Yes, wind is cheaper. Provided your expectations are lower.

Well, it makes a great headline.

“Renewable energy now cheaper than new fossil fuels in Australia”

Vive la Revolution! From now on, fossil fuels will not be built, renewables will because they are cheaper! Hurrah!!!

This headline comes from new analysis from research firm Bloomberg New Energy Finance  who have foundelectricity can be supplied from a new wind farm at a cost of AUD 80/MWh (USD 83), compared to AUD 143/MWh from a new coal plant or AUD 116/MWh from a new baseload gas plant”.

This may have caught people by surprise, not just that the wind is cheap(ish) but that the new fossil is costly.

But if you have already spotted the flaw, you get a star. This analysis, and the breathless headline accompanying it, are a great demonstration of how to be correct and irrelevant all at the same time.

These figures compare the cost of electricity production between:

  • Marginal costs of introducing incremental new wind, which will be intermittent with capacity factors around 30%
  • Marginal cost of introducing new modern coal, which would be baseload and quite large, with potential capacity factors above 85%
  • Marginal cost of introducing new baseload gas, which would be combined cycle plant, again quite large, again with potential capacity factors above 85%

All three produce ostensibly the same product (electricity), but they do not provide the same service. A new wind farm, with the well understood intermittency does a poor job of meeting our requirement for baseload, being the minimum electricity demand required at all times.

If you want to compare these sources fairly, you need to set them the same challenge, namely that of providing baseload.

That’s what we did in Zero Carbon Options. The challenge was to replace two small coal plants of 740 MWe in South Australia, with success defined as reliably producing 4,650 GWh per year. We compared a renewable option as proposed by renewable proponents with a nuclear option.

logo only

It is when you ask renewables to respond to this specific challenge that the full story is revealed.

The renewable option could not be wind only, as that fails the reliability test. It was a hybrid proposal of solar thermal with storage (760 MWe) and wind (700 MWe). So that’s 1,460 MWe to replace 740 MWe.

The capital cost is horrendous at over $8 billion, most of which is the solar thermal. The price of electricity is telling: wind cheapest, then nuclear, then miles of daylight, then solar thermal.

The renewable option had 2,810 GWh of the electricity coming from solar thermal at (low estimate) $250 MWh, and 1,840 GWh coming from the wind at (low estimate) $90 MWh. So the average cost of the electricity production was $186 MWh. As you can see, that is way, way higher than either the coal or the gas as priced by Bloomberg.

Even with this, the reliability test was poorly addressed. The level of storage was unspecified, but this solar thermal proposal had a capacity factor of 40%. Two wrongs don’t make a right, and two intermittents don’t make a baseload. To get a higher capacity factor would mean more storage, which would mean more cost.

The bigger you scale this, in trying to replace many different baseload plants in many places, the more the ancillary costs mount, as the ability to move huge quantities of power over great distances becomes critical for reliability and stability.

The nuclear option performed much better, with about half the capital cost, very high capacity factor, no question mark on reliability and electricity for (low estimate) $101 MWh. That is to say nothing of the advantages in lifespan, land use, water security and material consumption.

2x 728 MWe CANDU units from Qinshan, China. Unit 1 went from concrete pouring to criticality in 51.5 months. This could be readily reduced to 48 months with capital costs cut by 25%. Click the image  to read about this
2x 728 MWe CANDU units from Qinshan, China. Unit 1 went from concrete pouring to criticality in 51.5 months. This could be readily reduced to 48 months with capital costs cut by 25%. Click the image to read about this

The point here is not that wind is cheap, expensive, good or bad. It’s that replacing old baseload with new baseload is costly, no matter the tech, but the renewable pathway is most costly of all. If we are serious about replacing our aging fossil baseload with zero-carbon alternatives, then we need all options on the table. If Bloomberg New Energy Finance are serious about informing, not exciting, they will provide a fuller picture next time.

Update: This piece has sparked quite a bit of discussion. Lots of folk seem to want to say I am wrong, or argue that I will be in time. Bloomberg have taken note. Their lead clean energy researcher offered this over Twitter…


…which is of course my point. Where is the discussion of the “low pen(etration)” in the article? The price at “low penetration” means the price if we don’t build too much, and don’t try to use it to fill the role currently filled by coal and combined cycle gas. This makes their comparison entirely redundant. So it may be “news for most ppl”, but if we want an intelligent energy conversation then “most ppl” should be given the complete picture.

