Azure Power Global Ltd (AZRE) 2022 Q1 法說會逐字稿

Ladies and gentlemen, good day, and welcome to Azure Power's Fiscal First Quarter 2022 Earnings Conference Call. (Operator Instructions) Please note that this conference is being recorded. I now hand the conference over to Mr. Vikas Bansal, Head, Investor Relations at Azure Power. Thank you, and over to you, Mr. Bansal.

Thank you. And good morning, everyone, and thank you for joining us. On Monday evening, the company issued a press release announcing results for the first quarter fiscal '22 ended June 30, '21. A copy of the press release and the presentation are available on the Investors section of Azure Power's website at azurepower.com.

With me today are Ranjit Gupta, CEO; Murali Subramanian, COO; and Pawan Kumar Agrawal, CFO.

Sir, sorry to interrupt you, your voice is breaking.

Ranjit will start the call by going through recent key highlights. Murali will then follow up with an update on our projects under construction, technological innovation and industry update. Pawan will then provide an update on the quarter, and then we will wrap up the call with Ranjit providing quarter 2 fiscal year '22 and fiscal year '22 guidance.

After this, we will open up the call for questions. Please note, our safe harbor statements are contained within our press release, presentation materials and available on our website. These statements are important and integral to all our remarks. There are risks and uncertainties that could cause our results to differ materially from those expressed or implied by such forward-looking statements. So we encourage you to review the press release we furnished in our Form 6-K and presentation on our website for a more complete description.

Also contained in our press release, presentation materials and annual report are certain non-GAAP measures that we reconciled to the most comparable GAAP measures and these reconciliations are also available on our website, in press release and presentation materials and annual report. It is now my pleasure to hand it over to Ranjit.

Thank you, Vikas, and a very good morning, everyone. As you all know, India faced COVID-19's ugliest phase in the first quarter during April and May. They were massive medical emergencies witnessed across the country during its peak around mid-quarter, which along with local restrictions and greatly restricted man and material movement.

The daily infection rates receded around end of June and in the meanwhile, our vaccination drive also picked up pace. As we speak, well above 600 million doses have been administered in India till date and about 1/3 of the eligible adult population has at least got one dose. At Azure, we are still by our employees and stakeholders in this fight against the pandemic through several initiatives focused on supplementing medical supplies, tracking the health and well-being of team members and their families, organizing awareness and mental wellness talks and providing whatever support was needed by team members.

We also organized 2 vaccination drives for our employees, and I'm happy to report that 99% of our eligible employees are now vaccinated. I know that some countries across the world are still battling the pandemic, and I wish them all the best in dealing with COVID-19.

Moving on, I'm happy to report that we began signatory to Human Global Compact this quarter. We will fully align ourselves to the strength and skills on human rights, labor, environment and anticorruption. Our ESG risk score by Sustainalytics has further improved and Sustainalytics now puts us in the low-risk category compared to medium-risk category earlier.

MSCI, the leading ESG rating agency rates Azure Power as AA, which places us in the top quartile of all global utilities they cover and probably the highest amongst our peers in the country.

We also recently retired verified carbon units, VCUs, to offset our Scope 1 and 2 emissions of 2019-2020. And we will continue to do so in future on our path towards carbon neutrality. We continue to strive hard to improve our ESG performance and demonstrate our leadership.

We have a couple of major organizational updates to report subsequent to quarter end. IFC and IFC GIF, which has been long time support with Azure since our early days and has a major role to play in our journey, recently sold their entire balance 19.4% stake in the company to OMERS, one of the largest Canadian pension funds with net assets of over CAD 100 billion.

The confidence that global long-term patient infrastructure capital investors like OMERS have placed in Azure demonstrates the strength of our company and cements our status as one of India's premier renewable energy power producers with strong governance and profitable growth track record.

Presence of CDPQ at OMERS (technical difficulty) an investment view also strengthens our position in terms of meeting capital requirement for our pipeline of projects. We also recently placed our third green bond in the debt capital market, primarily to retire our first green bond.

The issue received tremendous response (technical difficulty) participation from top global asset managers with book building in excess of 5x in diversified order book across geographies. The issue closed at the lowest-ever coupon in the high-yield segment for any business out of (technical difficulty) from the lowest offering from any Indian renewable energy company till date and shall result in over 200 basis points of annual savings in landed interest cost for our 611 megawatt of underlying assets.

