CleanTechnica is the #1 cleantech news & analysis site in the world. Subscribe today!The future is now.

Clean Power

Published on October 3rd, 2014 | by Giles Parkinson


Solar Power Costs Headed Toward 4c/kWh

October 3rd, 2014 by  


The highly conservative International Energy Agency predicts the cost of solar energy will fall to around 4c/kWh in coming decades as the sun becomes the dominant source of power generation across the world.

As we reported on Monday, the IEA now expects solar to become the biggest single source of energy by 2050 and has now doubled its forecast capacity for solar PV.

Rooftop solar, it says, will now account for one half of the world’s solar PV installations, because as a distributed energy source the technology is “unbeatable”.

On costs, it says all solar technologies will fall dramatically in coming decades, with solar PV falling to as low as 4c/kWh, utility-scale solar to around the same level, and solar thermal with storage will fall to as low as 6.4c/kWh.

As this graph below shows, the minimum price tends to occur in regions with great sunshine, and it also assumes a low capital cost of around 8%.

iea solar lcoe

Indeed, IEA executive director Maria van der Hoeven said capital costs were a key element in bringing down the cost of solar technologies. To do this, she said, it was critical to have stable and long-term policies.

This graph below illustrates how financing costs have a critical bearing on the overall cost of solar – any technology for that matter, but particularly one where the bulk of the cost is in the upfront capital outlay for installation.

iea cost capital

Van der Hoeven said policy makers needed to make clear, credible and consistent signals, which can lower deployment risks to investors and inspire confidence.

“Where there is a record of policy incoherence, confusing signals or stop-and-go policy cycles, investors end up paying more for their investment, consumers pays more for their energy, and some projects that are needed simply will not go ahead.” The implications for Australia, where large-scale investment has ground to a halt because of policy uncertainty, could not be any clearer.

The forecasts from the IEA are not the most dramatic that can be found, but they are significant because the IEA is essentially a conservative organisation that was created in the 1970s to defend developed countries’ access to fossil fuels

It has a history of underestimating the impact of new technologies such as solar, as we pointed out in this article – even though it has doubled its forecast for solar PV deployment in just the last few years. Other agencies, such as IRENA, have a much more bullish forecast for solar.

The IEA insists that its figures are not forecasts, but what the world should be aiming for. The deployment of solar could be much higher – assuming costs come down faster than thought. The IEA’s base model still relies heavily on “baseload” generation.

If the IEA figures are right, solar PV’s share of global electricity will reach 16 per cent by 2050, a significant increase from the 11 per cent goal in the 2010 roadmap.

This will require 4,600 GW of installed PV capacity by 2050 – more than half of it in China and India. This requires the installation rate to nearly quadruple to 124GW year. It notes solar PV will be a highly effective abatement tool, avoiding up to 4 gigatonnes (Gt) of carbon dioxide (CO2) annually on its figures.

The IEA says variability is an issue, but it can be overcome with interconnections, demand-side response, flexible generation, and storage. The overall cost will require an increased investment of $44 trillion – but this will generate savings of $115 trillion in fuel costs.

The IEA says the dramatic cost falls in solar PV means that the sector is around five years ahead of where it thought it would be.

Solar thermal with storage, on the other hand, was lagging because its “dispatchability” was not being fully valued.

That would change, however, because solar thermal with storage would play a critical role in the energy systems of the future and would be a match for solar PV because of its ability to store energy for use at night or times of peak demand. In some countries, solar thermal deployment is expected to be greater than solar PV. In others, the opposite will be true.

iea regions

Paulo Frankl, the IEA’s solar expert, says half the large PV deployment considered in this roadmap would take place on buildings or nearby (such as over parking lots). More than half of this would be on commercial buildings rather than residential.

The forecasts rest, in part, on the concept of grid parity – when the cost of distributed PV generation is equal or below the per-kWh component of retail electricity prices – and on self-consumption. “At the utility level, solar PV has many competitors. At the distributed level, solar PV has a competitive advantage and is unbeatable.”

Source: RenewEconomy. Reprinted with permission.

Complete our 2017 CleanTechnica Reader Survey — have your opinions, preferences, and deepest wishes heard.

Check out our 93-page EV report, based on over 2,000 surveys collected from EV drivers in 49 of 50 US states, 26 European countries, and 9 Canadian provinces.

About the Author

is the founding editor of, an Australian-based website that provides news and analysis on cleantech, carbon, and climate issues. Giles is based in Sydney and is watching the (slow, but quickening) transformation of Australia's energy grid with great interest.

  • Mark

    “Photovoltaic power is extremely reliable”
    Have to disagree here. Photovoltaic production can quite immediately change as a function of the clouds coming out. Photovoltaic production must be backed up by spinning reserves or very high speed interruptible power sources. Additionally, because there is no mechanical inertia associated with a spinning generator or other means of energy storage, photovoltaics are actually de-stabilizing for the grid in periods of transient instability. Having said that, there’s a lot of work that needs to be done. I believe the potential of high speed interruptible users of power has not been appropriately tapped (ie: data centres) as stability issues become increasingly significant.

    • Bob_Wallace

      Clouds don’t cover an entire region at once. Solar is going to be widely distributed, not localized as happens with a large thermal plant.

      Because sunsets and cloud movement are highly predictable there is no need for spinning reserve. Spinning reserve is needed for large thermal plants which can go offline with no warning.

  • JamesWimberley

    Where are the IEA getting their baseline from? They quote the minimum LCOE of solar now (2013-2015) as $119/mwh, average $177. As readers here well know, substantial capacity is being installed now in the US Southwest near $50/mwh, <$80 before ITC. IIRC utility prices in India and China are similar or lower. The IEA's minimum will only catch up with reality in a decade.

    To recap: the solar pv industry industry is highly competitive at all stages, so there is strong pressure to converge on best practice rapidly. The products are virtually commodities, with minor technical differences at the innovation edge. At any one time, the fastest-growing countries will be the ones with the lowest prices, and they predict the future better than the laggards. Are these strange indices weighted by volume?

    Where do they get such data from? Remember that in an environment of very rapid change, "respectable" official sources like the EIA are always going to be out of date. Peer-reviewed academic publications will be even worse. We are better off here, as we can listen to what the EPIA, Jigar Shah, Mark Bolinger and Ryan Wiser are saying. "Anecdata" possibly, but more accurate than last year's EIA tome.

    • Matt

      James while I think you are correct. We have to at least give them credit for opening their eyes. For the last 13 years, every year they said PV was going to stop growing and totally ignored roof top solar. Now even while they assume cost higher than they already are, they are saying PV will take over the market. Conservative organization tend to be slow to learn or change direction. This a a major sea change for them. I guessing in a few year the prediction will move from 2050 to 2040. So while way to conservative and deploy rate way too low; this report pulse the legs out from under those who have been using EIA to “prove” solar was not ready for prime time and should stay in the lab.

      • JamesWimberley

        IEA is not the same as EIA. They used to sing from the same hymn-sheet, but no longer. I suggested the IEA relies on EIA data.

    • Bob_Wallace

      GMT reports utility solar installed for as low as $1.60/watt. That would mean ~$70/MWh in SoCal. The average installed price of $1.81/watt works out to ~$80/MWh.

      China is installing for around $1/watt. About $50/MWh.

      Those are non-subsidized prices. The IEA is not presenting real world numbers.

      • Burnerjack

        Ironically, as I hesitate and ponder my own possible installation, the costs continue to drop. Yes, I do pay more per month as I wait, but it seems that wait could be quite beneficial in the long run. I guess my question is not whether to pull the trigger, but when. At some point the curve should begin to flatten as the development and surpluses reach towards an effective apex. I suppose that would be the time.

        • Bob_Wallace

          I suppose if your goal is to maximize your savings then you could guesstimate the annual decrease in system cost and jump when annual decrease is roughly the same as your annual electricity bill.

          Take a look at this graph. The average cost of residential solar fell 22% in the last year. That might give you some basis for estimating the rate of drop.

          But don’t forget that there will be some disruption when the federal subsidy goes away.

          • Burnerjack

            Good advice. But the world is not ruled by geometry, but by calculus. As the slope progresses towards the ideal it will decrease and approach flat (adjusted for inflation). While the subsidy issue is true, it smacks of “supplies are limited. Call NOW!!…)

          • Bob_Wallace

            Yes, the rate of decrease will likely slow as prices reach the minimum possible. But we’re probably not close to that point at current $3.50+ average prices.

            Germany is now installing at about $1.60/watt and their price decrease slope look fairly stable. You can grab the numbers here and do a line of best fit if you wish –

            Supplies are likely to be limited as we approach the end of the federal subsidy. Lots of people are likely to jump in and tie up installations. Not trying to “sell you”, just thinking about how prices are likely to play out over the next couple of years.

            At the end of 2016 the 30% fed subsidy drops to 10%. It’s hard to predict how that will impact prices but if you’re in a market with good competition then it might take a few years to make up for loss of the 20% subsidy. If you’re in a higher priced market then prices might keep falling due to increased competition.

          • Jim

            So when the annual rate of depreciation equals your annual cost saving, buy the system. 2016 seems like a nature time to buy given the 20% discount you will get from the federal government as opposed to 2017

Back to Top ↑