Notes on a factory uprising in Yangon

Yangon H&M factory
Yangon H&M factory

Soe Lin Aung examines a recent strike-cum-riot at a Chinese-owned H&M supplier in Myanmar, digging beyond the headlines into its local context and broader political significance.

Submitted by Steven. on May 10, 2017

A Riot Foretold

Last month, a strike turned sour on the outskirts of Yangon, the former capital of Myanmar and its leading production base, when workers briefly laid siege to a Chinese-owned garment factory. On February 23, hundreds of workers descended on the Hangzhou Hundred-Tex Garment factory, damaging factory vehicles, breaking windows, wrecking machinery, attacking the Chinese manager, and taking seven Chinese supervisors hostage. After a police rescue operation initiated by the Chinese embassy, the hostages were freed, and five workers were arrested. This riot, as media accounts put it, followed a months-long strike for 15 months of unpaid overtime that had resulted in the factory’s union leader being fired. A worker reported that in the throes of the uprising, “most of the workers could not control themselves.” An outside union leader said the workers “saw that nothing was happening – it wasn’t going anywhere – so they exploded.”
Accounts of what happened are not fully consistent. Multiple reports state that footage shows “dozens” of women workers assaulting the Chinese manager, yet workers dispute that account, denying as well that they destroyed production machinery. A few things are clear, however. Amid an expansion of Myanmar’s garment industry after post-2011 economic reforms, militant worker actions have periodically convulsed Yangon’s industrial zones. These largely autonomous wildcat strikes tend to focus on wages in their formal demands, and in some cases they have lasted for months. After roughly three strike waves up to 2015, a collection of more violent incidents has taken place since then, including this riot in February. Labor unions, newly legalized since 2011, are struggling to catch up, with civil society more broadly pushing to centralize these struggles – namely under hierarchical union structures designed for tripartite labour arbitration. Worker distrust of Myanmar’s major labour federations runs high, however,[1] as does frustration with the new minimum wage law, which workers say employers are circumventing by cutting overtime pay and bonuses.
One possibility to consider, then, to paraphrase the union leader above: workers are exploding, because they’ve seen that nothing is happening. Or differently: they’ve seen the contradictions that are emerging, and they are intensifying their struggles accordingly. The riot last month follows at least two other industrial conflicts described as riots over the past couple years. In March 2015, at the tail end of one of the longest strikes begun earlier that year, about 100 workers from three factories attempted to escalate by marching to Yangon’s city hall. Blocked by police, they instead began a sit-in that ended with a clash between workers, police, and hired vigilantes – civilians wearing red armbands bearing the word “duty” (da-wun) in Burmese. 14 workers were formally charged with rioting. The subsequent year, workers from a factory in Sagaing, in Myanmar’s dry zone well north of Yangon, marched for weeks at the height of the hot season to the periphery of Naypyidaw, Myanmar’s capital and an army stronghold. A confrontation ensued when police intervened to stop workers entering the city, and here too workers were charged with rioting.
These events suggest continuities with struggles farther afield. In China and Vietnam, observers often attribute the prevalence of industrial riots to these countries’ lack of civil liberties and independent unions, which are expected to facilitate more orderly forms of collective bargaining.[2] The coexistence of such liberal institutions with increasingly frequent struggles of a similar nature in neighboring countries such as Myanmar and Cambodia[3] compels us to look beyond such explanations toward broader dynamics of contemporary capitalism in the region—and, perhaps, beyond. How does last month’s factory riot in Yangon relate to what some have called a new “era of riots” at global, regional and local scales?
Locating Hangzhou Hundred-Tex

The Hangzhou Hundred-Tex (HHT) factory is situated in Hlaingtharyar township on Yangon’s northwest periphery, across the Yangon River where speculation is driving up land prices as the area industrializes rapidly. Although Hlaingtharyar is often spoken of as an industrial zone itself, it is also a township that contains many smaller industrial zones, like the Shwe Lin Ban Industrial Zone where the HHT factory operates. The other major industrial zones around Yangon are in Shwepyithar and Hmawbi, also to Yangon’s north and northeast.
One study estimated that wages in these zones averaged $95-125 USD per month in 2015.[4] But a subsequent survey of 26 workers estimated base pay closer to 30,000-40,000 kyat per month ($21-32) and overall pay at about 95,000 kyat per month ($69) – a striking indicator of the importance of bonuses and overtime to workers’ overall take-home pay.[5] Among those workers, the survey also found that working hours came to around 10.5-11 hours per day, six days per week; that no workers had any form of contract, much less health or social insurance; and that only six of those workers said they could save up any money after meeting their daily needs. One worker said she usually breaks even, but the remaining 19 described debts of 30,000-500,000 kyat. 17 of those in debt are paying 20% interest or above. As the survey makes clear: “High-interest debt is a problem for anyone, but more pronounced for young women who are expected to remit to support household social reproduction, according to numerous workers interviewed.”[6]
Hlaingtharyar officially hosts the highest and densest population of any township in Myanmar, but this is not obvious when you drive through it. Along the main factory thoroughfares, every so often a rolling iron gate signals the existence of a factory behind it, but the scale and type of the factories themselves, which are almost always single-story structures, are nearly impossible to discern from outside. A small shop or two might be selling something at the gate, perhaps tea mix packets, betel nut or simple salads for workers on their breaks. But at least in the larger factories, meals – and accommodation – are provided inside the factory grounds. In contrast to the austere, relatively quiet, and almost modernist feel of major swathes of the zone, pockets of busier, more frenetic activity are concentrated around market areas where food stalls, beer stations, small goods shops, rickshaw stands and the like spill out seemingly at random. Informal housing settlements tend to cluster around these areas, with migrants often building thatched shelters on grassy marshlands that punctuate the zone.
Although individual factories are, no doubt, very much gated, it is hardly clear when one has entered or exited the industrial zone itself. Around 2012, to cruise the zone in a car or a rickshaw was to be met, at intervals, with sudden vistas opening onto the distance – visions of rice fields beyond the zone or green patchworks within where, presumably, it was only a matter of time before migrants occupied the space or a new factory was built. Sometimes the one follows the other, and migrants’ homes are cleared away, as in one struggle over a patch of land in Hlaingtharyar that developed into a violent confrontation between workers and police. But whatever fissures might have splintered the zone in the past, perhaps lightening its carceral burden, now surely its industrial character has become much more tightly consolidated. From above, it appears today as an expansive grid of small rectangles, a latticework of aluminum roofs, a modernist dream that, as ever, remains subject – as the February riot makes clear – to instability and excess.
Indeed, Myanmar’s garment sector has changed considerably since political and economic reforms beginning in 2011 ushered in a series of trade liberalization measures. Starting from a low base due to US and European sanctions during the military period, garment exports jumped from $349 million in 2010 to $1.56 billion in 2014. In 2014, new garment factories were opening on average once per week, mainly in the zones around Yangon like Hlaingtharyar. As of 2015, about 210 export-oriented garment factories were operating in Myanmar, employing around 195,000 workers, with a further 100,000 engaged in small-scale and/or home-based subcontracting outside of larger factory enterprises.[7] One study estimates that 93 percent of garment workers are women, particularly young, single, “unskilled” women from rural areas along the Ayeyawaddy littoral, from the dry zone in the north to the delta region in the south.[8]
There are various ways of classifying Myanmar’s garment factories, but a 2013 report cited by the International Labour Organization (ILO) works well as a guide, however schematic.[9] Three categories are in play. First, there are the largest factories, employing over 1,000 workers with either full foreign ownership or a joint venture agreement. Then there are the basically medium-level factories with between 500 and 1,000 workers, usually joint ventures or locally owned. Finally, there are the smallest factories, operating with fewer than 500 workers and likely full local ownership. In 2013, the smallest kind of factories was the largest by number, with about 160 operating at the time. Only around 30 of the medium-level factories were operating, and about 20 of the largest kind. Despite what seems like a preponderance of locally owned factories, “shadow companies” make up some 20-30% of those factories, with a Myanmar citizen signing on behalf of an essentially covert foreign owner.[10]
Foreign direct investment (FDI), in fact, has provided the stimulus for the expansion of Myanmar’s garment industry. Since the end of formal military rule in 2011, the fraction of fully foreign-owned factories almost doubled to about 50% by 2015, with the other half either joint ventures or locally owned.[11] East Asian capital dominates here, unsurprisingly, with Korean FDI leading the way, followed by China, Taiwan and Japan. Firms from these countries continued to produce in Myanmar during the US and European sanctions regimes, giving them a strong advantage now that sanctions have been lifted (in 2013 by the EU, and then in 2016 by the US). In 2014, with the total export value of garments pushing $1.6 billion, about 70% of all exports went to Korea and Japan. With these firm East Asian export markets, the low-end manufacturing that dominates production in Bangladesh and Cambodia – jeans, t-shirts, and other ready-made garments – has been largely absent in Myanmar. Japanese firms sourcing in Myanmar, for instance, have prioritized smaller and higher-end orders of items like woven products. This may change, according to a recent study, as exports to the EU by companies like H&M increase.[12]
The HHT factory attacked by workers in February encapsulates much of what has been reported about the growth of Myanmar’s garment sector. Opened two years ago by its parent company based in China’s Zhejiang province, the factory operates exclusively as a supplier to H&M, as media reports have stressed since this fact came to light. Fully foreign-owned, apparently, the factory is in that middle range in employing about 500 workers, almost all of whom are women – at least if some limited footage of the strike, shown by Al-Jazeera, is any measure.