Both these events have further solidified Azure's position as the destination of choice for both equity and debt capital investments. We also reported last quarter on our agreement to sell Rooftop portfolio to Radiance Renewables which is the first-ever asset sale in Azure Power's history and signifies our commitment towards capital discipline while recycling capital into higher return committed projects.

We are in the process of obtaining the content of off-takers and lenders and see the transaction closing over the next few months. I had mentioned in my previous remarks, how we are looking to increase our addressable market by foraying into wind and solar/wind hybrid space. Continuing with that, we participated in a few auctions in the last couple of months and have done some capacities for which we are awaiting letter of award.

We firmly believe that as the industry moves towards providing dispatchable renewable energy to the grid, wind and storage will be 2 important technology additions we have to plan for our portfolio. I had mentioned in my previous remarks how green hydrogen and planting storage cost have the potential to disrupt our industry. We continue to monitor developments in both the reciting technologies and will keep you posted as we take steps to deploy them to improve our returns.

On the 4-gigawatt projects, for which we have letter of awards from SECI, but are yet to sign power purchase agreements, we had a positive update from SECI informing us that they have signed power sale agreements with a couple of distribution companies for a total of 800 megawatts. This is start of the first tranche of 3,000 megawatts of PSAs that SECI is looking to close as far as the manufacturing-linked scheme. We expect to have PPAs signed for about 1/3 of the 800 megawatts soon.

As the second wave has eased, we have seen renewed interest in buying power from distribution companies. Despite the pandemic, power demand recovery in India has clearly bounced back with peak demand clocking 200 gigawatts last month, which has encouraged DISCOMs to invest in buying power for their future needs.

Today, we are 23% more megawatts operating than we did at the same time last year, excluding the Rooftop portfolio. There has been an 11% year-on-year increase in EBITDA from operating assets and a 12% increase in cash flow to SAB from operating assets during the quarter period. We continue to see steady improvement in this metric.

The government continues to support the renewable energy sector in India. India recently achieved 100 gigawatts of installed renewable energy capacity in the country, making us the fourth largest in the world. Honorable Prime Minister at his independence day speech from the grand past of Red Fort reiterated government of India's mission of 450 gigawatts of renewable energy capacity by 2030. And more importantly, announced the path towards India's self-reliance in energy by 2047, i.e., 100 year of Indian independence.

This is a significant announcement given that the climate change imperatives, coupled with energy's self-reliance target greatly enhances the renewable energy base for us. Most of India's oil and gas needs are met through imports and the only way towards energy independence is to bank more heavily on renewable energy.

In another significant positive development, a recent landmark judgment from the Appellate Tribunal for electricity in the country allowed compensatory tariff for solar power curtailment in the state of Tamil Nadu. This has provided a tremendous boost to investor confidence in the sector and bodes well for our growth trajectory towards 450 gigawatt installed capacity in the country by 2030.

For the first time in India, the Appellate Forum for Electricity has laid down the law that the developer will have to be compensated on account of illegal curtailment even in absence of a compensation clause in the power purchase agreement. We continue to look for suggestions from our investors and stakeholders on how we can further improve our disclosures and make it easier for you to understand and value our business.

With that, I would like to turn it over to Murali.

Thank you, Ranjit. As we last reported, the second wave of COVID at its peak impacted our projects under construction, not only disrupting the supply chain, but also impacting construction activity at several of our sites. Subsequently, however, both COVID and the supply situation has improved significantly. As of today, we have completed and commissioned 400 megawatts in our 600-megawatt Rajasthan 6 project and another 100 megawatts has also been completed and is now awaiting commissioning.

The remaining capacity in this project is expected to go live in the next quarter. Thanks to the MNRE's notification granting extension to all projects with commissioning due dates on or after 1st April 2021, we don't expect to incur any penalties for delays.

Construction work on 300 megawatts Rajasthan 8 is now underway full swing and construction activity in 300-megawatt Rajasthan 9 project has commenced in right earnest.

Work in Assam that had picked up after the initial COVID-related delays again took a hit due to the second wave of COVID and inclement monsoon weather. Despite this, after the initial 25 megawatts, we have commissioned another 12.5 megawatt capacity in May. The next 12.5 megawatt is currently under commissioning.

