Navigating the UFLPA: Geopolitical Risk in the PV Supply Chain
How Will the PV Supply Chain be Impacted by UFLPA and Other Global Legislation?
On June 21, the U.S. Government's Uyghur Forced Labor Prevention Act (UFLPA) went into effect, with major implications for the composition of the global supply chain, including raw materials for the solar industry.
Xinjiang is home to 50% of the global supply of polysilicon, an essential material in conventional solar modules. How much of this supply will be affected by UFLPA and other pending global forced labor legislation? The key to assessing what UFLPA will mean for the solar industry at large comes down to understanding how the law is written and, more importantly, how it will be implemented.
In this pv magazine Webinar, CEA outlines how the law is written, how it will be implemented, and dive into the larger market implications of UFLPA. Christian Roselund, Senior Policy Analyst at CEA, also covers techniques that module buyers can institute to help avoid product detainment, seizures, and rejections under UFLPA. CEA also explores other potential markets that could emerge to replace supply lost by UFLPA implementation.
In addition to UFLPA, The European Parliament has adopted a resolution condemning crimes against the Uyghur people in China and calling for a ban on the import of products made by forced labor. This legislation carries its own ramifications, as it will further restrict where companies operating in Xinjiang can export their raw materials.
Martin Deak, CEA's Associate Director of Supply Chain Management, joins the conversation to provide insight and perspective into how implementation of UFLPA and pending legislation in Europe will affect global supply chains, and analyze which markets across the world could emerge to fill the void in supply.
Webinar content:
A deeper understanding of the UFLPA and how it works
Insights into the larger market implications of UFLPA
Pending European legislation that may work in tandem with UFLPA to address concerns about forced labor in solar supply chains
Restrictions on imports associated with forced labor
Techniques for module buyers to help avoid product that could be detained/seized/excluded under UFLPA
Markets that could emerge to replace supply lost by UFLPA implementation
Q&A
Webinar Transcript
Tim Sylvia:
Good morning, everybody, or afternoon if you are joining us from Europe, or earlier morning, if you are joining us from the West Coast. I hope you are having a lovely day, regardless of what time it is, where you are. I know I am, and it's only going to get better in a few minutes here. My name is Tim Sylvia, I'm an editor with PV magazine USA, and I will be your moderator for today's webinar, navigating the UFLPA, that's the Uyghur Forced Labor Prevention Act, and geopolitical risk in the PV supply chain. And joining me today are a duo of excellent presenters, Christian Roselund, Senior Policy Analyst at Clean Energy Associates, and Martin Deak, Associate Director of Supply Chain Management with CEA.
The topic at hand is the implementation of the Uyghur Forced Labor Prevention Act, what this will mean for us imports of PV modules and sort of the PV market at large.
Understanding the UFLPA and how it works
So, on December 23rd, 2021, the United States took a major step towards addressing allegations of forced labor in the Xinjiang Uyghur Autonomous Region of the People’s Republic of China, with president Joe Biden signing into law the Uyghur Forced Labor Prevention Act, the UFLP Act.
Now after over six months of speculation, the law has gone into effect with major implications for the composition of the global supply chain, including raw materials for the solar industry. Xinjiang is home to roughly 50% of the global supply of polysilicon and essential material in conventional solar modules. How much of the supply will be affected by UFLPA and other pending global forced labor legislation? There are similar legislations being kicked around in the EU. The key to asserting what UFLPA will mean for the solar industry at large comes down to understanding how the law is written and more importantly, how it has been and will continue to be implemented since it was first implemented a few short weeks ago. To help get a better idea of how the legislation will be enacted, what we've seen so far and what it means for the industry at large, I will hand it over now to our first presenter, Christian Roselund. Christian, the floor is yours.
Christian Roselund:
Thank you so much, Tim, it's a pleasure to present to all of you today. So, to start, let's have a little bit of context. UFLPA didn't come out of nowhere: for several years now, there have been increasing reports about human rights abuses against the Uyghur People, an ethnic minority in the Xinjiang region of Western China. A watershed moment in this was the publication of the report In Broad Daylight, by Sheffield Hallam University, that for the first time connected what is going on in Xinjiang to the solar supply chain, specifically to the supply of silicon metal for polysilicon. Following on this, the United States government, in June, 2021 created a withhold release order. That is an import ban on Hoshine Silicon, a metallurgical-grade Silicon maker in Xinjiang. And in the summer and fall of 2021, customs seized at least 140 MW of solar panels from at least four suppliers. The industry over time did adapt to this though. It took three to eight months to get most of these shipments released, and some shipments were never released, they were either kept or re-exported.