Try this headline:

“New wind power at low penetrations cheaper than new baseload fossil in Australia”

Subtly different, but now it is actually in context, informative and useful in building understanding of energy.

Like what you see? Please subscribe to the blog, Like Decarbonise SA on Facebook and follow Ben_Heard_DSA on Twitter.


  1. I wonder where SA goes from here if they increase installed wind capacity from 1200 MW to 1800 MW with the Ceres wind farm (possibly backed up burning straw!). I understand SA wind penetration has hit 33% for periods but averaged 24% in 2012. However that high penetration is made possible by RET assisted interstate exports and the flexibility of 44% gas fired generation. Industry insiders seem to think the east Australian gas price will double by the time LNG export from Gladstone gets into full swing

    If we hit the national RET before 2020 there is no guaranteed market for all this wind power unless it makes up a low cost portfolio of generation. That is no mandate to take more windpower when Vic brown coal power should still be available for $70 a Mwh or whatever. I think SA has enough windmills and should think of some other energy choices. For example the ability to export baseload when the rest of Australia needs it.

  2. It is high time that people realised that the only sustainably renewable technology is in fact the ability of a suitably designed reactor to renew the supply of fissile nuclides, using neutron capture by uranium-238 or thorium-232. The former produces fissile plutonium-239, the latter, using an element three times as common as uranium, produces uranium-233. Neither one of these, if operated for civilian energy, produces bomb-grade material. Both consume all the trans-uranic isotopes, and create amazingly small quantities of nuclear fission waste, some of which even decays to useful stable elements like neodymium.

  3. Hydro-electricity is exceedingly useful for spinning reserve, and emits no greenhouse gases. It is the only energy of solar origin that currently produces significant amounts of reliable energy. But any person who cares about the life cycle of migratory fish, knows that the environmental impact of dams is quite serious. What baffles me is that my fellow environmentalists do not seem to suspect that the impact of wind turbines upon flying wildlife could easily be as bad. So far as I can ascertain, measured in proportion to the delivery of useful power, it is considerably worse. And the visual impact alone is horrible unless you have the most boring landscape imaginable.

  4. Hydro is not very good for spinning reserve at all. Hydro just cant quickly speed up and slow down. That flowing water has huge momentum. If you disrupt its momentum too fast you get a water hammer.

    Natural gas jet turbines are the best units to supply spinning reserve and regulations as they have the ramp rates.

    1. Depends on where you live. On flatlands, hydro is of the run-of-the-mill sort. So completely relies on the seasonal water flow, especially in dry countries like Australia. But once you have some mountains, it becomes impossible to beat with dammed lakes: you can get ~1GW in 2 minutes. You can also do this with a pumping station and a hill.

      1. Makes sense, and yes pumped storage is well known. We have big stored hydro in NSW mountains of course, gets a bit iffy in drought years! New Amazon hydro has surprisingly low capacity factors due to seasonality.

  5. The problem here is that wind has an edge: it is the cheapest renewable, and meets little to no opposition in most countries. So for utilities, the most alluring mix is probably something like gas+wind or coal+wind, where wind acts as a fuel saver about 1/3 of the time. Of course, complete decarbonisation is not achieved in that case.
    But in fact, it’s the proposition against which nuclear power is tested, be it in terms of cost, or popular opposition. To make matters worse, most electricity systems are not vertically integrated, which means the costs of intermittency are not internalized. If wind is cheap enough (or subsidized), some firms will just go around building wind turbines, while some other companies will then claim (and obtain) capacity payments for their fossil fuel plants, because, let’s face it, fossil fuel plants have the lowest capital cost when it comes to spinning reserves.
    I do not have any solution to this predicament.

  6. Ben

    Very good post.

    The BNEF report reads more like a piece of PR spin from the renewables industry. Some of the statements however beggar belief, e.g. “New wind is cheaper than building new coal and gas.” I am not fully aware of the situation in Aus, but the UK needs new capacity in the next few years to meet annual demand peaks. Quite clearly wind cannot do this. So, it is incredibly disingenuous to say new wind is cheaper than coal and gas. If wind cannot do what the coal and gas plant is required to do then it is absurd to say it is cheaper. It’s like train taking me half to London, but costing me less than one that goes the whole way.