Striking workers at Yangon HHT. Photo: DVB
However, one trend the HHT factory reflects that has not, it seems, been widely considered in Myanmar is that of deindustrialization in China and the consequent southward flow of Chinese capital. That is, as one of the first reports on the strike noted, the HHT factory’s owners set up operations in Myanmar in search of reduced labour costs. (The same article also cites mounting labour unrest in China and South Korea as a related reason for this new push south.) Media reports emphasize that Myanmar’s minimum wage law, roundly disparaged by workers and employers alike, sets the basic wage at 3,600 kyat per day, or about $2.62. Among garment producers in Asia, only Bangladesh offers lower wages. As noted in the survey discussed above, it is also clear that base pay at many factories may be below the mandated minimum, with only bonuses and overtime pay lifting wages to around $70 per month.[13] This helps make clear the very real stakes of the HHT riot, which originated in a strike to regain fully 15 months of overtime pay – in other words, a massive portion of what workers would have expected to earn in that time.
To be sure, capital in general tends to move towards lower wages and fewer worker protections, in articulation with specific histories of colonialism, postcolonialism, and related state claims on capitalist development.[14] Yet the particular flight south of Chinese manufacturing capital has been especially noteworthy – if not yet in Myanmar apparently. Investment in Cambodia and Vietnam has shown clear signs of this push, as well as beyond Southeast Asia. But the still relatively small scale of Myanmar’s garment sector, and the fact that Korea outpaces Chinese investment within it, may explain the more limited discussion of this potentially transformative process in Myanmar.
Riot Time
But we should return to the question posed, if read a certain way, by the union leader commenting on the February riot. Are workers exploding? There were the strike waves in 2012, 2013, and 2015. And then there were the three messier, more violent incidents like the riot at the HHT factory: the riot itself, of course, as well as the violence that erupted at the end of the 2015 strike wave and after the long workers march the following year. These events suggest the possibility, at least, that worker struggles are intensifying in ways that may be difficult to predict.
Several potential directions for analysis emerge here. First is the link, currently tenuous, between two changing landscapes in Myanmar: that of worker struggles and trade unions. As discussed, it may be the case that worker struggles are beginning to push beyond strikes. Each of the three incidents described as riots followed strikes of differing lengths of time. Furthermore, even the strikes that have intermittently paralyzed production in Yangon’s industrial zones have been largely autonomous and decentralized. Unions, NGOs, and civil society groups have tended to see these actions as regrettably disorganized, lacking in leadership and coordination, and like Myanmar’s fast-evolving labour movement more broadly, in need of more centralized authority – unions, in short – to effectively engage evolving tripartite arbitration processes. Upwards of 2,000 unions have registered since a 2011 law repealed a military-era law that prevented the formation of independent unions.[15] Among the large federations to have formed are the Confederation of Trade Unions of Myanmar (CTUM), with roots in Myanmar’s exiled political movement; the Myanmar Trade Union Federation (MTUF); and the Agriculture and Farmer Federation of Myanmar (AFFM-IUF).
