Clean Power

Published on April 15th, 2016 | by Cynthia Shahan


Futurist Ray Kurzweil Predicts Solar Industry Dominance In 12 Years –Trajectories Are Exponential

April 15th, 2016 by  

How easy is it to be confused about the global transformation to renewable energy? For some clarification about solar energy, in particular, it is time to listen to futurist Ray Kurzweil explaining exponential growth and why he predicts solar industry dominance in 12 years. Remember those first huge, now historical and way too expensive, computers? Time flies, and society now bursts with tiny, affordable phones sporting computer systems. Things change.


As reported by Paul Dvorak, founding editor of Medical Design & Outsourcing, futurist Ray Kurzweil stimulated understanding of the subject of growth in the energy sector — in particular the growth of solar energy — in this recent talk. “Kurzweil began with the comment that technical developments form predictable trajectories, and those trajectories are exponential.”

SolarPowerWorld reports more of the Kurzweil predictions/explanation: “Consider the progress of the computing industry, he said. He spoke on his cell phone, which he said is several billion times more powerful per dollar than the computer he used as an undergraduate at MIT.”

“I went to MIT because it was so advanced that it actually had a computer in the late 1960s,” Kurzweil said. “It took up the floor of a building. Still, this cell phone is thousands of times more powerful, and million times less expensive. That’s a several billion-fold increase in price performance. It’s also a tiny fraction of the size.”

Kurzweil turned everyone’s focus to solar, mentioning that he and Google founder Larry Page were asked by the National Academy of Engineering to study emerging energy technologies. The men selected solar due to its exponential growth. Kurzweil said solar has been around for over 25 years, and its market share has doubled every 2 years.

“In 2012, solar panels were producing 0.5% of the world’s energy supply. Some people dismissed it, saying, ‘It’s a nice thing to do, but at a half percent, it’s a fringe player. That’s not going to solve the problem,’” Kurzweil said. “They were ignoring the exponential growth just as they ignored the exponential growth of the Internet and genome project. Half a percent is only eight doublings away from 100%.

“Now it is four years later, [and solar] has doubled twice again. Now solar panels produce 2% of the world’s energy, right on schedule. People dismiss it, ‘2%. Nice, but a fringe player.’ That ignores the exponential growth, which means it is only six doublings or [12] years from 100%.”

And of the sun … two years ago, Kurzweil responded to a question from the Prime Minister of Israel that went like this: “Ray, do we have enough sunlight to do this with a doubling seven more times?” Kurzweil” “Yes. After we double seven more times and meeting 100% of the world’s energy needs, we’ll still be using only one part in 10,000 of the sunlight that we have.”

That comment brings to mind this chart, which is a visualization of annual renewable energy potential from different renewable sources and total energy potential from total known fossil fuel reserves (from a handful of years ago):


Source: Perez & Perez, 2009a

“’It’s not true we’re running out of energy,’ Kurzweil said before moving on to another topic. ‘We’re only running out of resources if we stick with 19th-century technologies.’ ”

Zach talked about very similar matters in his presentation at an institutional investment conference in India in February:

Read Related Stories:

Embracing ‘Exponential Growth Of Solutions’ For 100% Renewable Energy

Cleantech Freakin’ Revolution Tour — 1st Event Wrapup

Pre-fab Solar Plan Could Slash Cost Of Solar Farms

Solar Electricity’s Future: Grid-Connected vs. Off-Grid

Top Image Credit: eschipul via / CC BY-SA

Check out our new 93-page EV report.

Join us for an upcoming Cleantech Revolution Tour conference!

Tags: , ,

About the Author

is a Mother, an Organic Farmer, Licensed Acupuncturist, Anthropology Studies, and mother of four unconditionally loving spirits, teachers, and environmentally conscious beings who have lit the way for me for decades.

  • GdB

    Growth in EV sales can easily absorb excess solar power. Perhaps new CSP should be optimized for mostly storage to future proof them.

  • neroden

    Given the following:
    — the also-exponential growth of wind power (slower rate)
    — the existence of hydropower
    — the “nighttime problem” for solar
    We won’t get to 100% in 12 years.

    We WILL get to 50% solar in 10 years (2035, since Ray is looking at *last year’s* data).

    The other 50% will be “nighttime power” and we’ll be focusing on the slighlty harder bottlenecks of batteries and transmission lines and so on and so forth for a few years. Meanwhile, daytime power will be crazy cheap.

  • Eric King

    So-for a glimpse of the future-Elon Musk has proposed a hyperloop train that would run partly on solar and with a partial vaccuum.

    Despite his claims this is apt to be expensive-so it would make sense to combine projects-we need superconducting power lines to move power around effiecently from sunny places to places with less sun-from Sahara to northern Europe for instance.

    Superconducting lines need to be cooled and energy losses from this would be minimized with a partial vaccuum-and you would only need one right of way.

    Then the true killer app for solar is a worldwide grid-make it a 24-7 energy source and have zero intermittency. But for this we need superconducting lines otherwise most of the power would vanish-so you could have a hyperloop train across the Bering straits say and to Europe and the Sahara and the Gobi-the superonducting lines would not need to be combined with the train the whole way-it would simply be a cost saving measure.

    A worldwide transportation and energy grid-the hyperloop could have solar panels on it as Elon suggested but could also use a bit of the power from the superconducting lines-you would need seperation to keep inevitable accidents from crippling the system-but subways already do this to a limited extent as they are powered by electricity-

    Eliminate the co2 from much of transportation and make solar always available and do both cheaper than either alone-

  • blasterman

    It’s great in theory but the trouble with Ray’s idea is that the world uses a lot of energy. He’s missing a couple of key points. First, we don’t have the industrial capability to replace literally all power generation in about a decade. Even if we could, we wouldn’t want to. We need redundancies in the system. Second, we are bound to have a recession (or two) during his predicted time period. During a recession less energy is used and less capital investment will be made. I have no doubt that we could achieve some remarkable levels of solar generation but 100% seems impossible.

    • neroden

      Recessions will accelerate the growth of solar. While they cause a reduction in capital investment, the more important effect of a recession is that it *lowers the return on capital you get by investing in ordinary businesses*, which makes the return on capital from switching from expensive fossil energy to solar a better deal.

    • Jens Stubbe

      Sure we have the industrial capacity.

      To produce as much electricity as humanity did in 2014 we need 660.000 Vestas 164 wind turbines at 50% capacity factor.

      That is doable without breaking a sweat.

      The solar guys are also not troubled with increasing capacity so doubling, tripling or more in a decade is simply put not a problem because of industrial capacity or restraints on supplies.

      I see no recession now or on the horizon.

      • blasterman

        Look, here’s the problem with capacity. If you count on doubling, that means every couple of years you have to produce an amount of panels equal to all the panels ever made. The final doubling that gets to 100% solar would be such an enormous amount of manufacturing that we do NOT have the capacity in factories to do it. Even if we did, it would be massive overbuilding because once you hit 100% solar, the need for future panels would only need to keep up with increased demand, meaning we have all these factories built out and not producing. No one will run their business this way. We will NOT reach 100% solar in 12 years.

        • Jens Stubbe

          The final doubling is pretty predictable both for solar and wind and will occur with unabated growth around 2035 when the capacity is sufficient to produce all energy the globe requires including having substituted all fossil energy sources for each and every use case.

          The hasty process that is changing nearly all parts of fossil value chains into stranded assets involves far bigger economics and resources than what will be the case 20 years from now when solar and wind is built out.

          Your speculation that all these factories built and no market is a figment of your imagination because the real case will be a huge decrease in energy cost and hence a huge upsurge in energy demand.

          To match wind cost wise solar will long before be based on thin film and the fabrication will be really lean (Think like First Solar) and the buildings and the labour will easily be usable for a lot of other purposes.

          Wind power will also be a lot leaner than it is today because ever decrease in weight and increase in productivity will continue, so you will need a lot less materials and people to produce and service wind power in the future. Wind factories will also be suitable for a lot of other purposes and the staff will also be qualified for many other positions.

          Finally the economic growth will probably continue and the service requirement will probably persist along with a replacement market. Vestas has about the same turnover for service contracts as for sales of new turbines and the same is the case for GE whereas their competitors do not understand how to run that side of the business.

          In view of that I think the market winners will probably force the competition out of the market in a gradual process and land a tranquil transition to a lower growth level.

          • Bob_Wallace

            “the real case will be a huge decrease in energy cost and hence a huge upsurge in energy demand”

            A decrease in energy cost, yes. An increase in energy demand in those parts of the world where relatively little energy is used today, yes.

            Do you see an increase in energy demand in the “developed” world? I expect a significant decrease. Moving from ICEVs to EVs and improving building thermal quality should be two large factors in reducing energy use.

          • Jens Stubbe

            I see a huge demand rise in the developed world and an even bigger in the developing world.

            First of all I know the extraction of raw materials is becoming more energy intensive simply because the easy ores are exploited.

            Secondly I think agriculture will be much more energy intensive than today because the economic models for hydroponic farming with artificial light is improving fast.

            Thirdly I think the world will be primarily driven by Synfuels.

            I am fully aware that just LED’s fully implemented will slash global electricity consumption by 30% and that other more energy efficient technologies could contribute about the same. Never the less I think that the development of greater wealth will replicate the more wealthy peoples energy consumption so I think even the new more efficient technologies will not lead to reduced net energy consumption.

            IMO expects global growth to continue along the trajectory from the last three decades, which means that the economy will tripple by 2050. I think the energy consumption will be a far smaller part of economy (this has been a very consistent trend since the days where we used whale oil to illuminate our homes) but still think that the energy usage will grow more than the economy.

            But all this is guess work and I know you are a firm believer in EV’s so your perception of the future will be entirely different from mine.

            Ps. I got to drive Tesla S last week and it was a pleasant enough car.

          • Bob_Wallace

            And I expect to see mining and farming largely turning to electricity which will drastically cut energy use.

            I’m not expecting to see much indoor hydroponic food farming. Outdoor hydroponic farming does not consume much energy. Probably less than farming in open fields. One is moving modest amounts of liquid short distance, not large farm equipment long distances.

            You need to remember that as we electrify transportation it becomes less expensive to move fresh food from where it is cheapest grown to market than to create artificial growing spaces.

            I’m unconvinced synfuels have a significant role to play in our energy future. I’d have to see someone producing at a competitive price.

            The efficiency of almost everything is dropping. Modern heat pumps use lots less energy that older fuel based heating and AC units. Electronics and appliances are becoming more efficient.

            Energy use growth is almost certainly going to be less than economic growth.

            Yes, I can’t see any competition for EVs. Liquid fuels, natural or synthetic are simply priced out of the market.

          • Jens Stubbe

            Indoor hydroponic agriculture is getting more and more energy efficient and the greatest advantage is the market proximity factor. My girlfriend works with algae research and they have already pinned out a couple of companies.

            The forthcoming development in biofuels with reverse photosynthesis breakdown of biomaterial will accelerate biofuels production and as a by product deliver mineral rich fertilizers for hydroponics.

            Your expectation that energy growth will be less than economic growth hinges upon the present energy scarcity situation whereas my expectation hinges on the presumption that energy cost will go down fast and consequently that the demand for energy will rise.

            The main difference is perhaps that you are less confident about the current cost trajectory for key renewable technologies. Also I notice that you accept incredibly high cost for HVDC in USA, which effectively limit market access for renewables and create protected market niches for storage.

            To be quite blunt grid scale storage is never going to happen in any important measure.

            Regarding the EV vs Synfuels discussion you once claimed a $2/gallon threshold. I think that threshold is arbitrary depending upon a lot of different factors, but even so I do not think $2/gallon is out of reach for Synfuels.

            If my optimistic view about the future price point of electricity is correct and I am also correct in assuming that US HVDC will settle closer to European price levels, then the over provision strategy will be the cheapest route to meet market demand and Synfuels will be used to absorb excess electricity.

            Seen from the perspective of solar and wind manufacturers the Synfuel scenario is almost like a wet dream because they get to sell a lot more equipment. Seen from the perspective of miners, battery produceres and EV producers low electricity prices will be a problem.

          • Bob_Wallace

            I highly doubt that we will greatly increase our energy use via indoor hydroponic farming. Again, transportation is likely to get less expensive, cutting the competitiveness of local indoor growing.

            Outdoor local hydroponic growing. There’s a chance for that.

            “Your expectation that energy growth will be less than economic growth hinges upon the present energy scarcity situation ”

            The developed world is awash in energy.

            ” I do not think $2/gallon is out of reach for Synfuels.”

            I’m assuming you have no evidence of affordable synfuels so all your statements about the future of synfuels are pure speculation and you need to indicate that in your comments.

            “If my optimistic view about the future price point of electricity is correct and I am also correct in assuming that US HVDC will settle closer to European price levels, then the over provision strategy will be the cheapest route to meet market demand and Synfuels will be used to absorb excess electricity.”

            I’d be very happy to see transmission costs drop, but your math alone does not convince me when I see US and European prices in the 2-3 US cents per kWh range. I’m having a lot of trouble with the way you mix your hopes and assumptions freely with facts.

            ” Synfuels will be used to absorb excess electricity.”

            You continue to fail to offer a reasonable scenario in which a synfuel plant could sit idle for thousands of hours per year waiting for the few hours per year of excess electricity.

            It’s time for you to quit fouling up this site with pink unicorn thinking.

  • A_Siegel

    1. Isn’t Kurzweil conflating energy with electricity? Solar is far from 2% of all energy use globally.

    2. Are we actually at 2% electricity supply already? EIA (finally counting rooftop/small solar pv) had solar top 1% of US electricity in 2015. As per a CleanTechNia article, there are claims that the global 1% supply threshold was crossed roughly a year ago ( With a 2-year doubling, we could expect 2% of electricity supply threshold to be crossed perhaps late in 2016 or sometime in 2017.

    • neroden

      Non-electricity energy use is insignificant. Seriously.

      We *use* a lot of non-electricity energy but we use it *incredibly inefficiently*, and electric technology is way more efficient. If we replaced our entire land vehicle fleet with electric vehicles it would only increase the electricity usage in the US by about 10%. There’s similar stuff going on with heating.

      Basically switching to electricity eliminates a lot of the non-electricity energy use; it was pure waste.

      • Jens Stubbe

        Not true really. There are several processes that are even above 100% efficient and a lot of energy usage that is based upon biomass that if left to rot would induce more GHG by oozing Methane.

        Electrifying the globe calls for several hundred gigafactories associated with mining, recycling facilities etc. that have to work for decades.

        The much simpler and faster solution is to remove GHG at the source by switching to biofuels and Synfuels.

      • A_Siegel


        Out of roughly 96.3 quads of energy use in 2014, the LLNL charting has 38.4 in the electricity sector.

        Now 2/3rds of that is ‘rejected’ (waste) which on-site / distributed solar reduces tremendously (nearly eliminates — though, depends on calculations — are inverter losses considered ‘rejected’ energy … )

        While there is potential to shift significant amounts of the 60% of the energy use that is not electricity to electricity (electric transportation (cars, trains, …); heating; some industrial processes), the points that I raised remain:

        1. Kurzweil certainly seems to be conflating electricity and energy. Solar is nearing 2% of electricity supply, not global energy supply.
        2. The 2% figure does not yet to have been reached.

        • A_Siegel

          PS: Of course, the LLNL is US only, not global. Easy chart to grab/use for illustrative/informative purposes. (note: would appreciate tip to equivalent global chart)

          • A_Siegel

            EiA data is 39% of global energy use is to produce electricity: Of 524 quadrillion British thermal units (Btu) in 2011 “World marketed energy consumption” (e.g, not counting biomass cooking / heating, for example; nor solar cooking; nor …), 204 (39%) were in the electricity sector.

  • AltairIV

    I have to say that it’s not the growth pattern of solar itself, or any other single technology, that interests me. What really thrills me is the growth curve for the renewable infrastructure as a whole. Solar, wind, storage, HVDC, smart grids, EVs, net-zero construction… there’s a point coming where all of these individual technologies will fully converge and start feeding each other in an accelerating spiral of synchronistic growth and new emergent organization. I am certain that the future is going to be one amazing place.

    • Carl Raymond S

      When the whole is greater than the sum of parts, the word is ‘synergy’ – like bees and flowers, renewable energy and demand managed home charging, gigafactories built for EVs that produce storage and are zero net emissions – yes, I’m feeling it too.

  • JD234

    Well, it looks exponential for now, but unfortunately all technology growth is actually logistic, or some other sigmoidal function. At some point the technology approaches some limiting ceiling and starts to slow at an equally exponential rate, before eventually leveling off. Hopefully (and I believe) we are still at the point in the sigma where it will remain exponentially increasing for a few more orders of magnitude. But the usual mistake Kurzweil makes is taking all locally exponential curves as permanently exponential, rather than the exponential portion of a sigmoid. It produces all sorts of absurd predictions (like nearly infinite energy in 50 years) which are clearly not true, and which means we have to be careful in considering just when we think the curve will start leveling off. But I remain optimistic that solar, unlike silicon-chip computation, is a long way from approaching its ceiling.

    • AltairIV

      No, Kurzweil doesn’t actually do that. He may be simplifying the solar growth curve for the sake of keeping it easy for his audience to understand, but he does clearly recognize that no single curve will ever be straight and continuous.

      He sees instead a bigger picture. Individual technologies and paradigms come and go, but every time one growth curve falls flat there are always new ones coming along to take over, thus keeping the overall path of progress moving in an upwards, and similarly exponential, curve. It’s this overlapping meta-curve built up from the combined effect of all the individual curves that are the true foundation of his thesis.

      He lays it all out here:

  • helen

    And if more people and companies do solar it will double faster

  • kallen

    Great article but what to me is interesting are the other .5% phenomenons out there that are not yet on the public’s radar…like space travel, genetic manipulation to extend life or save threatened species and our ability to unravel the identity of distant worlds and their habit-ablity through better astronomical systems.

    • neroden

      Biology does not follow the same sort of exponential curves as technology does. There’s gonna be some crazy amazing stuff discovered with genetic manipulation but it’s gonna lead to *waaaay* weirder history. And a lot of it is gonna be bad. Sorry.

      • kallen

        @neroden – boy are you wrong: better get yourself informed. Maybe start by googling CRISPR. March’s Scientific American has a great article on it. You’re Welcome.

  • Jeff

    I can’t wait! All that is needed is reliable, cheap, and large energy storage. They have been working on this for over 30 years. They need a breakthrough! Please, get it going. Additional Wind and Solar is great, but too expensive at the moment.

    • jeffhre

      Et Tu Brutus?

      Why is that no one says – Natural Gas – “All that is needed is reliable, cheap, and large energy storage.”

      There is a gazillion (this is a technical term with no financial equivalent) dollars worth of natural gas peaking capacity. Which is used about 5 -10% of the time during any given year. NG peakers are ubiquitous! But somehow things like redundant renewable capacity, storage, and demand management ie renewable “peakers” are verboten on the grid? And wouldn’t it be possible to keep using NG peakers for that last 1% – for a while?

      Bill Gates keeps saying we need an energy miracle…in the face of renewable energy exponential growth. That’s a lot like…

      • Ronald Brakels

        The good news is, if natural gas generators are only providing a small portion of the electricity used on the grid, then becomes easier to remove the CO2 they release from the atmosphere and sequester it. Right now I could produce electricity from natural gas and pay to sequester the CO2 released at a far lower cost than new nuclear power.

        But the good news is once storage reaches a certain level, it will eliminate the need for less efficient peak natural gas plants, and more efficient combined cycle natural gas power stations can be used instead. This is because the peak generators’ role will be taken over by storage and the combined cycle plants, which can be 60% efficient will make sure the storage is charged up prior to a period of high demand / low renewable output.

        • jeffhre

          Multi-cycle combined heat and storage from gas turbines can be more efficient than that, in cold climates! Can they be used as peakers though?

          But, yeah, that’s the whole point – to do things that are sustainable.

          • Bob_Wallace

            Batteries do the peaking stuff. When battery levels drop enough (and no solar or wind is predicted to come to the rescue) then CCNG plants are turned on in sufficient amounts to replenish the batteries.

            There’s some chance that we can use CCBG (biogas) plants if we get the need low enough. Or biomass plants. Doesn’t matter if it takes hours to fire them up, we just need enough storage to carry the load until they can come online.

      • Carl Raymond S

        We’re going to have a fleet of EVs anyway and the additional cost of demand management of those home chargers is a bit of code in the software i.e. essentially zero, so might as well do it.

        • jeffhre

          Very true – and every time you mention it some blog surfer will say; sure that sounds great for the grid of 2050. And; today nothing is realistically going to replace the 2/3 of the grid powered by fossil fuel.

          • Bob_Wallace

            Show them this graph and tell them to look forward to a big improvement in 2016 as we close a lot of coal plants.


    • Ronald Brakels

      Solar is the cheapest source of electricity available to most Australians, and new wind power is cheaper than new coal or natural gas capacity. The state of South Australia electricity equal to over 41% of the state’s consumption is generated from wind and solar, and that might go up to 50% by the end of the year, and we have no storage. So while cheap storage would be great, it’s not needed to replace coal and an awful lot of natural gas use.

      The rapid decline in the cost of wind and solar has caught a lot of people by surprise.

  • Ronald Brakels

    Personally I think Ray Kurzweil is a few doublings short of a singularity, but I’m not going to pick on him on account of how he could probably shatter my skeleton with a single blast from his Moog synthesizer.

    However, I will mention that South Australia is expected to meet all electricity demand with rooftop solar alone at times next decade. It is so cheap now that it basically makes sense for every building to have at least a small rooftop solar system.

    There is no good reason why other countries can’t match or beat Australia’s installation costs. After all, we do have some of the highest wages in the world, and while other locations may not have Australia’s solar resources or high retail electricity prices, provided the cost of solar continues to decline at its current rate, it won’t be very long before most countries find themselves in Australia’s position.

    So unless there are surprising breakthroughs in other areas, and provided solar continues to decline in cost, Australia will get the majority of its electricity from solar, as will most countries. But this does not mean that other sources of energy such as wind or hydro will not be used as well.

    • neroden

      If you can explain to me how to get Australia’s installation costs here in upstate NY, by God, I’ll do it. Because installation costs are keeping solar much less competitive than it should be here.

  • Eric King

    Another thing people don’t understand about exponential growth-when solar energy gets cheap enough the market will find a use for it-this means that we will not have a ceiling of 100 percent of the energy we use now-instead we will keep using more solar for different things so that at the end of the century solar will provide something like three times our current yearly energy usage.

    One obvious use is to desalinate water and pump it inland to fight global warming by growing forests-and hunger by growing food at the same time-you could also pump the water uphill and have power storage at night.

    Then there is the hydrogen economy-splitting water with solar to create abundant hydrogen to replace fossil fuels for certain applications.

    Wealth equals the work done with energy-which means the input of cheap solar energy will be a huge wealth creator.

    • eveee

      Thats right. The demand market will not stand still with EVs growing by leaps and bounds among other things. Plus, markets are dynamic. When an excess of solar appears, people will find uses for it, and demand will shift to times when it appears.

  • Michael Bartell

    Should I point out the obvious Tesla model 3. Thank you Elon Musk

  • Tov

    Any reference to that weird 2% claim?

  • vensonata

    The problem is the manufacturing facilities. Imagine if you get to manufacturing capacity for 25% of the world energy needs in PV. Now if those factories operate for 2 more years they must shut down because the market is exhausted. You can’t build factories with 2 year lifespans. That is why exponential is not going to happen. What I think will happen is a factory manufacturing base of 10% of the worlds PV electrical/energy needs, and then 10 years of production at that rate. That makes sense economically.

    • eveee

      IMO, it’s incorrect to assume factories shut down the moment 100% solar is reached or that solar reaches 100%. Neither happens. There will be a saturation, but at what point? Detractors are talking saturation before the real growth takes off at the tipping point. There are other factors that slow growth, like installation, and infrastructure. Still, that didn’t stop cell phone growth. One has to analyze that.
      And there is a wear out, damage, obsolescence factor, population growth, and new market growth to consider.
      So the factories won’t just stop. We should note that % energy figure is a proxy for total absolute output and they are not the same, to be accurate. But for shorthand…
      If we got to manufacturing capacity for 12% of the world’s needs, and solar was doubling every 2 years, we would produce as much new solar as already existed in 2 years. That means we would already be at 12% and the factories would produce the next 12%, bringing us to 24%. If factory output plateaued adding 12% every year, it would only be six more years to 96%. That’s unlikely, too, because solar isn’t in a vacuum. There’s hydro, wind, and so on. And the need for transmission, storage, etc. plenty of practical realities.
      I suspect the real reason factories will keep growing is that the market is becoming electrified. EVs are coming soon. That means growth will actually increase.

      • juxx0r

        All over by 2025. Agreed.

      • Bob_Wallace

        At some point we’ll see the market for new solar (wind, batteries, etc.) drying up. There will be a continued market for replacement hardware but the new market will be over. (That’s assuming population growth stops and we don’t invent “must have” things that suck down huge amounts of electricity.)

        Two years, five years, some years before that point is hit company XYZ is going to meet and consider whether to open a new plant. They’ll look at the end of the “new market” coming to an end and realize they’ll never recover their investment.

        There will be enough existing capacity to pump out a huge amount of new solar/whatever per year. A lot of that capacity will no longer be needed. If solar panels have a 40 year lifespan then we can close most of our solar panel manufacturing plants and just keep enough to replace 1/40th of existing panels per year.

    • tibi stibi

      you are right its not about the install its about the factories. as soon as they hit the saturation point the cast of panels will go down to the cost of the materials (mostly sand).

    • Jens Stubbe

      A 100% renewable grid requires over provision just like any grid up until this point in time.

      The difference with renewable energy is that over capacity has become cheaper and will deliver electricity based upon free free energy from the sun, wind, water, biomass etc.

      There is no point in limiting renewable to produce electricity. The fossil age will not end because fossils run out but simply because they become too expensive.

      The most logical choice is to produce Synfuels based upon excess electricity.

      Ps. I agree that solar is not likely to be the sole provider of electricity and I think we should let wind and solar slug it out.

      Ps. ps. Wind is only three years slower than solar if the historic growth in wind continues unabated

      • eveee

        If those rates continue, solar will overtake wind in about ten years.

        If we assume 0.6% solar and 4.6% wind, say wind is 8x solar now.
        If solar doubles every 2 years and wind every 5, in 10 years, solar will be 2 to the power 5 or 32x. Wind will be 4x. That eliminates the 8x advantage.

        • Jens Stubbe

          It will be great to see how this projected future unfolds.

          Wind has kept a very stable 7 fold increase in capacity for four decades now but slightly more when it comes to kWh production increase due to steady capacity factor increase.

          This is still considerably slower than doublings every second year.

          During the now four decade long expansion of wind the industry has absorbed the blows dealt by stop and go policies, increased fossil subsidies, increased nuclear subsidies, commodity price fluctuations, Fracking gas rush and solar power uprise.

          In EU the trend is towards protective measures for classic power generation both instigated by national governments and through EU policies. They claim classic power plants are indispensable for supply security and at the same time block market access for renewable energy simply by smoldering the legislative process with all kinds of procedures that can block for years.

          • eveee

            Yes. I gave the simple math as an illustration, but its more complicated. Capacity factors are rising for wind, and solar has a more limited capacity factor, but can have a very high load matching. A limited analysis ignoring load matching doesn’t do the subject full justice.
            And different locations will mix various resources to match load over time. As James Wimberley pointed out, wind and solar are complementary seasonally, if not daily, in many places. The diurnal pattern of wind puts more wind at night in CA for example, and more solar in the day.
            In Germany, wind in winter, solar in summer. Its easy to see the match between air conditioning loads and solar.

    • neroden

      Vensonata, you are correct on a WORLD basis, which is why there will be a slowdown in the exponential curve once the factories are pumping out about 10%-20% of the world’s total energy needs per year. But you’re missing something: what is the WORLD total energy need going to be in 20 years? Africa and other third-world areas would sure love to use way, way more energy than they are using now…

      HOWEVER… the countries which can afford to pay more will get first grab at the solar panels. As a rich country, the *US* will probably be 100% solar well before that slowdown happens.

      That slowdown will affect Africa and other third-world countries, which will have a voracious appetite to install solar power to replace *lack of energy*. We’re going to outbid them until we shut down all our fossil fuel plants.

  • Doltsbane

    Ray Kurzweil is also predicting physical immortality by 2045. The man is slowly poisoning himself with 150 dietary supplements per day. Just because someone tells you what you desperately want to be true and uses a bunch of pseudo technological gibberish doing it doesn’t mean he should be taken seriously.

    • eveee

      That doesn’t mean you shouldn’t listen either. Listen to the reasoning not the source. What’s correct about the exponential growth? Why has it followed that curve ? Pay attention to the ideas.

      • Doltsbane

        His ideas are nonsense, Solar currently produces .4 percent of the United State’s electrical power, are you really going to take seriously the proposition that in only 12 years it’s going to become the dominant source of power? Even if you believe that solar energy is practical, there is neither the money nor the manufacturing capacity to reach that situation in anything even remotely resembling such a time frame.

        • eveee

          There are limits that have to do with economics and other factors. I would be more enthusiastic about declaring him an idle dreamer were it not for a few facts like these:

          “Solar, wind power accounted for 68 percent of new capacity”

          “In yet another record-breaking year, the solar industry in the United States installed 7,286 megawatts of solar PV in 2015”

          While wind added 8.6GW.

          Solar won’t be 100% of capacity, because of wind and other sources among other things.

          “More than half—51%—of the US’s electricity generating capacity was built before 1980. About 74% of all coal-fired power plants are at least 30 years old, and the average life of such plants is just 40 years, according to the National Association of Regulatory Utility Commissioners”

          Those plants will be replaced. How much coal? Its 33% of our generation. If solar and wind are the dominant replacements, its just a matter of time before all the coal is replaced by them,

          Maybe Kurzweil is wrong about 12 years. Idk. But a large portion of the current crop of coal plants will be dead in 10.

          Natural gas additions are lagging wind and solar. That means renewables will pick up more of what coal drops going forward. At todays rates, it looks like close to 20 years or a little more. But if wind and solar prices keep dropping, its going to just kick NG out of the new market altogether on economics.

          I wouldn’t worry about investment.
          The race to wind and solar will become a stampede as investors realize their FF interests are toast.

          The exponential nature of growth or the point of saturation are not so much the issue here, as when the tipping point is reached. Thats not quite here, but impending. Real growth has barely started.

          • Doltsbane

            You’re confusing rated capacity and actual output. a Gw of solar doesn’t actually produce a Gw of power continuously, whereas a Gw natural gas turbine or nuclear reactor more or less does. The large percentage of new capacity is a reflection of how little new capacity has been built. 8 Gws might seem like a large amount, but that’s only the equivalent of 2 typical conventional plants (for comparison, China built that much in coal fired plants each week last year). The figures you cite have more to do with pressure from environmental groups, the Obama Administration’s hostility towards non-renewables, and the general downturn in industrial energy consumption in the US. Take a look at what’s happening to the solar industry in Spain. Cut off the government subsidies, and the economics of solar collapses. Real game changing technology doesn’t require propping up.

          • eveee

            I know the difference between rated capacity and output. I know the capacity factors of all three.
            NG is running about 55%, increasing as NG is used displacing coal. Wind is about 33% and climbing. Solar is much lower in the 20s, for utility.
            I beg to differ about subsidies and renewables. FF still enjoy rich subsidies.
            Wind is cheaper than all other sources except NG without subsidy already and solar is dropping rapidly and cheaper than most other sources already.
            NG would have to give up a lot of subsidies to figure out where it really stands today.

          • Personally, I include externalities as a subsidy. We are subsidizing these as a society. There is a lot of momentum to price some of these externalities more. If we are being realistic, wind is the cheapest option already, and solar is #2.

          • eveee

            Yes, Externalities are the elephant in the room for FF. With them, its not worth getting them out of the ground. We are starting to see some people wake up to the liability of developing FF resources going forward. Their payback will be crippled, rendering them useless. So many already are being shut down.

          • nakedChimp

            You’re witnessing the start of a game changing industry..
            Where do you draw the line on a child.. when it’s 1 year old? 7 years? 16 years? 21 years?
            When do you stop supporting it, when does it start to run on it’s own and even begin to support others?
            Solar isn’t a grown up adult in it’s 20’s yet..
            Be fair.

          • Doltsbane

            The point of my posts is not to criticize the long term viability of solar power, rather it’s criticism of Ray Kurzweil for making grandiose predictions that will end in disappointment in the technology. Souring people on solar energy by making claims about it’s short term prospects that are highly unlikely to pan out will actually retard their adoption in the long term. It particularly annoys me because Mr. Kurzweil is doing it merely to A: Puff up his own ego as “the great technological sage” and B: To shove the large amounts of money that gullible people give him for such speeches into his already well stuffed pockets.

          • dkdk999

            In my opinion, it’s always bad to “personalize” a debate.
            I agree with you that Ray Kurzweil is probably a lunatic in certain ways, but you have to each of his positions individually, he can be wrong about one thing and right about another.

          • Doltsbane

            Kurzweil “personalized” the debate himself when he chose to weigh in on a subject that he isn’t qualified to offer an opinion on. He’s a computer scientist, what makes his thoughts on solar energy technology any more valuable than anyone else’s? Short answer, because he’s RAY KURZWEIL!. Why should anything he says be taken seriously when he talks nonsense so much? Because he occasionally stumbles across some glittering generality that might be construed as having been correct? Even a stopped clock is right twice a day.

          • The point is what is dominating new growth (and note that the rest is basically just nat gas that is unlikely to stay competitive for long).

            As more old power plants have to retire, they will be replaced, and largely by renewables. There are many on the dock.

          • Freddy D

            What happens as we reach 100%? One idea is that the market will use excess capacity instead of storage because it might be cheaper. Imagine a day when PV is so cheap that you install twice the capacity you theoretically need because it brings the output during shoulder times up to meet demand. Then one needs a smaller battery pack.

            Other ideas have also been mentioned – electrifying hot water, cooking (inductive stoves have become good enough to outperform gas), space heating, etc.

          • eveee

            Yes.Any time there is cheap electricity, things like aluminum smelting rush right in. We don’t even know what applications might be used yet. They will be developed.
            Point is markets are dynamic. In the past, daytime electricity rates were at a premium. The wholesale rates are negative or low during the day where solar reaches a large percentage.
            There should be low daytime retail rates in those places if PUCs do their job.
            Many people have mentioned that West facing panels or articulated panels could increase late day output, too. Truly, there are a host of ways to provide energy. CSP with thermal storage, residential and commercial thermal storage, and on it goes..

        • AltairIV

          Most people tend to grossly underestimate the speed at which a given technology can achieve penetration, once the conditions are right for it to take off.

          The first ford Model T’s started rolling out of the factories in 1908. Do a Google image search for “1920’s street scenes” and see what city life was like only 15-20 years later.

          The Wright Brothers made their first historical flight in 1903. A 10 year-old witnessing that event would’ve been able to fly non-stop across the U.S. in his 40’s, and travel across the Atlantic on the supersonic Concorde before he died.

          How about this image of how the Manhattan skyline has changed over the last 137 years?

          The world wide web was first opened to the public in 1993. By 2003 the internet had already become an indispensable part of life for the average person in the western world.

          The iPhone was released in 2007, the first device that really combined all the features of the modern smartphone. Less than a decade later, smartphones are so ubiquitous that the very fabric of society is being altered yet again.

          Here’s a graph of the historical growth curves for a few other important technologies. Notice how the vast majority of them achieved 50-60% penetration rates within 20 years, and often within a decade or less.

          So will we reach solar dominance in 12 years? Well, the time frame may be a bit optimistic, but historical precedent says that it’s within the range of possibility. What IS pretty much guaranteed is that a fully-renewable energy infrastructure is coming, and it will be here well before the average person expects it to be.

          • neroden

            I remember when the WWW was introduced. Never thought I’d be using it for most of the things I use it for.

        • Bob_Wallace

          It’s actually about 1%. Half a percent utility solar and half a percent end-user.

          Two year doublings maintained over 12 years just isn’t feasible. We couldn’t build plants and silicon processing plants that rapidly.

    • John Moore

      Physical immortality by 2045? This has some really painful implications for Cleveland Browns fans, especially young ones.

      • jeffhre

        Fortunately Cubs fans are immune already.

    • neroden

      Kurzweil’s a nut. He’s also correct about solar power, as anyone who follows the industry knows. Ad hominem arguments are not valid, you know.

      • Doltsbane

        Pointing out that someone consistently and persistently talks crap is not an ad hominem attack, you’re the one that labeled him a nut. Kurzweil’s problem is that he insists on applying Moore’s law to all technology, rather than just computers (And Moore’s “law” wasn’t handed down from Mt. Sinai micro-etched on silicon nano tablets, it’s only true until it’s not.)

  • nordlyst

    Yes there’s a lot of energy in sunlight. But it’s not as simple as that. Totally ignoring intermittency doesn’t lend credence to the man’s projections, since anyone who’s paid any attention knows that’s where the challenges are. Solar is already cost-competitive with fossil-based power, beaten only by wind.

    I’m sure it’s solvable, but I’ll happily accept bets that solar will NOT be 100% of the energy production in 2028. Norway for instance will still have a lot of hydropower. At least some aircraft will still use liquid fuels (though it might be biofuel it won’t be from photovoltaic). And so on. But if solar gets to 50% by then I’ll be positively surprised.

    • JamesWimberley

      There is massive complementarity between renewables. Wind and solar are variable, but inversely correlated seasonally, and in the short run randomly. so the variance in output of a mixed fleet is less. The variables are complementary to the usually more expensive despatchables: biomass, hydro, geothermal, load shedding, imports and storage. Tidal is in between – it’s cyclically variable, but 100% predictable on a long timescale. An optimised fleet will therefore be mixed, unless solar gets so cheap that massive overbuild is always the cheapest solution.

      • Richard Poore

        The problem with balancing solar and wind is that of distribution.

        Wind power transmission is already running into significant resistance. Just getting power from Iowa to Chicago is proving to be difficult, not to mention the fierce resistance against power lines in the north east.

        Eventually long distance power transmission will expand, but looking at current problems… 2028 isnt realistic.

        • eveee

          While US transmission expansion has been poor, its changing. Texas has CREZ, and recently we have this:

          “The other big development is the Energy Department’s announcement that it will get behind the proposed 700-mile megawatt Clean Line Plains & Eastern transmission line, designed primarily to transport electricity from Oklahoma and Texas wind farms through Arkansas to Tennessee and points east.

          At 4,000 megawatts, Plains & Eastern counts as the biggest renewable energy project so far in US history.”

          • Richard Poore

            Yes, its good that project has strong support. But please note that project is atm only the announcement of a proposal, its a long way from even being at the planning stage.

            What Im pointing out is that current actual projects are running into a buzz saw of nimby opposition.


            And thats in the midwest, where there arent the fears for tourism (ie scenic beauty etc) that are hitting the north east.

            Current projects are being delayed. We have to believe that future projects will also face delays.

            That is what makes an early date seem unrealistic.

          • Karl the brewer

            I have a feeling that if / when the true horrors of AGW can no longer be swept under the carpet we will see installations pick up again very quickly. But too late?

          • eveee

            Yes. Its been a slog. I agree it needs to be pushed ahead a lot more. I just don’t want to overlook progress that has been made in places like Texas, altho that has not been HVDC, which is really needed. Its almost scandalous how poor US transmission has been compared to China.

        • neroden

          Transmission is certainly a bottleneck. Here in the snowy Northeast, we could *really use* some major transmission lines from those giant solar farms in the Southwest…

      • Freddy D

        It’s easy to imagine a world where “massive overbuild” is cheaper than storage. In other words, assume PV is so cheap it’s nearly free, like CPU power is now. Instead of exactly matching my PV system to my annual needs, I’m going to put double the theoretical PV I need and a much smaller battery pack. I cover the entire roof with solar and on cloudy days and morning and evenings it’s covering 100% of the needs. Smaller battery pack covers nighttime. Mid-day, excess capacity goes unused.

        It changes the calculation on how I optimize the system. I gave an example of a single homeowner, but same principle applies to utility scale. Just like we don’t run every gas and coal plant full throttle 24/7, we may not run every PV plant to its full potential every minute.

        Wow, that was a fun mental exercise!

        • Bob_Wallace

          Our gas peakers run only 5% of the time. Coal and CCNG plants run 50% to 60% of the time.

          I’d love to see some real world data on how much we’d need to overbuild solar in order to get 100% of our needs met on a cloudy day. (Obviously it would vary with location. Would be pretty hard to pull off in northern Alaska in the winter. ;o)

        • smslaw

          My rooftop PV production drops to almost nothing on overcast days and early evenings. Doubling or tripling the number of panels wouldn’t do much. For example, in March, which was a pretty sunny month where I live in Maine, we still had 9 days with less than 10 KWH. In February we had 15 such days. We’ll need significant storage to come close to replacing all NG.

          • Bob_Wallace

            Does your metering system record panel production per day? Actual output from the panels – not net gain/loss.

            If so, do you have data to share?

            My system is older and I have no way to measure panel output, only net gain/loss. I’d like to gain a better understanding of the percentage of ‘normal’ one gets out of panels on cloudy days.

            It’s not an issue of sunny/not sunny day, as you know. There are varying amounts of clouds/sunshine and thin/thick clouds. What I think would be useful is a simple chart list of number of 100% days, 90% days, ….30% days.

          • smslaw

            Bob: Yes, my monitoring portal gives me actual production per day. I get net from my electric bill which gives me readings for both in and out meters. I only started monitoring in October, so I’ve been a net user for the most part. I think April will break even or maybe a bit better.

            Here’s system output since October:

            Oct. 637 kwh
            Nov. 438
            Dec. 286
            Jan. 239
            Feb. 370
            Mar. 678
            Apr. 435 (through yesterday)

            My system is nominally 6.6 KW. On sunny days, maximum output varies seasonally. For example, a bright sunny day in December has a maximum output of about 20 kwh, whereas in mid-April max output is about 45 kwh. I’m not sure what I’ll get in June, but it will obviously be more.

            The highest instantaneous output on a sunny day this time of year is about 6kw, generally in earty afternoon. At noon today, which is very sunny, output was about 5kw. It’ll be a bit higher in an hour or two.

            While not all or nothing, typically I get either close to max (say 80% or more) or very little (maybe 20% or less.) If we get snow, output is zero until it starts to melt, which is a function of roof material, roof pitch, air temp and amount of sun. We have a dark metal roof, so that helps melt, but we had some sticky snow twice in February and had 12 days with output between 0 and 4 kwh.

            Our roof is oriented about 15 degrees west of due south. Half the panels are on section of roof that has a 8/12 pitch and half on a 3/12 pitch. We’re at about 43 degrees north latitude.

          • Karl the brewer

            How does the output compare to your usage?

          • smslaw

            I’ve only been monitoring since October, so my use so far is much higher than my PV output. The next six months should be the reverse.


            Use 5923 kwh
            pv output 2648

            The house was new last year and is 100% electric. I expect total use for the year should be around 9-10,000 kwh and output about 8,000 kwh. In Maine, our net metering deal means any excess output over 12 months doesn’t benefit the homeowner. So we tried to estimate total use when we decided how many panels to have installed. If we had opted for a heat pump water heater instead of a regular electric resistance water heater, we’d be pretty much net zero. The problem is that HPWH tend to be noisy.

          • Karl the brewer

            Interesting data, thanks for the response. Are you thinking of adding more panels in the future to cover the shortfall and accept that you will overproducing in the summer? And a comparison for you…I’m in the UK in a hundred year old solid brick house. Our electric for the year is ~ 3000 kwh. Our space heating and water heating is natural gas and is 17,000 kwh per year !! And the majority of that is space heating.

          • smslaw

            Karl- we probably won’t add any panels for a while, at least until we have more data. Of course if we ever get an electric car, we’d surely add enough panels to keep it charged.
            In a small, tight, well insulated house, kwh for space heating is typically about the same as for heating water.

            Natural gas was never an option here as its availability is limited in Maine,especially in rural areas. Propane is available, but expensive and in any event, we wanted to avoid any combustion sources in the house.

          • Bob_Wallace

            Here’s the part that interests me –

            “I get either close to max (say 80% or more) or very little (maybe 20% or less.)”

            If you have daily production data I’d like to know the number of days where output it 80% or more, 60% or more, 40% or more, 20% or more, less than 20%. Something like that.

            I’m planning on doubling my panel wattage in order to cut down on cloudy weather generator use.

          • smslaw

            Can I upload an excel file here? If so, how? If not, if you give me your e-mail I’ll send an excel file that shows daily output

            Will doubling your panels do you much good on really overcast days?

            Are you off grid?

          • Bob_Wallace

            Thanks. Send it to

            I am off the grid. I’m trying to guess how much value I’d get from more panels.

            They may not cover my 24 needs on a really overcast days but I have a lot of days in the winter which are cloudy, but still on the bright side. On most cloudy days I get enough production to cover refer, stereo, computers during the day but not enough to store up for the night. I think adding another 1.5 kW or so might cover a lot of my evening/night needs.

      • Steven F

        “Wind and solar are variable, but inversely correlated seasonally,”

        It depends on where you live and the local weather. In California Wind and Solar output both peak at about the same time. In the summer. Solar peaks at about noon while wind peaks for the day about 5 hours later. In the morning both are at about zero production.

        In the winter fog and clouds reduce solar. Wind most of the time is not present unless a weather front comes through

    • Tony V.

      Have a million electric cars that offer spare capacity, inttermittancy problem solved.Only thing you need is two-way chargers.

      • dmeand management

        Just altering the time of charge can have a big impact, without having to feed the energy back to the grid.

        • Frank

          Right, more chargers where people park their cars when they go to work, or shop.

      • smslaw

        Electric cars could be a real game-changer if we can use the batteries for backup power.

        • Bob_Wallace

          I think that’s a low probability event. Car owners will purchase their batteries at retail and then rent them to the utilities at a high enough rate to cover their costs plus demand some sort of profit.

          Utilities can purchase storage at wholesale prices.

          The way EVs are likely to be game changers is by serving as dispatchable loads. Utilities would offer EV drivers a better rate if the utility is allowed to determine the actual charge times.

          Say you have a 200 mile range EV. Your daily drive is 30 miles. You want a 50 mile range “emergency” buffer “just in case”. So you set your normal daily/nightly charge minimum at 80 miles.

          If you get home with less than 50 miles left the utility charges you up to your emergency minimum. Overnight (day for those who plug in at work/school) the utility charges you up to at least 80 miles but does it when supply:demand ratio is highest.

          The extra 120 miles of range is available for the utility to shift load over time. Wind is up? Charge all EVs to the max. Then during a period of low wind that extra charge can be used up by not charging those EVs (as long as they stay above their set minimum).

          • smslaw

            Lots of possibilities, many not at all dependent on any new technology.

      • Carl Raymond S

        I once had that thought, then came round to thinking it’s even easier than that. Build a slightly over capacity RE grid, then demand manage the home chargers on the big battery cars. It’s the perfect relationship – girlfriend hopelessly non punctual, boyfriend unbelievably patient.

    • Aircraft may shift to some exotic battery chemistry having higher cost but also high energy per weight. Perhaps chemical power cells that, like Aluminum-Air batteries need to be remanufactured for each use.

  • Roger Lambert

    How long has “futurist” Kurzweil been this bullish on this topic? I think most folks here beat him to the punch.

    • I think for several years. iirc (and maybe i don’t), he was one of the few to expect the quick price drop of the past ~5 years.

    • eveee

      Kurzweil has been making the same predictions since 2011 at least. His voice was pretty lonely back then. Still is.

      • Karl the brewer

        This article dates back to 2008 – In the article he states 5 years to be cost competitive which takes us to 2013.

        • Freddy D

          Cool. In 2007 I ran models based on the same premise as kurzweil. My excel spreadsheets had a fascinating “prediction”. 2027 would be the year. (Remember this was 2007). Three things would happen in the same year:
          Solar would hit “100%” of generation
          Wind would hit “100%” of generation
          EVs would surpass ICEs in market share at 51%.

          Obviously can’t go above 100%, suggesting market expansion, cheaper overall energy, and market consolidation.

          Those 2007 spreadsheets need very little bias correction to this day.

        • Calamity_Jean

          “In the article he states 5 years to be cost competitive which takes us to 2013.”

          Which makes him pretty damned close to correct.

          • Karl the brewer

            Exactly. Which is why Kurzweil shouldn’t be dismissed out of hand.

  • Mighk Wilson

    But doubling of computer power did not depend on doubling of physical resources. Still, a hopeful message.

    • JamesWimberley

      A bit less than doubling. Solar efficiency increases by a few percent each decade. The panels get thinner and lighter. Utility solar has largely gone over to cheap single-axis tracking.

      • nakedChimp

        Still waiting on front AND back glass covered panels for the consumer market..
        Anyone seen them yet for JoeAverage?

        • Ronald Brakels

          Yeah… They’re available now. Not sure who sells them offhand, but I would guess they still have a considerable premium.

          And if no one carries them in Australia, getting some might be very expensive.

        • Steven F

          They are called Bifacial solar panels. they are available but cost more than regular panels.

  • Frederik

    We’re only 6 doubles away from 100% solar. I ask myself, does this mean that other renewables like wind energy will also become obsolete?

    • nordlyst

      Yes. But it’s wrong.

      • neroden

        Specifically because wind operates at night and solar doesn’t.

    • nakedChimp

      No, Kurzweil has rosy glasses on.
      Read up there the opinions of James and John.. more realistic.
      Solar will dominate, sure.. but not in the timspans Kurzweil paints on the wall.

      • Well summarized. Agreed.

      • Ross

        If the exponential cost reductions of solar continues as expected we could see more existing generation forced into early closure. It might be a fairly stretched S-curve.

  • JamesWimberley

    As a view from 30,000 feet, it’s fair enough. If you look closer up, it’s a bit more complicated. At high solar penetrations, diminishing returns come into play. The load curve of electricity demand is variable. Peak load – winter evenings in Northern Europe, summer afternoons in the US southwest – is a multiple of minimum 3 a.m. load. So some form of capacity has to be idle most of the time. If it’s solar, the capacity factor has gone right down – so the cost per kWh is rising. Now it’s possible that cheap storage will come to the rescue. That is not a sure thing.

    We are also seeing political resistance from fossil fuel incumbents. If they can’t stop the revolution, their plan B is to get compensation for being forced out of business. This pushback has already slowed the growth of renewables in the UK and Germany, and halted it in Spain. Technological revolutions are not clean and tidy affairs with everybody a winner, but messy conflicts.

    • Yes. I keep wondering if the bird’s eye view of Kurzweil & Seba & such will end up being more accurate than the one that tries to pick out these hurdles and estimate. How quickly can the industry innovate to keep growth on this disruptive curve? We’ll see.

      • JamesWimberley

        That’s a very good point. Greenpeace’s famous predictive success was based on simple extrapolation, while the Very Serious Statistical Persons at the EIA, IEA etc were getting it completely wrong with fancy models based on wrong assumptions about costs and policy sensitivity.

        I will go out on a pretty stout limb and accept the VDMA claim that the solar leaning curve is good for another decade, based on incremental – not disruptive – innovations in technology. The doubts are all on the demand side.

        • I’m with you. But who is VDMA?

        • neroden

          Why would anyone buy more expensive power when they could install cheaper power?

          At the moment the main competition for solar power is *next year’s cheaper solar power*. The moment the prices start stabilizing, the demand skyrockets.

          Seriously, what’s cheaper than new-build solar?
          — old-build solar
          — old-build wind
          — new-build wind
          — old-build hydro
          — some but not all old-build coal

          — some old-build and new-build natgas, but natgas prices are set to skyrocket

          Once the learning curve passes the last coal plant, that’s it.

          That’s IT. When solar is that much cheaper, only idiots will be buying the alternative. The competition will be “utility solar or rooftop solar?” which is already the analysis I’m making!

      • Ian

        While I understand and agree with the assertion that technology (and markets for technologies) advance exponentially, I think there is a basic logical/mathematical error in projecting exponential growth of solar and concluding solar dominance of electrical generation in 12 years. I think Kurzweil is conflating the market for new generating capacity with the market for electricity. The logical way to look at this is that solar generation capacity installed will grow exponentially until it dominates the market for new electrical generation capacity. Already solar is well well beyond 1% or 2% of new capacity installed annually. In the US solar was 14% in 2015. So solar can only double less than 3 more times before it reaches 100% of the market for new capacity, at which point growth of solar electric generation should go linear and growth in capacity installed each year should go near flat. I do think though that solar does have the ability to create new markets that do not now exist, so there may be more upside available. But still, from a 30,000 foot perspective you have to take into account that only a certain small percentage of the world’s generating capacity is replaced each year and demand for electricity is not growing exponentially. And for solar to start to be able to outcompete the operating cost of pre-existing gas and coal plants, it would have to fall in cost by about 75% more.

        My futurist prediction: Solar installations double two or three more times until it comes to dominate the market for new generation capacity by about 2022-2026. It then would come to dominate electrical generation by about 2045, after all the new coal, nuke and gas plants that have been built in recent decades reach their (slightly early) retirements. In the meantime solar PV will also find quite a lot of other markets outside the existing utility sector, which might allow it double one more time.

        • John Moore

          Many goods points. But I have disagreements, and questions.
          First, you don’t seem to give any weight to the possibility that there will be a new sense of urgency on the part of the public/governments to convert to clean energy due to the catastrophic consequences of global warming. I believe that the public will come around, as temperatures increase much faster than the current models suggest.
          Second, where do you get the 75% cheaper figure? Are you suggesting that solar has to drop to 1/4 of what it costs now? This is at odds with everything I think that I know. If that figure is correct, then the transition just may be as slow as you suggest. If not, you are way off.
          Third, I don’t think he is conflating the market for new generation with all generation. I think that in the 2020s, the vast majority of existing, operating coal plants will be closed, whether they are old or not. Coal is becoming such a pariah, that the momentum for getting rid of it altogether will increase until it is gone baby gone. Now I don’t have any facts to back this particular assertion up, but I’m guessing that coal will become everyone’s favorite whipping boy. I think (and hope) that it will become very much politically in vogue to persecute coal.

          • jeffhre

            OK. So let’s just call it exponential growth – without the dominance. Wink 😉

          • Agreed. And… natural gas is next in line behind coal to see its plants shut down simply because fuel & operational costs are higher than the cost of electricity from new solar & wind power plants.

          • Bob_Wallace

            If we develop economically acceptable fill-in sources to replace NG.

          • neroden

            Natgas fuel costs will probably double within the next two years.

          • Ian

            You are right in pointing out that old coal and gas plants are apt to be retired early due to climate concerns. But it’s hard to gage exactly how early. These plants have a typical life of 30, 40, 50 years or more. So if the plants being built today are phased out by 2045, that would be decades earlier than their planned life. That 75% cheaper figure is very ballpark and based on the observation that about 2/3 of the cost of coal generation is capital and only about 1/3 is fuel and operational expenses. If solar cost the same as coal now, than to compete with just the operational cost of existing coal plants, new solar cost would have to drop by 2/3rds. I bumped it up a bit on the assumption that solar is still a bit pricier than coal.

            But back to your point about early retirements, I agree that there is a real possibility that a lot of the world’s fossil fuel plants will be shut down early. Modern societies have grown increasingly intolerant of any risk to life or health. (witness seat belts, anti-smoking campaigns, etc.) I don’t think most societies would tolerate the terrible depredations of the fossil fuel industry if they had any other economic alternative. As solar, wind and batteries continue to prove themselves up to the job of powering the grid, folks are going to become much less tolerant of the fossil fuel industry and demand more closures. But still, it will be economically very difficult for a lot of countries to just write off their energy investments before their useful life is over.

            Actually, it’s not coal that worries me, it’s natural gas. Horizontal drilling and fracking are making natural gas very cheap.

        • Yep, the power plant industry is a slow-moving industry.

          That said, super cheap solar and wind can force power plants out that aren’t “old enough for retirement.” FPL bought a power plant it had very recently signed up to a 20-year PPA with because it realized that doing so and shutting down the plant was cheaper than sticking with the 20-year PPA. We’ll see this kind of thing happening from cheap wind & solar fairly soon, imho.

          • jeffhre

            That is what makes Kurzweil’s continued doublings feasible. Lower prices for solar, lower prices for installation and in the case of distributed, no cost for distribution means solar goes beyond replacing new generation by replacing old plants.

            It will continue to pressure the CF of existing thermal plants. At every level of solar decrease, new thermal plants will be economical for fewer hours of generation.

          • Bob_Wallace

            Continued doublings until we reach the point at which we can’t use more solar without storage. Because the wind blows more hours per year than the Sun shines it’s going to be cheaper to slow solar around 30% to 40% and let wind provide 30% to 40%.

            Both wind and solar should settle in under $0.03/kWh. At this point we don’t have an inexpensive storage solution. It’s going to be cheaper to use solar-direct or wind-direct than stored wind or solar.

          • jeffhre

            We have a lot of NG peaking. We don’t reach the point that we can’t use more renewable energy until NG peaking ages out. I think you also once said we could get to 1% NG use, and at that point it’s difficult to add more solar, wind, hydro, geothermal, biomass and demand management in some combination without great expense. As storage becomes cheaper on grids where electricity is relatively expensive, I see storage having a larger role sooner. But in some places NG peakers could still maintain the same functions…at only 1% of generation?

          • Bob_Wallace

            Sure. That last 1% is the hardest to supply with renewable energy. We need something with low installed cost (CCNG is as cheap as wind and solar) and cheaply stored fuel/energy.

            The last 1% might not be natural gas but biogas burned in a combined cycle plant or biomass burned in a converted coal plant. Or it might be massive storage tanks of flow battery chemicals. Or pump-up hydro.

            This is a ’20 years from now’ problem. We’d be very fortunate to be in a position where we’re dealing with the last 1% even 30 years from now. We’d be reaching 0% CO2 from electricity well before 2050.

          • jeffhre

            Even after we reach a point where the last 1% or as Nerodon notes interconnections, or demand response or another function scales to threaten remaining FF use…Nothing will stop property owners and plant managers from saying – look I’m paying $.13 per kWh and paying $200 a month (or $20,000) and that keeps going up. I can put in some solar with my good credit and pay $140 a month which is the same as $.09 a kWh for only 9 years.

          • neroden

            It’s really really slow to build transmission lines, but I see them as the substitute for NG peakers. Those peakers are only needed regionally; there’s usually a surplus of power somewhere else in the continent.

          • jeffhre

            Good point. There are efforts, but consensus seems to be the odds are not good.

          • Calamity_Jean

            Not just new thermal plants but old ones also. Every roof full of solar panels strikes a (tiny) blow toward making a coal plant unprofitable. Coal is dying the death of a thousand cuts.

        • neroden

          If solar is cheap enough, large numbers of *existing power plants* will be retired to replace them with cheaper solar plants. Solar will continue its exponential growth until all these retirements have happened.

          The key point is that *new solar* has, in many areas, an LCOE which is cheaper than *fuel & operations costs only* for existing fuel power plants. This is more true in the areas with better sun, so areas in the far north may take longer to dismantle their old power power plants.

      • Ross

        Kurzweil’s argument is that despite all the hurdles the exponential growth curve continues to deliver.

        • Yeah, I know. But my understanding is that he also isn’t examining them minutely and trying to determine what will happen each step of the way, but is just saying that it is destined to follow and exponential growth curve and all things will work themselves out accordingly to fit this natural pattern.

          • Ross

            Yes, exactly, fuller exposition of his position.

          • AltairIV

            From what I’ve read of Kurzweil he doesn’t just believe in exponential growth curves, he believes in continuing overlapping sequences of them. Whenever one curve begins to flatten out or hits some insurmountable hurdle, man always seems to find some other way to keep the overall upwards trend going by simply starting over on a new path.

          • Ross

            Yes, I’ve heard him say that.

          • Steven F

            I agree with him on this. Renewables will stay on an exponential curve untill limitations slow its growth to a linear rate. the solar limitations are mainly going to be about storage and distribution of power. Once we hit that point the storage and distribution markets will go exponential.

            The storage market is already showing signs of growth which in my opinion will be obvious by 2020. by 2030 utilities will only be signing renewable contracts that include storage or are for renewables with capacity factors of 80% or higher.

      • Ross

        I think it will for the simple reason that after it has occurred a best fit exponential or s-curve will be matched to the historical data and people won’t quibble over minor variation above or below an exponent = 2.0 and/or they’ll vary the doubling period up or down a bit from 2 years.

    • Bristolboy

      Fully agree with this. UK PPAs for exporting electricity directly to the grid are already falling for solar projects due to the greater integration of solar; despite greater wind deployment the declines in PPAs for these projects are less as wind better tracks seasonal electricity demand.

    • Adrian

      Speaking of “paid to shut down.” I wonder, with prices at < $2/share, could environmental groups make a viable takeover bid for Peabody Energy, for the purposes of shutting it down responsibly?

      • nakedChimp

        “They” should sell for symbolic $1 and be glad about it and the taxpayer not too grumpy as he has to foot the bill in either case.

      • Bob_Wallace

        Peabody will likely reorganize and re-emerge as a smaller company.

        • neroden

          Peabody will be tied up in the courts for years over questions such as:
          — whether the management is wasting the assets
          — whether the management failed in their duties to stockholders
          — how much they owe the federal and state governments for reclamation (nobody will trust them to “self-bond”)

          Meanwhile, their coal mines are mostly considered worthless.

          I see them sitting in Chapter 11 for several years, then going into Chapter 7 and liquidating.

    • John Norris

      JamesW, I think it was you who pointed out that solar CAGR (compound annual growth rate) has slowed recently. My numbers show that the 5 year CAGR was 50% in 2008, peaked in 2012 at 59% and is projected to be 35% for 2016. A two-year doubling requires 42%…

      • JamesWimberley

        Only a few months off.

        Thanks for the calculations. However I am not backing my slowdown observation as a prediction. There are as many shocks tending to speed things up (Mexico, India, Paris) as others to slow them down (UK, Nevada, SunEdison). The short-run growth potential is constrained both by demand at prevailing prices, and the capacity of the supply chain. The latter is strongly dependent on the “animal spirits” (Keynes’ term) of the solar manufacturers and their bankers. They had their fingers burnt in 2011-12, and will be inclined to caution for a while.

        It is also worth remembering that the spread in the estimates of annual installations is wide and getting wider, as well-ordered markets like Germany are supplanted by less tidy ones like Brazil. At the limit, who’s counting solar panel sales in Somalia? Nobody.

        • eveee

          Yes. That variation is noise. A longer period is needed to see if it continues.

      • eveee

        Are you talking annual global solar growth year over year? I think it was too small and too noisy to make much out of the data. Could need to be smoothed. You have the math relationships correct. Can you show your data and math? That would be quite interesting.

    • neroden

      Nope, you’re making a miscalculation. The capacity factor of solar is constant (I mean, it depends on the time of day, but it doesn’t depend on the market). Capacity factor is not useful for analyzing power markets….

      What happens is that you have some excess, or “curtailed” solar. We already have “curtailed” wind.

      People *already* overbuild their solar and wind installations to deal with clouds and calm days. The fact that some of the power is wasted due to being curtailed is irrelevant to the investment decision because you’re already factoring in intermittency. Sure, it means the effective price per kwh is a little bit higher, but
      not meaningfully or significantly: a 10% difference is nothing once
      you’re already “cheaper than coal”.

      We know exactly what happens when “curtailed” solar or wind shows up.

      Someone finds a way to use that free power. They buy batteries. The batteries are being filled with free power, which means they are profitable at quite a low price. Or they build an industrial facility of some sort which works “when the power is free”.

      Now, of course, they overbuild the batteries. (Yep, everyone does this, for security.) So with the overbuilt batteries, there’s a market for yet more solar…

      You get the idea. The solar installs will routinely be overbuilt, which will improve the economics for battery installs. The battery installs will routinely be overbuilt, which will improve the economics for further solar installs. It’s when I figured out this dynamic that I realized that this is is going to go *very very fast*.

      • Carl Raymond S

        10 things to do with surplus energy (off the top of my head)
        Dry food
        Charge cars
        Make ice
        Make bricks
        Drone farming
        Treat sewage

        • Bob_Wallace

          Need to think about cost of infrastructure of those activities and whether they can wait for surplus electricity.

          Take treat sewage. Would it make sense to build enough treatment plants to process all the sewage a city makes during only 5% or 20% of the time? (Not sure how many hours per year you’re expecting a surplus.) Would it be reasonable to build storage for the raw sewage that would have to wait for a processing opportunity?

          At this point in time very little wind is being curtailed. Most curtailment seems to be a transmission issue and gets solved. (Check graph)

          Very low and (rare) negative electricity prices are caused by thermal plants selling at a loss in order to avoid an off/on cycle. Those plants are going away.

          In the future we are likely to see solar/wind farms not being willing to sell for a price that does not cover variable operating expenses and some profit. Because, “Why bother?”.

          And we aren’t likely to see utilities move power from wind/solar farms to waiting users for free because “Why should we?”.

  • Shiggity

    We’ll have to wait until solar is fully dominant for the nuclear and fossil fuel boys to believe it.

    No matter how many statistics you give them, they are locked in denial.

    • Ha, too true.

      • nakedChimp

        Just mentioning here right at the top that this article AND especially DISCUSSION is the best that happened on CT within the last 3 months for me.

        BIG THANKS to everyone involved and for sharing your thoughts!!

        • Ha, thanks! 😀 I enjoy a lot of wonderful discussions on here, but they have largely been about EVs lately. Honestly, the solar & wind industry have become so mature and predictable that things have gotten less interesting, imho. But big-picture, long-term stuff like this is still a question mark and fun to discuss. 😀

          • Todd McKissick

            I wouldn’t worry too much. Solar is primed for a major tech leap as off-grid, residential CSP with CHP and cheap storage hit the market next year. And with excess storage in play, and owned by the consumer (not the utility) the debate will turn to real-time smart grids. Consider that the limits to ‘how exponential’ the growth is are far less for CSP than for PV so it’s growth will be even faster than PV has peaked at.

            And after all that, there’s the fact that oversupply is effectively free money in some form or another to every owner so the economics quickly turns into a positive feedback, not a negative factor to overcome. After all, its really only university campuses that are both a producer and a consumer. Producers won’t stand a chance against zero-down, lower payment installations at any scale.

            Looking at your 9th slide, those products were all purchases to buy a new product that the masses needed convincing to buy. They were not investments in money-making equipment like unlimited surplus electricity will be. Just consider the ways to make money if you could get nearly unlimited, very cheap power.

            It’s like you said in the presentation, people in the 70’s couldn’t envision why they might want a cell phone in the car.

    • Shane 2

      Many of the environmentalists over on are also in denial. There are a bunch of doomers over there who say that with oil gone, starvation, war and disease will lead to an 80% decline in population.

      • John Moore

        Very odd. It seems to me that the 80% figure is nonsensical. Imagine population going down by 5 or even 10 percent (700 million people) without a nuclear war happening. It’s just not possible. But five and a half billion people dying, and nobody starts a war? It’s preposterous. I am assuming, here, that in a full scale nuclear war, the survival rate will be somewhere around, say, zero.

        • Ross

          Even after a full scale nuclear war a small number of people would likely survive in less hostile niches. Think of humans as a kind of vermin, some of us would likely survive longterm along with the rats and insects.

      • Ronald Brakels

        Some people want to make themselves appear more saintly than they are by making the enemy they fight appear more diabolical than it is.

      • Ross

        I encountered that sort of thing on Facebook over the Tesla Model 3.

        They couldn’t see the bigger picture of stimulating the transition away from ICE vehicles to get to a world where people have their material wants but decoupled from environmental destruction.

        They were accusing Model 3 reservers as being (relatively) wealthy, privileged, people trying to salve their consciousness while continuing to destroy the planet.

      • Steven F

        It;s not only Risilience. Take a look at In short any site that posts about peakoil frequently is full of people that say renewables, energy storage, conservation, and EVs won’t have any effect on the problem. If you go to pro nuclear sites you will see the same thing.

        • Bob_Wallace

          It’s rather amazing. Some people simply will not accept data that contradicts their beliefs. One sort of expects that with religious beliefs but many of the oil/nuclear people are engineers or have science backgrounds.

          • Doug Cutler

            Or . . . they really do know what’s happening and are simply attempting to prop up stock prices for as long as possible.

          • neroden

            The nuclear nuts are simply religious fanatics worshipping at a plutonium altar.

            The back-to-the-land doomers just don’t want to believe that the problem will be solved. They want humanity to fail.

            The oil-industry peak oil doomers are a little different; they’ve just spent *so much of their life* in the oil industry that they cannot get their heads around the economics of solar. They just think that there *must* be some resource which is depleted when solar power is used. But there isn’t.

          • Yes, it really is concerning how illogical people are, even/esp. when presented with strong evidence contrary to their worldview.

    • jeffhre

      Generating station executive offices, 2030.
      Go home boys.
      Solar does this now.
      Clean out your desks and go home.

    • Ross

      The nice aspect is that there’s an inverse relationship between the growth of solar versus the time spent rebutting the deniers and the nuke fanboys.

      • hahaha, so true! 😀

        the past couple of years have been much more enjoyable than 2010–2012 😀

Back to Top ↑