So obviously the next step here is the signing of the UFLPA in December, 2021. UFLPA by law creates a rebuttable presumption that goods mined, produced, or manufactured in Xinjiang are made using forces labor, and directs customs to prevent import of these products. That was a fairly broad remit. And then over the next six months, we had the Forced Labor Enforcement Task Force, which was a multi-agency task force, come up with an actual strategy for implementing UFLPA. We got the first glimpses of this on June 13, when customs put out an operational guide for importers, and then on June 17, only four days before the law took effect, the Forced Labor Enforcement Task Force published their strategy, which includes the list of prohibited entities. Now, notably, the government is warned that implementation would be stricter than under the Hoshine WRO, so let's look at what that meant.
On June 13, in the customs operational guide for importers, on page 16, at the very end of the report, they had three sections for the three priority areas of UFLPA, which is cotton, polysilicon, and tomatoes. So on page 16, there was some specific guidance for polysilicon warning, that if importers bring in product that comes from factories that process both Xinjiang and non-Xinjiang inputs, that these would be at greater risk of detention. They also noted that traceability needs to go back to the level of the court supply. Now, this is different than the Hoshine WRO. Under the Hoshine WRO, you basically had to have traceability all the way back to silicon metal, because that's the part that was prohibited, to make sure that no silicon metal from Hoshine was integrated into the polysilicon, which is then made into the ingot and wafer, and so on and so forth. But for UFLPA, we're moving back another level, to the level of the courts, and our sources indicate that this is not an idle threat, but that for the shipments that have been detained, traceability to courts is being required. And I would note that these are important distinctions here. You know, unfortunately, some of the largest polysilicon makers were recent customers of Hoshine Silicon and or procured met-grade silicon from Xinjiang.
Insights into the larger market implications of UFLPA
Christian Roselund:
Now on June 17, we got the UFLPA strategy, and that listed the prohibited entities for the solar supply chain. There were four significant entities: Hoshine is there - surprise, surprise - but also Xinjiang Daqo, Xinjiang East Hope and Xinjiang GCL. Now that's notable in that those are the Xinjiang subsidiaries of these three companies, and not the parent entities. We saw this as being a targeted approach, focused really on polysilicon in Xinjiang, but there were still concerns about implementation in terms of how this would be put into effect. Now, UFLPA is not the only policy which has been enacted, which is being created to address the concerns around forced labor in Xinjiang. On June 9, the European parliament adopted a resolution on human rights in Xinjiang and specifically called on the European Commission to propose an import ban on products made by forced labor. At this stage, this looks a lot like Uyghur Forced Labor Protection Act for Europe. However, there aren't as many details. Basically, it was about a paragraph in this resolution. However, it does direct the European Commission to introduce these import control methods for products made by forced labor. And, uh, this is set to be defined in September of this year.
So, what do you do if you bring in product and it gets seized? Well, there's two paths for importers to get shipments released. One is an out-of-scope decision, basically, that's where you prove that what you brought in has no material inputs from Xinjiang and has no material inputs from any entity on the list. That requires providing a complete supply chain map and documentation that confirms the source of these materials. Again, going back, where are your cells from, where are your wafers from, where's the ingots from, where's the polysilicon from, where's the metallurgical-grade silicon metal from, where is the courts from all of that? It's a lot of documentation and, you know, it's a lot of work to get and produce these things, have them all translated in English, and have them presented neatly to customs. However, there's potentially high route here to get a shipment released if you're able to provide this documentation. The other route provided by the UFLPA strategy is to get an exception, and an exception would happen if you had material that was produced by Xinjiang and/or by one of the listed companies, but you want to prove that it was not produced with forced labor.
Now, the Federal Government has explained how you could do this. You could provide labor audits, you could go back to the suppliers and get them to provide statements that they do not participate in any of the Chinese government labor programs, which to the Chinese government are labor programs. and to the U.S. government are evidence of forced labor. Theoretically these things are available, but in practice, this is basically impossible. And here's why, it’s the China Anti-Foreign Sanctions Law. On June 10, the Chinese government enacted its third blocking law to counter foreign sanctions, and under the anti-foreign sanctions law, Chinese citizens consume court for damages caused by another company's compliance with foreign sanctions. But it also calls for the creation of a list of individuals who've complied with foreign sanctions, and gives authority to government officials to prohibit their visas and/or to seize their property and the property of their relatives.
So, no independent labor audits like the U.S. government are asking for are currently available in China, and anyone attempting to provide documentation that would allow for such an exception would be eligible for prosecution under the China anti-foreign sanctions law. So, this route is basically, as I said, impossible. Now, in terms of the timelines for getting stuff released from the UFLPA, there's some better news. As you recall, shipments seized under the Hoshine WRO were held for three to eight months before being released. Under the UFLPA there's some definitions for timelines. If shipments are seized and the importer wants to challenge that detention, they need to do so within 30 days and provide the documentation. Additionally, the customs is then required to process that documentation and make a decision whether to detain or release the product within another 30 days. So, we're looking at maximum 60 days. Now, customs can decide that the documentation was incomplete, and that the product will remain detained, potentially subject to forfeiture. And at that point, it would have to go to the courts. But there is a tighter timeline here for actually getting shipments processed and either held by the government or released than there was under the Hoshine WRO.
So, I'll now take a few quick questions here, Tim, before we turn it over to Martin.
Tim Sylvia:
Yep, absolutely. So, yeah, we had a few, probably three or so quick questions before we turn it over to Martin. The first thing that I would like to ask about is, you know, we've seen that in early kind of application of UFLPA, the issue has been this court's documentation. What I've heard is that this court's documentation is something that suppliers, you know, do not have at this time. We're not prepared to have at this time but is not necessarily impossible to put together. It is something that could be created, it could be provided, but it will take time. Do you have any idea what that timeline is like or what the feasibility of providing court's documentation, you know, in any sort of term is?
Christian Roselund:
Well, I, I think to start the documentation exists, right? These metallurgical-grade silicon (MGS) makers, they got quarts, you know, either it was produced likely on site, but if it was produced by another company, there's a record of its sale, there's a record of its shipment, there's a record of it being held in warehouse before it's turned into MGS. All of these documents exist. It's a matter of getting the MGS supplier to hand them over. So, you know, you basically have to go through the polysilicon maker, who has to tell the MGS supplier, “Hey, you know, we need these documents” and the MGS supplier has to produce them. Now, certainly this is going to be easier if the MGS comes from Brazil or the United States or Germany then it will be if this comes from China.
And you know, there's reasons for that. Uh, I would note that while this documentation exists, it hasn't been asked for before, you know, in there's been no previous circumstances where cell and module makers were going all the way back upstream to the met-grade silicon. Well, under MGS, certainly for the Hoshine WRO, but back to the level of the quarts, no one asked for that yet. So, um, it exists, it can be produced, but it requires going through these levels of the supply chain and, you know, this is likely gonna be easier in Western countries.
Tim Sylvia:
Right, right. Another question here came from the aspect of the suppliers. So, is the implementation, as it's been so far, a bit of a gray area of suppliers, as in, they know how it's saying it's going to be implemented, they kind of have some sort of rough idea of the supply chain history of their product, but they're not finding out about possible detention? You know, until it actually arrives in the U.S., or did they have any sort of, kind of inkling of possibility of detention, you know, prior to shipping, when are suppliers finding out they might be at risk?
Christian Roselund:
Well, I don't think anyone ships product thinking that it's gonna be detained. So I think that, you know, in each case, when it's detained in port, this is, um, <laugh> at least something of a surprise for the shipper and for the company trying to send the product. Certainly, you know, I think that they're all aware of the risk of detention and particularly the risk that comes from commingling within factories, because so much of the solar supply chain, it's not, you know, we don't have dedicated factories for the U.S. market going back to the polysilicon and MGS level. First off, this has never been needed before. Second, there are selling module makers who have factories that are serving the U.S. market. They may have some control over their ingot supply, but these are not the same companies who are making the polysilicon, and the polysilicon is sold to a wide variety of customers. So, each level you go back in the supply chain, the more difficult it gets.
Tim Sylvia:
Mm-hmm <affirmative>. One last question here before we hand it over to Martin. Is just kind of how broad the scope of implementation is, and what manufacturers will be subject. We had a question here that kind of stated that suppliers that they've spoken to out of China and Southeast Asia have been reporting that their shipments to the U.S. are not required, not being subjected to this sort of compliance filing, or kind of implementation. So, do we have any idea why some suppliers would be subjected to like a little bit more, uh, intense scrutiny than others?
Christian Roselund:
That's a good question. I think what we've seen so far is the larger suppliers have been detained, and that's true under the Hoshine WRO and it's true under UFLPA. We have yet to see the mini-detentions from the smaller, or any detentions from, you know, companies outside the big five or six. So, um, does that cover it, or is there another part?
Tim Sylvia:
Yeah, I think that helps.
Martin Deak:
I think, if I may interject there, I think what we can add as well is that customs has made us aware that they have a risk-based methodology that they're applying, but that methodology has not been made public. They are referring to, I think, trusted, you know, quote unquote trusted sources in a public or private, but none of this process is really made available to public. What we can infer from the data that we already have is what Christian just said, that it would appear that the list of detainees correlates quite well with the top suppliers by shipment volumes, the United States market, at least over the last year or so. This might change in the future. I think customs have also let themselves be heard that, in the future, midsize suppliers and small suppliers might come under scrutiny, but I think that's the next wave.
Tim Sylvia:
Right. Great. Well, I appreciate your perspective on that question. I'll hand it over to you now to get into your portion of the presentation. Thank you, Martin!
Pending European legislation that may work in tandem with UFLPA to address concerns about forced labor in solar supply chains
Martin Deak:
Perfect, thank you. So, in the second part of the presentation we'll talk about impacts of UFLPA on the silicon-based supply chain. And then, at the end, a little bit about the measures and prevention mechanisms, preparation mechanisms that both suppliers and buyers can implement. So first, what we are looking at here is a comparison between policy and capacity for the next three years, end-of-year numbers, GW equivalent and projected big module installations for all regions except for China. The comparison that we are trying to make here is, and this goes back to the anti-foreign sections law that Christian just discussed, the anticipation or the current state of affairs is that we should expect that it is much more difficult for suppliers in China, and especially in Xinjiang, to provide documentation with respect to traceability and labor standards and so on and so forth.
Then it is for suppliers in the overseas market. So, uh, the distinction that we are making here in a polysilicon capacity. The dark green figures here represent the estimated capacity of manufacturers in the overseas markets outside of China, and in the light green, the light green figure here is polysilicon capacity inside of China, but in regions outside of Xinjiang and owned and operated by suppliers without any affiliation to ship to Xinjiang. So, if we may just make a quick comparison of these two figures, for example, 140 GW total in 2022, with PV module installations projected based on what we know today in order at the end of second culture, there should be enough polysilicon capacity to supply good traceable raw materials to the United States market and to the European market if that market virtually requires a hundred percent traceable trace bar materials, and also to a large portion of the other global markets outside of these two regions.
I would just like to underline, we are talking about polysilicon capacity here, not necessarily polysilicon production, and also PV module installations do not equal shipments. So, there is some flexibility in this chart, but overall, the trend is clear here, and that is that these two types of polysilicon by source should be sufficient for U.S. and Europe in the next few years. If we look a little deeper at how UFLPA is impacting polysilicon capacity on the global level, this is the picture that we can get: Hemlock, Walker and OCI are the primary polysilicon suppliers that are supplying product also to the PV market, and REC Silicon should be coming online again at the end of 23 or 2024.
Overall, what we see from this picture is not much impact in terms of expansion of polysilicon capacities outside of China that we can directly relate to UFLPA implementation. So, in the previous chart, in 2022 we are estimating around 40 GW of overseas polysilicon capacity to be available. And then the REC Silicon can add another five or so gigawatts. The one change that is on the horizon is that India should start supplying relatively substantial quantities of polysilicon to our industry in 2025 and beyond. This is linked to the government subsidies for domestic manufacturing in India. There are three nominees, or three awardees, from the first round of these production-linked incentives, Adani, Reliance, and Shirdi Sai. The timelines for all three and the capacities are not a hundred percent confirmed yet, but we should expect around 30 GW of new polysilicon capacity coming online from this first round of production-linked incentives in India in 2025 or after 2025.
The impact of UFLPA implementation is a little more visible in China. This is the polysilicon capacity distribution end-of-year capacity in the next three years, and the major trend here that we can talk about is the decrease in the proportion of polysilicon made in Xinjiang between 2022 and 2024, a decrease of about 20% absolute. And also the second trend that we see here is an increasing of capacities in other regions in China, especially in Mongolia, in the North, the dark green color, in 2023 and 2024, but also in Sichuan, in some central provinces in China.
What we are considering here in these charts are publicly-announced capacities, but based on our analysis, we are just including those where we feel or believe that there is quite a good likelihood that these plans should move forward because we have also seen many new entrance in the market recently. Uh, two more points to mention here: One is that several of these new polysilicon factories, especially in Mongolia, but also in other regions, will come with an adjacent capacity for silicon metal. So this will improve traceability as polysilicon and silicon metal will be made under one roof by one supplier, with better access to documentation. Also, I think it doesn't come as a surprise, but these regions have a relatively lower cost of electricity, in China and especially in the South. There is a lot of hydropower in the energy mix in the North, there is some wind, still a lot of coal, but also several of these new complexes.
These new polysilicon factories will come furnished with, again, adjacent solar and wind power plants to supply electricity for the energy-intensive manufacturing processes. So we are seeing more impact, direct impact from UFLPA in China than we are seeing so far around the world, at least based on the public announcements. So as I think we are all aware, and as we talked about today already, the larger the supplier, the easier should, be at least in theory, to obtain the necessary documentation to satisfy CBP requirements, given that, typically, the larger suppliers have more vertically-integrated control, more of the supply chain nodes under the same roof, and therefore are able to provide this recommendation. And at the same time, they’re also in closer proximity to the polysilicon node, silicon metal node and quartz node. So, so far what we are seeing, implementation here is that larger suppliers should have an easier time. And this is also reflected in some preparation by mid-size suppliers, mid-size suppliers and smaller suppliers that we have seen so far that is perhaps a little laxer in this regards. Based on what Christian also discussed and based on the implementation strategy from June 17, CBP has also made us aware that they will be looking at factory segregation. So, ideally, factories that are involving the supply chains for the United States should be only using raw materials that are traceable or outside of injunction.
Of course, again, for large suppliers it is easier for them to dedicate factories. And also, based on the history that we discussed, the larger suppliers have more experience so far. I guess looking at how the factory segregation might work more downstream at the ingot and wafer level, we can have a look at this view here, which shows us wafer capacities in China and in different parts of the world. And you can see that we are seeing quite a substantial growth in Southeast Asia inhouse wafer capacities between 2022 and 2023 from around 11 GW to 26 GW or already these 11 GW is quite an increase also in 2022 in comparison to 2021. So, at least hypothetically, the suppliers that are more vertically integrated outside of China who have, in some cases, not all cases, segregated factories utilizing only overseas polysilicon should find it much easier to comply with the requirements.
Restrictions on imports associated with forced labor
Tim Sylvia:
Uh, Martin, before you move on here, we had a couple of questions pertaining back to slide 13. Would you mind going back to the slide 13 and just fielding one or two here quickly? The first question was that, to the best of your knowledge, are the identified risks of forced labor, human rights violations, everything that we're talking about today, are those strictly held to Xinjiang, there's no reports or any sort of reason to believe that any of these other regions would be subject to any sort of similar scrutiny in the future?
Martin Deak:
Yeah, I think that's a great question and something that we'll probably have to deal with as an industry in the future. But so there have been reports, I think we are aware, all of the reports of labor transfer programs between different provinces in China. So that is definitely a topic. However, in terms of you know, more in depth comments that I could provide right now, in terms of verification, I don't think there is much we can say. We will get to it in a moment, but, again, back to the anti-foreign sanctions law in China conducting any type of labor employment audit, social audits, this is basically impossible or at least highly not advised. So, any type of actual verification is very difficult. And you know, as I said, there are some reports of such correct this is.
Tim Sylvia:
Okay. Yeah, we can wait for the Q&A for the other ones. I just wanted to get on that real quickly.
Martin Deak:
Okay. I think one more comment to make here is that, if we go back to these slides, you can see a trend here. So, this in gray, we have non-Xinjiang based, but Xinjiang-affiliated capacity. So, essentially, all the three suppliers that go ETOP and GCL that are listed right now are investing quite substantial resources into expanding outside of Xinjiang. You can see it's quadriplus the capacity outside of Xinjiang between end of year 22 and 23. So this is perhaps the first topic that we'll need to figure out. We know that the UFLPA Entity List only lists the Xinjiang subsidiaries on these three companies. So, there might be a different risk profile as end by CBP if suppliers listed in Xinjiang produce outside of Xinjiang as well, right?
Christian Roselund:
So, I just wanted to note here that the UFLPA text itself mentions not just Uyghurs, but Uyghurs, Kazakhs, Tibetans, and other ethnic minority groups, so, there is the potential application outside of Xinjiang, for sure. And I think that this is something that was looked at for other industries, you know, notably UFLPA doesn't just cover polysilicon, it also covers cotton. So, there's been some documentation of labor transfer regions outside of Xinjiang for cotton. So, you know, there's a potential there, but I would say that for polysilicon the focus has been to date. I have not seen anything that suggests any sort of implementation outside of Xinjiang yet.
Tim Sylvia:
Right. Appreciate that clarity. All right. I will let you get back to your presentation, Martin.
Techniques for module buyers to help avoid product that could be detained/seized/excluded under UFLPA
Martin Deak:
Okay. So, in the last two or three slides I'm gonna talk about how to de-risk UFLPA in module procurement agreements. I think back to the topic that we just discussed. So, potential implementation in other regions, potentially we might see implementation also for other raw materials or other module components as silicon metal is used also for other applications. But the main message here is there are a few things that should be added to contracts, these days in 2022, and the first category, raw materials. So, the bombs should be expanded not to end at sale, but to include wafering, the polysilicon ideally, also silicon met, and factories and suppliers should be listed for all of these.
So, as early as possible in the process, the supply chain map, so to say, should be figured out and confirmed. As we just discussed, ESG or any type of social audits, labor center audits are practically impossible to do in China. The workaround that is probably best right now is: suppliers should be requested/required to comply with their own ESG policies, codes of conduct, corporate social responsibility policies. Ideally also, buyers, policies of the same sort, as these typically list references to international standards or look at the preferred or, you know, excluded labor standards. So, this is something that already exists, and it should be easier to implement in contract. The important point is that the upstream suppliers are the vendors should be subjected to the same standards here.
Again, these are workaround. Ideally, it should come also with a well-specified grievance mechanism and some operational guidance on how to deal with potential breaches or suspected breaches. Then the other topic here is detention and inability to import. So, supply buyers should keep track of information or activities actions by CBP in real time, as much as possible, and treatment and remedies, should be defined in advance if import or suppliers are unable to import and in other preparation. So buyers, as soon as possible, should review traceability documentation, should require traceability documentation. You can follow the CR protocol or CBP guidance or both, and if this is found to be not sufficient and audit should be conducted.
And that brings me to the last point. I think Christian already touched upon that a little bit, but administrative ruling is another process that can be applied for future shipments or prospective shipments. This process exists outside of UFLPA, outside of the WRO, it predates both. Typically, it's used for tariff classification, determining country of origin, and valuation of merchandise, but it can be in a sense also used to review compliance with any applicable laws. So, both buyers and sellers can ask CBP to review their documentation related to traceability prior to even shipping. So, while the merchandising is still in Southeast Asia in most cases and determine whether the documentation would be sufficient or not.
I think I leave it at this here, maybe we can get to it in a more detail just for later if there are questions. So, the use is that quartz documentation is required. So far customs appear to be focused on large importers, very similar to the WRO, but this could change and probably will change. The detentions are very fresh, a couple days or a couple weeks old. Initially, we have heard relatively optimistic feedback, but I think we still have to wait and see if importers can actually comply with civic requirements.
Tim Sylvia:
Yep, great. Well, thank you for that presentation. We can get into our Q&A here for the next 20 minutes or so, and we can kick it off with our first submitted question here. Given the political sensitivity of Xinjiang and the forced labor allegations within China, as well as the law that makes assisting with U.S. sanctions illegal, is there realistically any way to rebut the presumption of, you know, for any module whose wavers are produced in facilities that have in the past purchased Xinjiang polysilicon. So, given the factors that you went over, I think in one of your last slides, Christian, is there any way to review it or is it just a matter of, kind of wait and see what's the situation?
Christian Roselund:
Well, I think the key to that, the answer here is that last part of the question, in the past. Yes, if it can be shown that the material itself was not produced, that it has no inputs from Xinjiang, and if it has no inputs from the listed companies, yes. However, if you are unable to show that, and you're instead trying to say, “well, this wasn't produced with forced labor”, no. But the two routes are very different, you know, and out of scope. Decision and exception are very different routes. An exception is effectively impossible, but an out-of-scope decision is very much possible. You know, yes, many of the large polysilicon makers are former customers of Hoshine, including those not located in China and with headquarters outside of China, but that doesn't mean that they are still using Hoshine MGS. And so, if this can be documented, then you have a route to getting out of detention.
Tim Sylvia:
There's a question here, and this is something that I think is interesting. I'm not sure that we have any indication of this going on presently, it's more of a hypothesis, that is: How do you prevent false documentation from being used to get through the customs process? Would it be pretty easy, do you think, for customs to tell off the false documentation, or what would the scenario be there? Or is it just kind of unrealistic?
Christian Roselund:
I wouldn't try to give customs false documents. <laugh> I think most of the people listening on this webinar aren't selling module makers, they're probably more buyers, on the buyer end. And I would say, get to know your supplier. And work with suppliers you feel like you can trust and that you have a good relationship with, because I think it would be a really bad idea to try to fake documents and get them through customs. Customs has already raised concerns about transhipment and said that they've seen transhipment from people swapping shipments under, you know, trying to pass shipments under false documents, and that they're cracking down on that. Mm-hmm. They didn't say in what industry, though, this was a customs webinar. This could have been in the clothing industry. This could have been shipments of tomatoes, you know, not necessarily in solar.
Martin Deak:
Yeah, I think one approach you can also try to implement at least is the way we conduct chain of custody or traceability audits, is that we try to understand the systems first. How the traceability system works and then we try to test it. So, we do sample checks on specific product and we review the documentation for that specific product. Trying to do this as much in real time as possible, I think can be helpful. You know, if I ask you to give me documents for this module, I don't want to wait a month for it. I think what we have seen so far from most suppliers, not all suppliers, I believe, is that they are not necessarily compiling all the documentation for all the batches of raw materials in the form that they will be submitting to CBP. So, assuming that this is true, you know, trying to do a real time check, how fast can I get documentation, could help you determine whether it's, you know, actual documentation.
Christian Roselund:
Yeah, I think also if I can make a shameless plug for CEA's traceability services here, I think another way to deal with this is to have a third party do a traceability study and, you know, go into the selling module factories, document this and the ingot and wafer factories and document this information, at least as far back as you can go.
Tim Sylvia:
Someone here references a recently released IEA report, that the entire solar supply chain China's share in all manufacturing stages of modules is over 80%. Do you think that the UFLPA will have the additional effect of diversifying the supply chain? You know, not just as it relates to polysilicon, but throughout the solar supply chain. And if so, how, what that would mean in terms of international pricing, just kind of any sort of impact on diversifying the supply chain as a whole that will come from this?
Markets that could emerge to replace supply lost by UFLPA implementation
Martin Deak:
Yeah, I can provide some comments, and Christian, you can add. Right now we don't think that GFO, you know, we have seen in the slides that we showed on the overseas polysilicon expansions, for example, there might be some additions there, but we are not seeing UFLPA being a driver of global supply between the different global regions, at least at this point of redistribution. I think in most cases what drives this type of regional capacity is some type of support from the government. And that is really the key from the local government. So, it might happen, and affordable will happen over the next couple years in different public regions. But right now, UFLPA doesn't seem to be the driver.
Christian Roselund:
Yeah. And I would note that it's a little early for that. You know, we're only a few weeks in, so to actual implementation of UFLPA. So, it's a little early to tell what the longer-term effects are. I would also note that the ingot and wafer space was already diversifying.
Per Martin’s slide, we're already seeing more ingot and wafer factories in Southeast Asia, which is significant given that it creates an alternative. If you look, there's that 80% figure, but I think it's more useful to look at individual nodes in the supply chain and of all of those, the one that China has the greatest control over is ingot and wafer. China has 97% of the global ingot and wafer capacity. And so, even though the scale of the factories going online in Southeast Asia is small compared to what's in China, it’s still significant, given that you're working from a very small base there. So, that's already happening.
I do foresee more demand for non-China polysilicon in future years, given that it's the easiest to trace, it just makes the job of tracing the polysilicon back to the level of the quartz easier. But it also takes multiple years to do a poly factory. So, you know, you're not gonna have polysilicon makers suddenly jump up and say, “Hey, let's spend the next four years building a factory”. If they don't have assurances, longer term assurances. We're also seeing a little bit of diversification there. In 2024, we have already silicon coming back online in Washington, but you're probably more likely to see these sorts of tentative steps until you can see a longer-term business case.
Q&A
Tim Sylvia:
So, we had one person here who had kind of a two-part question. The first was due diligence for buyers, and I know you did plug third party audits and getting to know your supplier, but if there's any other due diligence you think should be taken on the part of the buyer, feel free to comment on that. And then the second was the scope of international legislation. You mentioned the legislation being introduced by the European parliament. Is there any else known, uh, of introduction for forced labor in Xinjiang or anywhere else in China or is it kind of UFLPA and the European parliament, at least at this time that's been introduced, considered.
Christian Roselund:
Well, I would say that I'll take that last one first, and Martin, feel free to jump in on the other parts, but right now, I think we've seen moves from other countries. You know, the UK is paying attention to the forced labor issue. Canada is paying attention, individual E.U. member states. They've all made noise about forced labor and about Xinjiang in particular, but what's happening is, over time, more documentation of the forced labor issue is coming out. So, you know, we saw the BBC release that Xinjiang police files and shortly thereafter we got the E.U. resolution. So, I think that there's really a dynamic back and forth between the human rights organizations and those who are leaking information from China and the policy. But it's definitely not inconceivable that we could see similar policies coming from Canada, Australia, Japan, the UK, etc.
Martin Deak:
Yeah. I think the first question was with respect to the diligence that can be done prior to shipments. This of course starts early on with understanding the supplier's, supply chain. So either this is a direct conversation with the supplier or, you know, CEA market intelligence materials can be helpful. It should definitely be a part of the RFP process. So ideally, you know, at least the silicon-based portion of the supply chain should be good in the RFP. And then the second powerful, maybe most powerful step, is including the right requirements in the contract. So, as I mentioned before, especially for the less experienced supplier so far, with WRO and UFLPA, and less particular integrated suppliers, many of them from other observation are preparing less that they could be. And this is for different reasons. Obviously, again, the proximity to the upstream is different, so there should be a discussion prior to contracting or prior to signature, and suppliers should know what kind of requirements buyer has and the work should start right away. So, you know, we don't want to wait until one day before production day to confirm the supply chain made or available traceability documentation.
Tim Sylvia:
We had a question here saying that they are hearing that there are certain pressures being applied from module suppliers in terms of imposing FOB freight onboard terms. Do you see this risk increasing going forward or what the situation is there?
Martin Deak:
That's another good question. So, I think right now we still need to wait and see. I'm aware of maybe one or two suppliers who are these days shifting to this incoterm. But we have had similar discussions over the last few months, especially with respect to AD CVD before the suspension, right? So it's never an easy discussion. If both parties are coming from very different angles, the supply is constrained today. I mean, in 2022, in 2023. But as much as possible, we strongly advise buyers to not take an FOB incoterm or any other incoterm that doesn't include customs clearance process.
Tim Sylvia:
We've had a couple of questions here about scrutiny for non-Chinese operating companies or subsidiaries of Chinese companies operating outside of China, or completely independent companies operating in, say, India, that have wafers that largely come from China. So, would we expect the same level of UFLPA scrutiny for these companies? Because even though they don't operate in China, they use Chinese wafers and materials, or is this implementation going to be across the board? You know, even if you're not in China, if we are to believe that any part of your supply comes from China, you will be subject.
Christian Roselund:
Yeah. You know, it's a good point there that the Indian manufacturers, a lot of them use Chinese cells. Those are being cut off by India's own set of trade barriers. But they've certainly purchased a lot and India is ramping up cell capacity as fast as they can, but then they're dependent on Chinese wafers. And very much so, I believe India, you know, there's been talk about some Indian wafer supply, but basically the wafer supply that's coming online is in that's ramping soon is in Southeast Asia. So that dependence on Chinese wafers does potentially make Indian modules in danger of UFLPA enforcement. That being said, we haven't seen that yet. What we've seen so far for enforcement as we've seen the large Chinese tier ones being seized. So that's what's happened so far.
Tim Sylvia:
We had a question here regarding CEA's process for conducting their traceability audits when it's possible that access to these factories can be difficult, kind of impeded in some situations, they're implying, I guess, dangerous. So, what is your confidence that suppliers would comply to traceability studies and kind of how is that compliance reached?
Martin Deak:
Yeah, I can briefly answer this. So yes, all the remarks are true. It's very difficult for essentially any third party, including buyers and especially non-Chinese parties, to connect these kinds of audits. And it becomes gradually more or exponentially more difficult the higher you move on the supply chain. What you can understand though from traceability audits is the current preparedness of a supplier to comply or not comply. And you can apply this process to compare across different suppliers. The preparedness varies between different suppliers. So, I think that this kind of understanding can be traceability. It can be helpful in this regard. There will be limitations and, especially as we move now to quartz, a heavier focus on silicon metal will have to learn to work through these limitations and improve how we do it as an industry.
Christian Roselund:
Yeah, though it is notable that CEA’s done what, Martin, hundreds factory audits today, and we've been in factory.
Martin Deak:
Audits, yeah, CEA’S quality factory audits. We are maybe on the order of 50 traceability audits, 50 to maybe 60, which is, and now I'm talking about, some of these include look at different projects. So, it's not necessarily on a project level basis but sort of supply chain level basis, specific supply chain. So, the company supply chain.
Tim Sylvia:
We're getting pretty close to running up on our time here for the webinar. We've gone over a ton today, and I think it might be helpful for each of you to give maybe just your high-level kind of top takeaways for buyers moving forward, just kind of how you see the situation evolving, kind of big takeaways and summation of what we've covered today.
Christian Roselund:
Sure. So, you know, UFLPA is forcing a new attention on the supply chain and it's forcing a traceability in the industry, which will make for a more transparent industry. And, certainly, there's some challenges for the industry in the near term, but I think this is something that's achievable in that the industry can overcome. I think also we're looking at some potential diversification in the supply chain, both within China and outside of China. And I think that that diversification of the supply chain is a good thing for the industry. It's not good to be dependent on any one region or any one nation. So, you know, there are definitely some near-term challenges, and we're going to see how these work out, but I think this industry can navigate this and continue to move on and grow as it has with past challenges, including past trade challenges.
Martin Deak:
Yeah. I, I think that's 2 points. One is that, based on the changed reduced timelines that Christian discussed at the beginning, we are looking at 30 days for the initial detention appeal process and then another 30 days for petitioning or protesting those decisions. So, this means that the next one to two months are very critical for the industry to figure out whether the current level or current status quo of traceability is going to work right now. So that's something that the next, maybe until the end of August, we should all keep our eyes building. And the second thing that I would recommend this or would highlight again is preparation, right? So, there's a lot that can be done early on. And again, understanding supply chains early on and having good contracts, doing traceability audit is very important.
Tim Sylvia:
Great. Well, I thank you both for taking the time to present today and to field all these questions. I know this provided me a lot of clarity and insight. I think it will anyone else who attended, so Christian, Martin, we appreciate your time and expertise here today.
