Finding Economies of Scale in Solar

AllianceBernstein L.P.

By Catherine Wood and Brett Winton

Advocates of solar energy have argued for years that the industry only needs subsidies to gain the economies of scale that would make it cost competitive. We think that day may never arrive.

Solar power is expensive to produce because the equipment is expensive and only works a fraction of the time—30% of the time in the optimal spots for solar energy, such as the Las Vegas desert. Nuclear or coal power plants, by contrast, can run almost round the clock, day in and day out.

We estimate that the cost of an installed solar power panel would have to fall from about $4.40 per watt today to $1.40 per watt to become cost competitive with a newly built natural gas–fired plant, assuming the gas plant had to pay $50 per metric ton for carbon emissions. It would have to fall to about $1.10 per watt to accommodate the cost of energy storage. But to get to that price, massive economies of scale would be required.

In the power business, there’s a rule of thumb: each time you double the cumulative installed base of power-generating infrastructure, the cost of a new installation should fall by 20%. Since 1976, the cost of solar power has declined somewhat less, by 18% on average, for every doubling of installed capacity. Over the past ten years, costs have fallen 13% for every doubling.

But let’s say the cost declines return to the 18% rate. How much solar power would have to be installed before incremental solar panels would be truly cost competitive? 9.6 million megawatts, or more than 100 times the capacity in place today, as the display below shows. With an area of 20,000 square miles, these panels together would be large enough to cover Vermont and New Hampshire, and produce annual quantities of electricity sufficient to exceed demand for the entire world through 2025.

Reaching Grid Parity Would Require Massive Capacity for Solar

So when is solar going to become cost competitive without subsidies? In three to five years? Try never. But if it did, society would have to pay out trillions of dollars to get there.

That doesn’t mean solar energy is a complete dead end. Solar power may indeed be the cheapest way to provide base electric power in rural India, or in parts of Africa devoid of infrastructure. Even in select markets with developed infrastructure, a small amount of solar power may prove valuable to provide electricity during the hottest parts of the day.

But should governments provide massive subsidies to support solar energy in places where electric power can be generated at a much lower cost?

I think not. Much of the money is likely to be wasted. Germany, for example, has subsidized the solar industry to the tune of $50 billion, yet it only gets 6% of its electricity from solar power. The marginal tax dollar would find a better home in the research labs of universities, where fundamental technological breakthroughs are more likely to yield a big increase in efficiency and corresponding decline in price.

The views expressed herein do not constitute research, investment advice or trade recommendations and do not necessarily represent the views of all AllianceBernstein portfolio-management teams.

Catherine Wood is Chief Investment Officer—Thematic Portfolios, and Brett Winton is a research analyst, both at AllianceBernstein.


  1. The cost of a “Solar Power Panel” as this article refers to, is well under $.80/W today and falling. I assume the author is referring to the price of an installed Solar Power Plant which currently runs around $2/W from major EPC”s. This figure has fallen over 50% in the last 3 years alone.

    I wish I could say this was the only inaccuracy in this article but suffice it to say that solar is competitive today in a larger variety of markets than this author realizes. A bit more research would”ve served this blog well.

    There”s a reason why the solar industry is the fastest growing industry today while providing more US based employment/watt than any fossil fuel based technology.

    Oh, it allows us to pass a living planet to our grandchildren.

    • Geoffrey Bramhall

      If solar power is doing so well, why does it need all the subsidies? Why are the European contries promoting it in such dire economic shape and now backing away from it. And lastly, why are so many of the solar companies going bankrupt even with government backing?

      • Why the subsidies ? Because it”s the only way to be able to get the investors doing their job against the lobbying of the fossil fuel industry.
        Why cutting the subsidies ? Because it looks costly to the consumerand the politicians want to get reelected … but nobody is telling the consulers that electricity price produced otherwise than by renewable sources is going to get much more expensive in the future.

