Decoding Sun Edison’s Record Bid for Selling Solar Power in India – Realistic or Wildly Optimistic? (Part 1)

The recent brouhaha over Sun Edison’s record low bid of selling solar power at Rs. 4.63 per kWh has shaken up the entire solar universe (which just 3 GW in size… by the way) in India. On November 9th an article in the Business Standard mentioned that

there is nervousness among investors in the segment and the supply chain. Experts say leading project developers are putting aggressive bids for 50-200 Mw projects, to be relevant in the market

I am personally quite surprised at the record low rates that have been achieved. It is easy to appreciate that there has been very little (if any) progress with the financial institutions lending to this sector. Banks don’t bat an eye lid when asking for an arm, a leg and your dog’s kennel as collateral or putting deals through committee after committee while the project owner suffers. There was a brief ray of hope for investors when panel prices dropped below 50 cents per watt but a rapidly devaluing rupee nuked the opportunity created and the panel prices remain firm on a rupee level.

So when the reverse bidding prices for the sale solar output start touching record lows without much changing on the input sides – it intrigued me to further investigate how these projects made financial sense for investors.

Instead of assuming a certain project cost and then getting an IRR based on the bidding done by Sun Edison I decided to and was encouraged to (by a senior executive at a leading EPC) to  reverse calculate the project cost by using the tariff as the final product and taking CERC or market accepted levels of return for such projects to find out the truth for myself.

First I refurbished a worksheet that I found online on www.IndianPowerIndustry.com and utilised by the website for “Calculation of Solar Power Tariff”. The sheet is quite comprehensive but I found a few errors both clerical & logical that I corrected and figured that for a bid of Rs. 4.63 to make sense the project cost would have to be about Rs. 4.40 crore per MW (all inclusive).

The assumptions made were (all figures are in lakhs of rupees) :

Project Capacity MW 500 Return on Equity for 1-10 years Per Annum 18%
Annual Energy Production Lac Kwh 8322 return on Equity for 11-25 years Per Annum 22%
Total Project Cost Rs. Lac 440 Depreciation till loan repayment 5.83%
Project life Years 25.00 Depreciation after loan repayment 1.54%
Equity to be Invested Rs. Lac 66000 Total O&M expenses for 1st year 4000
Loan Component Rs. Lac  1,54,000 Escalation in O&M Expenses 5.72%
Loan repayment period Years             15 Discount Rate 9.88%

Please note: The discount rates input of “Return on Equity”  was 16% (post-tax) as per CERC guidelines. So the final calculation was (70% x 11% x (1-33.99%) + (30% x 16%) = 0.0988

The final calculations were (all figures are in lakhs of rupees):

Year 1 2 3                4                5                6                7                8                9             10             11             12             13             14             15             16             17             18             19             20             21             22             23             24             25
Energy Sold (degrading by 0.84% per year) 8322  8,252.10  8,182.78  8,114.04  8,045.88  7,978.30  7,911.28  7,844.83  7,778.93  7,713.59  7,648.79  7,584.54  7,520.83  7,457.66  7,395.01  7,332.90  7,271.30  7,210.22  7,149.65  7,089.60  7,030.04  6,970.99  6,912.44  6,854.37  6,796.79
O&M Expenses (includes OPEX) 4000.0 4228.8 4470.7 4726.4 4996.8 5282.6 5584.7 5904.2 6241.9 6598.9 6976.4 7375.5 7797.3 8243.3 8714.9 9213.3 9740.3 10297.5 10886.5 11509.2 12167.5 12863.5 13599.3 14377.2 15199.6
Principal Repayment 10266.7 10266.7 10266.7 10266.7 10266.7 10266.7 10266.7 10266.7 10266.7 10266.7 10266.7 10266.7 10266.7 10266.7 10266.7 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
Interest on Term Loan 16422.4 15293.1 14163.7 13034.4 11905.1 10775.7 9646.4 8517.1 7387.7 6258.4 5129.1 3999.7 2870.4 1741.1 611.7 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
Return on Equity (Pre-Tax) 11880.0 11880.0 11880.0 11880.0 11880.0 11880.0 11880.0 11880.0 11880.0 11880.0 14520.0 14520.0 14520.0 14520.0 14520.0 14520.0 14520.0 14520.0 14520.0 14520.0 14520.0 14520.0 14520.0 14520.0 14520.0
Total Outflow 42569.1 41668.5 40781.1 39907.5 39048.5 38205.0 37377.8 36567.9 35776.3 35004.0 36892.1 36161.8 35454.4 34771.1 34113.2 23733.3 24260.3 24817.5 25406.5 26029.2 26687.5 27383.5 28119.3 28897.2 29719.6
Cost per Unit of Electricity 5.1 5.0 5.0 4.9 4.9 4.8 4.7 4.7 4.6 4.5 4.8 4.8 4.7 4.7 4.6 3.2 3.3 3.4 3.6 3.7 3.8 3.9 4.1 4.2 4.4
Discount Rate for Net present Value 0.099
Discount Factor 1.0 0.9 0.8 0.8 0.7 0.6 0.6 0.5 0.5 0.4 0.4 0.4 0.3 0.3 0.3 0.2 0.2 0.2 0.2 0.2 0.2 0.1 0.1 0.1 0.1
NPV of Cost per Unit 4.65
Therefore Tariff (Rs./Kwh) 4.65

I also devised an alternate sheet of my own which wasn’t as comprehensive as the one created by  IndianPowerIndustry.com but was based on IRR versus NPV. In this approach I estimated the investment cost by charting out the net cash-flow at the project level and hitting a target IRR (14% post tax) at the project level.

Year 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25
Generation 8760 8686 8613 8541 8469 8398 8328 8258 8188 8120 8051 7984 7917 7850 7784 7719 7654 7590 7526 7463 7400 7338 7276 7215 7155
Tariff ₹ 4.63 ₹ 4.63 ₹ 4.63 ₹ 4.63 ₹ 4.63 ₹ 4.63 ₹ 4.63 ₹ 4.63 ₹ 4.63 ₹ 4.63 ₹ 4.63 ₹ 4.63 ₹ 4.63 ₹ 4.63 ₹ 4.63 ₹ 4.63 ₹ 4.63 ₹ 4.63 ₹ 4.63 ₹ 4.63 ₹ 4.63 ₹ 4.63 ₹ 4.63 ₹ 4.63 ₹ 4.63
Revenues ₹ 40,559 ₹ 40,218 ₹ 39,880 ₹ 39,545 ₹ 39,213 ₹ 38,884 ₹ 38,557 ₹ 38,233 ₹ 37,912 ₹ 37,594 ₹ 37,278 ₹ 36,965 ₹ 36,654 ₹ 36,346 ₹ 36,041 ₹ 35,738 ₹ 35,438 ₹ 35,140 ₹ 34,845 ₹ 34,552 ₹ 34,262 ₹ 33,974 ₹ 33,689 ₹ 33,406 ₹ 33,125
O&M -₹ 3,000 -₹ 3,172 -₹ 3,353 -₹ 3,545 -₹ 3,748 -₹ 3,962 -₹ 4,189 -₹ 4,428 -₹ 4,681 -₹ 4,949 -₹ 5,232 -₹ 5,532 -₹ 5,848 -₹ 6,183 -₹ 6,536 -₹ 6,910 -₹ 7,305 -₹ 7,723 -₹ 8,165 -₹ 8,632 -₹ 9,126 -₹ 9,648 -₹ 10,199 -₹ 10,783 -₹ 11,400
OPEX -₹ 1,000 -₹ 1,057 -₹ 1,118 -₹ 1,182 -₹ 1,249 -₹ 1,321 -₹ 1,396 -₹ 1,476 -₹ 1,560 -₹ 1,650 -₹ 1,744 -₹ 1,844 -₹ 1,949 -₹ 2,061 -₹ 2,179 -₹ 2,303 -₹ 2,435 -₹ 2,574 -₹ 2,722 -₹ 2,877 -₹ 3,042 -₹ 3,216 -₹ 3,400 -₹ 3,594 -₹ 3,800
Cash Before Tax ₹ 36,559 ₹ 35,989 ₹ 35,410 ₹ 34,819 ₹ 34,216 ₹ 33,601 ₹ 32,972 ₹ 32,329 ₹ 31,670 ₹ 30,995 ₹ 30,301 ₹ 29,589 ₹ 28,857 ₹ 28,103 ₹ 27,326 ₹ 26,525 ₹ 25,698 ₹ 24,843 ₹ 23,959 ₹ 23,043 ₹ 22,095 ₹ 21,111 ₹ 20,090 ₹ 19,029 ₹ 17,926
Effective Tax Rate 10% 10% 10% 10% 10% 10% 10% 10% 10% 10% 10% 10% 10% 10% 10% 10% 10% 10% 10% 10% 10% 10% 10% 10% 10%
Tax ₹ 3,656 ₹ 3,599 ₹ 3,541 ₹ 3,482 ₹ 3,422 ₹ 3,360 ₹ 3,297 ₹ 3,233 ₹ 3,167 ₹ 3,099 ₹ 3,030 ₹ 2,959 ₹ 2,886 ₹ 2,810 ₹ 2,733 ₹ 2,652 ₹ 2,570 ₹ 2,484 ₹ 2,396 ₹ 2,304 ₹ 2,209 ₹ 2,111 ₹ 2,009 ₹ 1,903 ₹ 1,793
Net Cash Flow(s) -₹ 2,00,000 ₹ 32,903 ₹ 32,390 ₹ 31,869 ₹ 31,337 ₹ 30,795 ₹ 30,241 ₹ 29,675 ₹ 29,096 ₹ 28,503 ₹ 27,895 ₹ 27,271 ₹ 26,630 ₹ 25,971 ₹ 25,293 ₹ 24,593 ₹ 23,872 ₹ 23,128 ₹ 22,359 ₹ 21,563 ₹ 20,739 ₹ 19,885 ₹ 19,000 ₹ 18,081 ₹ 17,126 ₹ 16,133
IRR 14.14%

In both models the total project cost I arrived at is between Rs. 4 – 4.40 crore per MW. Now comes the really important part… is that cost realistic or attainable?

So the few things that are constant in such projects are:

Interest free Performance Guarantee Deposit Rs. 10 lakh per MW
One-Time Solar Park Development Expenses Rs. 42 lakh per MW
Service Tax & Other Taxes which I will estimate at 5% of project cost Rs. 20 lakh per MW
Panel investment* Rs. 297 lakh per MW
Inverter Rs. 20 lakh per MW
Total = 389 lakhs or 3.89 crore per MW

*Panel costs (this is the tricky piece) for a project of this size should be quite low.. Below market quoted prices for sure. I estimated them to be $0.42-0.44 per watt. However with 30% of the capacity under the DCR category it is my estimation that the average per watt procurement cost will be $0.45 per watt. You can argue that the panel prices may/will fall 10% in 12 months i.e. when they are needed. I would say that the rupee will most like devalue 10% by that time as well negating the gains made in panel procurement cost. Lastly one should also note that a significant percentage of these modules are to be sourced from Indian Panel manufacturers who charge  a premium to international rates.

Even if I take the average of the “derived project cost” I get 4.20 crore per MW as my project cost.. Deducting the break-up above I get a balance of 31 lakhs per MW to cover the remainder of my expenses which are:

  1. Balance of Plant Investment
  2. Transmission Investment
  3. Labour Costs
  4. EPC Margin (this goes to Sun Edison if they do their own EPC)
  5. Project Monitoring Costs
  6. Legal Costs
  7. Equipment transportation
  8. Financing costs like:
    1. Processing fees
    2. DSRA
    3. IDC
    4. Bank Guarantees
    5. EMD Costs
  9. Stamp Duties & other taxes

Do you think that 31 lakhs can cover all this?

Advertisements

7 thoughts on “Decoding Sun Edison’s Record Bid for Selling Solar Power in India – Realistic or Wildly Optimistic? (Part 1)

  1. Pingback: essayforme
  2. i ditto this article. Speaking on ground reality of handling 40 MW plant at separate location in Gujarat and managing part of other 45 MW, my experience says its dead deal for SunEd right now. The only thinkg they may be banking on may be is as overall electricity price will be increased in INDIA as spurted by union energy minister in few recent functions, if this happens, This will give edge to SunEd to go regulatory commission again to hike their tariff rate Rs. 4.40 to new higher FIT in future citing non-viability to run operation. This will get their tariff hike as Its 100% possible. You can take case of Adani power. and Voila…

    in all other case, if someone can ask me I can vouch leaving couple, All Gujarat based solar project running last 3 year have lost per MW generation significantly compare to day 1 generation figure that most analyst project sitting in their offices. In some project it is more than 30% due to panel insurance, inverter failing and financial issue for managing O & M for project that is rapidly declining generation.

  3. It might not be a very lucrative IRR in the long term for Indian investors but in countries like USA and several European countries where the interest rates are too low the offer seems quite lucrative as the investors by investing in solar have a safe investment with fixed returns in the long run. Also the land prices will appreciate quite a lot in the span of 25 years which is an added value. So it seems an overall good deal for foreign investors but not so much for the Indian investors. You can see this trend as the recent low price per unit have be done by foreign companies like Sunpower(Malaysia), SunEdison etc.

    1. Hi Shubham,

      1. Sun Edison’s WACC in the US is just over 5% even if we add a hedging cost of 7% for INR we get a 12% WACC for India (I have considered 10%). Sun Edison’s share prices have dropped over 65% in the last month as the yields provided have been lower than anticipated and the FED increasing interest rates increases the yield for Sun Edison too. Therefore (and I will address this in the next post) it would be better for Sun Edison to do a project in the US versus coming to India where the risks are higher and the returns not the same – my thoughts on it anyways.

      2. The land is on lease from the government in this project so there will not be upside from the sale of land – expect scrap value for Plant & Machinery.

  4. Nice Article Anirudh.

    Simply playing the Devil’s Advocate here:

    1. Single Axis trackers may yield higher generation for a marginal increase in cost. SunEdison owns NexTracker, hence pocketing a discount on prices and minimal technology risk. (not sure though if trackers are allowed)

    2. SunEdison’s cost of Debt should be cheaper than we have assumed. Large Indian Developers have raised 15yr debt for less than 11% post hedging.

    3. Plus SunEd actively trades on the market on USD-INR, so they may be taking an aggressive bet on generating some profit through trading instead of hedging upfront.

    4. from EPC, E&P costs are literally zero for this size of a project, gives SunEd’s experience. ‘C’ will literally be labour cost as SunEd would squeeze every cent from EPC players, who would one-way-or-another agree (multiple possible reasons).

    1. 1. Single Axis trackers require more area and increase costs not only on the CAPEX front but also on the O&M fronts. Their use for a project of this size would be a first in India and whether those costs can be covered in what is already a very thin margin product is highly unlikely.

      2. This question is going to be the crux of part 2 of this post. However a a preview:
      1. SunEdison has been raising Indian debt. 11% is basically the lowest cost Indian debt I have seen so far as the PLR is between 9.20% to 10.20% and assuming even an average of 9.70% + 100 basis points I am close to 11%. Borrowing in the international markets with the threat of FED increasing rates & the volatile movement in the rupee will threaten already thin margins.

      3. I shall check whether they have a large enough export to India to have a natural hedge but considering point 2.. I believe it would be best for them to keep their borrowing in INR

      4. Even if the work is done in house there are going to be significant labour costs that SunED EPC will incur – the cement wont pour itself nor will the land level itself or for that matter the engineers needed for a project of this scale are going to cost significant money.

      My questions is whether the 31 lakh delta that remains is enough to cover that. Even if they outsource to SW/LT/Juwi they will ask for BGs etc from them which have a cost that will be recovered by them through money charged to the project. Lets not forget IDC which will eat into that delta.

      Without adequate safety margin how will this project achieve returns that add value to SUN Edison share-holders?

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.