The full 90 megawatts Assam project is expected to be completed and commissioned by the end of this calendar year. We have provided some highlights of our ESG accomplishments on Page 6.

As Ranjit mentioned earlier, we received a AA rating from MSCR for CSG and our ESG risks score by Sustainalytics has improved the low-risk category. We highlighted our ISO-45001 certification earlier this year, which demonstrates Azure's focus on occupational health and safety.

I'm happy to report we have also recently won the Greentech Effective Safety Culture Award for 2021 from Greentech Foundation, which signifies the efforts we have put in to ensure safety culture is embedded across our project locations and sites.

Our carbon-free generation has avoided about 1 million tons of CO2 equivalent this quarter, bringing the total to 10.5 million tons equivalent since inception. We remain net carbon neutral. We have planned for planting 15,000 trees in the immediate vicinity of our 600-megawatt Rajasthan 6 project site during this year. We also remain actively engaged with the communities where we operate and provide proactive support towards medical and health facilities, especially on the pandemic front.

We are now looking to roll out our sustainability chart of this fiscal and shall continuously strive to implement best practices to enhance our sustainability. On the technology front, Azure continues to be an early adopter. We were among the first companies in India to install a large scale project base on mono PERC panels, and we have just started construction of another large-scale project using bi-facial tracker technology, where we expect yields in excess of 30% for our Rajasthan 9 project.

These are industry-leading efforts to ensure our projects are built and operated with the best returns metrics. Looking at industry and regulatory update on Page 7. India's achievement of 100 gigawatts installed renewable energy capacity has been a big milestone, and the country continues to offer solid growth opportunities along with adequate confidence measures for investors in the sector. The Government of India announced recently details of the USD 40 billion reforms-linked package for distribution companies, which will improve the health of the financially weaker DISCOMs.

It is also expected to help clear the backlog of PSAs that the power supply agreement to be executed with DISCOMs who have not been signing PSAs, which was accentuated by the second COVID wave and falling tariffs. The good news is that overall power demand in India has started to grow now as the country emerges from the second wave. We are pursuing new opportunities such as wind and hybrid, and we assure you that we shall only bid for projects at commercially viable tariffs.

We continue to believe that we would be able to obtain the 4-gigawatt PPAs at value-accretive tariffs. This would add to a contracted pipeline and provide returns above the cost of our capital. With that, I turn it over to Pawan to discuss the quarterly results.

Thank you, Murali. Turning to Page 9. As of 30 June 2021, we were operating 2,052 megawatts on a PP or AC basis, which is 23% higher than what we were operating a year before. Our portfolio of 6,955 megawatts remained stable from the previous quarter. While these portfolio megawatt numbers exclude Rooftop portfolio, which is in the process of getting transferred to Radiance, our financials number continue to consolidate Rooftop till the transfer process is completed.

On Page 10, looking at the quarter, our revenues continue to increase as we construct more projects. After adjusting for stock compensation expenses, our EBITDA has been USD 50.7 million or 14% higher against 13% increase in revenues from the same quarter in the prior year.

Turning to G&A. On Page 11, our G&A increased marginally by 5%. Excluding noncash items, the G&A was flat year-on-year. As we remain focused on controlling our costs, we continue to expect our FY '22 cash G&A to rise above 10% from FY '21 level.

As already shared, we've recently issued a green bond of USD 414 million at the lowest-ever coupon in high-yield segment out of India. All our recent refinancings both in the domestic as well as overseas markets have resulted in substantial savings in interest rates, thereby improving our equity returns.

Refinancing at lower costs reflects improved credit profile of the group supported by strong sponsors such as CDPQ and OMERS. We continue to expect lower interest rates for our ongoing as well as new financing and refinancings.

Turning to stock compensation expenses. As the sales price rise, our stock compensation expenses will increase thereby increasing our G&A. For first quarter of fiscal 2022, we had SCR expenses of USD 1.3 million. Despite the challenges in the past few quarters, our DSOs have been fairly consistent at around 120 days on an average in the recent quarters.

We believe there will be further improvements in the future, with commissioning of projects with high counterparty and as we continue to focus on improving our collections. On Page 12, you can see that EBITDA from operating assets increased about 11% year-on-year and that cash flow to equity from operating assets rose about 12%.

Net debt for operating assets was about INR 1.19 billion and EBITDA for the last 12 months was about INR 182 million, resulting in net debt-to-EBITDA ratio for operating assets of around 6.6x as of 30 June 2021. Finally, looking at Page 13, providing balance sheet information, we had about INR 90.6 million of cash and cash equivalents, and our net debt stood approximately at INR 1.34 billion.

As a reminder, for those that are calculating our debt ratios, the hedging assets of INR 105.8 million included in other assets on our balance sheet should be netted against our total debt as this is directly linked to the foreign exchange hedges we have put in place related to our green bonds.

During the second quarter, we have used part of these assets related to our first green bond to reduce the leverage on the green bond portfolio. With this, now I pass on to Ranjit to provide some commentary on our guidance.

Thanks, Pawan. I'm very happy to report that despite major disruptions during the quarter due to COVID, we have been able to achieve upper end of both our revenue and PLF guidance for this quarter provided during last quarter. Even though we have just started on the recovery path from the second wave as of now. We would reiterate our numbers for the current fiscal, but we keep the market posted in our upcoming updates. For second quarter '22, we expect the revenue to be between INR 3,600 million and INR 3,800 million and the PLF to be between 20.5% to 21.5%.

With this, we will be happy to take questions. Thank you very much.

(Operator Instructions) The first question is from the line of Philip Shen from ROTH Capital Partners.

This is Justin Clare on for Phil today. So I first wanted to start off. So there's the 800 megawatts of projects related to the manufacturing-linked tender that have recently had the PSA signed. And I wanted to understand the next steps in this process here. It sounds like you could sign a PPA in the near term, but are you in negotiations right now for that PPA? And could this happen in the next week or month or what's the time frame expected?

Justin, thanks for the question, and this is the most important question for us. We are very, very happy that SECI has signed this 800 megawatts of PSA. So 1/3 of the 800 megawatts, which is our allocation, 2/3 will go to our peers. So 1/3 will come to us.

The 1/3 allocation of 800-megawatt PPA, there is nothing being negotiated or anything of that sort. SECI is going through their internal process to get approvals to sign the PPA. And it's a question of today, tomorrow, day after, then they will invite us to sign the PPA is what we have been told. So SECI is going through their internal process for approval, and then they will invite us for the power purchase agreement.

Okay. Got it. So then I wanted to understand, it sounded like you might have to take a markdown on that PPA or at least there was that possibility. Could you talk about that potential still? And I think in the past, you could be 10% to 15%. So what is that likelihood for the markdown? What could the amount be?

And then is there a negotiation going on, on that? Or is it really that SECI is going to come to you with what the PPA will be?

Yes. Justin, there is no negotiation going on whatsoever. The PSAs have already been signed at a tariff of INR 2.61 with the distribution company. So the PPAs will be signed at the INR 2.61 minus the SECI margin. And there is no negotiation involved whatsoever.

Got it. Okay. That's clear. And then -- so there's the 800 of PSAs that have been signed so far, what's your expectation for the remainder of the capacity related to the manufacturing-linked tender? And then is it that you will get 1/3 of that capacity? So each time there's a PSA signed you would expect to get 1/3 of that? Or is there some other way that, that might work?

So Justin, whenever a PSA is signed, we expect to get 1/3. Apart from the 800, there is another 1,150 which has been approved by the respective Boards of the distribution companies. So we are expecting those PSAs to be inked soon. That will take us to around 2,000 megawatts. And another 1,000 megawatts, there are 2 other 500-megawatt distribution companies that SECI is in talks with, where the initial approvals are being taken as a distribution company's Board. So those will take a little bit longer, maybe a few weeks. But the first 1,000 megawatts, 1,150 megawatts that I'm talking about should happen faster than that. It should happen in the next week or 2.

(Operator Instructions) The next question is from the line of Puneet from HSBC.

Yes. Congratulations, Ranjit, you moved the first step. My question is, when do we have to start investing on the manufacturing side after the PPA gets signed?

So thanks for the question, Puneet. So as far as the manufacturing is concerned, right, there is a manufacturing contract agreement that needs to be signed with SECI. So as soon as the manufacturing contract agreement is signed, then I think we have 2 years from that date to build the plant.

