ChinaTalk Podcast: China’s Solar Industry

CEA’s Founder and CEO, Andy Klump, and Director, ESG and Traceability, Paul van Brenkelen, join ChinaTalk host, Jordan Schneider, to discuss how Chinese firms came to dominate the industry and with recent US import bans, how the global solar supply industry will adapt.

Video Transcript

Andy Klump:

The real reason that Chinese manufacturers succeeded was because the supply chain advantage. Not only could they convince a glass manufacturer to locate next door to their module manufacturing facility, but they could also get half a dozen of their suppliers within a hundred mile radius to be competing on very low pricing.

Jordan Schneider:

On June 23rd, the day we're recording this podcast, Bloomberg ran a story indicating that the White House was planning to quote, block some solar goods made in the Xinjiang region. How did Chinese firms come to dominate over 80% of the global solar supply chains? And what are some companies trying to do to ensure that those supply chains are clean? To discuss we have on Andy Klump, founder and CEO of Clean Energy Associates, a company that provides quality assurance to the solar supply chain and has recently expanded into ESG audits. Joining us also today is Paul van Brenkelen, director of ESG and traceability at Clean Energy Associates. They're both based in Shanghai. Andy and Paul, welcome to ChinaTalk. So Andy, first off, how did you get into China solar game?

Andy Klump:

My path to China started roughly 20 years ago when I took a trip to Southeast Asia and I was just fascinated with Asia. I made a decision during business school that I would shift my career towards China. And in 2002 started my career working with Intel after finishing business school. I joined joined Dell computer in 2003 in Beijing. So I've been living and working in China for the last 18 years and have been spending the bulk of my time over the last 15 years in the solar and energy storage supply chain. So that's when I first worked at Trina solar for two years as their vice president business development. And the last 15 years as the CEO and Founder of Clean Energy Associates.

History of Chinese Solar Industry

Jordan Schneider:

Let's go back to those early days that when Trina Solar and Suntech were the names, how did these firms start? And at what point did Beijing realized that this is something that could potentially be really important to the future of China's economy?

Andy Klump:

So I will say the Trina was actually one of the the early stage pioneers in the solar industry. And first established itself in 1997, they moved into mainstream solar module manufacturing in 2004 and 2005. And I became involved at that early stage when the company was not in the same realm as Suntech, it was, that was still a Tier 3 company, many people have never heard of. They just had about 500 workers in the company. In those two years, I spent there, the company grew about 10 X, almost every metric. And I was instrumental in really helping build out their overall supply chain strategy. At that point in time, Trina was not known not only by buyers of solar modules, but also of the suppliers. And so one of the biggest challenges that Trina faced in 2006 was procuring polysilicon supply. So that was my primary focus for the 2+ years I was there. I led a team that was involved with the strategic procurement of all long-term polysilicon agreements. And we had a very large effort to get the company on the mainstream stage.

Overview of Global Polysilicon Industry

Jordan Schneider:

Over the past few months, there have been increasing reports of forced labor in the polysilicon supply chain, which is what has caught the white house's attention and led to today's Bloomberg story. So, Andy, how do you buy polysilicon in 2006? And how has that changed In the past 15 years?

Andy Klump:

History of Polysilicon Pricing

Back in 2006, there was an oligopoly of effectively five suppliers controlling 90% of the supply chain. And these were American, German, Italian and Japanese facilities with virtually nothing made in China.

At the time, the cost was roughly about $30 per kilogram and the selling price at the time was about $150 to $200 per kilogram in the spot market. The accelerated interest in solar in 2007 to 2008, led to a polysilicon spot market pricing that accelerated all the way up to $475 per kilogram, so 90%+ margin in the product, and it was highly sought after, by every single major manufacturer of solar and solar modules

To understand this full supply chain, it's a polysilicon is the key upstream material. Then ingots, wafers, cells, and modules are manufactured. So Trina being vertically integrated from ingots all the way through modules, they had to buy polysilicon. So that's why I was engaged, but it was a very convoluted process. Many of the short term buyers had to go through traders and various other intermediaries, but it was a complete mess.

Polysilicon Industry Growth

The industry has matured and the total industry size has grown by about a 100 X in the last 15 years. And the size of the plants are just at a complete different scale. When we talk about 50,000 metric ton facilities, that the whole industry was roughly that size back in the early days.

And I will also mind you, I just highlight that the industry was focused on semiconductor wafer players as the major off-take. So solar was just literally 1% of overall polysilicon supply. Now the whole industry is flipped and it's dominated by solar.

