A Free-Market Energy Blog

W. S. Jevons (1865) on Windpower (Memo to Obama, Part I)

By Robert Bradley Jr. -- January 28, 2009

The most important book ever written on energy economics was published in 1865 by William Stanley Jevons, The Coal Question (London: Macmillan and Company). This classic is out of print but available in its entirety on the Internet. It is well worth reading. The book marks the birth of an entire discipline, and Jevons’s remarkably sophisticated treatment of energy sustainability remains pertinent today. In a real sense, the Obama approach to energy was refuted by the insight of W. S. Jevons almost 150 years ago.

Jevons makes four points regarding windpower.

1) windpower is not new

“When in 1708 windmills were wanted to try and drain certain Scotch coal-mines; John Young, the millwright of Montrose, was found to be the only man in the country who could erect windmills” (p. 75).

2) windpower is intermittent and unsuitable for modern work

“The first great requisite of motive power is, that it shall be wholly at our command, to be exerted when, and where, and in what degree we desire. The wind, for instance, as a direct motive power, is wholly inapplicable to a system of machine labour, for during a calm season the whole business of the country would be thrown out of gear” (p. 122).

“Before the era of steam-engines, windmills were tried for draining mines, ‘but, though they were powerful machines, they were very irregular, so that in a long tract of calm weather the mines were drowned, and all the workmen thrown idle. From this cause, the contingent expenses of these machines were very great; besides, they were only applicable in open and elevated situations’” (p. 123).

Civilization … is the economy of power, and consists in withdrawing and using our small fraction of force in a happy mode and moment” (p. 122).

3) windpower is land constrained

“No possible concentration of windmills … would supply the force required in large factories or iron works. An ordinary windmill has the power of about thirty-four men, or at most seven horses. Many ordinary factories would therefore require ten windmills to drive them, and the great Dowlais Ironworks, employing a total engine power of 7,308 horses, would require no less than 1,000 large windmills!” (p. 123)

4) windpower for transportation did not work

“Richard Lovell Edgeworth spent forty years’ labour in trying to bring wind carriages into use. But no ingenuity could prevent [wind carriages] from being uncertain; and their rapidity with a strong breeze was such, that … ‘they seemed to fly, rather than roll along the ground.’ Such rapidity not under full control must be in the highest degree dangerous” (p. 126).

“A wind-wagon would undoubtedly be the cheapest kind of conveyance if it would always go the right way. Simon Stevin invented such a carriage, which carried twenty-eight persons, and is said to have gone seven leagues an hour” (p. 125).


  1. Ed Reid  

    Windpower and sunpower are “source of opportunity” power, usable when available, but requiring backup when not available. Hydropower is also partially “source of opportunity” power, more available when there has been ample rain and/or snow. The trick with hydro is accurately identifying the reliable fraction, as BPA has demonstrated.

    The “bridge” from “source of opportunity” power to reliable power is storage. Hydro involves inherent storage, though it is important to understand the fraction of the potential storage which is reliable.

    Wind and solar both have capacity factors of ~25%. Therefore, achieving reliability of any given level of power output requires the installation of generating capacity approximately equal to four times the desired reliable power output level plus storage approximately equal to three times the desired reliable power output, assuming 100% efficient storage. With real storage, generating capacity must be five to six times the desired reliable power output and storage must be four to five times the desired reliable power output. The economics of such systems are unknowable, at present, because the necessary storage technology is not commercially available.

    Solar, at least, has the advantage of being available when it is most needed. Wind, on the other hand, tends to be most available when it is least needed.

    Hopefully, the new Administration will assure that it’s proposed “solutions” actually work before requiring their implementation.

    “Don’t begin vast programs with half-vast ideas.”


  2. Donkatsu  

    Intermittence is not the only problem for wind and solar. They are both poster children for the second law of thermodynamics. Because the heat gradient is small for both (or the work potential in the case of wind), you need a very large surface area (heat exchanger) to extract much useful work from such a low grade source. This requires that the cost of capture be greater than for energy sources with higher heat differentials. Pretty much the opposite of oil – but then good things always come in small packages.

    I will be damned if I can get a solar or wind proponent to understand that. They always believe that the next round of technology will undo the second law. Good luck with that.

    The only thing that really works for wind is to co-operate the wind farms with dispatchable (storage) hydro. If you do it right you can augment the wind plant factor by using the hydro to buffer the down periods. On the other hand, if you have load following combined cycle plants you can just use the hydro as nature intended, to meet peak demands. That’s hard to explain as well, especially since the solar people are also the ones who want to breach the dams.

    BTW, that Jevons download had one of the fastest speeds I have ever experienced. Good for Google.


  3. Danny Shahar  

    I largely agree, but a few notes about wind:

    1) As energy storage develops (whether in the form of flywheels, batteries, electrolyzed hydrogen, or something else), so wind energy will come to make more sense.

    2) Demand response technologies and energy management techniques may eventually be leveraged as sources of dispatchable load, which could increase the capacity factor of wind projects and increase their profitability.

    3) In the event that climate change legislation is passed (a probable development which I suspect you oppose), natural gas facilities may displace some coal output. Since natural gas generation can ramp its production up and down much more quickly than coal generation, this may lead to lower levels of stranded wind production.

    4) Even without wind power, our transmission infrastructure is in serious need of an upgrade after decades of marginalization. With expanded transmission capacity, wind projects may become more feasible.

    Of course, when I see negative energy prices in response to the production tax credit, I get frustrated too, and the current topology of our electricity infrastructure makes wind a questionable solution. But I think it’s exciting to think about how these technologies could be used in ways that would actually benefit humanity. I agree that right now, wind isn’t the best solution, but hopefully the power of human ingenuity will eventually find a way to take advantage of the tremendous amount of available power that can be captured from this resource.


  4. Donkatsu  

    Some interesting points. I agree that we are probably not at the end of energy storage technology development, and that will certainly improve the prospects of intermittent renewable energy technologies.

    Your mention of grid improvement is interesting. This, I believe, is not primarily a hardware issue, but one that revolves mainly around antiquated state level regulation. At present, it is hardly worthwhile for investors to build the smart grid, since most of the benefits require pricing changes that are currently illegal. Without a thorough modernization of state level electricity regulation, not only will the smart grid but also most new generation technologies, wither on the vine. We will be left to meet the future with the tools of the past. As VP Biden said, “China can build all the clean coal plants.”


  5. Ed Reid  

    1) Storage effectively reduces the CF of wind generation, as the result on in and out losses. However, storage can increase the value of wind power by making it dispatchable.
    2) The CF of wind turbines is a function of the turbine design and local wind conditions.
    3) Wind power generally peaks at night, while grid demand is at its lowest. The issue of surplus wind is the result of the minimum net output of plants which must be kept operating around the clock, because of the time required to bring them on line, exceeding the demand on the grid to which they are connected. This situation has triggered negative power prices in the PJM grid, even without the availability of wind power.
    4) Yes, the grid needs to be expanded and modernized. That is only significant here because it would permit surplus wind power to be spread over a larger customer base. However, as wind power generation increases, the problem will persist until storage is available to time shift the wind power out of the demand trough and onto the demand peak.

    With regard to your comment about natural gas replacing coal as the result of climate legislation, I would encourage you to think about the implications of the proposed “80% reduction by 2050”, particularly in light of an anticipated increase of ~50% in energy demand and consumption over the same period. Ask yourself which fossil fuel end uses will be permitted to survive and emit the remaining 20% of current emissions levels. Also, ask yourself whether an 80% reduction (even if accomplished globally) would be sufficient to halt the increase in atmospheric CO2 concentration which began in about 1750, when global annual CO2 emissions were ~1/2000th of current levels.

    You may find the papers below interesting:
    (Registration is required, but free. Or, e-mail me and I will send pdfs.)


  6. W. S. Jevons (1865) on Waterpower, Biomass, and Geothermal (Memo to Obama, Part II) — MasterResource  

    […] S. Jevons in his early day recognized a central problem of windpower for powering machinery–intermittency. The wind does not always blow, and it cannot be known […]


  7. Danny Shahar  


    On 1) That’s only true if wind electricity is not being dumped. If wind production is higher than the amount of generation needed from controllable sources, or overwhelming transmission capacity, it will often end up stranded. Accordingly, storage could actually increase the capacity factor, even if energy is lost in the storage process. See a parallel example in discussing solar: http://www.nrel.gov/pv/pdfs/39683.pdf

    On 2) See my answer to (1). If wind production is being stranded, then dispatchable sources of load could allow for higher levels of wind-generated electricity usage, even if production doesn’t increase.

    On 3) That’s precisely why points (1) and (2) are important. With energy storage and dispatchable load, negative prices would not be necessary, even with the tax credits that give rise to them in the first place.

    On 4) I largely agree, but reiterate points (1) and (2).

    Regarding the issue of climate change policy, I’m not endorsing any approach; I’m only pointing out that some policy seems somewhat inevitable in today’s political climate, and any such policy would favor natural gas over coal simply because of the former’s relatively higher CO2 efficiency.

    I’ll try to take a look at those papers when I get a chance; thanks!


  8. W. S. Jevons (1865) on Energy Efficiency (Memo to Obama, Part IV) — MasterResource  

    […] The Coal Question (London: Macmillan and Co., 1865) made the case that renewables (windpower; waterpower, biomass, and geothermal) could not substitute for coal. (Jevons underestimated the […]


  9. Wolfgang Flamme  

    @Ed Reid (#1)

    “Solar, at least, has the advantage of being available when it is most needed. Wind, on the other hand, tends to be most available when it is least needed.”

    The accordance between system load and photovoltaic generation frequently appears overrated to me.

    E.g. based on hourly data in 2006 from a reserch project in germany the statistical r2 between load and PV production turned out to be around 13% – that leaves ~87% of the PV production being of out of sync with demand.

    For wind power, r2 was only around ~3%, very small (Betz’ law) but significant due to seasonal variability in demand and production.

    If interested in more details (R-project code, sources) please mail me:
    wflamme at web dot de


  10. Windpower: Yet Another Texas-sized Problem (Hurricane Risk) — MasterResource  

    […] (the late Ken Lay, the current T. Boone Pickens). W. S. Jevons was right in 1865 when he concluded that windpower was unsuitable for the industrial […]


  11. Texas Wind Power: Reality vs. Hype (despite burdensome state mandate, only a 1.2% share projected for 2014) — MasterResource  

    […] William Stanley Jevons,  The Coal Question (1865), p. 122. Texas has repeatedly been lauded as a leader in wind power development. Some of […]


  12. Windpower Is Not an Infant Industry! — MasterResource  

    […] century economists unmasked wind as an inferior energy. Here is one example–see my post on W. S. Jevons (1865) for […]


  13. OVERBLOWN: Windpower on the Firing Line (Part I) | Institute for Energy Research  

    […] William Stanley Jevons, The Coal Question (1865), p. […]


  14. Girlie Man Energy | Institute for Energy Research  

    […] at the beach, he needs to bulk up on his understanding of energy density through the writings of William Stanley Jevons, Vaclav Smil, and Robert Bryce. His message might change for the […]


  15. Climate Scientists Try to Rescue Renewable Energies from 'Planet of the Humans' - Master Resource  

    […] This harks back to W. S. Jevons, The Coal Question, published in 1865, who said this about windpower: […]


Leave a Reply