Renewable Energy, what is the cost?

By Andy May

What are the costs of using renewable energy? The sun and wind are free, does that make wind and solar power free? Biofuels require power to plant crops, make fertilizer and spread it, harvest the plants, make and transport the ethanol. Solar and wind require power to produce, transport and install the equipment. All renewable energy sources require lots of land per megawatt of electricity produced. We will not be able to determine a cost for renewable power in this essay, but we can discuss the components of the calculation and provide some context.  A key question to think about, do renewable fuels decrease fossil fuel use, or do they increase it?

The energy return on investment (EROI)

Energy used to make a fuel, like a biofuel, has to be much less than the energy output for the fuel to be useful. The metric commonly used is known as “energy return on investment” or EROI. It is computed by dividing the usable energy in a fuel by the energy consumed to make the fuel. Thus, an EROI of one means just as much energy was used to make the fuel as we can get out of it. This calculation is complex and as a result, estimates for any given fuel can vary a lot. Often local factors dominate the calculation so comparing one fuel to another using this metric is tough (see figures 1 and 2). Notice the range of values for oil and gas, they depend very much on location, the type of crude oil or gas, the time period used, local regulations and infrastructure. It is very hard to pin down a value for a fuel because local economics, quality and timing have such a huge impact.

Hall, et al., 2014 describe four different EROI calculations. The standard EROI is computed at the point of origin, the well head, the ethanol plant or the factory making the wind mills or solar panels. This can leave out a lot of energy required to utilize the fuel and backup intermittent sources, like wind and solar. The second calculation takes place at the point of use, this includes refining and transporting the fuel, the Extended EROI includes the energy required to use the energy, such as maintenance, friction losses, etc., finally they describe the societal EROI which includes the actual gains from a fuel for society. This latter has yet to be achieved due to its complexity. None of the calculations include (except possibly the last one, which has not been done yet) the cost of fossil fuel backup for solar and wind generation. Often the values calculated do not include capacity correction factors to account for downtime.

Figure 1, source Hall, et al., 2014.

Figure 2, source Hall, et al., 2014.

So, EROI calculations are not standard or complete and are difficult to compare to one another. It is especially difficult to compare technologies to one another, for example natural gas to solar or wind. But, even so, most calculations show that corn-based ethanol has an EROI of only 1.25:1 (Kiefer, 2013). When we consider the loss of energy when using ethanol in a vehicle, this is a negative return. Cars and trucks are not perfectly efficient, so breakeven has been estimated to be around 3:1 by Hall, et al. (2009). According to Kiefer, even ancient Rome did better with grain for slaves, oxen and horses, their maximum EROI was about 4.2:1. In building the colosseum the EROI was 1.8:1. Kiefer also reports that, because modern society requires so much energy, when the overall EROI drops below 6:1 the economy contracts and we are at risk of recession.

Much later, coal-fired steam engines came along with an EROI of 10:1 or more. In the early coal mining days extracting coal was very easy since it was accessible at the surface or very near the surface. This was a good thing because early steam engines were very inefficient. However, as coal fired steam engines improved and proliferated they replaced slave labor so the social and economic benefits were large. Some historical published EROI values for coal are shown in figure 3 from Hall, et al., 2014.

Figure 3, source Hall, et al., 2014.

Our civilization depends very much on EROI as the surplus energy (energy not used in obtaining and using fuels) defines our affluence. Because we currently enjoy a large energy surplus, we can spend our time doing things other than simply growing food and gathering wood for shelter, heat or cooking.

Energy consumption or energy efficiency?

It is well documented that wealth and standard of living are closely related to energy consumption. This is worth repeating and emphasizing, it is energy consumption that is strongly related to our wealth and standard of living. Obviously, energy consumption is related to price, the cheaper energy is, the more we consume and the better off we are. But, price is secondary to consumption. Timothy Garrett (2011) has shown that every additional 9.7 milliwatts consumed increases our global wealth by one 1990 US dollar. Other documentation of the intimate relationship between energy consumption and wealth can be seen here and here. The quickest way to raise people out of poverty is to supply them with energy and the quickest way to throw more into poverty is to take it away or make it unaffordable. See this article in Spiegel Online for more details.