Jordan Schneider:

Sure. One of the things that happened as that industry grew a 100 X was those, those German, American, and Japanese firms really lost out to Chinese suppliers. So what about the Chinese firms and the Chinese market allowed them to compete, or I guess out-compete of Western polysilicon?

Advantages of Chinese Polysilicon Manufacturers

Andy Klump:

I would highlight maybe it wasn't just Chinese manufacturers, but also other Asian suppliers who entered this space. The Koreans were extremely large for a period of time, but eventually yes, that the market did migrate to China, but there still are a number of polysilicon facilities that are operating in the US and in Korea as well, but there are, but the market is still 85, 90 plus percent centered in China.

And to answer your question, but a lot of the costs from polysilicon is really in the electricity. And so having a lower cost of electricity is a key point of competitive advantage, but also the supply chain is extremely robust in China.

Now at that time, there was actually very few manufacturers of the actual polysilicon or the core equipment around the bell jar that actually changes the gas to a solid. And so a lot of the core equipment around the supply chain was really just concentrated with very few manufacturers.

And that industry supply chain has become much more advanced. And that's where I think the Chinese suppliers have a lot of the long lead time equipment are now in a major position to grow and scale.

But they're still, at that time, there was about a three-year lead time to set up a new polysilicon facility. That's now been shrunk down to 18 months and maybe some cases 15 months, but, you know, I'd say the supply chain advantages of China have been really what would help the industry accelerate along with lower costs of forms of electricity.

Jordan Schneider:

So is that 18 month timescale available to non-Chinese manufacturers?

Entry Barriers for New Polysilicon Manufacturers

Andy Klump:

For the existing incumbents who have the core tech process technology and the teams and the know-how. It can be, however, for someone who is entering the business in a brand new capacity, it is likely going to take longer than 18 months.

So you have to think, and the context of what a polysilicon facility is like, I would think of this as a giant petrochemical facility with giant pipes and distillation towers that are 10 or 12 stories tall, you're talking about a minimum CapEx of a billion dollars just to get to a basic facility of around 20,000 metric tons. So the amount of money and the resources needed is quite substantial. And so that's where it really is only very large well-established well-capitalized firms who can really enter this space, but those are the barriers to entry that certainly existed back then, but I'd say the barriers to entry have gone up since.

Overview of Global Solar Industry Growth

Jordan Schneider:

Then, let's take a step back. One of the reasons you said that the Chinese polysilicon industry has grown so dramatically was because the rest of the industry was grown so dramatically. So let's do the first five years of maybe your involvement going from 2006 to 2011, what were some of the key developments that started this industry down an exponential growth curve?

Andy Klump:

So I would back up to 2003, really, when you saw the first true policy support that came out of Germany with the the EEG, which is effectively a law that promoted the adoption of renewable energy in Germany. And it started a very small scale way, but then the German market really took off in 2004, 2005, and then many others in Europe supported with pretty strong subsidies that allowed the market to peak at very high levels in 2008.

Impact of Subsidies on Solar Industry Growth

Jordan Schneider:

So how are those subsidies structured?

Andy Klump:

It was effectively a feed in tariff or a FIT, which would effectively was paid. The owner of the project would essentially get a very large payment for the production of solar energy. And so these are for the developers of the systems that's going to the solar systems or the rooftop owners, but that solar module benefit went throughout the whole value chain and caused massive growth of the industry. Particularly during the 2005 to 2008 era. And then going into this five-year time horizon, you were suggesting to 2011, there was then a massive drop in with the global financial crisis in 2008 and early 2009, the industry fell off a cliff. The largest market at the time was Spain and effectively went from 50% of the global market to zero almost overnight. And then the market rebounded once pricing dropped, and that's where polysilicon pricing I mentioned was at $475 per kilos at the peak, and then dropped all the way down to roughly to a hundred dollars per kilo back in 2009 and 2010. And then all of a sudden the growth in Italy in 2011 actually took the polysilicon pricing back up to the mid one hundreds. And there was a resurrection of higher price polysilicon, but since 2011 until now, or I should say in early 2021 pricing dropped quite dramatically. And that's where price effectively dropped like 95% over that period,

Jordan Schneider:

Which is incredible.

History of Solar Manufacturing Costs

Andy Klump:

Oh, absolutely. It was an artificial bubble that was driven by a lot of excessive subsidies. And there was this oligopoly that existed within these kind of five core suppliers. And so with the limited supply and limited expansion, it led to this aberration and pricing. Once there were new players that jumped into the industry in 2007 and 2008 led by the Koreans, but then the Chinese players, some major players like GCL got in the business, then the markets started to correct. And there was a lot of high cost manufacturers who still were making money at a 100 or $150 per kilo. But a lot of these players then went out of business after 2011, 2012, when the markets started to correct again, and then pricing dropped all the way down to the sub $10 per kilo level at the depths of this period. But industry has streamlined its costs and really reduced both the CapEx and OPEX to get to a much more price sensitive level. And so even when pricing was in the 10, $12 per kilo range, some folks were making some money, at least you're able to cover their cash costs. Now we've seen polysilicon pricing jumped back up and that's, that's where we stand today.

Jordan Schneider:

Yeah, Andy, I got to say, I'm very happy. My livelihood doesn't depend on commodity super cycles because I can't imagine how many heart attacks have been caused by this one and many others. So how did the Chinese solar industry more broadly fair coming in and out of the global financial crisis?

Record of Chinese Solar Industry

Andy Klump:

Coming out of the global financial crisis? One has to think about the solar market in the context of what's going to this kind of constant supply and demand shift and struggle. And so part of the interesting dynamic with solar is that the cost being so exorbitantly high, many people wrote it off and never thought it was really a mainstream form of energy, but with the massive drop in costs. So it is the market applications open up and more and more folks adopted solar.

Jordan Schneider:

So there are incredible charts of the cost of solar starting out in the early two thousands being like 10 X, what you could get for coal or oil. And now it's lower. What in your view were the main drivers of this truly, you know, remarkable and likely world changing cost curve?

Solar Industry Cost Curve Overview

Andy Klump:

So if I look back in my early days at Trina, 2006, seven and eight, during that period, polysilicon was responsible for over 90% of the cost of the solar module. And so when all of a sudden your core commodity then drops in pricing by over 95%, then a lot of other things start to become more important. And so then the other components of the solar module involving glass, EVA, backsheet the non-core silicon substrate, which is derived from polysilicon. Then some of these other components became a higher point in attention. What's called silver paste is also one of the key components that's on the solar cell, but over the course of the last 15 years, all of these components have gone through massive pricing pressures. And that's where the overall China supply chain advantage has really helped to allow the industry get to a cost level that makes it appealing for many forms of energy. So you're absolutely right. Solar used to be 10 X, the cost of coal and now solar on a non-subsidized basis is actually less expensive than many forms of thermal energy. So the positive silver lining in all of this is that the market for solar has grown massively just to show the hundred X number. I was not joking with the time I entered the industry. It was 1.6 gigawatts for the entire global industry. And now here we are more than 160 gigawatts. So that's, what's led to the growth.

Jordan Schneider:

We talked about this sort of German subsidies to kick off the industry in the early two thousands. How important were Chinese government subsidies to keep the industry running?

Chinese Subsidiary Impact on Solar Cost Curve

Andy Klump:

The Chinese market in 2000 prior to 2010 was all sub one gigawatt. So it was a very small market that was inconsequential. So there was effectively no subsidies in place to support the deployment of solar on a large-scale basis, really until 2014, 15 and 16, when all of a sudden solar kind of jumped into the mainstream. And during that period, China went from nearly being, not even the top 50 places of importance to all of a sudden becoming the world's largest market. And it peaked in 2017 with 53 gigawatts of overall deployment. So at that point in time, this subsidy scheme in China was driven by getting as many large utility projects in place and drove the market to become not only the largest market from a production standpoint, but also from an end demand standpoint, the global financial crisis specifically was not as detrimental. And in 2008 and 2009 as what one would expect, many manufacturers did not go out of business as thought there was not a massive consolidation in the industry, many struggled and were not profitable for a period of time, but they basically lowered their utilization, hunkered down, waited through the downturn. And then the market rebounded in Europe led by Italy in 2011, then Germany in 2012 and 13. And then when these larger, higher subsidized markets in Europe went offline after 2013 of 14, that's when China stepped in and started to build an in-market from a demand standpoint. There, I will say the second part of the equation is on the manufacturing side. There obviously have been local government incentives. You have various tax breaks or supportive land acquisition to build out manufacturing facilities. But I wouldn't say that's just China. I would say that's a global phenomenon. We've seen that in the US and other places that also have manufacturing subsidies to incentivize local manufacturing job creation. So that's something that's really happened globally. It's not just, just in China.

Evolution of American Solar Industry

Jordan Schneider:

Andy, Bell Labs invented this stuff in the first place, like for the first kind of 30 plus, who was it? Jimmy Carter had some solar panels on his white house. There were 30, 40 years there where it was basically American and to a lesser extent, Japanese, R&D really driving the increase in inefficiency of photovoltaic cells. So how did the American market and US manufacturers evolve over over the subsequent decades?

Andy Klump:

Looking at Bell Labs back in the 1950s, the first solar photovoltaic effect was talking to the Einstein era. This is the knowledge of the importance of solar is not something that started with our modern day generation, but it wasn't really until PV cells were made in the 1950s. And they were really driven by some niche applications in space in the sixties and seventies on a per watt basis, which is how the solar module market defines its costs. We were talking about 30 or 40 us dollars per Watts, and these were just used in space applications for satellites once again, a very niche market. And that's all the, the, the way the market operated.

Creation of Mainstream Renewable Energy Efforts in the US

Then in the seventies, late seventies with Jimmy Carter, putting solar panels in the White House, there was this rebirth of the importance of, of solar energy and looking at alternatives to oil given the driven by the oil crisis. And so out of that period, there was a lot of innovation in the US and major companies. Solar X were acquired by BP. So actually one of our vice president technology, Paul Wormser has been in solar for 45 years. He actually started working at Exxon mobile in their solar division. So there actually were major oil companies involved in solar exploration, the seventies and eighties, and for awhile, the US market did grow. And actually just as another small diatribe, I will also say that there was a lot of growth of the plant industry in the U S and a lot of these are driven by off-grid solar farms because they wanted to be undetected. That actually was, that was the origin of the industry back in the seventies was space and marijuana, but the market then it was very small in the eighties. And then the Japanese started to pick up on solar in the nineties.

Jordan Schneider:

Have you told that story to Chinese manufacturers? What is there, what sort of responses have you gotten?

Andy Klump:

No, no, actually I've not told that story about that part of it, but that actually is true. And so there were a lot of drug dealers who were involved in using solar and deploying solar. So if you talk to folks at the industry in the seventies, like that was it, they were selling off grid, solar modules to folks who are growing Mary Jane in the back alleys of Colorado and California. So that, that is a true story. But the interesting part is actually China did have some small solar industry in like the seventies, but it was just, it was very small R&D labs just hardly anything. Yeah. I think that'd be a stretch to go through and translate all that to the Chinese. So I've actually never told that story just back to the seventies and eighties. And so, you know, solar module manufacturing was a few limited players and we measured the off-take of their facilities in almost kilowatts per the whole overall industry was just a few megawatts during this era, but with the advent of solar X being acquired by BP Solar.

Explanation of Energy Watt Measurements

Jordan Schneider:

And sorry for the ignorant among us, can we get the explanation? What's a kilowatt and megawatt.

Andy Klump:

Okay. Yes, yes. A watt for the sake of those are not familiar with the industry, watts of energy from a solar panel is basically the unit of electricity that folks will equate to. So we measure the solar modules or the solar panels that is in terms of Watts. And then what's going to thousand Watts equals one kilowatt, a thousand kilowatts equals one megawatt, and then a thousand megawatts equals one gigawatt. I know the US is not big in the metric system, but this is the foundation for the industry. So the overall industry was just, as I mentioned, a hundred X growth I was talking about when it reached the 1.6 gigawatt level. But when you look at the seventies and eighties, I think the industry was literally less than 10 or 20 megawatts during that early era.

Jordan Schneider:

Oh, sorry. Like a megawatt can power a 10 houses? Can we do that one too?

Andy Klump:

Just a base form of reference, roughly 5,000 Watts or about five kilowatts is the size of a maybe average east coast home installation. So, yes, if you're looking at a one megawatt, then there's potentially 200 homes that size. California and others may see a little bit higher deployments of 7,000 Watts or seven kilowatts, but it's more or less than that range.

Jordan Schneider:

Okay. So we were not very small bore niche in the seventies and eighties. How did US manufactures fare as Chinese firms starting to getting into the game in the two thousands and 2010s?

Impact of Chinese Solar Growth on Global Industry

Andy Klump:

Yeah. So if you have to then look into the nineties, I would say Japan, then outstrip the US and getting more attention. That's where companies like Sharp and Panasonic got involved in solar. Japan was actually supporting industry. So there was actually a lot of growth in the late nineties with Japanese firms. And there were subsidies in place to support Japanese solar deployment because Japan being an island nation had very high cost of electricity. Prior to 2003 or 2004 China module manufacturing was less than 1% of the overall industry. So very rudimentary, very small shops and Suntec didn't even exist before 2002. And Trina hadn't been in module manufacturing at that point in time. So the time that the EEG was announced in Germany, that subsidy drove the industry to suddenly pay attention to solar. And once Suntech went public on the New York stock exchange in 2005, then everyone realized, there firms that can go make money at this. And for a while, Dr. Shi, he became a multi-billionaire in China. And then everyone said, hey, I want to do that too. And then you saw hundreds of companies that dove into this industry and the competition really emerged. So this is in this Suntech being founded in 2002, three and four, and starting to ramp up during that time was the early stage of that. By the time I entered the industry in 2006, module manufacturing was still less than 5% of global module manufacturing came from China.

