The revealing numbers on solar employment in the USA.

Once again the green media are being transported into flights of ecstasy over the fact that the US solar industry now employs more people than the US oil, gas and coal industries. The data, however, show that the solar industry contributes virtually nothing to US energy supply, which is still filled dominantly by fossil fuels. Reviews also show that the problem of accurately estimating annual US solar generation has still not been solved.

First we will take a brief look at the US solar industry. How many companies are involved in it? It seems that no one has ever counted them, but as illustrated in Figure 1 they probably number in the thousands. (An interesting feature is the absence of solar companies in much of North and South Dakota. Could this have to do with the fact that these are the most productive areas of the Bakken Shale?):

Figure 1

Second, where are the existing solar installations? Rooftop arrays are present in varying amounts almost everywhere, but Figure 2 shows where the commercial arrays are (data again from SEIA.)

Figure 2

Figure 3 shows a map of US solar potential. Comparison with Figure 2 shows that except for Southern California and a few arrays in Arizona and New Mexico almost all of the commercial solar arrays are outside the areas of maximum solar potential – an example of subsidies in action.

Figure 3

Figure 4 now shows the employment statistics that have driven the greens into a tizzy. In 2015 there were 185,000 people employed in the coal industry, 190,000 employed in the oil and gas industry but 210,000 employed in solar, up from 170,000 in 2014 (data from Bloomberg):

Figure 4

And Figure 5 shows what 170,000 solar industry employees had achieved through the end of 2014 – a huge increase in US solar installed capacity and a factor-of-three decrease in costs since 2009:

Figure 5

But here’s the catch. Despite its job creation record the US solar industry still produces no energy worth speaking of. Before we can get into this question, however, we have some data issues to resolve:

The US Energy Information Administration (EIA, Table 10.1) gives the following US energy consumption data by source for 2015. Note that Mtoe = million tonnes oil equivalent:

  • Oil consumption: 884.3 million tonnes
  • Natural gas consumption: 708.0 Mtoe
  • Oil & Gas consumption combined: 1592.3 Mtoe
  • Coal consumption: 390.3 Mtoe
  • Total US consumption: 2438.2 Mtoe
  • Solar consumption:  ? Mtoe

Why is there a question mark where solar consumption should be? Because other sources give different numbers. Here are some examples for US 2014 solar generation, with details as to how I converted them into Mtoe:

EIA: 420 trillion btu . Crude oil 5.9 million btu/bbl = 71 million bbl = 71/7.33 = 9.7 Mtoe (the number missing from the list above).

EIA: 18.1 TWh (from Wikipedia) Using the BP 4.4 conversion factor = 4.1 Mtoe. (The derivation of the 4.4 factor is not given but BP uses it to convert all generation sources from TWh to Mtoe.)

National Renewable Energy Laboratory ( NREL, same source as above): 32.5 TWh = 7.4 Mtoe using the BP conversion factor.

BP 2015 Statistical Review = 18.5 TWh and 4.2 Mtoe (separately tabulated).

Obviously the problem of accurately measuring US solar generation has not yet been solved.

So which of the four estimates did I use? None of them. I made my own estimate based on installed capacity and average US solar PV capacity factors, which are listed by region in my earlier post on solar load factors. Figure 6 shows where different areas of the US plot in relation to other countries and regions for reference:

Figure 6

I estimated 2015 solar installed capacity by averaging the 2014 and 2015 end-year capacities given by Greentechmedia and got (25,634 + 18,348)/2 = 21,991MW (say 22,000MW). I have some confidence in this estimate because it includes both distributed and non-distributed capacity and because the Greentechmedia annual installed capacity estimates since 2000 are closely comparable to those given by BP. I then applied an annual capacity factor of 16.4%, the weighted average of some hundreds of years of operating data from 84 solar installations of all types and sizes in the US lower 48, again using data from the solar load factors post. The average is weighted to allow for the fact that about 75% of total US solar generation comes from California and the Desert Southwest.

With 22,000MW of capacity and a capacity factor of 16.4% we get 31.6TWh of solar generation in 2015, which according to BP represents 7.2Mtoe. This number is bracketed by the four 2014 estimates shown above, so I ran with it.

Having put the data problem behind us we can now calculate the energy, in terms of tonnes of oil or oil equivalent, that an employee in the solar industry generates in comparison with the energy generated by employees in the oil/gas and coal industries (the EIA data are used for oil, gas and coal):

  • Oil & gas (1592.3 Mtoe, 185,000 employees) 8,607 tonnes/employee
  • Coal (390.3 Mtoe, 190,000 employees) 2,054 tonnes/employee
  • Solar (7.2 Mtoe, 210,000 employees)  34 tonnes/employee

We see that the average US oil and gas industry employee produces 250 times as much annual energy as the average US solar industry employee and that the average US coal industry employee produces 60 times as much. The US solar industry still contributes less than 1% of total US energy demand, compared to 36% for oil, 29% for gas and 16% for coal, and gas and coal power are of course available on demand, which solar isn’t. The solar industry’s job creation record is impressive, but based on these results one has to wonder whether the jobs were worth creating.

And how much has the US paid to fill less than a percent of its energy supply with solar power? I can’t find a total expenditure number (maybe someone else has one) but it’s reported that taxpayers paid more than $150 billion over the last five years alone financing solar and other renewables projects, and the installation cost of the US’s plus 25,000MW of solar capacity (it continues to increase) would have been on the order of $100 billion. So we can assume that total solar expenditures to date are somewhere up in the low hundreds of billions range.

But why should any country in its right mind spend so much to add less than 1% to its annual energy production? The answer, of course, lies in the misguided efforts of the US federal and state governments, prompted by the green lobby, to stimulate the development of this “resource of the future”, which solar is still believed to be even by people who should know better (NREL sees 40% solar penetration as a realistic goal). Moreover, solar stimulus efforts have now reached the point where they can only be described as being out of control, as evidenced by the following quote:

According to the Government Accountability Office, federal government support for solar energy is massive, with over 345 different federal initiatives covering over 1,500 projects in 20 federal agencies–the Pentagon has 63 solar programs, the highest among the agencies, followed by the Interior Department, with 37 programs and the Energy Department (DOE) with 34 solar programs. States also have also heavily subsidized the solar industry by offering tax breaks and 538 rebate programs. Twenty states have personal tax credits related to solar products, 18 states have corporate tax credit and deduction programs, and 14 states and Puerto Rico offer taxpayer-funded grants to support solar electricity.

Clearly the US solar industry needs to be reined in. But there is hope. As noted in the recent death of rooftop solar post one of the underpinnings of US rooftop solar – net metering – is coming under attack and at the end of this year the federal solar tax credit is scheduled to expire. US taxpayers should make sure it does.

This entry was posted in Energy and tagged , , , . Bookmark the permalink.

117 Responses to The revealing numbers on solar employment in the USA.

  1. singletonengineer says:

    Very timely and, as usual, well explained.

    I have shared the same thoughts for a couple of years, during which I have been ignored, shouted down or been fed inconclusive junk statistics by the pro-solar crowd.

    Congratulations and thanks for publishing this.

    • Willem post says:

      I think this comparison is flawed.

      One should compare steady state versus steady state.

      Because oil, gas, coal production changes are small, employment is near stagnant, as the Bloomberg graph shows.

      That is not the case with solar.

      A very small percentage of total employment is involved with existing system O&M, replacements, upgrades, etc.

      A very large percentage is involved with adding capacity.

      The small percentage workers should be compared with the oil, gas, coal workers.

      • Yes, the comparison is indeed flawed from that standpoint, But I didn’t make it. The solar industry did.

      • oldfossil says:

        As any economist will tell you, jobs are a cost not a benefit. Read Chapter 1 of The Wealth of Nations with particular reference to the nail-maker. Wealth comes from productivity. Also, because the high cost of solar is borne by individuals, i.e. a kind of tax, solar jobs are effectively welfare recipients.

        • Willem Post says:

          old fossil,

          “Wealth comes from productivity”.

          Wealth accumulates due to production/business costs being less than sales prices, which means a profit.

          It is the accumulation of profits, wisely invested, that creates net worth, a.k.a. wealth, as a business, and as a nation.

          Implementing solar in Germany (which happens to have a lot of wealth to play games with), is an unwise/stupid use of wealth.

          Germany should use its wealth to invest in places with abundant sunshine.

          The US Great Plains states and west Texas have abundant wind and almost all of the US wind turbine capacity.

          Private investors are not stupid.

  2. Gaznotprom says:

    Excellent analysis Mr Andrews. This was alluded to in the comments recently, thank you for boiling it all down!
    In a nut shell, solar is largely a vanity, virtue-signalling, shrine for greenies too kneel-before.

  3. Beamspot says:

    Great work.

    Let me do some back of the envelope calculations.

    2445.4 Mtoe of total energy consumption in US, by 34 tonnes per employee, will mean 71923530 employees if all energy were solar, about 72 million.

    IIRC, US has a population in the range of 300 to 350 millions, where about 100 millions have a job.

    It’s strange (maybe this is only a question of time) that some green dreamers didn’t give this number from the top of their lungs: 72 more million of jobs!

    Of course, this will raise a little question: who’s gonna pay them? 100 millions of salaries have to pay for 72 millions more? Or the government has?

    Will the economy work under those conditions?

    If we do the ass-umption (I’ve read this elsewhere) that only 10% of those workers will be required for O&M once all the solar has been installed and that’s all, that still accounts for 7.2 million jobs, while currently coal and oil stands for about 0.375 millions, about 5% of the estimated 7.2 million jobs. (And what will happen meanwhile those solar sources are being deployed?)

    Will the economy still work under those new conditions?

    I guess not.

    Peak oil (and gas, and coal) is not a problem with energy.

    Peak oil is a problem of economic viability of energy.

    Venezuela is a damn good example: it has huge reserves of extra-heavy oil (asphalt), but they can’t earn any money on it, neither can they use it to generate their own energy (probably, because ERoEI for extra-heavy oil is well below 1:1).

    In this case, Gail explanations are quite right to the point.

    • colin says:

      I think if US oil output relied on the sweat of 185,000 workers and no one else then a $50 oil price would present no problems. Or indeed a $10 price. There’s a logistical tail unaccounted for here, not to mention all these people making wellhead, drill pipe, rigs, instrumentation etc. The list is bewilderingly long. Multiply 185,000 by 10 for a more realistic figure.
      Needless it is still a far more efficient use of human capital than solar.

  4. Serphin says:

    Have you got the same statistic (MTOE/employee) for Nuclear?

    • robertok06 says:

      Civil nuclear workforce is, according to this…

      … 57thousand… if anyone wants to do the math, calculate the equivalent TOE of 780 TWh generated by the ~100 nuclear US reactors then your wish will be satisfied….

      • According to MacKay 780TWh is about 65Mtoe so for 57K workforce that’s a bit over 1K toe/employee (which is under half the productivity of even coal, which seems wrong – can anybody sanity check this?)

        • According to BP US 2015 nuclear production would have been around 190 Mtoe, giving about 3,300 toe per employee. I suspect there’s a problem with Mackay’s conversion chart.

          • BP is using the primary energy equivalent:
            “The primary energy value of nuclear power generation has been derived by calculating the equivalent amount of fossil fuel required to generate the same volume of electricity in a thermal power station, assuming a conversion efficiency of 38%”

            But they do the same for solar (and other direct electricity generating renewables) too so there’s no problem with MacKay’s chart (which is a physicist’s-eye view of actual energy units) but it doesn’t give a valid comparison in this case. My bad.

            Even so I’m surprised that it’s so much lower than oil & gas and not hugely more than coal.

          • robertok06 says:

            @john Stumbles

            “Even so I’m surprised that it’s so much lower than oil & gas and not hugely more than coal.”

            The problem is that the comparison is not fair: nuclear generates electricity (could generate a lot of distributed heating, like done in few places, like in Switzerland and France), while oil and gas (especially oil) generate a lot of energy in forms other than electricity. Same for coal, which is used also largely by other forms of industrial production not related to electricity.
            The fair comparison could be obtained only after the fraction of energy generated by oil, gas and coal under the form of electricity is found, and then the number of workers of the respective power stations is added to the fraction of workers in the extraction field of the respective fossil fuel.
            Alternatively, one should take the total energy generated by nuclear and divide it only by the “upstream” part of the nuclear fuel cycle, extraction, refinement, transport, enrichment… which is probably 1/10 of the total.

            Example: Olympic Dam mine, Australia, workforce ~4000 (co-extraction of gold, copper, silver and maybe other minerals)… total yellow-cake production 6700 tons (5700 t U metal, total world consumption is… if Iremember correctly, 55000 tons U… so about 10% of the world’s uranium comes from there).



          • robertok06 says:

            Roger: the discrepancy comes from the “usual” 3x factor due to the primary-energy conversion, 33% of thermal efficiency)…
            On some statistics the renewables’ values are multiplied by 3.

      • robertok06 says:

        Correct, I checked the numbers myself, same result.
        Now one should find out how many of these “nuclear” jobs are really jobs, and not only paperwork to keep track with the latest NRC “safety” rules and changes… I’d guess at least 1/3 of them, if not more.

        Anyway, here is another example for another country:
        A typical 4×900 MWe french power station has a full-time staff of 1000 people, plus temporary workers during refuelling/maintenance, for a totla of 1700 …

        … so this is for 3.6 GWe, out of a total of 63 GWe, which makes a reasoble guess of 1700×63/3.6~30 thousand workers.
        The 63 GWe generate 415 TWh… or 35.7 Mtoe… for a ratio of 1200 toe/worker.

  5. gweberbv says:

    Comparing the numbers of employees is misleading (as long as PV is still ramping up capacity). Because the FF industry is in business since many decades, but solar just started a few years ago (with respect to installation numbers and also significant employee numbers).

    Most people working in solar are occupied with installing *new* PV plants. Which will generate energy *in addition* to the already existing fleet. Thus, even without increasing the numbers of employees, PV will double/triple/etc. the output. To the point, when the employees are mostly occupied with replacing existing, but aged PV installations (this would be more or less comparable to the modus operandi of the employees in the FF industry).

    Be aware that Germany now operates more than 40 GW of PV capacity with less than 40.000 employees in this sector (2014 data) and a lot of these people are just part-time working for PV.

    • Beamspot says:

      As some readers here have direct experience with managing and running solar PV plants, maybe some real data on full time employees for already installed and running PV plants are needed per MWp. This will help on the lower boundary. I guess it is still too big (as I pointed before, at 1/10 of actual numbers).

      But replacement of old plants, as well as manufacturing the new infrastructure should add on the upper boundary, and one part (PV and electronics manufacturing) is totally fogotten, as well as the other one (installation staff) may be too large for today, but too short just to keep replacing the old PV’s, so this also needs to be clarified.

      And, besides that, the ethernal unanswered question that is the weakest point of all this: balancing, but let’s forget about this now.

    • 40GW of PV capacity in Germany will generate about 42TWh/year, or about 10Mtoe. Divided by 40,000 gives about 240toe per year per employee, still far below US oil, gas and coal.

    • Gw

      I posted the info w.r.t. Germany. At peak installation rates (where the USA may be at now), there were 120,000 workers in the industry according to Fraunhofer. In 2013 this fell to 75,000 with relatively few installation rates. Take that on face value and assume that 2013 rates are enough to provide replacement, then you might, just might have to double the figure given tonne/employee. No difference really.

      However one of the reasons given for the decline in solar workers in Germany is also the collapse of solar manufacturing. Now I have not plotted the number of workers per installed capacity for every year but if that were significant, we would see a step change. Perhaps you could do the analysis?

      • gweberbv says:


        I just found a German study (from 2010) that predicted about 8000 people to be permanently employed for O&M of 52 GW of PV (this installation number is the long term goal of the government). But to end up with such a number one has to make an assumption about the typical sizes of PV plants. Clearly, operating a gigantic fleet of 10 MWp.

        • gweberbv says:

          Now correct (hopefully):


          I just found a German study (from 2010) that predicted about 8000 people to be permanently employed for O&M of 52 GW of PV (this installation number is the long term goal of the government). But to end up with such a number one has to make an assumption about the typical sizes of PV plants. Clearly, operating a gigantic fleet of 10 MWp.

          • gweberbv says:

            Clearly, operating a gigantic fleet of *smaller than* 10 kWp rooftop installations will be more labour intensive than having the same installed capacity concentrated in a few large-scale plants with *bigger than* 10 MWp.

  6. OpenSourceEnergy says:

    The numbers do not add up. If the numbers would be correct, 1kWh electical from oul would have to be at about 10% of the price as electricity from solar power, so the electricity from Oil and Gas power plants would have to be sold somewhere around 0,5-0,7ct/kWh. This is obviously not the case. So obviously there is some error in the input data – most likely in the way of counting employees in different areas. If only employees are counted to bring the oil to the mouth of the well, the correct number of employees in the solar industy up to this point is 0. The sun operates without employees. If all employees are counted till the oil has become a useable product (gas filld in in the cars tank, electricity in the grid, etc. ) the number involved somehow with Oil and gas is surely several times higher. That’S why I usually ignore such numbers, and wonder why some people are so much eager on them.
    The cost of energy matters, not the number of employees – because behind every cost somewhere in a other part of society there is a employee hidden, if it’s not directly within this industry.

    • If the numbers would be correct, 1kWh electical from oil would have to be at about 10% of the price as electricity from solar power,

      I agree. 🙂

    • robertok06 says:


      “If the numbers would be correct, 1kWh electical from oul would have to be at about 10% of the price as electricity from solar power, ”

      Spot on, pal!… average cost of German PV kWh?… 30 Eurocents… present value of 1 kWh on the electricity market?… 3.2 Eurocents….

      I leave to you the math.

    • Euan Mearns says:

      Rigged market where price and cost are totally disconnected.

  7. Hugh Sharman says:

    Good work as usual, Roger!

    However, this story crossed my desk this morning about ENEL in Chile., fairly respectable source!
    28% capacity factor is claimed at a total cost of about $1.6 million/MW turnkey which is impressive!

    In the inevtable Hallelujah follow-up story at, in the comment under mine there is a tweet claiming PV efficiency of 28.5%

    I remain in almost passionately in favour of PV + storage for Africa’s 500 million children of 14 and under who are highly unlikely to have any access to electricity in the foreseeable future unless this is PV…or they head towards Europe and other OECD countries!

    • Hugh. I worked in the Antofagasta area back in the 1980s and remember quite a lot of low clouds and sea fog. So I think the claimed capacity factor – which indeed works out to 28.5% at 160MW capacity and 400GWh annual generation – is wildly optimistic, particularly for a fixed PV array. Anyway, we’ll see what happens.

      On Africa. A solar PV array backed up with a few lead-acid batteries that gives a few hours of light each night is only a first step on the road to the eradication of poverty, which is one of the major goals of the UNFCCC and Kyoto. To make any real progress Africa will have to go through the “fossil fuel stage”, as India and China are already doing. It won’t get very far with intermittent wind and solar.

  8. Euan Mearns says:

    This post links nicely with my recent posts on ERoEI and brings into focus the need to understand where system boundaries are drawn. Who is counted as an oil and gas worker? And who is counted as a solar worker? This relevant to OpenSourceEnergy’s point. In oil and gas I guess we are talking about workers up to the well head. Building and operating power stations and pipelines and a power grid all adds economic costs, energy costs and jobs to make the electricity.

    A broader boundary for oil and gas would probably reveal a lot more jobs created in that industry. For example coal mining jobs, steel jobs, drilling rig and pipeline manufacturing jobs etc. The USA will have the full oil and gas supply chain inside its borders.

    For solar, we have to assume these are jobs in retail (?), installation, operation and maintenance. And we have to assume further that the main employment for manufacturing panels is located in China?

    Energy scarcity creates employment. We’ve seen it here in Aberdeen 2005 to 2014 when there was a jobs explosion that managed eventually to halt the long-term decline of the North Sea production. We’ve seen it in the tar sands and shale industries. These are low grade deposits that require a lot of labour and energy to exploit. But when scarcity turned to abundance the jobs evaporated.

    Of course jobs creation in the energy industries is a good thing. But politicians and Greens need to be wary of promoting this virtue too far since too many jobs will lead to too expensive energy. Boasting about loads of jobs in an industry that produces little is surely a sign of delusion.

    • Hugh Sharman says:

      No offence, Euan, but North Sea depletion, give or take a blip during some years, has continued all the way from 2000 through 2015, despite the (welcome) explosion in jobs in Aberdeen in recent years!

        • Hugh Sharman says:

          Thanks Euan!


          Yes, you are right!

          But I stand only slightly corrected! I am not about to clash with a giant of charting like you but it seems (just eyeing the chart) that Stavanger contributed rather more than Aberdeen to this miniscule and (clearly?) very temporary upturn.

          It would be interesting to learn how, as ERoI in the oil extraction (not “production”) industry declines, oil industry employment per barrel extracted has fared.

          However, I am NOT lobbying hard for you to use your scarce and hard-pressed resources to answer that!


          • Euan Mearns says:

            to halt the long-term decline of the North Sea production.

            I said production had stopped falling which is different to it rising.

            It would be interesting to learn how, as ERoI in the oil extraction (not “production”) industry declines, oil industry employment per barrel extracted has fared.

            International rig count is one good measure of employment:

    • Willem Post says:


      Figure 4 now shows the employment statistics that have driven the greens into a tizzy. In 2015 there were 185,000 people employed in the coal industry, 190,000 employed in the oil and gas industry but 210,000 employed in solar, up from 170,000 in 2014 (data from Bloomberg):

      Should be:

      Oil, gas 185,000, and Coal 190,000, to be consistent with Bloomberg

  9. Doug M. says:

    No offense, but this is a silly response to a silly claim. The US is still building its PV solar capacity, so of course the job numbers are inflated. It takes a lot more labor input to build a PV solar plant than it does to actually run it once it’s built.

    By way of analogy, it took about 5,200 people to build Hoover Dam, but only about 250 people work there today. If you did an output-per-man-hour estimate based on the numbers employed during the construction phase? You’d be off by a factor of 20.

    Doug M.

    • robertok06 says:

      Correct analysis… so this multitude of jobs created by the green industry is only going to last a short time… unless the Ponzi scheme of “incentivizing” ad vitam aeternam the intermittent renewables goes on forever… which is obviously economically impossible.

      • Doug Muir says:

        Yes, just as the “Ponzi scheme” of Hoover Dam jobs only lasted a few years.


        Doug M.

        • robertok06 says:

          Wrong example, Doug!… the Hoover dam is a low-cost-kWh baseload facility with a lifetime of >100years, while the Ponzi scheme source PV is a highly intermittent and seasonal, high-cost-kWh with a lifetime of 20-30 years max.

          Sheesh! 🙂

    • Euan Mearns says:

      Doug, if you read my comments here and posts else where you will realise that you are on the correct logical trajectory. But it’s the RE industries and politicians that need to appreciate this, not us.

  10. Doug M. says:

    Also, I’m afraid you might have misread Figure 2. You say “Figure 2 shows where the commercial arrays are”. However, clicking through, those colored circles are not commercial arrays — they’re retail outlets, distribution centers, corporate headquarters and manufacturing facitilities. (Look at the legend in the lower right hand corner.)

    In the next paragraph, you shake your head about subsidies. But since you’ve got Figure 2 wrong, then that’s wrong too. It’s not exactly surprising if the major manufacturing and distribution facilities are concentrated in the country’s industrial and commercial areas. Did you really think we were building massive commercial arrays in New England?

    FYI: we’re not. There are basically no large (>150 MW) commercial solar installations in the Northeast or the upper Midwest. They’re almost all in Arizona, California and Nevada — in the red areas of Figure 3, exactly where you’d expect. All ten of the country’s ten largest solar installations, both PV and thermal, are located in those three states.

    Doug M.

    • thinstoomuch says:

      I can agree to a point. Problem is why are those suppliers in those places. Because that iswhere the rooftop istallations are happening. Which is where a lot of those jobs are occuring. Be aware North Carolina is talking about adding 1.5 MW of capacity this year.

      Posting from my phone but the EIA monthly report lists where capacity is and where it is adding. Table 6.2.b. ithink.

      Also second note US population to employment is aronud 58 percent I thonk. So we are talking an employment number of around 160 million or so.
      Third note the subsidies got extended to 2022. Which is a bad thing. No worries they will extend them again then. The poloticians always do.


      • Doug M. says:

        1.5 MW? That’s adorable.

        The biggest PV solar installation in the US is currently Solar Star in California, which nameplates at 579 MW. By 2017 it should be surpassed by Westlands, also in California, which will nameplate at around 2.5 GW.

        Doug M.

        • Thinkstoomuch says:

          Glad to see you have such a tolerance for typo’s.

          How much PV is installed in the US in Utility Scale units?

          How much is planned for the The 12 months between Jan 2016 to Dec 2017?

          I did remember the number wrong. It was only 762.8 Net summer Capacity in NC alone. I did the work for you as I did give you the wrong table as well. It is monthly report Table 6.5.

          Of course the 222.5 MW being installed by FPL doesn’t count either.

          Heck the sites you give don’t even equal the estimated distributed total MW for the US RIGHT NOW.

          Stop looking at the headlines and see the rest of the picture.

          Which actually probably outnumber the number of installers that the SEIA is talking about as all you quoted were power numbers not the people installing it. Took what a 1,500 to install Solana which is a lot more manpower intensive being CSP.

          So I agree with Roger’s follow up comment. More people are working out of those retailers and contractors in that figure than work the big projects. Very inefficiently but that is the way the US Government set up the subsidies.


    • You’re quite correct. I did misread Figure 2,or to put it more accurately my failing eyesight prevented me from reading the microscopic legend. And being unable to read the legend I naturally assumed that the “commercial systems” prominently displayed in the title were what the map showed. But where you have distribution centers and retail outlets you can be sure that solar systems aren’t very far away.

      • Doug Muir says:

        1) Actually, no — the solar systems /are/ quite far away in many cases. Manufacturing and distribution are concentrated in the Northeast, commercial installation is overwhelmingly in the Southwest.

        Seem strange? Well, by way of comparison, about half of the steel used in the US is made in China. New Jersey – Arizona may seem far, but it’s not much compared to Shanghai – Arizona.

        2) I have to say, it’s kind of amusing that you looked at that map and instantly assumed we really, truly had more commercial solar installations in Maine and Vermont than in Arizona. Because subsidies.

        Doug M.

  11. clivebest says:

    I just heard on Radio4 ‘Today’ the following statistic quoted.

    “In 2015 the world invested £200 billion in renewable energy resulting in 147 GW of new ‘capacity’ – or the equivalent of all the power needs of Africa. ”

    However if we now apply an average load factor of say 20% we get an ‘average’ continuous power yield of 30GW. Therefore the cost of renewable energy is about £7 billion/GW. (Lets assume we can magically link up all this renewable power into some global grid to iron out variability)

    Hinkley C construction cost is £18 billion for 3GW which is about the same as the equivalent amount of (global) renewable energy (£21 billion for 3GW)

    So is there a catch ?

    The lifetime of Hinkley Point is 60 years, whereas the lifetime of wind/solar is 20 years. So the conclusion is that it is still 3 times cheaper to build Hinkley C rather than cover ~1500 km^2 of the UK with wind farms.

    • Beamspot says:

      Not to mention the elephant in the room: balancing production and consumption.

      • clivebest says:

        Yes this the show stopper for solar energy anywhere north of 50 Lat. Solar has exactly the inverse production cycle to that required by energy demand. i.e. Little energy in winter and none at all after 4pm when annual UK peak energy demand reaches > 50GW.

        Pursuing solar in UK is barking mad.

        • Euan Mearns says:

          Pursuing solar in UK is barking mad.

          I agree. And so did David MacKay. But its twice as mad in Scotland as it is in England. So how do you describe the stance of Scottish politicians who support this and Scottish Universities that have PV deployed on every available surface?

          It’s the stance of the universities that pisses me off the most. They have become an affront to scientific reason in pursuit of squalid money grabbing policies brought about by national bankruptcy brought about to large extent by the failure of academic reason.

          • PhilH says:

            Is what they’re doing effectively saying “We want to do something, and this is the only thing we can do, given that we can’t have a micro-nuclear power station on campus, and the powers that be aren’t doing what we want for us”?

            Actually, the other thing they could be doing is implementing a program of energy efficiency improvements to their building stock. Each UK campus I visit, as well as a few examples of modern, efficient buildings, has several old, horrendously energy-hungry looking large buildings that are just crying out for a makeover.

        • Beamspot says:

          You are absolutely right in this.

          I wonder why there are many important questions unanswered. The good point in this blog is that at leas one of the most important is taken into account: when?

          Also the Where? question is also in the mind, like in your reply.

          But there are more questions: What? and How?

          After thinking a lot about that, I find that there is a question that ties together all of them in the same issue, and that I almost never see questioned:

          Why electricity?

          Our energy needs are quite diverse, being heat the most demanded one, and with ‘electric only’ down to less than 11% of the real needs (mechanical is different, by ‘electric only’ I refer to light, IT and communications, probably control only).

          Trying to respond to that question rises many point that, IMHO, has to be adressed, but they are never questioned.

          Luckiliy, there is a secondary point that is easy to understand and that makes many people realize that maybe we have to think in other different ways.

          I miss a good debate and informed discussion about this topic, but perhaps that is a flavour of my skewness…

  12. Jim Brough says:

    For many years the Greens shouted “Solar, not Nuclear. How does nuclear compare with solar?

  13. Regarding the installation jobs v mature jobs critique

    The series “Development of renewable energy sources in Germany” has some interesting employment data to 2014 (not included in 2015 issue).

    They put peak solar jobs around 2010-2011 at approx 120,000. These were the peak years for solar installation. For 2013, this had dropped back to 75,000. The investments and capacity added in 2013 compared to 2011 are well less than half.

    So I think there is some merit to the installation versus mature jobs critique. Maybe double or triple the solar tonne/employee figure to be safe (ignoring capacity expansions in coal, oil or gas etc).

  14. Roger, you seem to be assuming that US oil & gas workers produce all the oil & gas consumed in the US. It’s true for gas but about half the crude oil consumed is imported.

  15. jos hendriks says:

    Thanks for this information. But the situation is more complicated. For instance you have to account for im- and exports. Also the differences in job types in oil/gaz/coal industry and solar industry is a problem. So I try to work around them. From figure 2 you can deduce that about 10% of installed solar was installed in 2015, From the figures in the article, that counts for 0,72 Mtoe.
    Suppose you want finally 30% of the at this moment used energy, by solar. According to the figures, again in this article, you need 2438/0.7 x0.3=1044 times the amount of installation in 2015. So install them in the next 20 years. After that period all starts over again. To do this you need about 50×200000=10000000 ( 10 million) workers. You need about 7% of the labor force of the USA.

  16. Grant says:

    Clearly in a blog called “Energy Matters” one should expect to be discussing energy and its sources.

    However, some energy providers also support secondary products through the extraction or mining of materials.

    Others don’t. In fact they tend to consume some of the traditional side products and one wonders how such consumption could be substituted if the formerly dominant energy sources were eliminated from the mix or became so rare and expensive that the costs of peripheral products, for example plastics or steel, became prohibitive for general use.

  17. disdaniel says:

    It is a little like groundhog day around here.

    Solar has completely caught up to fossil fuels with respect to price over the past decade. Remarkable, but you would never know that by reading the posts here. Keep telling yourself it is too puny, too expensive, too silly looking to possibly work–I’m sure that is what the buggy whip manufactures thought of those loud, smelly, dangerous automobiles.

    Over the past decade plus, Wind has shown how states and countries can integrate rather high amounts (between 25%-50% of supply in spots) of low cost intermittent capacity onto the grid without undo difficulty. Actually it might make for a more robust grid, but folks here continue to talk as if the world is one sparrow’s fart away from grid collapse because of renewables.

    Greater energy efficiency keeps showing how society can grow/expand using only about as much energy as the prior year. Yet people here dismiss the benefits of efficiency, as if it were some parlor trick not to be taken seriously. (not something proven to work literally decade after decade for half a century in sprawling, populous California)

    Energy storage costs have fallen ~75% in 7-8 years. We are just a few global storage capacity doublings away from putting the utility owned power plant business model and the ICE (internal combustion engine) completely out of business. Yet again people here scoff that since today (this instant) there is no way to store enough energy to power every single thing one could imagine wanting to power–completely and all at once–that really storage can’t do anything.

    There is an enormous amount of work left to be done in all these areas, but to pretend the EROI of solar is less than one, and that building out solar will require 72 million workers is simply goofy.

    It would make for a more credible site if you bothered to acknowledge the real advances of wind and solar every now and again.

    • clivebest says:

      I would love to believe you but unfortunately the facts don’t support you, There is a very new report – Renewables 2016 Global Status Report which is written by the industry itself.

      The world invested $160 billion in new solar energy in 2015. This resulted in an extra 50GW of capacity. So the net cost is just over $ 3 billion per extra GW without energy storage. Load factors are 15% at best, but lets be optimistic and assume 20%.

      Power delivery costs are then $15 billion per GW power. This makes nuclear seem very cheap.

    • If we could find anything good to say about power sources which

      1. After years of effort and the expenditure of over two trillion dollars still supply less than 2% of the world’s energy..
      2. Work only when the wind blows and the sun shines
      3. Because of rigged markets are displacing the dispatchable generation that keeps the lights on.

      We would.

    • Doug Muir says:

      disdaniel, yeah. I find this site a weird mixture of good analysis and cane-shaking crankery. I’ve learned stuff here, no lie — but, yup, many posters and commenters seem frozen around 1995 or so. Start talking about the crashing costs of installed PV solar or technological prospects for large-scale storage and well, lies, Green hippy LIES, subsidies subsidies subsidies blind politically correct politicians here come the blackouts! why won’t they see that nuclear is the truth and way? Greens!

      Again, there’s stuff to be learned here — Euan’s series on the island is good, and if you squint hard there’s useful discussion of potential policy responses. You just have to wade through stuff.

      Doug M.

      • Greg Kaan says:

        Doug, you are doing a good job of finding some issues with Roger’s interpretation of installation distribution. Plus your point about the number of installers being transitional is defensible, although the relatively short lifespan of PV implies the cycle that jos hendriks described.

        But what about the energy consumption – ie solar generation – figure? Do you dispute that solar generation is extraordinarily ineffective at actually providing power, given the efforts and resources used for deployment? If so, what are your figures?

        Also, you talk of “technological prospects for large-scale storage”. Fine – these are desperately needed, especially to fulfill disdaniel’s vision. Please provide some information of what these prospects may be. Roger posted an article about this very subject and there were no prospects that presented themselves. Discussion of these is probably almost as pressing as energy sources, especially if wind and solar are to play a significant role in industrialized nations.

        So feel free to speak up but this site does focus on the numbers as in the end, wishes mean nothing without the numbers to back them up.
        I have made errors in presenting quantitative information and was called on it – I was wrong and I accepted that but if you have positions on the subjects being discussed, you need to present the figures that they are based on (and be prepared to be critiqued).

      • Roberto says:

        ‘Start talking about the crashing costs of installed PV solar’

        Yes, let’s start talking about it… even at zero Euro/dollar/whatever per Wp the intermittency and non storageability of PV makes it’s cost out of bound of any modern industrialized country.
        Proof is that EVERYWHERE ‘incentives’ have been reduced or removed, new installation rates have halted our dropped… Spain, Italy, Germany, you name it.
        Whoever likes the freaking panels should do only one thing… live on them , get disconnected from the grid, and show the world how it can be done.
        Question: are you one of them? I bet not.
        If not… why are you not one of them, if it is so good?
        Please answer.
        Nice try, though, a bit more articulate than most of those I’ve read so far.

    • Greg Kaan says:

      We are just a few global storage capacity doublings away from putting the utility owned power plant business model and the ICE (internal combustion engine) completely out of business.

      A “few doublings” equates to 8 to 16 times, maybe 32.
      Can you please provide the figures for current utility and EV storage capacities so we can compare these numbers against current global oil and coal usage? I’m quite happy to ignore nuclear generation for the purposes of this comparison.

      Any claim of exponential growth requires at least some basic investigation.

      • disdaniel says:

        What does current coal and oil usage have to do with it? Especially since neither industry is exactly firing on all cylinders given current economics? They are going to have big financing problems going forward. In another 15-20 years they will have a very hard time competing with fully paid off wind and solar. Do you think markets will wait until year 14 or 19 to react?

        As energy storage scales up 10x in a few years, costs will fall 20-30%. Repeat that a few times and you quickly see that ability to store power will not be the limit for renewables looking out even 10-15 years. I include non-Li and non-chemical storage (i.e. gravity, thermal and pressure based storage) in my definition of energy storage. Obviously Li is very attractive for the EV market.

        • As energy storage scales up 10x in a few years, costs will fall 20-30%. Repeat that a few times and you quickly see that ability to store power will not be the limit for renewables looking out even 10-15 years.

          Energy Matters has published a number of posts on energy storage which are based on numbers rather than green pipe-dreams, including:

          Read them and you might learn something.

          • Ampere says:

            which is why anyone in science would solve this problem by extended grids. If I add up all grid extensions under construction I know you’ll have in 10 years a grid connection from Aberdeen to Kapstadt and from Aberdeen to Hanoi. How many hours by day there is no sun in this area, how likely is it that there is now wind, and how much hydropower storage is in this area?
            The grid in 10 years will not be able to handle all wishes you could have to it, but it will not stop expanding then.

          • Euan Mearns says:

            which is why anyone in science would solve this problem by extended grids

            This I’m afraid is bollocks. The best solution for me by far here in Aberdeen is to have a 1.2 GW nuclear reactor just down the road at Torness. With backup from the Gas fired power station at Peterhead just up the road. That is proper distributed generation. And I think the whole system is connected by 400kV cables.

            For a Scotland – Vietnam interconnector to have any consequential value it would need to be of the order 100 GW or even more. And we’d have to have the same Scotland – Morocco, Scotland – Russia and so on. Now the grid already accounts for about 50% of the cost of our electricity and you are proposing to increase that by a factor of over 100%. This is just Green pipe dreaming.

          • Roger Andrews says:

            The Aberdeen – Hanoi interconnector might help smooth out wind a bit, although I doubt it, but if you want to smooth out seasonal solar variations it won’t do you much good. For that you will need the Melbourne-Madrid interconnector. Or maybe Christchurch-Copenhagen.

          • Euan Mearns says:

            The only question that remains is do you go over the surface or call up Arne Saknussemm and take a shortcut through the core?

          • Ampere says:

            Euan @ Roger, start calculating and not producing nuclear pipedreams to avoid green pipedreams.
            A gid is expensive, but it supports _everybody_ in it’s reach. so per head the costs are low, no matter how big the grid is.
            Moving 100GW is much? Not really.Maybe compared to the scary grid interconnector Scotland – England this looks much, for the chinese east-west interconnector this is daily business.
            The North-South interconnector in germany will have>40GW in 2050, on an width of 500km – the corridor between Aberdeen and south east Asia is about 10.000km wide in the middle – to transport 100gW here you need just a 3,2 GW 400kV corridor every 300km, or 1 6,5GW 800kV AC line every 600km, or a 10GW 1100kV HVDC Line every 1000km.
            So compared to the existing density of the grid in Europe hat is required for this task is nothing. Ignoring basic electric mathematics in such things does not get you any further.
            And do you need to balance seasonality of a few inhabitants in northern Europe exactly W by W on the southern hemisphere? How many people live north of the 50° and how many live south of it? What happens if in a grid with 20.000TWh yearly consumption (2050), a seasonality of northern European countries solar power production of 500TWh remains? A blackout? Not really. Some fractions of a cent higher wholesale costs in winter in northern hemisphere than in Summer? – more likely. This would make PV in the Kalahary a bit more economical than in the Sahara.

            And never forget – The bottleneck in the power Transfer between Hanoi and Aberdeen today is between Calais and Dover, and the next bottleneck is at the Scottish-English border. All other parts of the grid are already much stronger. But there is no real cause to supply Scotland from Hanoi, it is just a example point on the map, easy to find. before someone starts to search for a configuration which does not work- remember it needs just one working configuration to prof that it works, and no matter how many non working configurations you find, this does not proof that it can not work.
            Sorry, here I have to rant a lot about your non scientific contributions.

            If you want to do this discussion at least on a back of a envelope calculation, then we could calculate how much a Power line +/-1100kV DC or +/- 1500kV DC With 6 Systems 10GW each on one Mast would cost between Hanoi and Aberdeen, with 10% of transmission losses on the cable. But then it would also be necessary to divide this costs by the number of people living in the stats threw which this power line passes, and which profit from the balancing effects of this powerline.
            And then this could be compared to the per-head-costs of Hinkley Point, Flamaville, Vogtle, etc. to get a comparison to sort these costs, in although a grid and a power production is naturally not the same.
            (And no worries, dimensioning of powerlines is my profession. If your knowledge about this is too small, I can help you out of this)

          • Euan Mearns says:

            And never forget – The bottleneck in the power Transfer between Hanoi and Aberdeen today is between Calais and Dover, and the next bottleneck is at the Scottish-English border.

            Well we have 3.5 GW of interconnection between Scotland and England to service 5 million people, 0.7 GW / million people. A direct result of Green crap. We have gone from being electricity independent to a dependency on imports which in my opinion is a bollocks policy.

            Europe has a population of 742 million and so according to you we already have at least 519 GW of interconnection between Europe and the East. Pray, tell me where all this grid is. A link to a map would be nice to see.

          • robertok06 says:


            Before writing such silly statements you should study a bit the consequences of the physical law that has your same name.

          • Ampere says:

            Euan, plaese reference to the text where I sayed that the grid has anywhere a capacity of 0,7 GW per Million Person. This is a making of yours.
            But if you do not want a discussion on scientific level, go on as you do, but without me.

        • robertok06 says:

          “In another 15-20 years they will have a very hard time competing with fully paid off wind and solar.”

          Most turbines won’t make it that far… 20 years… there’s already a large “revamping” program in Germany, to substitute “older” turbines installed less than 10 years ago.
          As far as PV is concerned, 20 years from now it will suffer from even bigger storage problems, not to mention the costs.

    • Beamspot says:

      Electrochemical storage doubling is pure wishful thinking.

      Lithium batteries hit the diminishing returns a long ago. From 2014, almost all price drops had been due raw materials costs dropping to harming levels for miners and such, not driven by any improvement.

      Doubling capacity for Li storage is phisically impossible (do the stochiometry by your self). And prices will stop falling, if not rising soon. Lithium carbonate battery grade skyrocketed from rought 6600$/T to 14$/T in january, and probably beyond by now.

      Although Lithium has a limited impact in battery cell cost, it remains to see what happen with LiPF6 EC:EMC battery grade (40 € cents per KWh returned just in battery costs, not to mention the cost of the 1.2KWh you have to put inside before, neither the costly electronics required to do so, and that usually lasts between 5 and 10 years (mostly in the 4 – 7 years range).

      And respect to BEV’s, there are no alternatives. Li – air had been proved to be not worth, and was the last hope standing in the field.

      Arrhenius is never considered, but is the main problem behind using batteries in warm and hot places, that just are the sunny places like Spain, coincidentally.

      • Beamspot says:

        Ah, and I forget: economy of scale doesn’t favour battery technology, and more specifically battery pack manufacturing for ‘large scale’ energy storage (like the case of BEV’s) as in other cases.

        Battery manufacturing scales very painfully.

        BEV’s are much cheaper at lower production volumes than ICE cars, but very expensive when volumes begin to rise beyond 100K units/year. It is also a simple question of math and knowledge behind this.

        Even Ford, together with Edison, find out that in the 20’s, and I wonder if those two were aware of the production lines and economy of scale issues.

      • Hugh Sharman says:

        @Beamspot, I agree, but ONLY as regards lithium. See any recent analysis of the industry by the Great John Petersen;

        Under the radar, there are lots of budding and almost sprouting technologies that will displace most large-scale use of lithiums as RE continues to succeed, whether by hook or crook!

        • RDG says:

          Great John Peterson: “Since I’m not a chemist…”

          Well, thats not exactly encouraging, is it.

          Here is a post from Reddit on why Lithium will NOT be displaced:

          “The periodic table itself holds the answer as to why this is unlikely.

          The weight of the material is overwhelmly determined by the number of nucleons. The number of electrons is going to be determined by the number of protons. As you go down the table, the more electrons and the more nucleons and nucleons grow faster overall than protons and electrons. So for each electron, you have one proton of weight and one or more neutrons of weight.

          The real kicker though is electron pairing. As soon as you jump down a step on the table, you have paired a bunch of electrons together that aren’t going to want to separate from each other. Basically, you created a bunch of extra weight that can’t do anything. So aren’t going to find what you are looking for after more than a couple of rows down the table.

          You might say, so what if there are fewer electrons per lb, what about the strength per electron? That does count too but going down the table again makes things worse. The most energetic electrons are closest to the nucleus. Go down the table and the strong electrons are paired up and won’t react in a battery. So your strongest electrons available for battery reactions is going to also be found in the first few rows.

          So now you are looking at hydrogen, a gas at all but exotic conditions, helium which is noble and not a candidate, and then lithium. Keep going and you are going to go backwards very quickly.

          The best option in energy density is found in optimizing everything other than lithium. Removing the heavy cobalt for Mn, Al, Ni, or others. Using polymers and complex organics to hold lithium in optimal configurations. Lighter anode materials. Or using air as one electrode. But lithium is still going to be the best electron holding material and I highly doubt we will see order of magnitude increases in energy density with batteries as we know them.”

          • Hugh Sharman says:

            @ RDG, energy density is only relevant for cars and handheld devices. Lithium is probably an on-going winner there but it won’t be cheap!

            For grid-scale applications, it’s cost, cost and cost!

            You obviously did not read John’s articles, or if you did, you did not understand them!

  18. Hugh Sharman says:

    Thanks Roger,

    All the western Aid Agencies working in Nigeria (US AID, EU, DIFID, World Bank, etc, with the notable exception of the weak and under-resourced African Development Bank) have a “Paris” agenda and so preventing any new, major investment into fossil-fired plant. At the same time, they are locked into support (theoretically $ one billion per year of “investment”) for the hopelessly under-performing Transmission Company of Nigeria.

    If fossil generation, India/China style is ever to take off, the whole mind-set I discovered will need to be replaced!

    Nigerians are taking charge of their own solutions, as they have for almost the lifetime of the Federal Government. There are literally millions of petrol and diesel engines. It takes no special genius to see that if microgrids, incorporating diesel, PV and battery storage (not wind in Nigeria, there is none!) can out-compete pure diesel, then that is the way forward.

    They are almost there!

    • Greg Kaan says:

      Sorry Hugh. By that logic, we should all be buying home diesel generators, battery banks and PV arrays for our homes. It can be done (one federal politician down here has done it) but it is inordinately expensive with current technology unless a normal grid connection cannot be established (or is similarly expensive).

      The undeveloped nations would still be better served with traditional grids and large centralized generators. The Nigerian situation is a reflection of the national government’s priorities leaving it to communities or households to install local generation but I cannot see it as being ideal.

      • Hugh Sharman says:

        No Greg!

        If it were not for our previous correspondence, I would find your remark a bit crass!

        I am only reporting the perfectly dreadful situation in Nigeria, made worse by the straitjacket of Western donors and the only logical, never-give-up response of the Nigerian citizens who need to figure out ways of reducing their 100% dependence on diesels and diesel fuel, even to get any light at all and no chance whatever of escaping poverty.

        I understand that most of Sub-Saharan Africa has similar issues, even in the large cities. But I am NOT an Africa expert, so can only report my limited experience of trying to help 170 million Nigerians, 25% of Africa’s population, half of which are under 15!

        Another option they have, of course, is to join the river of migrants trying to get into Europe by all possible means!

        • Greg Kaan says:

          Sorry Hugh. In that context (which I failed to take in), I do agree with you – the Nigerians are doing the best they can and PV with storage helps extend their limited fuel supplies as does sensibly applied wind generation for remote islands. It is supremely ironic that oil producing nations like Nigeria and Venezuela cannot provide for the energy needs of its people.

          Ideally, the industrialized nations would step in and make the Nigerian government allocate its (considerable) resources properly so that the population gets the infrastructure they need. But realpolitik doesn’t seem to allow for that in Africa (unlike the Middle East),

          Again, I apologise for any offense you took from my comment – it was not intended,

      • OpenSourceEnergy says:

        In many places there are no railroads or rivers / channels which could supply a central power station with the needed fuel and heavy components.

        And you have difficulties to operate such a power station at places with little to no water, even if there is a railroad to supply fuel.
        And in the villages you have the problem that the grid operater has neither the finances nor the interest to build a power line anywhere near to your place.

        So the energy supply in such places is, for the years to come, limeted to what a light 4×4 truck can transport on a earth road.

        Which leaves Diesel generator and expensive Diesel (long bad road transporrts), PV, small wind generators which fit disassembeled on a usual truck, batterys as possible options.

        Unless someone constructs a tiny nuclear power station, coal power station, a CCGT-power station, which can be transported with a 4×4 pickup truck.

        A fridge +PV+ a litttle battery which keeps the fridge running long enough that it does not become too warm before sun goes up in the morning extends available food in African villages by approximately 30%. As simple as that.
        But this is another topic.

        • Greg Kaan says:

          Actually, my position is not really different from Hugh’s if you take into account my statement “unless a normal grid connection cannot be established (or is similarly expensive)”.

          My error was taking Hugh’s final statement “They are almost there!” as being the ultimate end point for Nigeria’s (and other undeveloped countries) supply of electricity to end users. Hugh was describing the situation in the context of the existing, politically created, infrastructure limits while mine was for the (idealistic?) case where resources were properly allocated for infrastructure development.

          • OpenSourceEnergy says:

            I have read your post which was published after my post was sent.
            But I think the other way round it becomes a shoe, as we say.
            There are no resources to be allocated, by whatever cause. But with part time electricity supply, the economy in the willages can grow by roughly factor 10, and this way make ressouurces for infrastructure available.
            But then you might get the same situation as with the telephone network. There the state of cable based telephone infrastructure was mostly skipped in africa.
            If all those villages get a part time supply with solar and wind, when you connect them to the grid to provide a more stable electricity supply those villages will
            – most likely supply a bit more electricity to the grid, which was so far unused, than they need at other times from the grid. So there is no base for traditional coal nuclear or gas fired baseload generation
            – the people from the villages will be very used to load management to keep a grid – be it local or later central, running.
            If the development follows this path, residual power will be needed in Africa later on, so e.g. in hydropower big dams with many turbines on a comparatively little river, maybe with the possibility to pump too. Or conventioal power with low capacity costs and high fuel costs, so OCG, Diesel generators and similar.
            As Euan in earlier posts has shown, for all countries +/- 30° around Aequator a mostly solar power supply with additions of other generations can work pretty well with diurnal storage / residual generation for unavoidable nighttime demand. This allows power supply for a few billion people, although it is no solution for Aberdeen, unless you conect aberdeen with a strong grid to the region around the aequator.

          • Hugh Sharman says:

            Thanks Greg!

            I quote “…idealistically… where resources were properly allocated for infrastructure development.” That sounds very hollow for anyone living in in Nigeria today. Its natural resources have been systematically pillaged during the last 50 + years by its “big” men who are among London’s and NY’s most enthusiastic “shoppers”.

            There an on-going insurgency in the Delta (Biafra) where most of its discovered oil and gas is extracted, dominated (highjacked?) by criminal (“big” men) elements! This insurgency has recently caused a sharp reduction in oil and gas extraction which is boosting global oil prices.

            Output of the country’s 10 GW of gas-fired power stations, for a country of 175 million fell to 1000 MW these last few days.

            It is against this background that I believe (imported!!) diesels of which there are millions + PV + storage is the only practical solution I see that can be implemented within a reasonable period of a year or two!

            It’s not necessarily a solution that works anywhere else….

          • Greg Kaan says:

            I have read your post which was published after my post was sent.

            The statement I made about grid connection cost vs off grid setup was in my original reply to Hugh at June 2, 2016 at 2:19 pm – your post was at 3:29pm. The comment number sequence also shows the order of the posts.

            As for building grids in undeveloped nations, terrain issues etc, have a look at this example.
            Bhutan is one of the poorest nations on earth but look at what they have managed to achieve.

            Of course there are environmental costs with their hydro generation but that is a separate issue.

            The problems in Africa are mainly due to the “big men” (strong men is the term the media used to use) who filled the void after the rapid colonial abandonment by European nations post WW2. That most of the European colonies were run in a totally exploitative manner set poor examples of governance for the local replacement leaders to carry on with (and simply continued the poor expectations for the general population).

        • Grant says:

          OpenSourceEnergy’s posts make some good points.

          There is no reason to pre-suppose that African countries will need or choose to follow the path of full industrialisation that the US and the “Western” economies, plus no China and perhaps India have adopted.

          The example of the telephone network going straight to mobile (and on the back of that the mobile payments and banking adoption) probably sets a precedent for approaches that will be taken during the area’s development.

          So is it logical to anticipate that power distribution development will occur in anything like the same sort of model that the industrialised countries have in place?

          I suspect not.

          The “developed” world seems to be insistent about the destruction of its current energy model. We seem to be talking about behaviour change as a major part of that movement. Realistically, if we are to rely in variable renewables resources, that will mean adapting to undertaking energy intensive activities only when energy is available.

          So, just as things were some generations back in time, the warmer times of the year with optimum temperature and light (away from the equator) were likely the most productive for both agricultural and “industrial” activity.

          With solar power as base load life based in industrialisation or semi-industrialisation would be likely adopt work patterns familiar to the ancients.

          Add in wind and there would be some potential for using lighting to extend active industrialisation into night time activities but not necessarily in a regular and predictable pattern.

          Whilst the “industrialised” countries have people with philosophical concepts that seem to require re-programming the population’s expectations for on-demand energy supply the developing areas don’t need to do that (at least not to the same extent).

          No doubt the great and the good want to ensure that the local population to not catch the “western habit” as the basis of their expectations.

          Skipping the responsibility of creating a centralised and co-ordinated energy infrastructure would remove the potential for the Western Habit to become culturally embedded. And it would probably avoid the potential issues of attempting to create a cohesive system in those parts where social cohesion may currently represent some deeper challenges anyway.

          I can’t currently imagine that any supranational organisation offering advice or finance would be contemplating a “western” stye infrastructure – unless some elements of it could be made to turn a profit by selling allegedly redundant “dirty” technology for redeployment as part of a “CO2 reduction strategy”.

          Much better to keep local expectations low and call it the “New Energy Model”. Then change the expectations of the currently developed world to match the “new energy model” as it develops. Or, perhaps the preferred option for some apparently influential thinkers, does not develop.

          • singletonengineer says:

            Are you for real, or just pulling my leg?

          • OpenSourceEnergy says:

            Well I can not follow the thoughts.

            For big parts of africa, the question is
            a) no electricity at all or
            b)electricity which is not allways available to fulfill all wishes.

            b) allowes a much faster development.

            But b also produces implications on later grid operation.
            A expensive baseload power station can not operate economically when there is already existing renewable generation in the market whichsupplies power at zero variable costs at tree quaters of the time or more, depending on the size of the grid. It only leaves room for residual power generation.

            NAturally for everybody it is better to have a stable grid, and also all spply becomes more stable when the grid becomes more powerful and stretches over bigger areas. The good thing: grids are expanded worldwide, allwing to transport power ever longer distances at lower losses. But in many countrries can’t wait decades till power comes to their place. And also a faster growing economy due to decentral produced electricity which is not continuously available, allows to get the resources to expand the grids faster.
            Simple basic economic equations.

  19. Roger Andrews says:

    For those commenters who seem to regard Energy Matters as a dyed-in-the-wool, stuck-in–the-mud anti-renewables site (I seem to remember reading the word “crank” in one comment) I would like to point out that we do occasionally make efforts to come up with workable plans for a low-carbon future. Here are a few examples:

    My rooftop solar system also cuts global CO2 emissions by about 1 ton/year, according to my calculations. A small but sincere personal contribution to the fight against climate change. 🙂

    • Nathanael says:

      You’re still behaving like you’re stuck in the mud. I’ve been doing investment-grade analyses, and solar is totally unstoppable.

      Look at Australia. At Australian installation prices and Australian sun output, the LCOE of *rooftop* solar is so good that nothing can beat it on price. And it’s unsubsidized. The utility companies are now trying to throw regulatory obstacles in the way of it to stop it.

      The financial situation makes nothing else even slightly competitive.

      US rooftop solar suffers from inflated profit margins and inflated marketing costs. Get rid of those and we can do as well as Australia.

      • Thinkstoomuch says:

        You seem to have forgotten that the price of electricity is 3 times as expensive in Australia than what I pay in South Florida. California IS trying to achieve your electric rates.

        So what is the price of a solar panel in Australia? Not subsidized of course. I know the feed-in tariff is 6 cents. Unless you got locked into the 44 cent rate that the less fortunate pay to you.


      • Grant says:

        Any imbalance that appears to be too advantageous to the “owner” of the benefit will eventually attract taxation of some sort.

        The exception being a subsidy that appears to offer some sort of political benefit in the giving.

        In the case of self contained “renewables” that offer a large personal costs benefit (or are thought to) the owners will probably need to ensure that they become fully self sufficient 24/7/365. Eventually the cost of them accessing off-site support through a grid connection may become very expensive when those who are unable to take advantage of the “free” electricity balk at carry the entire cost of the grid – and indeed the electricity share they need to buy, if available, from other forms of reliable generation that have become less and less economically successful as economy of scale declines.

        I note that people are beginning to discuss how some sort of occasional the grid connection could be mass funded and some sort of insurance policy has been mooted.

        When you get an insurance company involved you can be fairly certain that they see an opportunity to turn a nice profit over and above the costs incurred.

        It sounds like an expensive solution to me – and will almost certainly be driven by legal requirements. If the Insurance industry is involved you can pretty much rely on that – even if you can’t rely on the power.


        • Ampere says:

          It’s as simple as always in economy, but politics has a tendency to ignore it: charge the costs as they are,not higher, not lower. Charge the price to provide a grid connection with the price i really costs to keep the connection up, and charge the price for electricity as it is at the moment.
          There is no cause today to charge the same costs per kWh at all situations in the grid, weather the supply is high or low. Food prices in the store also go up and down with season.
          This will apply to those who also produce their own power, and to those who don’t.
          If it is the case that when those who produce their own power draw power from the grid the electricity is difficult to produce, they will pay their fair share. If it is just a rumor, they will pay accordingly less.
          But some people dream that those who produce power for their own needs also pay the power for those people who don’t .

          • Grant says:

            In this case it is not so much a case of “those people who don’t pridyce power ” as “those people who cannot produce power”. People who live in situations where direct production for their own local purpose is not possible. Apartments in cities for example.

            The market could eventually learn to overcome that limitation. However where the state is effectively mandating the use of a certain type of energy – electricity – as part of its social control strategy the market will inevitably be skewed.

            There is nothing especially unusual in the existence of a skewed market. However there may soon come a time when a ledge enough sector of society feels so far detached from the most affluent sections that persistent problems over a wide scale become normal.

            It does not take much for civilised human groups to become far from civil very quickly.

            Would the idea of supranational electrical grids survive widespread social upheaval across the occupiable continents?

      • Greg Kaan says:

        I’ve been doing investment-grade analyses, and solar is totally unstoppable.

        Could you please provide your analysis?
        It would be interesting to see the figures

  20. Grant says:

    I don’t see Reply buttons to the replies to my last posts so I’ll dump a response in here.

    The processes through which widespread power might be brought to Africa and other locations that are currently “energy lite” will need some significant shift in politics, internationally and locally (including what we might generally summarise as “tribal” positions) if it is to be successful.

    I do not see that as a short term process even if, for some reason that might not be based entirely on economics or eco-activism, there is a huge effort to engineer an internationally heavily subsidised effort to achieve such an end.

    For the rest of this century, or so it is predicted, the population of African is expected to increase by something like 3 billion people, maybe more. That pressure alone is unlikely to make the path the political, economic and social stability any smoother – unless something miraculous happens.

    In my opinion.

    Currently, looking forward, there seem to be two options that are likely to become dominant although one should never exclude the possibility of disruptive influences that may emerge and develop rapidly.

    Under the current “controlling” “Western Industrialised” regime the trend seems to towards Global Government – perhaps developed under the auspices of the United Nations and a few Supra-National banking and finance organisations.

    IF such a development continues and becomes mature then one might envisage that it would imply that all nations and therefore all peoples would be co-operating in the global interest and political, social and financial special interests would mostly be managed for global rather than local benefit.

    Quite how that would work yet retain any sort of status quo is difficult to imagine.

    However if we assume that such a Utopia could somehow be achieved that one might infer that the sharing of energy sources on some sort of worldwide energy web (assuming that the technology come into existence) would be quite easy to decree and deploy.

    The people who tend to be influential in such global matters also seem to be of the type that like wield influence in ecological matters as well. Currently they seem keen to reduce “consumption” of everything on a per capita basis. For energy that would mean re-educating people and introducing demand management – the start for that being smart meters and, perhaps, the Internet of Things.

    Of course that means having reliable electricity supplies available to manage the system end to end ….. but that’s another matter to discuss.

    At the more extreme end of the views associated with such Global Control philosophy we see those who would wish to constrain the Global Population and then see it greatly reduced. Whether such views might become influential is difficult to foresee.

    On the other hand there are, some might say, signs that the currently empowered “Western Industrial” grouping is “going soft”, as all empires eventually do, and might be usurped by other interests.

    So if, for the sake of the point, one felt that some form of, perhaps, “religiously associated” social and political force might gain significant power globally the ability to dictate global energy sharing across territories and local “tribal” interests might be very much less likely. Indeed it might not be of any interest at all to those with influence. Security of supply could be challenging.

    (Note that personally I suspect that are more than enough social psychopaths around the world to make and large scale global agreements possible let alone secure and cost effective in terms of return on investment and continuing servicing.)

    Another possibility might be that China, through size, influence and its investment activity worldwide, become the primary driving force for everything taking over from a declining “Western Industrial power base”. Or perhaps by being presented with that base?

    India might also be a player in that type of scenario.

    Ultimately “predictable” Energy through “renewable resources” (i.e a solar based global network although I’m not sure how the Pacific fits with that) will not be just about technology. It will be about the will to obtain agreements and security of supply that are guaranteed for the long term as well as ensuring that the fiscal viability is understood and accepted by all parties.

    For example, northern industrial Europe’s success for the past few centuries has been based on the availability of certain RAW materials and then, as things progressed, the a plentiful supply of generally good energy sources (coal) and food production availability.

    Once you eliminate coal the energy advantage is gone.

    If you then constrain oil (and some chemicals based on oil) there is little left to work with that can be guaranteed to be available when the “renewables” don’t work.

    Nuclear would be an option but not if it is deemed to be socio-politically unacceptable for some reason.

    So all Northern Europe would have to offer to the global energy creation industry would be intermittency. On balance it would a net importer of electricity from sunny countries in the mid latitudes and once in that position it would make no economic sense for Northern Europe to have any significant solar generation of its own, especially if the returns on investment could not compete withe the mid-latitude installations.

    In summary I would suggest that unless you can be certain that a global (or near global) agreement can be reached within a wide area of socio-politically and economically stable and secure co-operation (basically that would have to be Asia, Africa and Europe including Russia) I cannot envisage the grand renewable dream coming into being even if it is technically feasible and economically viable.

    But then so called miracles do sometimes happen – or so it is claimed.

  21. Thinkstoomuch says:

    Not that it really makes much difference. But total solar is available from the EIA site.,0,1&fuel=0047&geo=g&sec=g&freq=A&start=2001&end=2015&ctype=linechart&ltype=pin&rtype=s&maptype=0&rse=0&pin=

    It shows 26,473 thousand megawatthours for utility scale solar ouput in 2015. Of which 3,241 MWH were from solar thermal.

    Also shows 12,141 from distributed PV. That number is an estimate of an estimate so not worth much but …

    I do think your 16.7% number is probably low for capacity factor. The utility numbers are as hard as it gets as it comes from submitted EIA-860 forms (I think). For some reason unable to find the installed capacity right now. But utility scale PV is at 28.6% capacity factot preliminary for 2015. Final number for 2014 was 25.9%. Capacity factor for thermal was 22.7 and 19.8% respectively.


    Like I said doesn’t really mean much but in the interests of accuracy.

    Thank you for all you do to educate this fool,

    • Hugh Sharman says:

      @Thinkstoomuch, thank you!

      These utlily capacity factors, almost certainly based on the sunny SW of USA, are indeed, if true, solid evidence that PV technology (light to electricity efficiency) has also improved massively during the last 5 years.

      Time for a re-assessment of PV for Energy Matters by a real expert who can also be completely objective, methinks?

      • Thinkstoomuch says:

        Actually not so much that has been addressed here in the past. What it does point out that installing PV in Scotland or Germany is not so smart.

        Far better to install half as much in Morocco and spend the saved money on connectors. But as Germany is having trouble getting one installed in its own territory (from north to south) shows that having power here does not make it possible to get it there.

        Another thing to bear in mind is it seems like a lot of utility scale show up then disappear. Like the one at Denver International Airport. Airport one on the EIA plant level data produced power for one year, 2014 then no more.

        Be interesting to see the rest of this year and next as the US is supposed to go from ~12 GW Net Summer Capacity to ~21 GW. for utility scale production since Doug M has disappeared.

        By the way Net Summer Capacity is already down 15% or so from array peak capacity. Thus an installation Space Coast Solar PV has a net summer capacity of 10 MW but 11.5 MWp DC array.

        So the 28.9% becomes 24.565% when compared to the array size..

        Probably should have explained better in my original post but communication, I am terrible at that. 🙂

        Plus I was on my phone as the solar setup for my motorcycle campsite is taking a beating from cloud cover and rain in Mississippi.

        Have fun,

      • Hugh:

        In the introduction to this post I stated that “Reviews also show that the problem of accurately estimating annual US solar generation has still not been solved.”

        The EIA numbers are another example of this. For 2015 they give a solar capacity factor of 25.9%. The 2015 BP data, however, give a capacity factor of only 13.1% (18.5TWh generation, average installed capacity during the year 15.2GW).

        The reason I went to so much trouble to estimate solar capacity factors from individual solar arrays in my post was that it was impossible to estimate them from national statistics, which were all over the map depending on the source of data. I presented the following graph as proof:

        To this graph I have now added the 2015 BP and DOE capacity factors. BP has gone up from 6.9% to 13.1%. DOE (the numbers are actually from NREL) stays the same as it was in 2012.

        And who’s right? The BP estimate is probably low but the DOE estimate is implausibly high. I don’t trust either of them. That’s why I went with my own estimate based on actual operating data,

        • Thinkstoomuch says:

          Additional data I went through and some results follow.

          For 2015 the CA ISO shows 14,388,109 MWH from Utility Solar PV alone. Though to be fair CAISO is where most is probably installed 70% of the total is a large number if you include both distributive in that area plus both plant and distributed in other areas.

          Also I agree with the EIA Number being high. Reasoning is contained in the post.

          Understand this is my more or less illiterate, fat finger data manipulation.

          I went through most of the Utility Scale Solar Power Plants for CA using EIA Data to determine the Capacity Factors for currently active units. I am on the road so the internet connection is slow to non-existent. So I am just using the annual data tables (plus annual is the only way to figure CF, IMO) from the Electricity Data Browser Power Plant sets(which are derived from sales provided on the form EIA-860A). I am not all that smart or efficient so it gives something to do over coffee in the morning and on rainy days. 

          2010 had 20 plant outputs, 2011 had 34. 2012 had 66, 2013 has 129, 2014 had 191, and 2015 had 23. To emphasize these are current operating plants not ones that have been retired or inop during the period covered. I may go back and see if I can find numbers and outputs for retired and inop plants.

          First thing that surprised me was the amount of one axis tracking systems installed. Almost 40% of the plants in 2014. Live and learn.

          Capacity Factor vs Tracking Types

          2010 CF; fixed 19.45%, single axis 22.47%, dual axis 28.53%
          2011 CF; fixed 19.51, single axis 25.71, dual axis 25.33%
          2012 CF; fixed 20.96, single axis 24.59%, dual axis 27.44%
          2013 CF; fixed 21.31%, single axis 25.90%, dual axis 23.98% ???
          2014 CF fixed 21.19%, single axis 25.84%, dual axis 22.13% ???

          I did the numbers for 2015 but I don’t trust them as the finalized figures won’t be out until October and the 2015 EIA Capacity Factor number is straight out of that limited data set. 2014 has 191 data points (out of 202 power plants) and 2015 has 23 (out of 276 power plants).

          Also the Dual Axis CF numbers must be lunch meat somewhere. Possibly by Crafton Hills College Solar Farm (58170) not doing its maintenance as they are around 12% and it is a small data set and that one unit is 1.3 MW of 7 MW and 13.5 MW. Just a guess.

          Total Capacity in the available data sets.

          2010 45.7 MW (F=32,SA=11.4,DA=2.3),
          2011 68.1 MW (F=41.9, SA=20.5, DA=5.7),
          2012 237.4 MW (F=178.9, SA=51.5 ,DA=7),
          2013 725.3 MW (F=489.8, SA=228.5, DA=7.0),
          2014 2,409.7 MW (F=1,447.1, SA=949.1, DA=13.5)

          To me Nationwide it would seem that 20% seems reasonable. Especially if the single axis tracking is in affect nationwide.

          That huge gain in efficiency is not apparent from the numbers that every body is talking about. ~10% gain in output really isn’t all that much over 5 years, IMO.

          Probably am going to noodle around for other states as they should be much easier but won’t post unless you are interested. I also might use the Solar Edge data for distributive systems. Again please tell me to keep it to myself if you think that is appropriate.

          For what little it is worth,

          I had fun doing it. Yep I am that messed up. 😉

          • Roger Andrews says:

            T2M. Thanks for all your hard work.

            I short-circuited the process by going to the EIA 2014 annual report


            Where I found the following numbers:

            Total US 2014 PV+CSP installed nameplate capacity 10,478MW
            Net US 2014 PV+CSP generation 17,691MW

            Giving a US capacity factor of 19.3%, biased towards the sunny Desert Southwest.

          • Thinkstoomuch says:


            To me it is not work. After your comments I went and looked at some of the stuff I was posting and thinking and decided I needed to do my own manipulation. Sort things out in my own mind. Plus I am learning things.

            I am not really liking the way the EIA does the CF and I want to look in more detail. So when I have a connection I get the data when I don’t I play with it at the picnic table over coffee.

            After I do more might post it and might not though looking around the CF of small installations is getting more interesting which gives my easily amused self more things to contemplate. Which I consider a good thing. Whether it is worth anything or my thoughts are worth anything is relatively unimportant.

            Thank you and the others for giving me stuff to contemplate,

          • Thinkstoomuch says:


            Just an info post of my latest contemplation’s.

            EIA data (after crunching) that proves your capacity figure is very close. Good job, shows why you are the smart one and I am barely literate.

            The following is for PV only! Only for plants whose operational date was prior to 2014!

            For 2014 for 611 Solar Electricity plants throughout the US rated at 6,110.2 MW. Capacity Factor was 19.34%.

            Fixed axis consisted of 414 plants rated at 2,444.1 MW. Capacity Factor 18.17%.

            Single axis, 161 plants rated at 1,320 MW. Capacity Factor 24.23%.

            Dual axis, 14 plants rated at 53.6 MW. Capacity Factor 21.75%.

            Notice none of those numbers are as good as the “official” values in the EIA table.

            Lost some plants in the breakout on tracking due to EIA indicating more than one tracking type at an installation just used them for the overall.

            Selected capacity figures in 2014 was located in California(2,675.5 MW), Arizona(1,258.5 MW), New Mexico (212.1) and Nevada (471.3). (Total=4617.4 or 76% for 2014).

            Though March 2016 EIA values North Carolina has installed ~25% (1,555.9 MW) of California’s installed (5,923.9). Southeast has more or less ~25% of the Pacific and Mountain regions, which includes most of the Southwest US plus some.

            So more capacity is being installed, relatively, in other less productive(?) locations over the last 2 years plus. Which would seem to indicate more installers accomplishing less MWH output, To be in keeping with your article’s main point.

            I am still thinking on why the EIA CF data is so skewed and I have come up with a few theories but I am going to let them stew a while in my head.

            I am still surprised at the number of single axis tracking installations. Though the 6% additional CF is a big selling point. Is that because better siting, just the tracker or combination of both?

            CA is a big State, so is North Carolina if you travel through them both with varying weather conditions and terrain. Which makes the determination very difficult. I did not look at individual locations, yet. May not that is much more difficult.

            If you are interested in any particular data for a state or group would be fairly easy to generate now for the years from 2010 to 2015(partial data set) let me know.

            Thank you for many, many hours of entertainment,

  22. alturium says:

    *little late*

    If 70% of China’s energy is coal, and china supplies almost 60% of the world’s solar PV, then we can infer that a type of subsidy based on cheap energy => expensive energy.

  23. alturium says:

    If China’s energy mix is 70% coal and produces almost 60% of the world’s PV, then we can infer that we are subsidizing a cheap energy => expensive energy.

  24. Pingback: Recent Energy And Environmental News – June 20th 2016 | PA Pundits - International

Comments are closed.