While Stephen Chu and the IEA may think energy efficiency will save energy overall, history says this is not so. Energy use is subject to “Jevons Paradox” which states that greater energy efficiency leads to lower energy costs and more energy consumed. This counteracts most of the efficiency gains, improves the economy, our standard of living and leads to still more energy use. Today, because of energy consumption, fewer people are needed to grow our food and build our houses. One might think in an affluent country, like the US, a market would become saturated with energy consuming goods, like cars, air conditioners, and refrigerators. This is not likely longer term, since new products, all using energy, will be invented and, if attractive, will be bought and used. When I was a young child we did not have a television in the house, now my wife and I are empty-nesters and we have three televisions, four computers, two computer tablets and two smart phones. They all use energy. It’s quite simple really, consumption goes up faster than efficiency improves.

While EROI is the ideal way to compare the quality of fuels and energy sources, the calculation is complex and fraught with problems. Another way to look at energy sources is through the levelized cost of electricity. Both the US EIA and the International IEA have attempted these calculations and while they come up with different numbers they mostly agree qualitatively. The EIA computes the costs for the United States and the IEA computes them for the whole world, this is part of the reason for the differences. The results are summarized in table 1 below in US$/MWh. Both sets of numbers are corrected for downtime, that is wind is corrected for the time the wind does not blow, solar is corrected for nighttime and cloudy days, and coal and natural gas are corrected for maintenance time. However, unlike regular maintenance, solar and wind downtime is not planned or controllable and these numbers do not reflect the cost of the required (nearly) 100% idling fossil fuel backup for windless nights, thus the solar and wind estimates are lower than reality. Given how dependent wind and solar are on the weather, we can only imagine the boom in meteorology that must be occurring!

For the most part the renewable technology used in all countries is the same, the difference in costs reflect differing prices (for example coal or natural gas prices), fuel quality or the cost of differing regulations. Modern coal power plants in the US and Europe have all had pollution control equipment installed for decades and as the table shows these plants produce electricity for $80 to $100 per megawatt-hour. But, “CCS” (carbon sequestration) to remove non-toxic carbon dioxide from a coal plant effluent is very expensive. Carbon dioxide is inert and not a toxic gas like sulfur dioxide or mercury and is very difficult and expensive to remove. This raises the cost of coal power plant electricity to $140 per megawatt-hour in the US.

Table 1, source EIA and IEA

Cost of renewable energy sources

Renewable energy sources have many advantages, but they tend to have low EROI values and their use raises electricity costs. In table 1 we can see that renewable power averages about twice the cost of the cheapest alternative. These estimates do not include the cost of fossil-fuel backup for wind and solar. Nuclear power is the cheapest source of power in Europe and natural gas is the cheapest in the US due to shale gas production.

The Wall Street Journal has reported that Germany’s electricity costs have risen 60% due to their subsidies of renewable energy. This has lowered their GDP, standard of living and competitiveness. BASF, SGL Carbon, Basi Schöberl GmBH and Siemens have all moved operations from Germany or are planning to, due to the high cost of energy.

Wind and Solar Power

Wind and solar are intermittent because the wind doesn’t always blow at the right speed and the sun doesn’t always shine. According to Hall, et al. (2014), good EROI calculations for solar and wind are not available at this time. This is largely because solar and wind are highly dependent upon fossil fuel backup and are essentially “subsidized by” higher EROI fossil fuels and exactly how to account for this is not known. According to Hall, et al. (2014):

Alternatives such as photovoltaics and wind turbines are unlikely to be nearly as cheap energetically or economically as past oil and gas when backup costs are considered. … Any transition to solar energies would require massive investments of fossil fuels. Despite many claims to the contrary—from oil and gas advocates on the one hand and solar advocates on the other—we see no easy solution to these issues when EROI is considered.”

A clear example of the problem can be seen in Germany, where some 32% of installed energy capacity is from wind and solar. Just because solar and wind are installed and can produce 32% of the required power doesn’t mean they actually do. Demand fluctuates and so does wind speed and incoming solar radiation. Electricity in 2016, cost Germans 39 cents per kilowatt-hour, versus 27 cents for the rest of Europe and about 11 cents for Americans, this has cost Germany jobs and businesses. Because of regulations that require all renewable power be purchased, there are times electricity prices are negative (paying people to take electricity) in Germany. Because 100% fossil fuel backup is necessary, the German and UK governments must pay for idle fossil fuel capacity, because emergency backup is essential for a stable grid. These payments are called “capacity payments” and can be thought of as subsidies for coal power plants. These payments are above and beyond the normal excess power generation capacity required in the absence of renewable power purchase mandates.

One factor about wind power generation that is often overlooked is the space required. In table 1 we see that offshore wind generation is 2 to 3 times the cost of land wind generation. Yet, onshore windmill farms produce less than 2 W/m2 of land, this means to make any significant contribution to our power grid an enormous amount of land must be taken up. For comparison, corn-based ethanol produces 0.315 W/m2 of land, which is even worse. A modern oil field, on average, produces 90 W/m2 of land. The late David Mackay in his ebook “Sustainable Energy” estimates that even if 10% of the UK were covered in windmills they would not generate enough energy for the country.

Commercial solar has a similar space problem (6 W/m2), rooftop solar does not. But, solar has a severe capacity utilization problem. In 2012, Germany had an installed solar power electricity generation capacity of 254 TWh (terra-Watt hours), but it only produced 19.3 TWh of electricity, 8% of rated capacity. In other parts of the world the usable electricity from solar is higher, but it almost never exceeds 30%. German usable wind power is about 17% of capacity. By way of comparison most fossil fuel power plants achieve capacity factors of over 80% and nuclear 94%. It is the uncontrolled and unplanned intermittency and the wasted power generation that results in high costs. Further, although the stated goal of using renewables is to lower carbon dioxide emissions, the fact is Germany’s CO2 emissions have gone up. The country’s fossil fuel backup is coal because they have shut down their nuclear power plants.


The US mandate to use biofuels, mainly corn or cellulosic ethanol, has been a disaster. When the Renewable Fuel Standard (RFS) was enacted in 2005 it was hailed as a step toward energy independence, but it has been fraught with unintended consequences. Ethanol has a lower energy density than gasoline and when they are blended, the resulting fuel gets fewer miles-per-gallon than the original gasoline. Using corn to make fuel raises food prices around the world, which hurts the poor. And because corn requires fertilizer, growing more corn can be an environmental hazard. Further, ethanol, whether from corn, sugarcane or cellulose requires so much energy to produce it either barely breaks even on EROI or is negative, in that it may take more energy to make ethanol than it delivers.

As Captain Ike Kiefer writes inTwenty-First Century Snake Oil,”

“Some prominent figures and pundits argue that biofuels will increase our domestic supply of transportation fuel, end our dependence upon foreign oil, reduce military vulnerabilities on the battlefield, and generally improve national security. Biofuels are further promised to reduce fuel price volatility, reduce polluting emissions, reduce greenhouse gases, and even stimulate the economy. These arguments all fall apart under scrutiny. … uncultivated biomass produces biofuel yields that are far too small, diffuse, and infrequent to displace any meaningful fraction of US primary energy needs; and boosting yields through cultivation consumes more additional energy than it adds to the biomass. Furthermore, the harvested biomass requires large amounts of additional energy to upgrade it into the compact, energy-rich, liquid hydrocarbon form that is required for compatibility with the nation’s fuel infrastructure, its transportation sector, and especially its military. When the energy content of the final product biofuel is compared to all the energy that was required to make it, the trade proves to be a very poor investment, especially in consideration of other alternatives. In many cases, there is net loss of energy.”

Ethanol is corrosive, it contains oxygen, it attracts water, and can cause steel to crack. This means that it cannot be put in normal pipelines and often has to be trucked. Engines, pipelines and storage tanks have to be specially modified for ethanol, increasing costs and lowering EROI.

Some claim that burning wood produces “good” CO2 because the trees cut down for fuel will be replaced by trees that will absorb the CO2. This may or may not be true, but either way the cost of cutting down trees, planting new ones, preparing and transporting the wood to a power plant is so high, the EROI of burning wood is very low. It is much lower than the EROI of burning coal, even with a full set of pollution scrubbers on the coal burning power plant.

As Senator Roger Wicker said last year:

Ending the RFS [Renewable Fuel Standard] would … save money for consumers. For example, biofuels like ethanol in the fuel supply mean drivers are paying for fewer miles per gallon of gasoline. The Institute for Energy Research says the RFS has cost consumers an additional $83 billion since 2007.”

“In the last Congress, I introduced legislation to prohibit gasoline blends with 15 percent ethanol, which have been shown to cause engine damage.”

Nicolas Loris of the Heritage Foundation has written:

“Enacted in 2005 and expanded in 2007, the Renewable Fuel Standard—the ‘ethanol mandate’— decrees that American oil refiners must include a minimum amount of renewable fuel each year, increasing to 36 billion gallons by 2022. Fifteen billion gallons may come from corn-based ethanol; the remainder must come from other biofuels.”

As of July, 2014 only 50,000 gallons of cellulosic ethanol had been produced for the year, although the EPA target was 17 million. The EPA has repeatedly cut the cellulosic ethanol blending requirement, but the actual production remains extremely low. The customer (the refiners and the public) cannot buy what does not exist, but is punished for not buying it. According to Nicolas Loris:

“… until 2012, no cellulosic ethanol had been produced because it was not commercially viable. In 2012, only 20,000 gallons had been produced—far short of the 8.65 million gallon revised target. Consequently, refiners had to pay millions of dollars in waiver credits or surcharges to comply with the EPA’s minimum volume requirements. Refiners pass those costs on to the consumer, further inflicting economic pain caused by the RFS. In January 2013, a Washington, D.C., Circuit Court of Appeals ruled that the EPA’s target was an “unreasonable exercise of agency discretion” and vacated the cellulosic ethanol requirement required by the RFS.”

Not only has cellulosic ethanol production fallen far short of the government mandate, it has provided fertile ground for con-men and hucksters of all sorts. KiOR was supposed to produce cellulosic ethanol in Mississippi and was provided with millions of dollars in subsidies to help them get going. In November of 2014 they filed for bankruptcy and shut down after producing very little ethanol. The Mississippi Attorney General Jim Hood then said:

“[KiOR was] one of the largest frauds every perpetrated on the State of Mississippi.”

The reasons behind KiOR’s failure were simple, they couldn’t get their technology to work and when they did manage it, yields were far lower than KiOR had claimed. The plant was supposed to produce 13 million gallons of biofuels a year, it produced 133,000 gallons in 2013, sold another 97,000 gallons in early 2014, and then shut down.

Another failed cellulosic ethanol plant was recently sold to Alliance BioEnergy by Ineos Bio, which could not get their process to work. Alliance BioEnergy has just finished testing their new “CTS” process for converting cellulose to ethanol. Time will tell if this process proves commercial. Other companies that tried to make cellulosic ethanol have also gone bankrupt, these include KL-Energy, Range Fuels and Codexis biofuels. Codexis still supplies enzymes to pharmaceutical companies, but has shut down their cellulosic ethanol company.

In March of 2014 Mark Peplow wrote an article on cellulosic ethanol in Nature. He showed a map of facilities in North America, figure 4 is his map, updated by the author.

Figure 4 (Source Nature, modified and updated by the author)

While the 2014 Nature article was very upbeat and optimistic, the Abengoa facility, which was to “start commercial production in the next few months” is shut down and Abengoa has filed for Chapter 15 bankruptcy. In May of 2016 Scientific American published a more up-to-date article on cellulosic ethanol, entitled “Whatever Happened to Advanced Biofuels?” They summarized the situation in this way:

“Cellulosic fuels’ main hurdle seems to be economic. In April, 2012 Blue Sugars Corp. of South Dakota produced the first batch of qualifying cellulosic ethanol, a little more than 75,500 liters, then promptly went out of business. In 2013 no cellulosic ethanol was produced but by last year—after several DoE-supported plants came online—all five of those biorefineries produced a total of 8.3 million liters of cellulosic ethanol, according to the U.S. Environmental Protection Agency, which administers the RFS. Already, Spanish multinational corporation Abengoa’s cellulosic ethanol plant—which opened in 2014 in Hugoton, Kans.—sits unused due to technology troubles as well as Abengoa’s bankruptcy. That plant consumed a $132-million loan guarantee as well as a $97-million grant from the DoE before idling.”

The EPA now requires 1.2 billion liters of cellulosic ethanol for 2017, which will never happen. Only 8.3 million liters were sold in 2015 and probably less than 10 million liters were sold in 2016. We are now 12 years past the ethanol mandate and there is not even one commercial cellulosic ethanol plant in production. It seems very unlikely we will see one soon.

Captain Ike Kiefer summarized the situation quite well in this WUWT comment on March 1, 2017:

“The technological feasibility of cellulosic ethanol is readily assessable for those willing to look and think. It is essentially the same challenge as a paper mill making paper from trees, except after extracting the cellulose, you are trying to make a very expensive additional conversion from solid to liquid that involves the steps of colonization and fertilization, fermentation, distillation, and dehydration. And the liquid product (ethanol) has a lower market value than the solid (paper). [This is a] pretty awful business model, especially when we know paper mills are barely hanging on. Also, when the EROI of corn ethanol is less than 2:1 … and it is chemically 5 times harder to hydrolize cellulose than corn starch, there is a pretty good chance EROI is going to be upside down for cellulosic ethanol.”

Finally, we need to consider that numerous studies show that more than 70% of the increase in food prices around the world are due to making biofuels according to Kiefer’s paper. Both the World Food Program and the Food and Agriculture Organization of the United Nations have called for all G20 nations to drop their biofuels subsidies. Grain prices have increased for the poorest consumers as much as 50%, increasing hunger and malnourishment for the global poor.


While it is clear that fossil fuels and nuclear are cheaper, more flexible and more reliable than renewables, it is less clear by how much. Kiefer reports that, since 2007, the military has spent $68 million on 1.35 million gallons of biofuel, over $50 a gallon. This does not include the cost of federal and state biofuel subsidies. Conventional fuel would have cost the military $8 million. As Hall, et al. (2014) concluded solar and wind cannot be used without lower EROI fossil fuels or nuclear power to back them up, they are not “base load” technologies. Corn based ethanol requires a lot of fossil fuel energy to grow and transport the corn and even more to make the ethanol. Cellulosic ethanol is not currently economic and may never be. As discussed in Kiefer’s paper and others, renewables are not displacing fossil fuels, they are accelerating their use. The only way to displace fossil fuels and nuclear power, is to produce renewable fuels that have a higher EROI than either fossil fuels or nuclear. This can happen if the fossil fuels or nuclear EROI goes down or the renewable EROI goes up, but it must happen. Otherwise renewables are simply a parasite on conventional power and as renewable use goes up, fossil fuel and nuclear will go up, just like in Germany and Denmark today.

Currently, the true cost and EROI of renewables is unknown. The only way to compute it accurately would be to find a renewable fuel that is made using only itself as a source of energy. As an example, this would mean a solar powered factory for solar cells and panels, solar powered vehicles and equipment for transportation and construction, backup batteries made in a solar powered, factory, etc. Would this facility be economically viable? What about analogous facilities for ethanol and wind? But, no such facilities exist, so we don’t know. Only when the dependence of renewables on fossil fuels and nuclear power is fully understood, will the calculation will be possible. To compute the true, backed up, EROI of solar, biofuels and wind requires deciding where to draw the line between costs that should go into the fossil fuel EROI calculation and costs that should go into the renewable EROI calculation. This is not a trivial problem. But, the final EROI for renewables, once calculated, are all likely to be less than 3:1, which is the practical lower limit and clearly in recession territory.

March 13, 2017

One thought on “Renewable Energy, what is the cost?

Leave a Reply

%d bloggers like this: