Addressing Forced Labor Concerns in Polysilicon Produced in Xinjiang

The solar industry is in trouble: there is mounting evidence that the polysilicon produced in Xinjiang, the first step in the supply chain for solar photovoltaics, possibly uses forced labor. Given that Xinjiang now produces maybe half of the world’s polysilicon, there are no easy options to deal with this problem, at least in the short term. But this does not mean that the United States is powerless; instead, it requires the adoption of a more patient, targeted, multi-year response.

Today’s Predicament

The U.S. government and the U.S. solar industry have grappled with how to deal with this problem right now. The U.S. government has previously sanctioned people and entities in Xinjiang, and there are efforts to understand whether similar measures might be employed to stop polysilicon produced with forced labor from finding its way into solar cells and modules. The U.S. solar industry, for its part, has promoted a “Supply Chain Traceability Protocol” to improve the ability of companies to understand where the various components that make a finished product came from.

These initiatives have merit, but they also speak to the difficult position that the United States and the industry find themselves in. Xinjiang is too important in the global supply chain, and there is no realistic, short-term way to get rid of polysilicon from Xinjiang without also slowing down the deployment of solar energy for several years, not just in the United States but around the world. There is some latent production capacity—several production lines have been shuttered in the United States and in South Korea—that could possibly restart under the right conditions. But this will take months and maybe years, depending on the condition that these mothballed facilities are in. Some have been repurposed already and may no longer be available to resume production.

The leverage that the United States has, on its own, is also limited. In 2020, the world added about 127 gigawatts of solar capacity, of which only 12 percent was in the United States. A country with a 12 percent market share cannot set rules for the whole world—not without coalitions. (And given the scale of the U.S. market, it’s theoretically possible to keep U.S. supply chains free from Xinjiang-based polysilicon without curtailing its overall production and merely triggering a rerouting to other markets.) Major allies like Germany, Japan, Australia, and South Korea are also big markets for solar: any successful effort to clean up supply chains must be made in partnership with them and others, which could be politically difficult and add time and complexity to the response posture.

Another major problem is that there is limited visibility into the actual conditions under which polysilicon is produced. In part, this is because Xinjiang is inaccessible, so it is hard to get verifiable facts on the working conditions inside factories. Much of the incredible research that has brought this topic to the surface has involved inference and triangulation. More crucially, the United States is absent from the next two stages in the chain: ingots and wafers. So U.S. companies come in contact with a product at a much later stage, making it harder to trace where the polysilicon used came from. Without independent verification, it will be hard to assuage fears. And it is difficult to see such verification being available today.

The reality is that the United States finds itself in this position in part because it lost a trade war with China. The United States first imposed tariffs on Chinese solar cells in 2012, to which the Chinese retaliated with tariffs against U.S. polysilicon production. These retaliatory tariffs crushed the industry, shutting it from markets in China. Unlike other parts of the industrial chain—manufacturing cells and modules—the U.S. polysilicon industry remained relatively competitive: it was undercut by tariffs, not by cheaper competition (see page 29 in this document, showing the relative position of the Moses Lake facility in Washington).

Tariffs were part of the problem but not the only ones. China also became a bigger market for solar than the United States, which helped develop an industrial supply chain (aided, of course, by cheap capital and other support). Until 2012, the United States had more solar capacity installed than China did; by 2017, China had three times as much installed capacity, and in 2020, China accounted for almost 39 percent of the world’s added solar capacity. In such a dramatic shift, supply chains followed, and the market grew to supply the Chinese market.

Developing a Response

So what can the United States do? First, it is important to point out what will not work: tariffs aimed at malign actors without a broader strategic concept of where this industry should go. So much of the history of solar over the past decade has been tariffs followed by unforeseen consequences—shifting production from China to Southeast Asia, retaliatory tariffs against U.S. polysilicon—that ultimately met none of the original objectives. Tariffs, and wishing for the best, is not a strategy.

Similarly, sprinkling around tax credits is not a strategy. Neither is invoking innovation as a panacea. If you sift through the ashes of the solar manufacturing industry in the United States, you will find plenty of tax credits. Tax credits are an element of industrial strategy, not a substitute for it, and certainly not how they are usually applied in the United States: disbursed over broad geography and several sectors, without any link to either a deployment strategy or a concept to develop dense clusters. Industrial policy that is scattered and haphazard does little good.

If the United States wanted to help develop a solar supply chain outside of China, it could use several tools. BloombergNEF estimated that it takes $15 million to build a thousand tons of state-of-the-art polysilicon capacity today. (For context, the Moses Lake facility in Washington had a nameplate capacity of 25,000 tons.) A private-public partnership to develop polysilicon capacity need not be overly costly.

Discussions to rejuvenate the 48C manufacturing credit could aid in that regard. But that money should be spent purposely—focusing on creating clusters and encouraging production not just in polysilicon but also in ingots and wafers. More importantly, this money needs to be supported by other tools: clear signals for demand and trade instruments to prevent a repeat of the 2010s. This is not about protectionism—it is about protecting investments that the U.S. government has made in the past and will make again.

Such a strategy should not be focused exclusively on domestic capacity. For example, Malaysia is an important polysilicon producer, and OCI is spending $55 million to add 5,000 tons of capacity at its plant in Sarawak. The Export-Import Bank of the United States and the International Development Finance Corporation have pumped hundreds of millions and often billions for hydrocarbon projects. A focused effort, together with allies, could scale up these efforts. Again, the point is density and networks, not nodes.

For investors to respond, however, there must also be a clearer demand signal. Everyone understands, in the abstract, that the United States must deploy more solar energy in order to lower greenhouse gas emissions. But there is little to anchor that faith and to reassure people that the demand will be there. The country has used tax credits to support solar in the past, and states have used renewable portfolio standards. But greater clarity at the federal level, including through a clean electricity standard, could help provide that long-term signal. Without a strong demand pull, no incentives for manufacturing can really succeed in delivering an industry.

Finally, the United States could also decide to encourage the transition of the solar industry to a new industrial structure rather than play catch-up with China. There are several alternatives to polysilicon-based production, which now dominates the solar photovoltaic market. Naturally, such alternatives can take time to develop, and they offer no immediate reprieve from today’s predicaments. But managing a challenge becomes easier once the scope of alternatives is enlarged, and a more targeted discussion to building a solar industry where China’s legacy advances might be muted is well worth having. It is only in the context of that broader response that individual measures—tariffs, credits, public support—can make strategic sense.

Nikos Tsafos is interim director and senior fellow with the Energy Security and Climate Change Program at the Center for Strategic and International Studies in Washington, D.C.

Commentary is produced by the Center for Strategic and International Studies (CSIS), a private, tax-exempt institution focusing on international public policy issues. Its research is nonpartisan and nonproprietary. CSIS does not take specific policy positions. Accordingly, all views, positions, and conclusions expressed in this publication should be understood to be solely those of the author(s).

© 2021 by the Center for Strategic and International Studies. All rights reserved.