Bizarre Payback Analysis Being Used for Alternate Energy

Check out this payback analysis that is being trumpeted for wind power:

US researchers have carried out an environmental lifecycle assessment of 2-megawatt wind turbines mooted for a large wind farm in the US Pacific Northwest. Writing in the International Journal of Sustainable Manufacturing, they conclude that in terms of cumulative energy payback, or the time to produce the amount of energy required of production and installation, a wind turbine with a working life of 20 years will offer a net benefit within five to eight months of being brought online.

So of all the scarce resources that go into producing wind power, if you look at only one of these (energy), then the project pays itself back in less than a year.  This is stupid.  Yes, I understand that there are some "green" energy sources (*cough* corn ethanol *cough*) that cannot even produce more energy than they consume, so I suppose this finding is a step forward from that.  But what about all the other scarce resources used in producing wind power-- steel, labor, engineering talent, concrete, etc?  This is roughly like justifying the purchase of an 18-wheeler truck by saying it will pay off all the vanadium used in its production in less than a year.

Environmentalists seem to all feel that capitalism is the enemy of sustainability, but in fact capitalism is the greatest system to promote sustainability that has ever been devised.  Every single resource has a price that reflects its relative scarcity as compared to demand.  Scarcer resources have higher prices that automatically promote conservation and seeking of substitutes.  So an analysis of an investment's ability to return its cost is in effect a sustainability analysis.  What environmentalists don't like is that wind does not cover the cost of its resources, in other words it does not produce enough power to justify the scarce resources it uses.  Screwing around with that to only look at some of the resources is just dishonest.

The one reasonable argument is that the price of fuels does not adequately reflect the externalities of Co2 production.  I don't think these are high but obviously there are those who disagree.  The right way to do this analysis is to say that wind power provides a return only if electricity prices are X (X likely being well above current market rates) which in turn reflects a Co2 cost of Y $/ton.  My gut feel is that it would take a Y -- a cost per ton of CO2 -- way higher than any of the figures that are typically bandied about even by environmentalists to make wind work.

Postscript:  I did not critique the analysis of energy payback per se, but if I were to dig into it, I would want to look at two common fallacies with many wind analyses.  1) They typically miss the cost of standby power needed to cover wind's unpredictability, which has a substantial energy cost.  In Germany, during their big wind push, they had to have 80-90% of wind power backed up with hot fossil fuel backup.  2)  They typically look at nameplate capacity and not real capacities in the field.  In fact, real capacities should further be discounted for when wind power produces electricity that the grid cannot take (ie when there is negative pricing in the wholesale market, which actually occurs).


  1. donald:

    another bizarre analysis...How can SRP make profit by by pumping water back up into the upper lakes around Phoenix to use as hydroelectric power? (or at least I was asked this question in engineering in the late 80's, assuming they still do this) The answer is that Palo Verde Nuclear Plant produces power cheaply at times when it is not needed. So they take that power that is not wanted or needed at lets just say 9pm to 5 am and use it to pump water back up the lakes. Then in the morning when we turn on the coffee pots and all the lights, they throw open the power generation at the dams and get that peak power they need. More or less the lakes are a battery or power storage device that can be charged by off peak generation.

  2. sean2829:

    The 5 to 8 month payback for the energy input to wind turbines, however incomplete and misleading that might be, is much better than for solar cells where the energy payback exceeds 3 years. Wind at least can deliver 25-30% of nameplate capacity over a year while solar is less than 15%.
    In response to Donald's comment, the pumped storage of hydroelectric has been used at power plant sites in the TVA and I always wondered why it wasn't more closely tied in with renewable technologies. (It apparently is in Portugal.) Perhaps it it because renewables often get priority and higher prices and the subsidies wind power tax credits makes it profitable even at negative cost for electricity.

  3. Richard Harrington:

    Here's a fairly balanced article about the costs:

  4. jdgalt:

    I would expect any analysis of energy payback to include the energy cost of manufacturing the product. I doubt that any hybrid or electric car has even gotten within a factor of 2 of energy break-even when its entire life cycle is considered.

  5. stevewfromford:

    Here is another uncovered cost. I live not far from the large windfarms spoken of in the article in Eastern Washington state where I was listening to a local radio talk show one day during a discussion of the economics of these very windfarms. An engineer working for Bonneville Power Administration called in and related that the giant turbines on the huge BPA dams and, for that matter at dams all over the region, were suffering greatly increased wear and premature failure due to the constantly shifting load imposed upon these base load generators. These turbine generators were designed to last for over 50 years and are, apparently, failing after only a few years because they cannot withstand the wildly changing demands as the wind changes. These costs are passed onto consumers of the power produced and are undoubtedly never accounted for when figuring the cost of windpower. I always make a point of counting how many of the windturbines are broken down when I drive through this area and the wind is blowing. Usually it is about 15% non functional at any given time.

  6. Canvasback:

    "But what about all the other scarce resources used in producing wind power-- steel, labor, engineering talent, concrete, etc?"
    They would just be used to build a different power source? Maybe it takes more engineering talent to build a wind farm, but less concrete, environmental review, and manpower than building a hydroelectric dam.

    But I'm with you, their energy calculation is unsatisfying.

  7. poitsplace .:

    Adding to your PS, They don't simply fail to count the cost of standby power or the negative spot prices when the wind blows. As their idiotic policies cause prices of standby power to rise, they count those costs against the price of those sources...and then use that increased price to say wind is that much cheaper. They're effectively doubling up on the error.

    So in reality, unless the region they're used in is know for unusually stable wind...wind turbines are little more than multi-million dollar, government funded monuments to a green religion.

  8. FelineCannonball:

    Hybrid and electric cars don't make energy they just burn it up. But I think you're pointing out that the total cost might wash out the energy savings. That's pretty much true (civic vs. hybrid civic), but it depends how far you drive it and how much your saved capital would be earning. A 1989 Yugo with 500k miles beats them all. Keep your car longer and don't get power windows and you'll save money. Well, actually, an urban vegan bicyclist has the lowest capital costs and operating costs. Even better if he glides through stop signs.

  9. FelineCannonball:

    There are a few places where they win on level-table economics. They beat out diesel generators in remote sites in the Gobi, Siberia, and small islands. But for the most part they're inferior in economics, base load, peak load, etc.

    Things are changing though. Design changes, manufacturing changes, global competition has driven down prices dramatically. Wind will be a part of the equation no matter what when natural gas gets scarce in a few decades and terminals make it's global trade viable.

    As for the pacific northwest, they really need to connect better to the rest of the countries power grid. Columbia River power (without the alumnium smelters that Bonneville scared off) is a huge surge every spring. And they can't economically shut down the turbines when prices go a little negative. It costs more in turbine replacement. Hydro+Wind is just stupid. Expanding lines to California or building a crap load of pump-storage is where the government should be focusing it's subsidies -- if anything.

  10. marque2:

    It actually looks like we have about a 3000 year supply of natural gas. And if we start using coal again (it can be liquified and gasified) there is an additional 600 year supply of energy. If nuke energy were allowed there is about 100 years of conventional and maybe 5000 years of thorium.

    The only thing we could run out of in its pure form is oil.

    So plenty of time to perfect those windmills.

  11. Mole1:

    Your criticism is misplaced. This is an academic journal article, and uses a metric that is also used in many industries, and is especially useful in the fossil fuel industry. There is a wikipedia page on it:

    There is nothing stupid about this metric. For example, it is very useful to know that on a straight energy balance measurement, oil has fallen from ~50:1 to ~5:1 over the last century. This doesn't take into account the engineering effort that went in to the advanced extraction techniques required to get at the more difficult sources of oil, but so what? Using this metric demonstrates in a straightforward way that the difficulty of extracting oil is increasing. In order to get to a complete analysis of the costs and benefits of wind, you have to start by counting the costs and the benefits. This is a start.

    You and many of the commenters seem not to understand how complex systems analysis works in the real world. You can't expect, for example, economists who are analyzing the utility of wind farms to do the energy balance calculations. They look to the literature for that, for analyses done by engineers. Large complex analyses are enabled by simpler analysis done at a lower level by subject matter experts, and published in the academic literature.

  12. rst1317:

    "2) They typically look at nameplate capacity and not real capacities in the field. In fact, real capacities should further be discounted for when wind power produces electricity that the grid cannot take (ie when there is negative pricing in the wholesale market, which actually occurs)."


    Consider the afternoon of August 2, when electricity demand [ in Texas ] hit 67,929 megawatts. Although electricity demand and prices were peaking, output from the state’s wind turbines was just 1,500 megawatts, or about 15 percent of their total nameplate capacity. Put another way, wind energy was able to provide only about 2.2 percent of the total power demand even though the installed capacity of Texas’s wind turbines theoretically equals nearly 15 percent of peak demand. This was no anomaly. On four days in August 2010, when electricity demand set records, wind energy was able to contribute just 1, 2, 1, and 1 percent, respectively, of total demand.

  13. rst1317:

    I'm not sure it's misplaced. That measurement doesn't for everything needed for that wind power. That is what the criticism is, correct? I don't think anyone is saying the measurement is wrong, they're pointing out we have a lot more to look at than just that.

  14. rst1317:

    Sean2829, I'm curious what wind projects are delivering more than 15% of their name plate let alone 25-30%. Do you know which ones are?

  15. Mole1:

    Did you bother reading past the first paragraph I wrote? Yes, that measurement doesn't account for everything needed for wind power, or oil extraction, or nuclear energy. So what? It is a standard metric used by engineers, and has utility in system-level analysis. Is the argument that engineers should keep their mouths shut until they've done all of the engineering, environmental, and economic analyses themselves, and integrated them into a system-level analysis? That would be asinine. Other people do the environmental and economic analyses, and publish those analyses. Then yet other people combine all of those analyses for a system-level analysis.

  16. sean2829:

    This document has an analysis of wind capacity factors in different countries.
    I suspect if you correlated wind power subsidy rates with capacity factor they might be inversely related. In other words, if the project needs to be paid for with competitively priced power, the factor is likely high but if the wind farm is more subsidy farm, the capacity factor does not matter as much. Interestingly, Germany has a very low capacity factor (17.5%) while the US is on the high side (25.5%).

  17. rst1317:

    I'm not sure the criticism is misplaced. The criticism isn't that the metric is being used or that it's not occrrect but that it doesn't show the entire picture.

  18. FelineCannonball:

    3000 years? That's quite the prognostication. You'd have to define currently economic deposits, future technological innovation, future economic deposits, population, economic growth, other energy innovations, etc.

  19. marque2:

    According to my teachers in the late 1970's we ran out of oil in 1991, which was OK because we would probably all die from nuke war before I graduted high school in 1985 and now 23 years later and we have 50+ years of proven reserves.

    3000 years is probably short based on current usage - and if you kept up on technical innovations you would know what I am talking about.

  20. herdgadfly:

    Warren wrote about the BPA phenomenon back in 2009. Nothing ever changes at government-run facilities.

    I had an exchange on my blog with Michael Milstein of BPA Public Relations. The exchange went like this:

    Milstein: We handle power from dams on the Columbia River and its tributaries, which are located more than 1,000 miles away in Oregon, Washington, Idaho and Montana.

    Our primary role in handling wind involves integrating wind from turbines owned by other utilities or companies into the power grid. This is a very important point: We run the transmission system that delivers energy from the wind farms owned by others to the utilities that purchase that wind power (which may be but are not always the same). We are responsible for making sure the amount of power flowing into our grid always equals the amount flowing out.

    Hydroelectric dams are ideal for helping to balance the ups and downs of wind because, unlike nuclear or coal plants, dams can rapidly adjust their energy output up or down by running their turbines faster or slower (letting more or less water through). When they slow down, the dams typically hold that water back so it can be used later. So the water is not wasted.

    This varies somewhat by dam because certain dams have more storage capacity behind them than others. Some can hold only a relatively small amount of water. However, the dam in our system that does much of the wind balancing is Grand Coulee, which is the largest dam (and largest power plant) in the country and has a large amount of storage capacity (it also provides flood control by holding back water when needed). So when the wind picks up, its turbines slow down and the water typically remain behind the dam until it is needed to generate power later.

    (There are other reasons water is spilled around turbines, most notably to help juvenile salmon make a safe downstream migration. In this case, we forego the energy that could have been generated, but for a very important reason.)

    The combination of hydro and wind power is free of any emissions.

    In addition, the utilities that actually receive the wind power delivered over BPA lines typically use that wind power to displace other fossil fuel resources, typically gas plants (which can adjust their output rapidly) and coal plants. So the wind does reduce emissions that way.

    I hope this helps clarify the picture.

    My Response: Thank you very much for the detailed information on BPA. ... I also understand that BPA is a federal power facility that we folks back here in the Midwest do not get to enjoy.

    The point that Warren Meyer was making about wind farm electricity is that it doesn't play nice in the power grid. There is no way to store it and big winds at night are generally worthless since there are too few users. So if you are not diverting water, then you are shutting rotors down. Like all renewable electrical energy, with the notable exception of hydro, the expense to build and operate is not competitive ... and if the government and the environmentalists would concede the obvious, that carbon dioxide does not cause global warming, then the old fashioned carbon-based power plants would do fine for a few more centuries.

    I was amused by your comment on salmon migration. I am sure that you have heard from the many Washington state nature lovers that man had no right to build those ugly dams which interfere with nature, and like Cape Wind, take away from the awesome beauty of your state. Now I expect that those wind rotors are chopping up spotted owls.

  21. jdgalt:

    You're almost certainly right about comparing the savings from better gas mileage with the higher cost of a hybrid car (= money break-even), but the comparison I was trying to make is whole-life-cycle energy use (the total energy needed to build the car, including producing all materials, and operate it until it's scrap, divided by the total miles driven, vs. that same ratio for a conventional car). If I'm right in thinking that number, total energy used per mile, is higher for the hybrid, then the greens have just scored an own-goal, and it's far from their first.

    I'll resist the temptation to comment on urban vegan bicyclists beyond the fact that they impose costs on other people (including traffic delay) that seem to me a lot worse than a little exhaust.

  22. FelineCannonball:

    This has been done. Basically a Tesla has a huge carbon footprint from cradle to grave. Other electric cars have results that depend in the source of electricity. A gas car getting 35mph does better than a Leaf in Kentucky where the Leaf basically burns coal.

    Some hybrids do quite well but it depends on their longevity.