Governing Global Garment Production (Pt. 2)

In my last post, I argued that the two Bangladesh fire and safety agreements reflect the failures of existing legal regimes, both public and private. The agreements intend to impose some basic standards on the Bangladeshi garment industry; rather than competing (almost) entirely on the basis of cost, as existing law encourages them to do, brands will collaborate to ensure that suppliers’ factories are safe. In this post, I’ll defend the Accord (signed between most major European brands and various unions) as substantively preferable to the Alliance (an agreement among U.S. brands), because the Accord has the potential to reshape the political economy of global garment production and governance.

To see why, note that both agreements invite a question common in the sociology of law: what explains the emergence of particular legal regimes at particular times? As Professor Alan Hyde has pointed out, this question is surprisingly understudied in the international labor field. Hyde proposes that transnational labor governance institutions emerge to solve coordination problems when all participants “would gain by cooperation but will be disadvantaged if their rivals defect.” That theory has some traction here. Nations have a common interest in minimizing child labor, but no nation can do so alone; brands have a common interest in ensuring minimally safe factories, but must cooperate to do so. Once the failures of existing governance regimes became clear after the Rana Plaza and Tazreen Fashions disasters, brands needed a new system. So they built one.

But such an account is incomplete—as Hyde would surely agree—for it disregards the need for a prime mover to build trust among disparate actors. I’ve addressed this question in the context of domestic labor organizing, arguing that organizers actively shape workers’ preferences and build solidarity, in part by creating moral crises (read: strikes) in which workers rely on each other out of necessity. As it turns out, unions and workers’ rights NGOs played a central role in the shaping of the Accord, and then, by extension, the Alliance.

The Workers’ Rights Consortium and other unions and NGOs have long argued that that the solution to basic standards noncompliance in the garment sector lay in binding commitments by brands to ensure safe factories and to pay more for garments. Indeed, as I understand, the basic terms of the Accord had been drafted well before the 2012-13 disasters, but the unions were unable to press enough brands to sign on until after those disasters.

This echoes Richard Locke and the NYU Stern School’s diagnosis of the problem, as discussed in the prior post: brands’ practices need to change for supplier working conditions to change. But the WRC and the unions’ approach differs from Locke’s and the Stern School’s. Locke argues that states should step in to regulate labor standards, especially freedom of association, while the Stern School proposes collaboration among the state, the Alliance and Accord, and other actors in Bangladesh to ensure safety. But neither Locke nor the Stern School considers in detail what political and institutional arrangements would make that collaboration possible.

The Accord, in contrast, reflects an approach to the political economy of labor regulation that a mentor of mine calls “the dialectic of strikes and standards.” Put simply, unions often build power through concerted action that pushes against or works around existing legal regimes; the power they build may then enable unions to enact favorable laws or to reach contractual agreements with employers/capital; such agreements and laws then require concerted action, or the threat thereof, to enforce.

Now, this can be a virtuous or a vicious cycle—the key is the interrelationship between law, concerted action, and unions’ and others’ power. Over time, as unions and capital play this game repeatedly, and depending on other background political dynamics, the result may be high-trust, collaborative forms of production (see Germany; Locke’s vision for garment manufacturing) or constant low-level battles amid low-trust, high-conflict forms of production (see U.S. labor law; most garment manufacturing today).

Viewed in this light, the Accord has clear advantages over the Alliance. It will likely be more effective in the long term at improving working conditions, and it may also help enhance global distributive justice. Three provisions of the Accord are particularly important.

First, the Accord requires brands to negotiate terms with their suppliers which “ensure that it is financially feasible for the factories to maintain safe factories” and to upgrade facilities where necessary. The Alliance has no such commitment. Rather, Alliance members commit to provide low-cost capital to factory owners to finance repairs.

Second, under the Accord, disputes between suppliers, brands, and/or workers’ representatives can be submitted to binding arbitration, enforceable in the brands’ national courts. The Alliance, while binding among the brands, has no such arrangement, making it more akin to a set of unilateral promises by brands.

Third, the Accord involves unions, workers’ NGOs, and the ILO. Unions are signatories with the power to enforce the agreement; the ILO serves as Board Chair; the NGOs are nonparty witnesses intimately involved in the implementation of the agreement. The Alliance, in contrast, is an agreement entirely among brands and local suppliers.

These three provisions give the Accord some of the characteristics of a collective bargaining agreement. Over time, to extend the analogy, the Accord may lead to more stable worker-employer-brand relationships, as the parties learn to trust one another—again, against the backdrop of possible industrial action, arbitration, or further naming-and-shaming.

In fact, it may be the kernel of a stable industrial relations system for the global garment sector. Going forward, unions and NGOs may seek to extend the Accord to cover other nations, and to cover issues other than safety, especially freedom of association. Labor standards then may begin to ratchet upward in a more sustainable fashion.

These structural features of the Accord may ultimately make it different in kind from the Alliance, which was apparently characterized as a “gentleman’s agreement” by its own Chair. Of course, only time will tell which approach will be most effective. Both agreements are just now getting off the ground, and another major factory disaster could have devastating consequences for either one. But past CSR efforts have relied on public outrage over particular abuses, outrage which is usually fleeting. Once public attention to Bangladesh wanes, Alliance members may have less reason to continue their governance experiment; but the Accord may well endure.

I may have more to say about all this in future posts, but I’ll again end with two broader thoughts. First, this strikes me as an interesting development in global governance generally. As in other spheres, public and private regimes are intertwined, as are harder and softer forms of legality. More importantly, here those regimes intertwine in particular ways: they help create institutions of worker representation, which are to date absent at the global level.

Second, I wonder whether the Accord could be the basis not just for a transnational collective bargaining, but also for a new transnational labor law. I’ve been skeptical of Gunther Teubner’s description of corporate codes as “constitutions” for firms’ social activities, simply because CSR initiatives are almost always non-binding and do not have external dispute resolution processes.

But the Accord is different in both respects; it even has a system of primary and secondary rules. If it begins to strongly shape social relations and production processes on the ground, would that be a sort of transnational reflexive labor law? To be clear, like many labor scholars, I have zero desire to fetishize legality for its own sake, though I would endorse new forms of legality that advance broader goals of justice. The Accord holds the potential to do just that.

You may also like...