Posted by: bmeverett | January 9, 2013

The state of play on wind energy

My readers may have noticed a series of excellent comments on this site by Christian Schiller, one of my favorite Fletcher students, who writes on the site globalberliner, which you can find at

In response to Christian’s comments on my statements regarding wind and solar power, I’d like to expand on my views. Let’s start this week with wind and we’ll address solar later. Before we crunch numbers, however, let’s agree on some ground rules.

First, we focus on cost. The wind lobby loves to talk about how many jobs it creates, how fast wind capacity is growing, how much its costs have fallen and how much property tax it pays. All well and good, but the critical issue is whether wind can generate electricity at lower cost than fossil fuels or nuclear plants.

Second, we look at the actual cost of wind power. Not just the cost of the wind turbines, but the full cost of installation and operation. When a wind farm buys turbines from GE, Vesta or one of the other vendors, the turbine is delivered, but then has to be erected on a prepared site and connected to the grid. These “balance of plant” costs are generally 20-40% of the cost of the turbines.

Since wind power is intermittent and unpredictable, it requires back-up generating capacity. Let’s say we build a 400 Megawatt (MW) natural gas-combined cycle (NGCC) plant, which is capable of operating 90% of the time, with only 10% required for maintenance. If we build in addition a 100 MW wind farm, the NGCC plant would have to be shifted to stand-by about 8% of the time to accommodate suddenly available wind power. Although the NGCC plant would save something on fuel and variable maintenance, it would still require more revenue to meet its investors’ return expectations. These costs must be included as a debit against the wind farm.

Since wind farms tend to be located farther from demand centers than fossil fuel plants, we also need to consider transmission costs. According to the Energy Information Administration, these costs are about 1.3¢ per megawatt-hour (MWh) for natural gas plants but 4¢ per MWh for wind. In states like Texas, where the wind farms are hundreds of miles from demand centers, this difference can be much higher.

Third, we exclude all the special taxes credits and subsidies in making our assessment. Subsidies do not reduce costs, but merely shift them from electricity consumers to taxpayers. These subsidies can be explicit, such as the 2.2¢ per kilowatt-hour (kWh) production tax credit, or they can be subtle. Renewable portfolio standards, for example, require utilities to purchase a certain amount of renewable energy regardless of its cost, which allows wind energy producers to charge an above-market price for their kWhs. “Net metering” regulations require utilities to pay the full retail for wind energy whenever it’s available, thereby forcing the utilities to bear the added cost of stand-by generating capacity. Government loans and loan-guarantees reduce the cost of borrowing to below-market levels. All these subsidies must be made explicit in our analysis.

Fourth, we use proper capital costs. Many studies assume very low returns on capital (4-5%). No investor, particularly a venture capital firm, will put his money at risk for such a return, unless he has government guarantees against loss. In any case, people who wish to offer opinions on renewable economics must at the very least be explicit about their capital cost assumptions.

Fifth, we use the proper comparison basis. When we buy a new car, we don’t compare its cost, efficiency, styling and comfort with the old clunker we’re replacing. We compare new cars with other new cars we could buy. By the same token, a new wind farm has to be compared with the lowest-cost new power source. The concept of “grid parity,” where we compare the cost of new energy sources to the average retail price of power from the grid, is meaningless. The existing grid is full of older power plants, inefficient transmissions systems, onerous regulations, taxes, fees and many other market-distorting factors. Today, NGCC power plants are the low cost option, generating power at about 4.2¢ per kWh. That’s the target, not the national average retail electricity price of about 10¢ per kWh (12¢ for residential customers).

Sixth and finally, we quantify externalities. Wind power does have advantages in terms of air pollution and carbon emissions. We must not, however, treat these external costs as infinite. The Climate Community tends to see global warming as apocalyptic, thereby freeing them from the need to put a price on carbon. We can’t make policy that way. It’s impossible to put a price on the impact of atmospheric carbon dioxide, because in reality we have no idea whether and how much it will affect the climate, but we’ve seen some political calibration. The most recent price for European carbon futures is about €8 per metric tonne or roughly $11. The Regional Greenhouse Gas Initiative has sold carbon emission rights at about $2 per metric tonne.

If we do not quantify the externalities, we have no way of knowing whether the higher cost of wind power is worthwhile. Talking in generalities about the need for low carbon fuels and bragging about the tons of carbon dioxide saved have no meaning. The question is are you getting a reduction in external costs comparable to what you are paying?

With these ground rules in mind, here are my assumptions for wind power economics: (a) turbines cost $1,200 per kW, based on the most recent averages, (b) balance of plant costs are 30% of turbine costs, (c) annual maintenance is 2% of original capital cost, escalated with inflation, (d) load factors average 31½% based on data from the American Wind Energy Association, (e) the wind farm consists of 50 X 2 MW turbines for a total of 100 MW, (f) the wind farm takes 1 year to build, (g) the wind farm is built in conjunction with a 400 MW NGCC power plant, (h) the NGCC plant takes 3 years to build, (i) per the EIA, the cost of the NGCC plant, including balance of plant costs is $1,056 per kW, with annual fixed maintenance of $15.39 per kW and variable maintenance of $3.27 per MWh, (j) the load factor of the NGCC plant is 90% without the wind farm and 82% with the wind farm, (k) the price of natural gas is $4 per million British Thermal Units (MBtu), (l) the loan/equity ratio is 80/20 on both the wind and NGCC plants with a 30-year loan at an interest rate of 8%, (m) investors earn 15% on equity, (n) the project lasts 33 years from ground-breaking, including 30 years of wind turbine operation and (o) the net lost earnings from the NGCC plant are debited against the economics of the wind farm.

These assumptions produce break-even generation costs of 4.3¢ per kWh for NGCC and 6.8¢ per kWh for wind, making wind roughly 50% more expensive than NGCC. A carbon tax of $11 per metric tonne, equivalent to the current European carbon market price, would debit the NGCC plant by only 0.3¢ or about 15% of the discrepancy. To fully close the gap, the carbon tax would have to be around $100 per metric tonne, a huge burden on the economy.

I highly recommend the article “Understanding Trends in Wind Turbine Prices Over the Past Decade” by Mark Bolinger and Ryan Wiser of the Lawrence Berkeley National Laboratory. This article tracks turbine prices from 1997 through 2011 and finds not a smooth decline curve but a cycle with turbine costs falling from 1997 to 2001, then rising from 2001 to 2009, then falling again. Overall, we see a drop of about 20% over 15 years or 1½% per year – not an impressive technology curve.

Wind power is still not competitive in the marketplace. Without subsidies, there would be no wind turbines built in the US. The rationale for these subsidies continues to evolve. The argument about externalities is weak, because the gap between wind power and NGCC is simply too wide. Implicit in the argument for subsidies is the assertion that heavy government subsidies can force a drop in the cost (not the price) of wind power. As I have noted in previous posts, the opposite is likely to be true. Manufacturers may gain economies of scale from increased turbine production, but subsidies diminish the incentive to improve the technology, since the government forces utilities to buy inferior technology as it is.

Another argument that’s making a run is that all energy sources have been subsidized and that oil was heavily supported by the government during its early days. Therefore, renewables are simply following the same path as other energy sources. This argument is becoming one of those myths that is widely believed because it’s repeated so often. It is simply not true, however, and I’ll address this issue soon in a future post.



  1. Dear Professor,

    it’s great that you spelled out the assumptions that you used in your calculation for wind power economics.

    I have to take issue with assumption k, namely natural gas price of 4 USD/MBTU. As you know that applies only to the US. In Europe is more 10 USD/MBTU, In Latin America, depending on the country, 6-8 USD/MBTU, in some parts of Asia (LNG imports) 12/16 USD/MBTU.
    Therefore, in most of the world the gap between NGCC and wind power is much norrower becasue the price of natural gas is at least double that of the US.

    Even in the US you might consider the cost of hedging…. but that is another story.



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 )

Twitter picture

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

Facebook photo

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

Google+ photo

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

Connecting to %s


%d bloggers like this: