What\’s desirable is often not practical. The National Solar Mission has set a target of 20,000 MW of solar electricity by 2020. This may be desirable, but at today\’s solar technology costs, it will be economic suicide.
Remember that just 700 MW of high-priced power from Enron in 2001 was enough to bankrupt the Maharashtra government, which therefore refused to pay. At the time, Enron\’s power cost Rs 4 per unit if run at 90% capacity, and Rs 7 per unit if run at less than half capacity, as was often the case. Solar power today costs Rs 9 to Rs 10 per unit in roof-top photovoltaic panels and other applications.
Hopefully, technological breakthroughs in the next decade will send costs crashing and make solar power economical. Rich countries are spending billions of dollars on solar R&D. The National Solar Mission Plan hopes to reduce the cost to Rs 4 to Rs 5 per unit by 2017-20 in order to make solar power competitive with coal-based power. But this represents a hope unsupported by any track record and grossly overestimates the cost of rival coal-based power.
Globally, no technical breakthrough may come. After the second oil shock in 1980, many hoped that amorphous silicon and other photovoltaic technologies would make solar power economical. Alas, photovoltaics remain hopelessly uneconomical today.
Meanwhile a breakthrough has come in concentrated solar thermal (CST) technology, using parabolic mirrors. Pilot CST projects in the US and Spain have raised hopes of solar power at Rs 5 per unit. But this is still far higher than the tariff per unit for India\’s ultra-mega power projects at Sasan (Rs 1.19), Tilaiya (Rs 1.77), Mundra (Rs 2.26) and Krishnapatnam (Rs 2.33). The first two are based on Indian coal, and the other two on imported coal. Even allowing for rising coal import prices and a heavy carbon tax, coal-based power looks much cheaper.
Now, some private power plants are selling limited amounts of electricity at Rs 6 to Rs 7 per unit to industries desperate for power. This encourages the government to think it can bundle expensive solar power with cheap coal-based power, and remain viable. Maybe, but remember, this was exactly the argument used for Enron — that expensive Enron power could be bundled with cheap power from MSEB stations — and it proved a financial fiasco.
Besides, solar power needs a lot of land. This can be neglected in a small pilot project, but not in large, commercial projects. The biggest CST projects in the US use 6 to 10 acres per MW of power. By this yardstick, even a pilot project of 100 MW requires 600 to 1,000 acres of land. A commercial project of 1,000 MW needs 6,000 to 10,000 acres. After the troubles of Tata Motors at Singur and Posco in Orissa, we must be cautious about land-intensive projects.
State governments rightly want companies to buy land at commercial rates, not ask for acquisition. The market rate in many states is now Rs10 lakh to Rs 20 lakh per acre. So, 1,000 acres for a small solar project could cost a whopping Rs 100 crore to Rs 200 crore, making it totally uneconomical.
Cheaper land is available in the deserts of Kutch and Rajasthan. But even in Kutch, industrialists have paid Rs 2 lakh to Rs 10 lakh per acre. Besides, CST projects need huge amounts of water for cooling towers, and the Rajasthan desert is inappropriate for that. However, CST plants in Kutch could use sea water.
So, while solar power is desirable, we should proceed cautiously. Global spending on solar R&D runs into billions, and any breakthroughs will come from abroad. We should not waste money duplicating global R&D. Rather, we should limit ourselves to pilot projects, testing the best global technologies in Indian conditions.
If global technical breakthroughs arrive, then we can scale up in a big way. That will not require fancy National Solar Missions. Private entrepreneurs will flock to build solar plants once they are proved viable. But if no breakthrough comes, we must not waste money on an arbitrary target of 20,000 MW of solar power by 2020. We must learn from the fiascos of Enron and Singur.
A small voice in my head tells me i may be on the wrong track. \”Swami, the government knows all this. But it needs to do something in global climate negotiations. The US will not come on board unless China and India are seen contributing, and without US participation the climate talks will fail. So, we have made fancy long-term projections — 20,000 MW by 2020; 100,000 MW by 2030 — getting good publicity. But our near-term target of 1,000 MW by 2013 implies no more than some pilot projects. This will keep climate negotiations going at little cost.\”
Is that the government\’s hidden agenda? If so, I need to think again.
13 thoughts on “Expensive solar power can mean another Enron”
brilliant article. puts things in perspective.
First of all a huge fan. I have been reading your articles for years and no one compares in terms of the sagacity and simple logic derived inferences.
With respect to solar power, the prices have been coming down quite rapidly with economies of scale being reached. A company called First Solar(PV thin film) can produce electricity at about 7-8 Rs/ Unit. Going forward this is only going to come down further.
Comparing solar power with coal at this juncture may not be right. Rather comparing it with diesel operated gensets(50,000 + MW in use) at 10-11 Rs./unit may be a fairer comparison.
The problem with the ENRON deal were many.
1) The price was fixed and the terms were unreasonable.
2) The fiscal health of the state was bad as it is(the State electricity was in negative even before this deal was made operational.
3) It did nothing to promise a better tomorrow.
Plus the fact that India of 2000 was very different from the India of 2010.
As things stand today, we can afford some of the more expensive solar power today and take a hit for while. It is likely that a few years down the line it may come down to more reasonable levels and taking a punt on clean energy may appear to be a visionary step.
A little like mobile telephony which witnessed such masssive correction in prices.
I believe the government is taking a genuine punt on cleantech, hoping that climate change negotiations will eventually end up picking up a part of tab.
Please note followings:
1. in India Solar PV projects require 4.5 Acre land per MW.
2. Capital Cost of Solar PV project is high, but operational cost is zero – as it requires not more than 10 people. Nothing else.
3. Solar PV will also give returns in terms of Carbon Credits (that is about 13 Euro/CER today, was not there in Enron).
4. As per PPA, fixed charges were constantly going up; whereas here the cost at which government is buying electricity is constant over 10 years.
5. The fuel was Neptha, and charges of the same were $ denominated, and government did not exercise any currency hedging options against fluctuations.
6. When this fiasco happened GoM had 3.5% of revenue deficit… in 2009-10 it’s 0.9%.
7. GoI has budget provision for JNNSM, what was not the case with GoM.
& many more such facts.
I need to communicate with you. For that I need your email id. My email id is firstname.lastname@example.org
If you could pls drop your email id.
Thanks and kind regards,
Cost and Management Accountants
When I read the headline and started reading this article, I kept wondering till I reached the last paragraphs, that how could Mr. Economist discourage solar fantasies. But the last paragraph holds the message, solar projects are the need of hour and as a fast growing economy we are key negotiator on climate change and we do need to take steps to set examples for rest of the world. At the same time we must take lessons from the past examples, Enron and Singur, as mentioned by writer. What we need to understand is that the sustainable development or the vision is not justified till it runs well and ends well.
As far as cooling of towers is concerned the government must encourage rainwater harvesting projects near the identified locations. As far as I am aware Jodhpur is identified for solar projects in Rajasthan and it receives good rainfall. Porbandar has highest solar radiation in the country and receives good rainfall and also have advantage of located on sea bank, sea water advantage applies for Kutch as well.
I agree with writer to proceed cautiously but we should not step back from fancy long-term projections — 20,000 MW by 2020; 100,000 MW by 2030
If writer has suggestion for alternate clean energy solution, then he must come up with one.
In the Silicon Valley, Pacific Gas & Electricity (PG&E) is encouraging its customers to have solar cells and wind turbines on their rooftops and provide the generated electricity into the grid. PG&E installs a bidirectional meter and the customer pays only if the ‘net’ is negative. I have a lot of friends who have installed solar panels on their rooftops and have credits running from PG&E, as their ‘net’ is positive. Why cant we replicate this model in India?. What are your thoughts on this?.
More information is available here – http://www.pge.com/myhome/saveenergymoney/solarenergy/nembilling/
Dear Mr. Aiyar,
I would like to thank you for critically evaluating the economic viability and practicality of the national solar mission, and highlighting the important issues to be considered. However, I believe that your apprehensions regarding the future of solar technologies (both PV and CST) and their applicability in India are not completely justified.
It is true that photovoltaic power is still uneconomical when compared with coal, but it is important to make note of the fact that the costs of silicon PV has decreased by a factor of 30 since 1980. The expectation of achieving grid parity in next 7-10 years is not based on hope for a major technological breakthrough, but on incremental manufacturing innovations that are expected to decrease the costs by a minimum of 10% per year through 2020. It is a prediction that is well supported by the trend observed for past 30 years.
The CST technology is even more promising for Indian conditions. As you have pointed out, parabolic mirrors have certainly offered hope for conventional water-based CST plants. Also, one must appreciate the potential offered by the Stirling dish technology, which does not require water and is regarded as one of the most promising developments in solar thermal space. Its non-dependence on water makes it particularly attractive for deserts in Rajasthan, where the solar irradiance is high and land costs are still affordable. Sterling dishes have obtained extremely high solar to grid conversion efficiencies (greater than 30%). Moreover, the modular and scalable design of the dishes makes this technology the one to look out for in the future.
India’s power demand will increase enormously over the next 10 years, and it is imperative for us to gradually reduce dependence on fossil based power generation. If India does not take proactive steps to match the world in promoting solar power, it risks lagging behind miserably in the future, which does not augur well for India’s long term economic and political aspirations.
Solar as base load makes no sense. Solar as off-grid power does. The best application is at remote sites of Ladakh and some deserts. Leave it at that! Oppose national(but not regional) solar policies. Let Solar fill in the niche locations, where the grids simply cannot reach economically.
what is the present cost to generate solar power per unit,and how much we have to invest to produce one mega watt solar power,what is the initial amout we have to kept,what is the total cost,how much loan we will get how much land is required,what are the climate conditions,what the problems we may face.please explain me
It seems rather strange that 700MW running at a 90% plf at 4 rupees/unit could bankrupt Maharshtra, considering that 700MW at 90% plf produces 55.18 lakh MWhr/year.
Considering a price of 4 rupees/unit, it works out to only 220 crores.
Even assuming underrecovery of 50%, a 100 crore loss is hardly going to bankrupt the Maharashtra discoms…
Secondly, considering that the government is planning 1000MW in this phase (maybe by 2013/14) at an average price of 12 rupees/unit and a plf of 18%, it works out to a total gross cost of 189 crores. They will, of course, pass this on to people through the RPO requirements all over India. Will hardly make an impact of 1/1000th of a rupee.
10 to 20 lakhs/acre?!? Do you think developers want to bring up solar plants in the middle of some village? Most waste land in Rajasthan, MP, TN etc works out to less than 1.5 lakhs/acre.
While most of your articles are well thought, you seem to be playing to the gallery here and seeing phantoms where they don’t exist.
Will it be possible to re-visit your article and do a re-assessment of the financial viability of Solar PV MW plants since this is four year old already? Nowadays we keep hearing about the fall in costs, courtesy the Chinese Solar PV manufacturers.Here is one such article.
It will help young entrepreneurs like me to assess the current situation and act accordingly.
Dear Mr. Swaminathan Anklesaria
I am writing in the context that Haryana Government has made it compulsory that all industrial units(and homes and commercial establishments) have to install 5% of their connected electrical load as solar power capacity. Did anyone consider what financial burden it places on industrial units?
The cheapest of solar power viz photo-voltaic modules costs Rs.80,000/- per KW of capacity. One KW of installed capacity gives at the uppermost 1400 units of electricity per year,which is at the non peak hours and therefore is worth say 1400×6=Rs.8400 per year at most. Take out the absolute min of Rs.400 per KW as annual expense on daily cleaning and yearly maintenenace of the solar unit. What the industrial units gets is an annual return of Rs. 8000 on an investment of Rs.80,000. That is if the unit has a shade-free space facing in the correct Southernly direction. In an average unit with tight land situation, the return may be only Rs. 6000 per annum on an investment of Rs. 80,000. for which an average indusrial unit borrows from the bank at 13%per annum. Consider the other handicaps which the government has already placed the industrial units under. I m talking of my actual case of the industrial unit based in Haryana-NCR-outside Gurgaon. The actual cost of power after adding all the extras for 3 month average is an astronomical Rs. 8 per unit as billed by state distribution utility DHBVN. Add to this the cost of generating power for 10% of the time using Diesel Geneating set, and the actual average cost works out to Rs. 9/ per unit. Plus the compulsory capital cost of the DG set without which you cannot run an industrial unit.
If the Government is serious about cleaner(or less dirty energy) a good step they can take is instead of forcing the money to be put on impractical(as of today) solar power, put the same amount of money in improving the defunct distribution system so that 24/7 power is available to the industry and they do not have to run the DG sets.
Incidentally DHBVN has written to the consumers that they would offer 4% discount to any consumer if the consumer deposits a years estimated bill in advance for which consumers like us will have to borrow the money at 13 % to advance to the obviously broke DHBVN at 4%. To sum up: Haryana Govt. has done all it can to make the state electricity utility broke through their policies of encouraging theft of electricity and now they are taking one step in the same direction by making it compulsory for industry to install 5% of their connected load as solar power. And please remember that the actual benefit of this 5% installed power will be only 5/5=1% of the time because the Sun shines only 1/5th of the time.
Can anybody tell the Government of Haryana that instead of encouraging MAKE IN HARYANA they seem to be bent on telling the industry GET OUT OF HARYANA if you can.
Sushil K. Jain
Article is a bit outdated now. Following is a link (dated – March 2014) which gives a recent estimate at Rs 6.5 per unit.