        Plus : all energy souces have been subsidized, ion one way or another … or do you know a country that financed its nuclear power plant developement by a Feed-in-Tariff ? no, that was with tax money.
        And for shale gas ? that will be subsidized in a certain point in time by the cost to “repair” the nature and the health of the nearby population.

        So , where is the difference ?

      • It also gives you some freedom that the grid doesn”t offer. If there is a power outgae you can still have some power. I a 50 watt panel and a small battery and charge controller. It isn”t much, but as long as there is sun it will work when the grid doesn”t. After the initial investment in hardware, I don”t have to spend any money on it. Gas powered generators are noisy and require spending money on fuel. Sun light is free. So is wind, but thats for a future project.

  2. 1) Peter – Totally agree with you. My company setup a utility scale solar PV project in India, sometime back. The modules were procured from America`s most famous thin film module manufacturer. The project cost came to about $ 2.8/watt with nearly half going to the module cost.Financing in India is expensive compared to the US but would, I guess compensate for the higher cost of labour in the US. Attached is a link to a DoE presentation(year 2010) which shows the cost of a utility PV system at $3.4/watt of which $1.7/watt is the module cost(This would mostly be a mono-crystalline module, the most expensive module type with the highest efficiency)

    2) Modules are currently selling at 70 cents/watt(Chinese).Adding the 30% tariff on chinese ones still brings the cost to 90 cents/watt. So that, as per our previous scenario has already brought down the project cost to $2.6/watt. Looks like this blog post is off the mark by $1.8/watt.

    3) Natural gas prices are at a historical low today and everyone in the industry agree that natural gas prices are bound to double from here to $ 4/mmbtu. Imagine a scenario where fracking leads to a major accident and the US government imposes some more oversight on the industry. Taking this into consideration, I am sure that the above blogger being an investment advisor would agree that setting up a gas plant is a far bigger risk compared to solar, where the raw material is free and the cost over a 25 year period are well known upfront.

    4) It surprises me that the blogger wants to divert government subsidies which help commercialize solar to colleges where breakthrough research happens. The blogger seems to be under the assumption that scientific research is the magic wand for achieving lower costs. In 2007, polysilicon, an important raw material used in making solar panels was going at $500/kg. 3 years later, the price had fallen to $30/kg(in conjunction with Solyndra and Evergreen bankruptcies). Evergreen`s annual report shows $50 mi in R&D expense in the 3 years from 2005-07.

    Although R&D is important, the solar industry in the last couple of years has been transformed due to government subsidies. Poly silicon costs have not come down due to R&D, rather it has been due to commercialization and economies of scale. This would not have been possible without German and Italian subsidies.

    The blogger probably needs to take a serious re-look at this blog post and more importantly, about investment recommendations for the solar industry.

  3. The world´s largest market for solar PV is still Germany (though China may well overtake it in 2013). So German system prices are the benchmark, not American. The <10kw price index is here. Latest is 1.75 euros/watt, or $2.24/watt. US residential prices are an anomaly. Australisn ones have now fallen to German levels, so there´s every reason to think that the inefficiencies will be ironed out quite rapidly in US residential, as they already have been in utility.

    What the German subsidy – actually cross-subsidy scheme, the cost is borne by consumers not taxpayers – has bought them isn´t 6% of their electricity supply, it´s grid parity. Feed-in tariffs are now below retail, so the cross-subsidy looking forward, as they expand to and no doubt overshoot the government target of 52 GW, is zero or negative.

    The German retail electricity price is admittedly high, twice typical US rates, including a premium of higher reliability and aesthetics (urban buried cables). On the other hand, Germany is as northerly as Canada and US insolation is on average a third higher, two-thirds higher in the Southwest. On reasonably optimistic assumptions about convergence of BOS costs to German and Australian levels, grid parity will arrive in the SW in the next two years.

Leave a Reply

Your email address will not be published. Required fields are marked *


You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>