell, Grace and Geert De Neve 2013, ‘Labouring for Global Markets’, Geoforum, 44: 62–70.
Chiarella, Carl and Peter Flaschel 2000, The Dynamics of Keynesian Monetary Growth, Cambridge: Cambridge University Press.
Chiarella, Carl, Peter Flaschel and Reiner Franke 2005, Foundations for a Disequilibrium Theory of the Business Cycle: Qualitative Analysis and Quantitative Assessment, Cambridge: Cambridge University Press.
Coe, Neil M., Martin Hess and Peter Dicken 2008, ‘Global Production Networks: Realizing the Potential’, Journal of Economic Geography, 8, 3: 271–295.
Coe, Neil and Henry Wai-Yeung 2015, ‘Towards a Dynamic Theory of Global Production Networks’, Economic Geography, 91, 1: 29–58.
Cumbers, Andrew, Corinne Nativel and Paul Routledge 2008, ‘Labor Agency and Union Positionalities in Global Production Networks’, Journal of Economic Geography, 8, 3: 369–387.
Dicken, Peter 2007, Global Shift, London: sage Publishing.
Dougherty, Michael L. 2008, ‘Theorising Theory: Origins and Orientations of Commodity Chain Analysis’, The Global Studies Journal, 1, 3: 29–38, <doi:10.18848/1835-4432/CGP/v01i03/40929>.
Dutt, Amitava Krishna 2016, Growth, Distribution, and Uneven Development, Cambridge: Cambridge University Press.
Fernandez-Stark, Karina, Stacey Frederick and Gary Gereffi 2011, ‘The Apparel Global Value Chain: Economic Upgrading and Workforce Development, Technical Report’, Center on Globalization, Governance and Competitiveness, Duke University, available at: <https://www.researchgate.net/publication/281845193_Workforce_Development_in_the_Apparel_Global_Value_Chain>.
Flecker, Jörg, Bettina Haidinger and Annika Schönauer 2013, ‘Divide and Serve: The Labour Process in Service Value Chains and Networks’, Competition & Change, 17, 1: 6–23.
Freeman, Linton C. 1978, ‘Centrality in Social Networks Conceptual Clarification’, Social Networks, 1, 3: 215–239, <doi:10.1016/0378-8733(78)90021-7>.
Frenkel, Stephen and Sarosh Kuruvilla 2002, ‘Logics of Action, Globalization, and Changing Employment Relations in China, India, Malaysia, and the Philippines’, ILR Review, 55, 3: 387–412.
Frenkel, Stephen J. and Carol Royal 1997, ‘Globalization and Employment Relations’, in Research in Sociology of Work: The Globalization of Work, edited by Randy Hodson, pp. 3–42, Stamford, CT: jai Press Inc.
Galanis, Giorgos and Ashok Kumar 2021, ‘A Dynamic Model of Global Value Network Governance’, EPA: Economy and Space, 53, 1: 53–71.
Gereffi, Gary 2014, ‘Global Value Chains in a post-Washington Consensus World’, Review of International Political Economy, 21, 1: 9–37.
Gereffi, Gary and Karina Fernandez-Stark 2016, Global Value Chain Analysis: A Primer, pp. 1–34, Durham, NC: Duke University Press.
Gereffi, Gary, John Humphrey and Timothy Sturgeon 2005, ‘The Governance of Global Value Chains’, Review of International Political Economy, 12, 1: 78–104.
Goodwin, Richard 1967, ‘A Growth Cycle’, in Socialism, Capitalism, and Economic Growth, edited by Carl Feinstein, Cambridge: Cambridge University Press.
Greenhill, Brian, Layna Mosley and Aseem Prakash 2009, ‘Trade-based Diffusion of Labor Rights’, American Political Science Review, 103, 4: 669–690.
Henderson, Jeffrey, Peter Dicken, Martin Hess, Neil Coe and Henry Wai-Chung Yeung 2002, ‘Global Production Networks of Economic Development’, Review of International Political Economy, 9, 3: 436–463.
Herod, Andrew 1997, ‘From a Geography of Labor to a Labor Geography: Labor’s Spatial Fix and the Geography of Capitalism’, Antipode, 29, 1: 1–31.
Hobsbawm, Eric J. 1952, ‘The Machine Breakers’, Past & Present, 1: 57–70.
Iliopoulos, Panagiotis Takis, Giorgos Galanis, Ashok Kumar and Lilit Popoyan 2022, ‘Sectoral Market Power in Global Production: A Theoretical and Observational Study’, Advances in Complex Systems, 25: 1–48.
ILO 2005, Promoting Fair Globalization in Textiles and Clothing in a Post-MFA Environment, Geneva: ilo.
Jackson, Gregory, Virginia Doellgast and Lucio Baccaro 2018, ‘Corporate Social Responsibility and Labour Standards: Bridging Business Management and Employment Relations Perspectives’, British Journal of Industrial Relations, 56, 1: 3–13, <doi:10.1111/bjir.12298>.
Jackson, Matthew O. 2008, Social and Economic Networks, Princeton, NJ: Princeton University Press.
Kalecki, Michał 1971, Selected Essays on the Dynamics of the Capitalist Economy, Cambridge: Cambridge University Press.
Katz, Harry C. 1997, ‘Industrial Relations in the US Automobile Industry: An Illustration of Increased Decentralization and Diversity’, The Economic and Labor Relations Review, 8, 2: 192–220.
Knorringa, Peter and Lee Pegler 2006, ‘Globalisation, Firm Upgrading and Impacts on Labour’, Tijdschrift voor economische en sociale Geografie, 97, 5: 470–479.
Kumar, Ashok 2019a, ‘Oligopolistic Suppliers, Symbiotic Value Chains and Workers’ Bargaining Power: Labour Contestation in South China at an Ascendant Global Footwear Firm’, Global Networks, 19, 3: 394–422.
Kumar, Ashok 2019b, ‘A Race from the Bottom? Lessons from a Workers’ Struggle at a Bangalore Warehouse’, Competition & Change, 23, 4: 346–377.
Kumar, Ashok 2020, Monopsony Capitalism: Power and Production in the Twilight of the Sweatshop Age, Cambridge: Cambridge University Press.
Kumar, Ashok 2024, ‘When Monopsony Power Wanes. Part One: Material Conditions’, Historical Materialism, 31, 4: 3–49.
Lakhani, Tashlin, Sarosh Kuruvilla and Ariel Avgar 2013, ‘From the Firm to the Network: Global Value Chains and Employment Relations Theory’, British Journal of Industrial Relations, 51, 3: 440–472, <doi:10.1111/bjir.12015>.
Lester, Richard 1967, ‘Pay Differentials by Size of Establishment’, Journal of Industrial Relations, 7, 1: 57–67.
Lund-Thomsen, Peter and Adam Lindgreen 2018, ‘Is There a Sweet Spot in Ethical Trade? A Critical Appraisal of the Potential for Aligning Buyer, Supplier and Worker Interests in Global Production Networks’, Geoforum, 90: 84–90.
Mahutga, Matthew C. 2012, ‘When Do Value Chains Go Global? A Theory of the Spatialization of Global Value Chains’, Global Networks, 12, 1: 1–21.
Majeed, Ishfaq 2001, ‘An Analysis of the Labor Process Theory: From Marx to Post Braverman Debate’, Arabian Journal of Business and Management Review, 11, 2, available at: <https://www.hilarispublisher.com/open-access/an-analysis-of-the-labor-process-theory-from-marx-to-post-braverman-debate.pdf>.
Marx, Karl 1990 [1867], Capital: A Critique of Political Economy. Volume One, translated by Ben Fowkes, London: Penguin.
Moody, Kim 2017, On New Terrain, Chicago: Haymarket Books.
Mosley, Layna and Saika Uno 2007, ‘Racing to the Bottom or Climbing to the Top? Economic Globalization and Collective Labor Rights’, Comparative Political Studies, 40, 8: 923–948.
oecd 2017, Skills Outlook, 2017: Skills and Global Value Chains, Paris: oecd.
Onaran, Özlem and Giorgos Galanis 2014, ‘Income Distribution and Growth: A Global Model’, Environment and Planning A: Economy and Space, 46, 10: 2489–2513.
Piketty, Thomas 2013, Capital in the 21st Century, translated by Arthur Goldhammer, Cambridge, MA: Harvard University Press.
Posthuma, Anne and Dev Nathan 2010, ‘Introduction’, in Labor in Global Production Networks in India, pp. 1–36, New Delhi: Oxford University Press.
Rangaraju, M.S. and S. Hanuman Kennedy (eds.) 2012, Innovation in Management Challenges and Opportunities in the Next Decade, Bangalore: pes Institute of Technology.
Raskin, A.H. 1982, ‘David Dubinsky – A Reporter’s Memoir’, New York Times, 19 September, available at: <https://www.nytimes.com/1982/09/19/weekinreview/david-dubinsky-a-reporter-s-memoir.html>.
Reinecke, Juliane, Jimmy Donaghey, Adrian Wilkinson and Geoffrey Wood 2018, ‘Global Supply Chains and Social Relations at Work: Brokering Across Boundaries’, Human Relations, 74, 4: 459–480.
Riisgaard, Lone and Nikolaus Hammer 2011, ‘Prospects for Labor in Global Value Chains: Labor Standards in the Cut Flower and Banana Industries’, British Journal of Industrial Relations, 49, 1: 168–190.
Rudrappa, Sharmila 2012, ‘India’s Reproductive Assembly Line’, Contexts, 11, 2: 22–27.
Sawyer, Malcolm C. 1988, ‘Theories of Monopoly Capitalism’, Journal of Economic Surveys, 2, 1: 47–76.
Seigmann, Karin Astrid, Peter Knorringa and J.J.S. Merk 2014, Voluntary Initiatives in Global Value Chains: Towards Labor-led Social Upgrading?, ciri Working Paper № 4, available at: <https://www.researchgate.net/publication/280935507_Voluntary_initiatives_in_global_value_chains_Towards_labour-led_social_upgrading>.
Selwyn, Ben 2008, ‘Bringing Social Relations Back In: (Re)conceptualizing the “Bullwhip Effect” in Global Commodity Chains’, International Journal of Management Concepts and Philosophy, 3: 156–175.
Selwyn, Ben 2012, ‘Beyond Firm-centrism: Re-integrating Labour and Capitalism into Global Commodity Chain Analysis’, Journal of Economic Geography, 12, 1: 205–226, <doi:10.1093/jeg/lbr016>.
Selwyn, Ben 2013, ‘Social Upgrading and Labor in Global Production Networks’, Competition and Change, 17, 1: 75–90.
Shaikh, Anwar 2016, Capitalism: Competition, Conflict, Crises, Oxford: Oxford University Press.
Silver, Beverly 2003, Forces of Labor: Workers’ Movements and Globalization since 1870, New York: Cambridge University Press.
Skott, Peter 1989, Conflict and Effective Demand in Economic Growth, Cambridge: Cambridge University Press.
Stockhammer, Engelbert 2015, ‘Determinants of Wage Share: A Panel Analysis of Advanced and Developing Economies’, British Journal of Industrial Relations, 55, 1: 1–33.
Tarrow, Sydney 1998, ‘Fishnets, Internets, and Catnets: Globalization and Transnational Collective Action’, in Challenging Authority: The Historical Study of Contentious Politics, edited by Michael P. Hanagan, Leslie Page Moch and Wayne te Brake, pp. 228–244, Minneapolis: University of Minnesota Press.
Taylor, Lance 2004, Reconstructing Macroeconomics: Structuralist Proposals and Critiques of the Mainstream, Cambridge, MA: Harvard University Press.
Tewari, Manish 2010, ‘Footloose Capital, Intermediation, and the Search for the “High Road” in Low-wage Industries’, in Labor in Global Production Networks, edited by Anne Posthuma and Dev Nathan, pp. 146–165, New York: Oxford University Press.
Tilly, Charles 1995, ‘Globalization Threatens Labour’s Rights’, International Labour and Working-Class History, 47: 1–23.
unctad 2011, ‘Non-equity Modes of International Production and Development’, in World Investment Report, New York and Geneva: United Nations Organization, available at: <https://unctad.org/system/files/official-document/wir2011_en.pdf>.
Weinstone, William 1946, The Case Against David Dublinsky, New York: New Century Publishers, available at: <https://www.marxists.org/subject/jewish/case-dubinsky.pdf>.
Wright, Erik Olin 2000, ‘Working-Class Power, Capitalist-Class Interests, and Class Compromise’, American Journal of Sociology, 105, 4: 957–1002.
Raskin 1982.
Weinstone 1946, p. 13.
Kumar 2024.
Anner, Bair and Blasi 2012, p. 22.
Anner, Bair and Blasi 2012, p. 5.
Much analysis of labour in the garment gvc assumes a high dmp and therefore a low structural and associational power for workers. Therefore, the only method deployed by workers has been to protest outside the workplace. Anner 2011, p. 16, explains: ‘High worker turnover rates result in weak structural and associational (organizational) power for labour (Silver 2003; Wright 2000). But workers are not powerless. First, as Sydney Tarrow suggests, resource-poor actors often turn to protest because disruption is a form of power (Tarrow 1998). At the same time, the harsh conditions in buyer-driven value chains allow grievances to be framed in terms of basic human rights that resonate with broader publics. Since control lies on the retail and brand-name manufacturer end of this commodity chain, left labour unionists will have an incentive to form alliances with activist organizations such as women’s groups, human and labour rights organizations, and student organizations and can maximise the shaming mechanisms as they pressure leading apparel firms through consumer-oriented campaigns.’
Marx 1990.
Braverman 1974.
Ibid.
See Kumar 2024.
oecd 2017.
ilo 2005.
Gereffi, Humphrey and Sturgeon 2005.
Henderson, Dicken, Hess, Coe and Yeung 2002; Dougherty 2008; Coe and Wai-Yeung 2015.
Frenkel and Kuruvilla 2002.
Riisgaard and Hammer 2011.
Lakhani, Kuruvilla and Avgar 2013.
Selwyn 2013.
Reinecke, Donaghey, Wilkinson and Wood 2018.
Jackson, Doellgast and Baccaro 2018.
Wright 2000.
Cumbers, Nativel and Routledge 2008; Flecker, Haidinger and Schönauer 2013; Selwyn 2012.
Riisgaard and Hammer 2011; Shaikh 2016.
Silver 2003, p. 13.
unctad 2011, p. xii.
Kumar 2019a, 2019b, 2020.
Coe and Wai-Yeung 2015, p. 66.
Iliopoulos, Galanis, Kumar and Popoyan 2022.
Silver 2003.
Jackson 2008, p. 62.
Freeman 1978; Jackson 2008.
Selwyn 2008, p. 157.
Galanis and Kumar 2021.
Majeed 2021.
Tilly 1995, p. 7.
Hobsbawm 1952.
Tilly 1995, p. 13.
Tilly 1995, p. 8.
Interview, 12 February 2019.
Marx 1990.
Katz 1997.
Mahutga 2012, p. 163.
Onaran and Galanis 2014.
Stockhammer 2015.
Dutt 2016; Chiarella and Flaschel 2000; Chiarella, Flaschel and Franke 2005; Carpe, Chiarrella, Flaschel and Semmler 2011; Skott 1989; Shaikh 2016; Taylor 2004.
Kalecki 1971.
Sawyer 1988, p. 55.
Goodwin 1967.
Coe, Hess and Dicken 2008; Selwyn 2012; Gereffi 2014; Knorringa and Pegler 2006.
Riisgaard and Hammer 2011, p. 5.
Herod 1997, p. 46.
Lester 1967; Brown and Madoff 1989; Burdett and Mortensen 1998.
Barrientos, Gereffi and Rossi 2011.
Gereffi and Fernandez-Stark 2016.
Posthuma and Nathan 2010.
Tewari 2010.
Carswell and De Neve 2013.
Selwyn 2013, pp. 83–4.
Seigmann, Knorringa and Merk 2014, p. 19.
Fernandez-Stark, Frederick and Gereffi 2011.
Frenkel and Royal 1997; Rangaraju and Kennedy (eds.) 2012.
Dicken 2007.
Sharmila Rudrappa describes the relationship between garments and the surrogacy industry in India: ‘In Bangalore, the garment production assembly line is the main conduit to the reproduction assembly line, as women move from garment factories, to selling their eggs, to surrogacy’ (Rudrappa 2012, p. 23). In a further indictment of garment-sector conditions, women in the reproduction assembly line describe it as ‘more meaningful’ and ‘creative’ than that of the garment factory (ibid.).
Interview, gatwu organisers, 24 January 2024.
Interview, 20 February 2019.
For example, in Pune the factory is represented by the far-right Shiv Sena union.
Interview with Jayram, 20 February 2019.
Kumar 2019a.
Moody 2017, pp. 49–50.
Moody 2017, p. 56.
Lund-Thomsen and Lindgreen 2018, p. 87.
Shaikh 2016, p. 751.
Marx 1990, p. 799.
There is much recent scholarship, namely Thomas Piketty’s Capital in the Twenty-First Century (2013), for which, contrary to dominant supply-side logics, the rate of return is greater than the rate of economic growth, which results in increased concentration of wealth. While this is a sturdy thesis upon which to base larger claims about capitalism and wealth disparity, the focus of my own work is on the bargaining power of workers at the firm level.
Kumar 2019b.
Selwyn 2008.
Selwyn 2008, p. 164.
Selwyn 2008, p. 194.
Selwyn 2008, p. 167.
Butollo 2014, p. 361.
See Mosley and Uno 2007, Greenhill, Mosley and Prakash 2009, and most recently Adolph, Quince and Prakash 2017 on the Shanghai effect.