Yet workers remain distrustful of the large federations. A leader of one of the 2015 strikes said the federations only come around during strikes looking to gain membership. “They just want to go to the ILO conferences,” he said.[16] CTUM in particular, headed by the controversial leader Maung Maung, has come under heavy criticism of this kind, as well as for its top-down, hierarchical, cronyistic approach more generally. On the other hand, workers have shown more support for enterprise-level unions, seeing them not as outsiders like the federations but as more democratic structures arising from workers’ own initiative.[17]
Still, militant incidents in and beyond Yangon have only limited relations to trade unions and are hardly restricted to strikes – as the liberal hand-wringing of unions and NGOs attests. Thinking with these struggles, we might see them as pushing the boundaries of what – and who – actually constitutes a labour movement, challenging assumptions about who represents workers and how, and what tactics might be necessary beyond the strictures of bureaucratic dispute mechanisms. It is tempting even to call into question tripartism itself, which comes out of a particular Western Fordist trajectory. Nonetheless, liberal public discourse, from the ILO to employer organizations, holds that the upheaval in Yangon’s industrial zones is simply a sign of the growing pains of tripartism. The message is clear: it will all settle down soon.
Discussions at a recent workshop in Cambodia help point up what is at stake.[18] A labour activist stressed that in Europe and the US, a kind of organic development of workers’ politics and institutions had been able to take place, with strikes, trade unions, and tripartism evolving to suit particular conflicts with capital. Workers’ movements in Asia, however – many of which are now well-established rather than being in any sense emergent – face a constant didactic barrage regarding the “correct” way to operate. Such disciplining comes willy nilly from global union federations, Western trade unions preaching solidarity, labour NGOs, labour movement donors, and no doubt, researchers and scholars of labour, labour politics, and labour history. Against that stifling pedagogy, the autonomous development of worker struggles in Myanmar – in all their visceral, chaotic “disorganization” – might suggest something more open, something more charged with possibility. In some ways they should be welcomed, and the rush to impose institutional forms, particularly suffocating hierarchical union structures, should be avoided. There may even be a certain historical justice in doing so. The riot, in this sense, may stand for another kind of labour politics within a more differential, more flexible approach to labour history.
And here perhaps is a broader, more conceptual direction to explore. That is, we could name this differential account of labour and labour history, and we might do well to begin with the tactic here associated with it: riot time. The phrase has a certain history. A recent Chuang article cites Jason Smith’s argument that a series of incidents over the last decade or so – the banlieu riots in France, the youth riots in London, anti-austerity strikes in Greece, the movement of the squares, and the uprisings across the Arab World – together signify a new era of political possibility. “We are awakening,” he writes in 2012, “from the neoliberal dream of global progress and prosperity: after forty years of reaction, after four decades of defeat, we have re-entered the uncertain stream of history. We bear witness to a new cycle of struggles; ours is a time of riots.” The claim echoes Alain Badiou, who in the same year argued that the Arab revolts announced the “rebirth of history” in these “times of riots and uprisings.” The Chuang article asks whether the proliferation of “mass incidents” in China could be legible in this vein, which for Smith suggests a return to the conditions that sparked the revolutionary struggles of the 19th century. Those conditions, in short, found socialist and utopian movements failing to address the direct needs of workers, who increasingly responded with violence to profound transformations of the production process.
This notion of a turn, or return, to the time of riots allows us to clarify something regarding the specificity of Fordism and its political appendages (the union, the strike, tripartism, and so on). The point is not that workers’ turn towards riots in places like Myanmar and China index a singularity that prevents common cause with workers elsewhere. Rather, one recognizes the provincialism of a Fordist trajectory as well, beyond which even in the West there had always been – and surely are today – a much wider range of political possibilities in play. These include the revolutionary movements that preceded bureaucratic tripartism and its restricted fields of dispute. Riots, then, allegedly deviant or deficient vis-à-vis more strategic and unified labour movements, may actually express something very central to capitalism’s transformations of production over time and in many places, namely the turbulence of those transformations and the limits of organized politics in addressing them.