So once the manufacturing contract that we have designed, then we will, I guess, the order placement, et cetera, will take place. And I would think that the investments will happen about a year or 15 months after the manufacturing contract agreement is signed, Puneet.

Okay. And would it also be proportional the way you signed it or you have to still invest a full 500 megawatt?

So this is a discussion that we are doing with SECI, and we will do it with our technology partner also, the manufacturing partners, to see what makes the sense for them? How does it go with their modular construction plan? So we will have to see it. I'm sure it cannot be done, for example, in 50-megawatt kind of base, so we'll have to figure out what is the right size, which we can actually set up.

So we -- SECI has also told us, and we have also told SECI that let's get the PPAs signed first and then we will discuss this. There is no clearing hurry. We are committed to building the manufacturing capacity. And therefore, SECI also understands that, that it is going to help us if we do the manufacturing capacity. So they are happy to discuss this first after getting the PPA signed.

Understood. And my second question is, you said even at INR 2.61 minus maybe INR 0.07, it is still attractive enough. But when I look at the project costs, which we disclosed this time, it was $0.52 on DC basis. Back in September 2019, it was about $0.46. So it is still worse-off and tariffs are low. What has changed favorably on this, if at all, anything else?

So it's a little bit unfortunate that the DC cost is being portrayed the way it has been portrayed. And the reason is that very few megawatts were commissioned in this quarter. And this number is actually an accounting number, given whatever on -- as on March 31 what was capitalized in our book, on June 30 what is capitalized on our book, we take that 2 numbers, divide that by the number of megawatts that have been commissioned, then we come out with the number.

So that number doesn't fully reflect the cost of -- when the numbers are small, when the commission's numbers are small, it doesn't fully reflect the actual cost of the project because if you see the AC cost per megawatt in this quarter is $0.53, right? And the DC cost is $0.52. So the AC cost and DC cost is almost the same. So it's just unfortunately, the way the accounting is done. So there are -- the costs are not at the levels of pre-COVID, yet the costs are slightly higher than that, but they are coming down.

Can I just add to that? It's Murali here. You see the other thing that's happening is, you can't just do a like-for-like DC cost because what constitutes a DC has also changed. Back in the day, it was a polycrystalline module with fixed tilt. Today, we are looking at mono PERC modules, we're looking at bi-facial, we're looking at tracker. So a lot of things have changed. So the yield per megawatt has also gone up, the temperature coefficients have improved.

So the overall performance has actually gone up. So the cost for what may be a metric if the technology is standard. But if the technology changes and the yields change, then there will -- we have to account for those. The second thing, of course, is that, as we all know, interest rates have dropped since what it was 2 years ago. But of course, that's a different discussion. So these are the other things that's changing in the environment.

So for the balance, 903 megawatts, which is still have to commission, how -- what kind of PLF should one expect?

I'm sorry, can you please repeat that question?

Say for the balance 903 megawatts which you still have to commission, what kind of PLF should one expect from those projects?

So the current project that has -- which is under commissioning of which 400 is commissioned, 100 is ready and another 100 is away. There, we -- this is built on 1.5x overloading. So we would expect a PLF in the range of 29% to 30%.

The next project is built with 40% overloading, so the PLF may drop a little bit. And then the next project, which is coming up with trackers and bi-facial will be well north of 30%. It may be in the range of 31% to 32%.

Okay. And the cost for bi-facial would still be the same or will that go up as well?

No. The cost of bi-facial would go up and because it's trackers, it would go up. And so that's why -- just looking at the cost per watt may not be a fully -- it probably wouldn't give the entire picture.

Okay. Understood. That's very useful. My last question is post your green bond issuance, what is the average interest cost? And how does the majority profile look like now?

So the interest cost of this bond that you've raised is that -- actually, the coupon is coming closer to 8 percentage, less than 8%. And then we -- as you speak, we are refinancing projects and how much in there and the kind of response we're getting from lenders is very encouraging. So in the process of refinancing, we are getting interest rates for our new refinancing as low as 7.5%. So it is ranging from 7.5% to 9% based on project to project or which -- like whether we are completely refinancing or we're in the process of refinancing.

(Operator Instructions) The next question is from the line of Maheep from Crédit Suisse.

Maheep here from Crédit Suisse. Just on the PSA for the rest of the 1 gigawatt, maybe if you can talk about what PSAs do you expect on that? And is it still in that INR 2.61 per kilowatt hour with that SECI margin of $0.07, sorry INR 0.07 or is looking -- how you're thinking about that?

Maheep, thanks for the question. So the first 3,000 megawatts tranche is likely to be at the INR 2.61 tariff. The 1,000 megawatts, 850 megawatts that I'm talking about approvals that has -- we are told they have been approved by the respective Boards of the distribution companies, are at the INR 2.61 mark.

Perfect. And for the future projects, like how are you thinking about the PSAs or PPAs for the future auctions? Because the solar plus wind hybrid which you won recently this month was signed at INR 2.35, right? So does that pressure that INR 2.61 for future auctions? Or how are you thinking about it?

So once the first 3,000 megawatt is signed, then we will -- then SECI starts the process of marketing the next tranche, that's when we will know because I guess the projects at that time, whatever is the solar tariffs around that time, let's say, 4 or 5 months down the road, that's how SECI will look at because that's what the distribution companies will ask.

The current tariffs that we are talking about are reflective of the pricing that we bear in a few months back. And when we do the new tranche, the second tranche, it will be reflective of the solar tariff around them.

Got it. And maybe in terms of the timing here, I know that all of these signings or auctions or PSAs are with SECI and there's obviously a lot when the government has to deal with right now. But in terms of timing or maybe from an equity point -- capital needs point of view, when do you expect to need that $150 million of equity for these projects? And how do you -- what plans do you have right now? And what are you thinking about potential options to fund the first gigawatt for yourself?

So Maheep, as far as the construction is concerned, right, as soon as close to 300 megawatts of the first tranche because 800 megawatts we'll get 1/3 gets signed, we will start the process of financial closure and so on. As I had mentioned in my last call, we have already started acquiring the land. So the land acquisition is well underway.

So we are in a very great position to start planning the project construction on the ground as soon as the power purchase agreements are signed. The -- as you know, we have 2 years to build this project. So that means that typically, the equity requirement happens about 9 to 12 months before the project is commissioned. So therefore, the equity need for us would come about a year from now, assuming that the power purchase agreement is signed this week, then we will meet the equity in about a year from now.

So well before that, we should go out to the investors, the shareholders, to the stakeholders and figure out how we want to raise the money. So we have plenty of time.

Got it. Got it. No, that's helpful. And maybe just one last one for me and then I will jump back in the queue. Like just in terms of the overall supply environment for solar and now you're increasingly getting into wind. So how are you thinking about the module supply on prices and turbine supply and prices for the next 2 years?

So as far as the wind turbine module prices are concerned, wind turbine -- sorry, wind turbine machines are concerned, the pricing is a little bit more stable. The technology is a little bit more stable. It is largely dependent on the price of steel, whatever little variation happens on a quarter-to-quarter basis on the turbine manufacturing is on account of steel.

So that is how we expect and it is easier to -- more easy to predict the cost of wind turbine. As far as the modules are concerned, we understand that they are heavily dependent on the cost of polysilicon at the moment. We are seeing some stabilization in polysilicon pricing. We are seeing that supply is easing up. And what we hear from the manufacturers is that they expect prices to return to preCOVID levels over the next couple of quarters. So that is what we have -- that's what we have to go on. We are going to -- I mean, in India, most of the modules which are going to be procured over the next year or so will be done through Indian manufacturers. So we are in touch with them too to figure out what their view is on module pricing. And everybody feels that within a couple of quarters as the policy refund manufacturing comes online, we will see a moderation in pricing.

That's really helpful. And actually, just a follow-up on that, if I may? What tariffs are embedded in the INR 2.35 PPA? And yes, and that's the last one for me.

Very difficult for us to answer that. I don't even know whether I should answer that. So I would like to pass on that, Maheep.

(Operator Instructions) The next question is from the line of Moses Sutton from Barclays.

Just continuing on the inverter loader ratio, what should we be using in our models? I think you noted project is down to 1.4, should we still assume 1.5 long term? Would you be doing 1.5 for bi-facial? And then even on top of that, are you even thinking of using symptom of modules down the road as First Solar has decided to start local manufacturing?

So the load is really a function of the specific site. It's a function of how much clipping that we see, right? So when we build the 600 megawatt Rajasthan 6 project, it's located in Bikaner, where the insulation is the 2% or 3% lower than a couple of hundred kilometers west where we are building our 8 project. So over there, we dropped it from 1.5% to 1.4%. That's one parameter.

The other parameter is, of course, the cost of everything involved. We see the -- we trade off the clipping losses versus the incremental investment for the yield, right? And then on your next question, which is on the bi-facial side, yes, the module costs are a little bit more expensive. We are also going through it, bi-facial and tracker. We've done some pilots. We've reviewed the performance of certain other pilots across the country, and we understand the gain that is possible with a tracker on bi-facial modules mounted on a tracker. So there, we are going in the range of 1.3 to 1.35 as the overloading.

When it comes to First Solar -- I think -- sorry, I thought there was a question on thin film. So on First Solar, I think let them come on. I think they are planning to come in with their Series 7. Let's see how that -- the facility sort of gets underway and when they would be ready for production. And at that time, depending on the market, we'll take a view on how that would sort of fit in.

No, that's very helpful. And should we think for the inverter load ratio, should we think of that 1.3, 1.4 range as a better assumption beyond April 2022 when a good portion of what you're building is going to get affected by the safeguard duties? Or should we not think of it that way?

No, I don't think we can think of it that way because it's largely driven by the site conditions and the yield and the cost of panel, bi-facial tracker, et cetera, right? It's the cocktail of all of these things, which would eventually lead to the decision in terms of whether it's going to be 1.3 or 1.4.

Whenever we have greater yields with bifacial and tracker, then the inverter loading drops. If you go to the southern latitude, say, in Karnataka or Tamil Nadu, you might still go with 1.5 because the overall radiation is a bit lower. So the effective energy which has clipped maybe a little lower, right?

Got it. Got it. And then for the solar wind hybrid projects, what's the IRR we should think of the hurdle rate for the business there? Is it a bit lower than the mid-teens so you can sort of get your footprint there? Or is it in line with the rest of the portfolio?

I think the IRRs would not be too different. They would all be very similar. Although I don't know how to model some of these things, but there is a phase that's commonly used the risk-adjusted returns. So I think the risk-adjusted returns would all be very similar. So what we do is in case of a wind project, we -- the way we model the wind, we build in for the fact that wind can have certain more uncertainties compared to solar. The deviation in Insulation is varies between just a few percent year-on-year, whereas in wind the deviation can be a little higher. So to account for this additional deviation, the energy yield estimates are slightly lower. That's already factored in the EYA study, the yield assessment study.

So once that is taken into account the construction risks are fairly similar. There is nothing more or less. In fact, solar, the land acquisition challenges can be a little more because you need continuous plan, whereas in wind the act of installing turbines and building the foundations are a little bit more tricky compared to solar.

So these things trade off each other, and they're not dealbreaker. They're not materially different from each other in the overall scheme of things. So therefore, once you account for the resource variability in wind, which is slightly higher than solar, then beyond that point, I would assume that the returns would be similar for a solar or a wind or a hybrid.

Okay. No, very helpful. And then last one on trackers. We've seen a lot of noise with steel and freight impacting the economics on trackers. CapEx in India, you're a lot lower than the fully stacked CapEx in some other countries. How should we think of the tracker cost either on a U.S. cents per watt or as a percentage of the total CapEx on the project. Of course, this is for projects that are in the upper range because they're using bi-facial already from the cost profile. But how should we think of the type of tracker and its cost profile within the broader stack?

Good question. The bulk of a tracker is beyond the IP or the technology. The bulk of the cost is actually coming in steel. And unfortunately, over the last 6 to 8 months, we have seen a very significant rise in steel prices. And therefore, this has pushed up the cost of trackers more than what you would have hoped say, same time last year, for example, right.

If I don't account for the increased cost of steel, then the cost of a tracker -- so even if I didn't install a tracker, I have a basic cost, right? So I'm only going to look at the incremental cost. So if I didn't account for the increase in steel, then the incremental cost would have probably been in the range of $0.05 or $0.06. With steel, it's probably be a $0.01 or $0.02 more.

(Operator Instructions) Ladies and gentlemen, there are no further questions. On behalf of Azure Power, that concludes this conference. Thank you for joining us. You may now disconnect your lines. Thank you.

Thank you. Thank you, everyone. Thank you very much.

Thank you.