Jordan Schneider:

So why weren't the American and German firms able to scale at the rate and speed that the Chinese ones were?

Advantages of Chinese Manufacturing Scaling Efforts

Andy Klump:

So during the 2003 to 2006 era, there were some modular manufacturers in the US and Germany that did grow in scale and were significantly larger than the Chinese, but the real ascension of the major Chinese manufacturers that we know today happened during this period, because they did start to ramp up. They did go public on the New York stock exchange and NASDAQ. They collected capital, and they put all their money into CapEx, which then led to a large, a large expansion of the supply chain. And the real reason that Chinese manufacturers succeeded was because the supply chain advantage, not only could they convince a glass manufacturer to locate next door to their module manufacturing facility, but they could also get half a dozen of their suppliers within a hundred mile radius to be competing on very low pricing. And so that competition drove others to say, okay, we'll give you just in time inventory right next door. And so you don't have to put in your working capital into the equipment or into the raw material components. Whereas US manufacturers still had a very small scale operation. They did not have sub components, suppliers that were as aggressive or able to lower their costs. And so, as a result, the industry shifted towards China for some parts of it. And that's where polysilicon, because of the high barriers to entry, the very high levels of process technology. And know-how that large scale investment of billions of dollars, as opposed to just tens of millions of dollars of CapEx, a capital expenditure, that led to a polysilicon not shifting over to China until later, but the core parts of the supply chain, as I mentioned, the ingots, wafers, cells and modules that then shifted to China more quickly at first.

Jordan Schneider:

So one argument you're coming out of some industry folks and think tanks is the fact that these Chinese firms heavily weighted to capex spending and spent less as a percentage of their firms on R&D. And the argument goes that the scale and speed at which these Chinese firms grew locked us into a technical scenario where the photovoltaic cells weren't as efficient as they could have been had firms spent more money firms like the Western ones spent more money into R&D, do you buy this line of thinking?

Discussion on R&D Developments and Implementation

Andy Klump:

So, first of all, I want to dissect the different facets of the industry, because we're talking about kind of mainstream commodity based manufacturing of some of the core components. And one has to understand that the industry always knew there would be a large commodization of the product and that commoditization drove the industry's cost down and allowed the industry to go through this a hundred X growth, but keep in mind that there were a number of US and European manufacturers of a lot of the core equipment that sold all their equipment to the manufacturers in China. That's what allowed the industry to grow and expand. And so there was a lot of R&D that did come in from Western firms, but some of this was on the capital equipment side. And that's actually what drove innovation that then went on a global basis. So many manufacturers of the core equipment still had an upper hand for a number of years and did make very nice margins for a long period of time as Chinese manufacturers expanded. The second part of the equation that did benefit US firms is the fact that this low cost solar panel or the solar module allowed the downstream part of the industry to grow massively. And so the amount of jobs in solar in the US today is well over hundreds of thousands because this low cost equipment could then be installed by many electricians, technicians, other workers, blue collar workers throughout the industry. And that industry still remains very robust, it's a major form of a new job creation within the US. So that, that is also a key part of the growth of the industry. There's also a lot of other services firms around this, a lot of areas related to financing and business model innovation. That's where the US continues to dominate the growth of that. That's not something that can be outsourced. So there's massive job growth that's resulted from the fact that the manufacturing base has shifted to a lower cost operation.

Thin Film Modules Production

The second major theme, I'd also say on the manufacturing side, which is noteworthy, is that not all modules are made using crystalline technology, although that is the dominant form, the industry, there've been some manufacturers of thin film technology, and that's a very different manufacturing process. So First Solar based in Ohio is one of the largest thin film manufacturers, but the thin film technology has been adopted in many markets around the world, but it's only been really deployed in scale by First Solar and First Solar have they have facilities both in the US as well as in Malaysia and also in Vietnam. So they have actually grown both with Asian-based facilities, as well as the US-based. So it's important to note, there still are some manufacturing jobs that are happening in the US and there's other smaller and mid-sized firms that are involved in solar cell and module manufacturing that do exist in the US and some of those have been Asian based firms. So companies like Kanwall and JINKO, solar have also invested in Georgia as well as in Florida with the US-based manufacturing facilities. So there still is a niche for some folks who manufacture in the U S as well.

Parallels between Solar and Semiconductor Industry

Jordan Schneider:

Andy, you spent some time in the semiconductor industry. What parallels do you see in terms of trying to building up fabs and what not from the solar to the chip space?

Andy Klump:

There are a lot of parallels and there's been a lot of shifts from the semiconductor industry into solar, the industry core technology. It does require a pretty dynamic supply chain, but there's a lot of innovation and ongoing R and D that's needed. But I would just say that the core where the I'd say where the technology differs is the level of complexity and solar is not nearly as sophisticated as what the semiconductor industry is. So that's where the barriers to entry arguably are much, much smaller in the solar side. So while there are a lot of parallels with respect to Moore's law and the doubling of capacity equates to lower manufacturing cost. And a lot of those, a lot of that industry, the economies of scale, industry growth have been achieved. And they're going to continue to be achieved. I would say that there is some limitation because the core technology is also quite different, but that's where I think a lot of companies who have been very large, even companies like Intel, tried to get into this space, GE was in the module manufacturing space for awhile. We saw BP go in and out twice, but I think the challenge for a lot of these large multinationals is the margins are very thin when the market's in a good era, but when there's any amount of oversupply, the margins turned to be negative and become quite challenged. So it's not a space for these typical high businesses companies like Intel we're not destined to be in solar for a long time.

Overview of Clean Energy Associates (CEA’s) Services

Jordan Schneider:

Yeah. You're not, you're not cashing in on, on three nanometers chips and whatnot. Coming to today's news first off, let's do some context on Clean Energy Associates. What do you guys do and how does the the sort of supply chain verification work? You guys done connected you to today's news?

Andy Klump:

We are an engineering services company. And so our core work is we help people to buy products. We inspect products in the factory. We then verify that they're installed properly in the field. And then we have a team that tracks all the data and market intelligence throughout all the work that we do. So with respect to supply chain transparency and traceability, our teams have the ability to go all the way upstream to the core materials, check where the materials are from, and then actually help our clients understand and validate the components that they're receiving.

Jordan Schneider:

Paul, how does this work?

Traceability in PV Supply Chain Services

Paul van Brenkelen:

So if you think about what this tracking and traceability in the PV supply chain, that goes back to our clients. So it would mean if a client is a module manufacturer, then we will go to the sites and start tracking and tracing throughout the premises of that company that using like incoming materials, documents, their housing, the production lines inventory, and the related systems being used. So we can check the computer systems. We can check documentary flows. We can use barcodes used inside the systems inside the client premises. but we will do like a walk around. We will speak with people, interview the different subjects in the organization and the site visit is important. So we can really see what people tell us is there, or what procedures have been written up and are being implemented. And we will take samples during the tour in the sense of, yeah, we are in a warehouse, we pull a few boxes or we pull the visual units out of those boxes and ask them to show us where they did end up in the final product, either up or down. So that is how such a process would work. And based on the capabilities and the know how of the staff inside CEA and their technical know-how and technical capabilities, they will be able to look at certain critical control points in the organization. It has to be sample based because we cannot inspect or assess anything and everything that we need to make a calculated choice to look into something.

Quality Assurance Concerns in Polysilicon Production

Jordan Schneider:

Sure. So, Andy, how has the industry responded to allegations of forced labor and how has that response evolved over the past year or so?

Andy Klump:

So our core customers are the end developers or IPPs, which is independent power producers, or the engineering procurement construction companies, EPCs. They're the ones who are often buying the modules and installing them. So their requests are for us to understand the traceability of all the sub-components of the core modules and the manufacturers themselves are not our clients, but we're doing the inspection work at their facilities. So there are many manufacturers who are willing to help support the requests, but they also were somewhat reluctant to give us all the information that we're looking for. So there's been a little back as we go through and talk through this, keep in mind that the manufacturer of a solar panel or a solar module, they may make the cells and the module, but they don't necessarily make the wafers, or in almost all cases, they don't make polysilicon. There's probably one or two exceptions of folks that are integrated from upstream polysilicon, all the way to modules, but most manufacturers are buying from other sub-component suppliers. And the problem that exists in the industry is that traceability requires them to go through and check with their suppliers. And that's where the reaction has been mixed. Some subcomponents suppliers are willing to provide the data and information and others are more reluctant to do so.

Jordan Schneider:

So what is the political economy of that reaction and not naming firms, but what makes it far more likely to be more gung ho about addressing this versus more reluctant?

Solar Manufacturer’s Role in Quality Assurance

Andy Klump:

Well, I think part of the challenge is that the manufacturers themselves require some amount of incentive to go through and do this extra work, because historically they've not been asked to do this. So I'll introduce a term called MES, which is manufacturing execution system. These are part of the traceability protocol that we've identified is to have a check at the manufacturing level of what they're MES states, where the components are from. And the problem that we've encountered is that some of the manufacturing execution systems that exist only show the product within the walls of that manufacturer. So that may be the cells and the modules, maybe it's all the way to the wafers or the ingots. But as I said, the polysilicon manufacturers themselves don't have an MES, which then connects with the downstream module manufacturers MES. So the polysilicon manufacturers themselves don't have the system in place, but they're willing to go through and do this in some cases, but they also want additional support in terms of the cost, because it does require more work on their side. So one of the gaps is that we've been hearing that these suppliers are asking for more money to this traceability work. I would also paint this in the picture that the large majority of the industry does happen in Xinjiang, we're certainly aware of that, but for the size of the U S market is while it's growing. And it's large, there's actually more than three times the amount of polysilicon supply in regions, outside of Xinjiang. So that's areas like inner Mongolia, Xi'an, and other provinces. They have polysilicon which can meet more than three X, the size of the U S market demand. So each manufacturer can go through and separate where they get their products from and give that traceable information to their customers. Some have not been preparing up to this point. And so they now need to invest further in their systems and their efforts to do this. Others have been doing this traceability work, but there is additional costs to installing these systems and operating them.

Jordan Schneider:

So it's doable. And it's annoying to create two streams of one with Xinjiang polysilicon presumably for domestic Chinese consumption and rest of world consumption. And then another one for the U S and the costs increases like somewhat is not going to make solar more screw up like solar economics more broadly. But it's a question of a few cents of who is going to end up putting the bill.

Impact on Downstream Costs

Andy Klump:

Correct. There are some numbers, it could be anywhere from a one penny, a watt to two and a half pennies. A watt. The question really is what's the scale at which this has done and how many folks are requesting certain items and what the depth of those of those requirements are. So a lot of this is something that the industry is working through right now, but that anywhere from a 4 to 10% price increase is a substantial uptick in costs. And particularly now in the industry is there've been cost increases in polysilicon already, as well as the logistics costs that are a big price increase that many buyers are unwilling to take as well.

Paul van Brenkelen:

The clients of CEA are the developers, IPPs, EPCs, and the financial institutions. So those are the people who contact CEA to deliver those services related to traceability.

Purpose of Solar Developers

Jordan Schneider:

What is a solar developer?

Andy Klump:

A solar developer is a much way that one thinks about real estate development. As solar developers, someone who secures land, they have to secure the permits. They have to have all of the proper steps to ensure that a solar farm or a solar plant can actually be installed at a certain location, but that developers effectively putting that whole financing package and a delivery package together to allow energy for the solar modules to be installed into a system that then produces energy to an off-taker. And that's anyone who's consuming energy. That's where you see large corporates like Google, Amazon, many others that want renewable energy from solar, they're working with developers to meet those needs.

Forecast of Global Solar Industry

Jordan Schneider:

So it may not just be the US which ends up taking some action in regards to Xinjiang and polysilicon. How do the dynamics change if more OECD countries end up signing up for some storm of import restrictions?

Andy Klump:

So there are certainly a number of other geographies that are interested in this inquiry. And so many European customers who are developers and independent power producers are asking for our traceability work as well. But the legislation in the US is really, I think is a little bit ahead at this point. So I think the US is likely to happen first, but then you're absolutely right. Europe and other regions are also likely to follow suit.

Global Changes in Polysilicon Manufacturing

But I think on the flip side, if you look at the growth of the polysilicon industry, many polysilicon manufacturing plans are happening outside of the province we're talking about. So many plants are happening expansions are happening right now in Xi'an, inner Mongolia, Yunnan, Shanghai. And a lot of other regions are seeing massive growth. And so if we look at the overall industry and what the industry is measured in metric tons, we're seeing the industry more or less double in less than two years. So of roughly 550,000 metric tons that existed in 2020, we're at 600,000 metric tons. Today, next year is projected to be over 900 hundred thousand metric tons of capacity. And by 2023, 1.2 million metric tons and vast majority of these expansions are happening outside Xinjiang. So even if the entire global production or customer demand comes and says, they want traceability, all these new expansions will be more than enough to meet the global market need in the coming two to three years.

Jordan Schneider:

Coming back to the sort of, what if everyone shuts off the Xinjiang polysilicon tap does that one to two and a half pennies, a watt price increase like the forward 5 to 10% for the end customer, does that increase or decrease because then more companies get with the program and then you end up having more like Xinjiang free supply.

Andy Klump:

I guess the answer is like over time, I definitely expect them to decrease, but I think that's maybe the initial ask because of the amount of work and effort that's needed. But I do expect that over time numbers will be less than a penny, a watt, but the answer really comes down to what are the process required by the US government and other regions that have traceability protocols in place. And so the more in depth, the inspection work or the more owners, the requirements, the higher that cost threshold could be. So that's where I will say over time as more and more manufacturers of modules prove that they're buying polysilicon from Xinjiang or other regions, then I think these costs will eventually become much, much smaller and less material.

Upstream Cost Impacts on US Solar Market

Jordan Schneider:

Sure. And there's also a potential future where it's not just the US government mandating supply chain verification. It's also them cutting off large Chinese manufacturers who have been accused of allegations of using forced labor to the US market. And I imagine when again, this is speculation, but once you pick a handful of enormous, of really large Chinese players in the supply chain, then this sort of like end cost to the US customer ends up increasing much more than just doing the verification.

Andy Klump:

Correct. So it's not that all these costs are born in terms of the verification costs. It is also what is the investment in the system and the internal resources needed to execute some of this work. Many of these manufacturers are very well versed in these type of topics. This is not the first time there's been a regulatory change. If one looks back to the Obama administration is early as 2013 and 2014, there were a variety of tariffs in place on solar. The industry had to adjust and adapt. There's a lot of paperwork and other legal and other regulatory preparation that's needed. So all these manufacturers will bear a lot of costs on how they prepare and develop their systems to meet the market need. So those are also factors that play in this broader equation, but also understand their sensitivities as well.

Global Solar Supply Chain Working Together for Traceability Efforts

The Chinese government has stated that manufacturers should not be adjusting their systems based on other government states. And so there is a period where there's a little bit of a stalemate where manufacturers can't openly acknowledge that this is their change of approach. They're also abiding by the Chinese regulatory rules, but obviously they also want to abide by American regulatory rules. Those are all private organizations that have a business interest in trying to meet the needs of the market, but there are solutions that are needed, but some of this is based on what regulatory frameworks and requirements are coming out, taking this back to the traceability work that Clean Energy Associates has done. We've worked closely for six months with the Solar Energy Industries Association and drafting a supply chain protocol. That protocol was announced and released in late April. So that is a very large document that CEA has spent a lot of time on. We have three PhDs, all with manufacturing experience throughout the value chain, as well as Paul Wormser, our vice president of technology, who helps support the drafting of this. But that's part of the 40 person team that CEA has in the US was working on these regulatory frameworks and trying to help support the industry growth.

Jordan Schneider:

Yeah, this was going to be my next question is because you had this very striking, awkward moment of the American industry associated and saying, this is a problem. We got to do something about it. And the Chinese industry association saying in Chinese, what are you talking about? We're doing just fine, don't criticize us. We're great. And figuring out how to thread this needle. Like basically it seems like there is a steady state equilibrium where America and potentially European and Japanese panel needs are supplied by supply chains, which don't go through Xinjiang. But that is, that is contingent on the Chinese government, letting these firms make the sort of adjustments that are going to be required for Western capitals and consumers to feel comfortable buying this product.

Andy Klump:

Yes, no, absolutely. It is an unusual circumstance and there's a lot of reluctance by either side to back down. But there's a effectively a different view on terms of labor practices that one considers to be ethical or acceptable versus others. And so that's where there are some macro differences of opinion. Everyone is very clear. They want to have growth in the solar energy industry. There's no doubt about it. And the industry will evolve towards the solution that meets that need. I, as I said, there's a lot of growth, polysilicon and other regions where this is not a concern. And so the market will evolve towards those of those regions that of meet those needs. So we do think that the market will adapt and change. We have to wait for some of these regulatory frameworks to come in place. Another point I'd like to make is that there has been a demand in the market initially driven by France, but then later, also in Korea for low carbon modules. And so a number of manufacturers have adjusted their supply chain they've sourced and ingots and wafers from Norway and other locations that have low carbon footprints or they're powered by hydro. So just as an example, like that has been a policy that's been out there for several years and a number of major manufacturers shifted, and they're able to sell their products at a premium to those markets that have that low carbon content requirement. So yeah, just another facet of ESG is certainly right there in terms of carbon content. So I think it's a great example of how the industry will shift and migrate to the solution that's needed by the market.

Podcast Hosted by ChinaTalk

About the Host

Jordan Schneider is a Senior Analyst at Rhodium Group focusing on China's technological rise, US-China tech relations and the impact of Chinese technology policy on the broader economy.

Jordan is the creator of the ChinaTalk podcast and newsletter, and is also a fellow at the Center for a New American Security. He previously worked at Kuaishou, Bridgewater and the Eurasia Group. Jordan received a master’s degree in economics from Peking University’s Yenching Academy and a bachelor’s degree in history from Yale.