After “Freedom”
Finally, from China to Myanmar and elsewhere in Southeast Asia, an apparently violent turn in labour politics invites comparison. We might usefully track, for instance, different levels of protection for civil liberties, especially freedom of association and the legality of independent unions, in China, Vietnam, and Myanmar. A widespread belief across the political spectrum seems to be that limits to such liberties go a long way toward explaining why labour struggles escalate as they do in China and Vietnam, and that consequently, the expansion of such liberties would stem this tendency for strikes to become riots.[19]
Yet strikingly, in Myanmar it is precisely in the context of sweeping liberal reforms that an already increasingly militant workforce is pushing the limits of union-mediated tripartism, and showing signs of more violent forms of struggle. That is, a sudden and thoroughgoing liberalization process has not only not stemmed the radicalization of labour politics, one could argue that exactly the opposite has occurred. Then again, liberal actors like the ILO argue it’s just a matter of time before the landscape stabilizes and workers fall into line. And to be fair, while the historical antecedents of Myanmar’s labour regime require serious reckoning, as Dennis Arnold and Stephen Campbell argue persuasively,[20] the novelty of its present form would be hard to overstate: a raft of new legislation, over two thousand new unions, no shortage of new investors, and probably most significantly, thousands upon thousands of new workers. Myanmar’s garment factories are filled with workers sending remittances to rural households, many of whom will be the first in their families to work for a wage.
But this all hardly means that “more freedoms” is a workable prescription for de-escalating labour struggles – a suspect program in any case – as if the reforms in Myanmar will ultimately produce a more pliant workforce than in China or Vietnam. Rather, one could just as easily read mounting worker militancy in Myanmar as pointing out that signature fissure in liberal polities, increasingly fraught since 2008: between the formal equalities of politics and the spiraling inequalities of the economy. That dissonance can be deeply felt, particularly when technocratic liberalization measures overhaul production processes – and upend the lifeworlds of workers – while eluding democratic control. Some argue that after neoliberalism’s eclipse now, that tension between being formally free yet relentlessly exploited has produced terrifying political movements. Pankaj Mishra, for one, writes that “Never have so many free individuals felt so helpless – so desperate to take back control from anyone they can blame for having lost it.” Here, maybe, is an altogether darker, more ominous reading of the cycle of events that Smith and Badiou, among others, saw in almost messianic terms. In other words, much has changed since 2012, not least in the Arab World, where the time of riots – history reborn – had reached its apotheosis.[21]
This essay only scratches the surface these questions, but one lesson we can draw for sure is an old one: the rule of capital produces contradictions, which in certain times may become plain to see – and feel. The violent transformation of the production process in Myanmar might be one of those times. Liberal freedoms can do little to mask, mitigate, or obscure the contradictions that are emerging. The form those contradictions take, the nature of their development, and the shape of the society to be forged in their wake are all eminently political questions. For better or for worse, those questions may be posed in riot time. Bending this process in one direction or another – towards militant labour struggles rather than revanchist right populisms – might be one of the more central political tasks in the present, determining how and on whose terms history is being reborn.
–Soe Lin Aung, 30 March 2017
Special thanks to Dennis Arnold and Stephen Campbell for allowing their forthcoming articles to be used here. This discussion draws heavily upon them. Any shortcomings remain my own.
[1] Arnold, Dennis (forthcoming). “Civil Society, Political Society, and Politics of Disorder in Cambodia.” Political Geography. 2017.
[2] This is a thesis of Eli Friedman’s book The Insurgency Trap: Labor Politics in Postsocialist China (Cornell 2014), for example, along with many statements by the China Labour Bulletin.
[3] Arnold (forthcoming).
[4] SMART Myanmar (2015). Responsible Sourcing in a Frontier Market. Yangon: SMART Myanmar.
[5] Arnold, Dennis and Stephen Campbell (forthcoming). “Labour Regime Transformation in Myanmar: Constitutive Processes of Contestation.” Development and Change. 48(4). July 2017.
[6] Ibid.
[7] Ibid.
[8] SMART (2015).
[9] International Labour Organization (ILO) (2015). “Myanmar Garment Sub-sector Value Chain Analysis.” Yangon: ILO.
[10] SMART (2015).
[11] Arnold and Campbell (forthcoming).
[12] Ibid.
[13] Ibid.
[14] For an influential contribution to studies of the relation between capitalist development and postcolonial politics, see Sanyal, Kalyan (2007). Rethinking Capitalist Development: Primitive Accumulation, Governmentality, and Post-Colonial Capitalism. New Delhi: Routledge.
[15] Arnold and Campbell (forthcoming).
[16] Ibid.
[17] Ibid.
[18] The workshop, held in Phnom Penh on February 16-17, 2017, was titled “Labor Rights and Economic Development in Mekong Asia.”
[19] See, for example, Friedman (2014).
[20] Arnold and Campbell (forthcoming).
[21] On the global rise of right-wing mass movements from the ashes of 2011 (with a focus on Thailand), also see “The Solstice” by Ultra.
Taken from