    There’s also the horrid headline: “RENEWABLE ENERGY NOW CHEAPER THAN NEW FOSSIL FUELS IN AUSTRALIA.” This is referring to wind, and wind only. Based on Michael Liebrich’s quote this seems to be a narrative they want to peddle. But it helps no one. The costs of renewables vary massively, with oneshore wind much cheaper than every other non-hydro. Saying renewables are cheaper than fossil fuels is simply a recipe for bad policy making.

    The other big problem with this narrative is that these people seem unwilling to put their money where their mouth is. Similar claims come out of the UK constantly, that wind is cheaper than gas. Yet, suggest reducing wind subsidies and they go off in a rage.

    The straightforward problem here is that fossil fuels are cheap, and these efforts to make it seem otherwise are really just getting in the way of serious attempts to change this equation.

    1. I could not agree more. My big problem with this piece is that it is not moving Australia forward to a more informed discussion, but rather building expectations that will then be rebuffed by experts like network planners, who will then be accused (again) of being somehow anti-renewables or “dinosaurs” , and the useless merry-go-round of ignorance continues…


      Reminds me of headlines such as

      “Renewable energy to become world’s second-largest power generation source by 2015”

      It’s great when you can count hydro, biomass, wind, solar, geothermal, tidal and wave as a single source yet split up fossil fuels into separate categories. Makes it seem so much better than reality.

      The spin from the 100% renewables crowd is astounding. I have very smart friends who are usually very good critical thinkers fall to pieces on this topic. They stoop to the very same tricks climate skeptics use yet are blithely unaware that they do so.

      1. Now, if the headline had instead been this:
        “New wind power at low penetrations cheaper than new baseload fossil in Australia” I could have endorsed it 100%! The difference is not big, yet it is huge. This headline would actually inform, and invite analysis and understanding that reflects the reality of the way energy systems work! But they did not go with that headline, which makes me query the impartiality.

        1. Ben

          I think a lot can be inferred from what BNEF defines as “Sustainable Energy” in this report:

          Astonishingly natural gas and biofuels are both in there. Nuclear is not. The report, like the Australian wind one, is packed full of spin. Just look at the final 3 pages, where they produce a graph of the percentage renewables and gas make to total nameplate capacity in US, Canada, China and Germany. Anyone with basic knowledge of electricity generation knows this is totally meaningless. The only reason for doing it is to make renewables and gas seem to be generating more electricity than they are.

          My only conclusion here is that they are just spinners for the industries they have clients in, and not producing “quantitative and objective” reports as they claim. Their reports should be stamped buyer beware.

          1. That’s very disappointing, but pretty consistent with what I felt needed to be critiqued by this post. Those calling themselves analysts need to do better. It seems like “agitators” is a more accurate description for BNEF.

  7. The issue of capacity payments would not arise if wind and solar did not get preferential treatment but had to bid unsubsidised prices that cover their average cost. Recently we’ve seen spot prices on the NEM hit an unbelievable $12,000 per Mwh
    While the trendy wind build continues and solar panels remain a middle class fashion statement the open cycle gas plants have become like gold mines. At some point consumers must ask if it could be done cheaper. In SA’s case energy must have been part of the ‘input costs’ that put the kibosh on the Olympic Dam expansion. That project was meant to hire thousands of people but is now laying off staff.

  8. In the UK, DECC finds nuclear to be the cheapest low emission technology on a straight LCOE basis in it’s most recent assessment:

    Click to access 6883-electricity-generation-costs.pdf

    And that’s based on FOAK EPR costs. 2012 estimates are:

    CCGT Gas: 80p/kWh
    Nuclear: 81p/kWh
    On-shore Wind (All UK): 93p/kWh
    On-shore Wind (England and Wales): 104p/kWh

    I remain skeptical of claims that wind is cheaper than nuclear in any universal sense.

  9. Does anyone have the full BNEF report?

    They do Nuclear analysis too. Personally I think it’s a bit brash to label BNEF the way they are being above before anyone has seen the full report. To slide down that slope ends up in Gundersen and Busby land. Be a shame if that happened.

    I’ve read a few articles analysing this report, and they appear to have points. But until a full report is seen…it’s just assumptive.

    At the end of the day, in Australia Nuclear has no LCOE. No investment bank will touch it with a 10-ft barge pole as it is illegal. That needs to be a major priority, all the tools are there. How long has it been since I wrote/organised a petition and had it sent to my local representative? Because if you meet a certain criteria petitions can be tabled in Parliament.

  10. I wonder if the Bloomberg piece has made a crucial omission by not adding the value of the LRET payment to the raw sale price of wind generated electricity. A couple of days ago the LGC was worth $33 per Mwh and if I understand correctly it is paid separately to any NEM transaction.
    If I’m right about the mechanism then if new wind can be sold for $80 we need to add $33 to get $113. That’s more like the price of new nuclear with 90% capacity factor.

    Notice if you browse the website the 1997 baseline swiftie with the LGC system. That’s so highly dependable 1960’s built hydro gets nothing.

  11. I’m eager to see a forthcoming report by Deloitte on how to supply water, energy and other services to new mines and extractive industries in west and north of South Australia

    Click to access Joint%20Press%20Release%20-%20RMIP%20(2).pdf

    Slight problem the person signing the cheque for the study is federal infrastructure minister Albanese, I recall him saying words to the effect ‘nuclear power is absolutely not needed in Australia’. It seems we’ll have ore crushing plants operating at 2 am powered by solar panels. More likely by our old favourite coal even if it’s 1,500 km away.

  12. Ben this is my latest comment to Giles Parkinson re the Bloomberg claims at

    Giles, you know I have raised this before but one has to question how Bloomberg determined the “coal” band for each of these graphs. The future increases for coal are going to be largely based on the carbon price as the cost of building the plants and O&M costs in Australia are probably not going to rise that much over time.

    The BREE estimates for the non-coal options are largely the same as Bloomberg’s. But they are very different for coal.

    BREE estimates the cost for new black coal plants without a carbon price at $84/MWh and $95/MWh for brown coal. To get the Bloomberg range for 2012 of $146 to $240 for coal means a carbon price of at least $51/MWh. At a volume weighted emission intensity for new coal plants of 0.85 t/MWh and a known carbon price of $23/t gives a carbon price of $20/MWh not $51.

    Based on BREE estimates the coal range for 2012 should be between $102 and $119/MWh depending on the coal mix using a carbon price of $23/t. Wind is around $116/MWh so lower than brown coal but not black coal.

    The only explanation I can come up with is that Bloomberg has assumed new coal plants will cost between $126 and $220/MWh without a carbon price. These figures are between 40% to 140% higher than BREE – which beggars belief.

    The error appears to have been repeated in other years. By 2030 Bloomberg has coal between $190 and $300/MWh in real 2012 AUD (the cost of the plants will not increase) so again the only reason is an increasing carbon price. For simplicity, let’s assume the volume weighted emission intensity for black and brown coal has risen to 1 t/MWh. We need a carbon price of between $95 and $205/t based on BREE plant costs to get the Bloomberg range.

    The Treasury estimates a carbon price of $55 in 2030 so the Bloomberg plant costs must have risen to between $135 and $245 in real terms – which beggars belief even more!

    1. Here’s a number of uncertain factors about future fossil LCOEs
      1) Combet concedes the carbon price may fall not rise however
      2) if the 2015 ETS eventuates EU type offsets will be limited to 12.5% of an emitters liability.
      3) should generous cash and free permits be factored in? Especially for brown coal.
      4) a policy of mandated high penetration wind power will mean coal plants generate fewer Mwh to spread their fixed costs.
      5) industry insiders tip the east Australian gas price to increase from $5 per GJ to $9+ after Gladstone LNG export commences.
      6) an Abbott victory may make the official carbon price $0 but direct action may impose an implicit price, as yet unknown.

      It would be so much simpler if wind and solar stood on their own merits without preferential treatment.

      1. Well, it might be nice if all energy sources had to do that. God only knows fossil fuels have had and continue to have their share of handouts, including apparent windfall gains from our carbon price as you point out.

    2. Martin, thanks for re-posting here. That’s great analysis, and easy to follow.

      The conclusion I am left to draw is that BNEF are not only misleading in their comparison, they are flat out, terribly wrong in their analysis.

      Between my high level piece, Tristan Edis’s considerations at Climate Spectator and the issues you have raised above, there seem to be real problems with both the work and the way it has been presented. The work has traveled far, but it also seems to be getting seriously interrogated, and that’s a good thing.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: