First Solar’s Easily Localized Module-Making Process Bodes Well for PLI: Interview

Module manufacturing costs in India would come down in 5-10 years


Unlike the cell-wafer-module process of polysilicon-based solar modules, First Solar’s thin film solar module technology allows it to easily identify local vendors for manufacturing processes, Sujoy Ghosh, Vice President & Country Managing Director India at First Solar, said in an interview on the sidelines of Mercom India Renewables Summit 2023.

Ghosh added that First Solar has a distinct advantage in achieving localization, one of the key aspects of the PLI program for which the company qualified earlier this year.

Here are excerpts from the interview.

First Solar was one of the winners under the Production Linked Incentive (PLI) program. What’s the production timeline?

First Solar is building a 3.3 GW per annum integrated thin film model manufacturing in Tamil Nadu. The factory will start commercial production in October this year. And then, it will take us at least 12 -14 months to ramp up the plant’s throughput to get to the nameplate capacity.

We expect that by the end of 2024 or early 2025, we will be at 3.3-3.4 GW per annum capacity of modules.

The PLI criteria are that you must demonstrate full-level integration and nameplate capacity. Whatever is being produced on the day of inspection by Solar Energy Corporation of India (SECI) will be extrapolated to annual production.

You also have to demonstrate the integration. So, they will check whether you are starting with the semiconductor and finishing with the module in one fully integrated process. The third eligibility criterion they have is localization. They will begin with a local value add of 68% in the first year, ramping up to about 90% by the fifth year.

As we start our factory in October of this year, we won’t be at those levels required under the PLI program because supply chain development and localization are time-consuming.

The critical aspect is to create the local component ecosystem because that’s where you eventually deliver a competitive product.

Is the vendor development process for First Solar different from other manufacturers, given the proprietary thin-film technology?

Most of our suppliers would also supply to the crystalline silicon industry. Our specs might be different for things like module junction boxes, cords, plates, and frames, but the companies would be more or less the same.

The vendors will have a slightly different product for us and a little different for the crystalline silicon module.

So a vendor is not just tied to you; can he supply to others?

The vendors are not exclusive to us. Some of them are, but there is a mix.

You’ve been in India for some time now. India’s core competency is not (module) manufacturing. How would you describe the trajectory of the last ten years?

Solar photovoltaics is a unique industry where 90% of the global value chain is consolidated in China. And that was perceived to be an advantage until COVID happened.

That advantage has quickly turned into a disadvantage. China’s cross-subsidization approach allowed them to export at below cost, which was perceived as an advantage until recently.

And that created a tailwind for the power tariffs to go down. But it has destroyed any kind of domestic manufacturing and domestic R&D initiatives outside of China for the whole decade between 2010 and 2020.

As you look at this decade, and when you start to deploy technology at scale, the developers want traceability of the supply chain.

And during the last two years of the previous decade, the pandemic exposed the risks of over-dependence on one country and one province within that country for 40-50% of the global supply.

So that was perceived to be a risk, which made certain other countries like India and the U.S. to start looking at how they de-risk their deployment of solar energy.

India has connected demand with a manufacturing policy and adopted a carrot-and-stick policy. I would say ring-fenced imports, incentivized with the tax rate and the PLI to perhaps equalize the increased costs, which one must go through when starting to manufacture.

Over 5-10 years, that cost is going to come down. So I think the government of India thought about it the right way, which is why you saw the result of the significant amount of interest from Indian corporates to invest in what will be the mainstream energy resource for the next decade.

Similarly, you saw the U.S. come out with an Inflation Reduction Act (IRA), where they had $350, or $365 billion of investment into these industries for clean tech, which was not just solar PV, but even energy storage and electric vehicles, and then there was a separate incentive for semiconductor manufacturing.

Another big change that we see happening is the creation of alternative wholesale, commercial, and industrial markets, where the ESG drives corporate procurement of renewable energy, and the decarbonization needs of corporate or the commercial sector are growing, despite the issues they have with the DISCOMs.

Today, the elephant in the room is that solar and wind tariffs are going down, but retail tariffs are increasing. And that’s a direct function of the inefficiency of the DISCOMS.

There’s a point beyond which industrial consumers cannot bear the cost of cross-subsidy because it will make manufacturing less competitive.

And you see industry switching wholesale to building large-scale renewable parks and consuming energy through the grid aided by some of the changes which the government has regulatory changes, like the ISTS waiver.

Given the manufacturing processes of First Solar is quite different from the technological aspect, does that impact market competitiveness?

If you look at the crystalline silicon value chain, the cell-to-module conversion costs, where you take a cell, and you put glass, back sheet, and frames to make a module, the bill of material cost is about $0.12-$0.13/Watt (W).

The cell cost today is around $0.14-$0.15/W. If you start making cells and import wafers, your wafer-to-sell conversion is about $0.03-$0.03/W. So, you add $0.13 and $0.04, which comes to $0.17/W, so the wafer cost is predicated on the cost of polysilicon. And if you look at your wafer cost, it could be anywhere between $0.06 and $0.08, depending on the price of polysilicon. That’s the hard cost.

In our case, we don’t have the wafer cost. The downstream components, which are glass and frames, are more or less similar to that of other companies.

Because suppliers are captive to us and we do believe in long-term global contracts, we bring in a different scale. We might have a slight cost advantage, or even if it is not there, we should be at least lower compared to any other manufacturing operation in India.

For us to still be fully integrated, our capital equipment side is much higher than the wafer-cell-module operation. Because we are also doing semiconductor creation, by the way.

Our plant and machinery are closer to full integration. But in a solar module business, the fixed cost is about 20 to 25% of overall costs, and the variable cost, which is all the components, is the balance.

So, as long as you optimize your variable cost, you can hit a more competitive cost profile. The other thing about our technology is that our Bill of Materials is very simple since we have no dependencies on China. It’s easy to localize; there are no entry barriers to that.

It’s mostly glass frames and junction boxes. Yes, we have time to qualify the vendors from a quality standpoint. But these are not high-tech products, unlike wafers.

So from a cost point of view, we are competitive. Globally, today, we have 68 GW of booking backlog. So we have booked contracts for 68 GW against a manufacturing footprint globally of 10 GW. So that’ll give you some idea of how competitive we are.

Where does India stand currently vis-à-vis China on manufacturing expertise?

It’s all a function of the industry’s ability to deploy CAPEX. And recognize the shifts in technology. So planning for technological shifts will mean you need to have sustained interventions. So, your plants must be up to date to start producing the latest technology.

It boils down to what kind of risk you’re willing to take as an investor in building manufacturing.

I think it’s a well-proven fact that China acquired this status of becoming the manufacturing powerhouse globally, not in one day; they’ve worked on it.

Yes, a combination of the government’s attractive subsidies to incentivize manufacturing and create long-term demand helped China build 100 GW per annum of solar.

That justifies their investments in terms of scale. So, the point is that China created the domestic demand to a scale where you could bring in manufacturing to cater to that scale.

China thought about this ten years ago. The difference between then and now is that people want to have a distributed risk around supply chains after a couple of Blue Swan events in the world.

And so there’s an incentive for governments to do what the Chinese did, which I think is what they’re doing.

(Note: Sections of the interview have been paraphrased for better reading. Check out the video for a full chat)