Windpower – Just another economical power source or a premium for clean energy?
After my recent tour of the Solano Windmills just outside San Francisco, I thought, OK this is a good time to finally start this blog. Nextera energy was so kind to host us and provide us with a great tour, and also thanks to PG&E and the folks at the CPUC (The California Public Utilities Commission) for setting this up. I remember years ago watching the movie “Less Than Zero” and at the very end, was a scene where the main characters in the film went on a road trip and drove by a huge wind farm. What a surrealistic vision that was; seeing these massive windmills spinning around, churning out clean energy; an awe inspiring utopian vision of a smart and benevolent government making the investment to take care of it’s citizens and the environment. Now years later I wanted to see it for real – and ask a lot of questions. When I visited Nextera’s High Winds facility at the Solano Wind location I was not disappointed. As I drove across the Bencia Bridge just outside San Francisco, I glanced over my right shoulder and could see the windmills from more than 10 miles away. This was exciting. I thought about all the work I had done in alternative energy, and now I was going to find the answers to many of my unanswered questions; answers I could only find by talking to the line people running these facilities. These were answers I would not find on the Internet or talking to energy experts.
As I approached the wind farm I thought of 2 questions on my mind. Firstly, each of the largest wind turbines are about 2 Megawatts (MW), enough to power somewhere around 200 – 750 homes depending on who you talk to. Can a single turbine power that many homes? Secondly I thought of the BIG question: Can clean electricity only be generated at a premium cost over fossil fuel (coal and natural gas) and nuclear energy and thus be a difficult sell to the public, and if so, what can be done to make it competitive. Upon entering the facility, all I could think about was the first question. I thought to myself, “Woe, these things are huge, who’s job is it to put these up and maintain them? I’m glad it’s not me!” I walked up to the base of one of the windmills and first noticed that not only was it huge, it was pretty loud – wind sound loud. And as I stood under it, of course the first thing I thought was, do these blades ever come flying off? My host assured me that this had never happened, except for occasional damage during lightening storms. As I stared at the blades swooshing around just above me, I could feel and hear the power. These things were huge! I could image that this single turbine could maybe power several hundred homes. I tried to imagine this huge spinning blade physically spinning hundreds of air conditioner compressors. As I settled past the shock of just being near these things, the real questions finally started. First I noticed that they were not spinning as fast as I thought they would. My host explained that they spin at a constant speed (I think it was 18 RPM) much like a high performance airplane prop engine. Constant speed RPM is optimal, so the blade twists to maintain a constant speed, and this also allows it to easily generate a 60Hz output, ready to go to the transformers for distribution. I said to my host, too bad you can’t turn the whole thing to always be facing exactly into the wind. I was surprised to find out that the whole assembly moves to always face into the wind! Wow, these are impressive, I immediately followed up with my first technical question – “How much do these cost?”. He responded $4M installed for a 1.8MW turbine with a 25 year life. That comes out to about $2500 per Kilowatt (KW) of installed capacity (that is $2500 to power ten 100-watt light bulbs for 25 years – if you assume the windmill runs at full capacity 24 x 7 which it doesn’t and doesn’t take into account capital, maintenance and distribution/transmission costs). Once he threw a number out at me, I was pulled back into remembering why I was here, and put my financial thinking cap back on. The real tour was about to begin. I thought of the key question I wanted answered; is wind energy economical against traditional coal, natural gas, and nuclear plants, or will wind and other forms of clean energy require the government to get involved in a big way. And of course the other question in the back of my mind which I didn’t expect the wind expert to answer is, can the US be competitive in a global economy if the government mandates a certain percentage of clean energy and competing economies do not. But that’s a question for another day. Let’s focus on wind energy first.
Taking all the information about electrical energy generation/distribution and wind power that I had collected from so many sources, and making sense of it in order to answer the big question is a daunting task. But after my visit, I had enough pieces of the puzzle collected to finally put together a reasonable hypothesis. I should mention that one of the pieces of the puzzle was getting acknowledgement from a prominent person in Washington who works in the energy field that serious discussion of energy policy is not happening; i.e. there doesn’t appear to be a readily obvious and reliable source that can answer the big question. So through a combination of powering up the spreadsheet and thinking through the information I had collected from people and various Internet sources here’s what I came up with.
The key to this question is converting all this information to the bottom line which is cost per Kilowatt Hour (KWH), and understanding what factors affect it (the amount of wind, time of year, utilization, market dynamics, transmission/distribution, etc.). The average cost of electric power in the US is $0.12 per KWH. There are lots of variations by state and time-of-day, but here we are going to focus mostly on averages in order to put wind power into perspective; we will deal with de-averages at another time. Firstly, how do you convert Megawatts of capacity into KWH? You use the load factor which is basically the amount of power generated on average throughout the year relative to the maximum capacity (i.e. how often does the wind blow, and by how much). The load factor of Solano is around 30% – you could actually only power three (not ten) 100-watt bulbs for 25 years for $2500. You can put a wind farm anywhere there is wind and real estate to do so, but you clearly want the load factor to be greater than some minimum amount to make it economical. I believe load factors for wind facilities are in the 20% – 40% range. So a 1.8MW turbine generates an average of 0.54MW, which comes out to almost 400,000 KWH per month. Sounds like a lot, what does this mean? Well the rest is pretty simple, if it costs $4M for a turbine, and it lasts 25 years, this comes out to approximately $0.09 per KWH (assuming a capital cost of 10%). Cool, less expensive than $0.12…… But we’re not done yet.
What about operating costs? Each of these turbines requires maintenance twice a year, plus an oil change every 2 years. Each of these activities takes about 2 skill technicians 2 days. Plus you have all the operations people, leases, property taxes (which in Solano’s case was the biggest operating cost of all!…), etc. This comes out to about another $0.01 per KWH, so we’re at $0.10 per KWH. This facility has 90 turbines, so if we had a larger facility would we save much? At the level of 90 turbines, it turns out most of the costs are variable (i.e. driven by the number of turbines), so let’s say this cost is pretty reasonable for large scale facilities.
What about transmission costs (i.e. the big towers transmitting power across long distances). These numbers are hard to come by, but averages in the US appear to be around $0.004 / KWH, and power losses of around 7%. But it’s not that simple because wind facilities are often located in remote locations relative to power plants, and the amount of power generated is less than a typical power plant, so the costs will be higher. By how much? Transmission lines are about $1.5M per mile, but the cost is driven much by whether you can exercise eminent domain (i.e. an unregulated private company may have higher costs than a Utility). Transmission losses appear to be around 6% plus 5% per thousand miles, but this number in particular is very difficult to come by because there are a variety of technologies and no single source appears to clearly indicated what this number is. Finally, a nuclear power plant for example, is around 1000 MW, whereas this facility I visited was around 700 MW – The Nextera facility was only around 162 MW, but there were other wind farms in this “wind location” which total 700 MW. But if you apply a load factor of 30% you are only at 210MW – i.e. less KWH to spread costs across, thus higher transmission costs. A nuclear plant’s equivalent “load factor” (referred to as capacity factor) appears to be around 90%. If you take all of this into account, you come up with a costs of somewhere around $0.004 – $0.02. But this is driven largely by whether there are already transmission lines there (e.g. A wind facility near a pre-existing power plant) and how far those transmission lines are. If you’re near a pre-existing power plant, $0.004 is a good number. If you have to build transmission lines for a facility similar to Solano (700MW with a 30% load factor), but you need 200 miles of transmission lines, it’s around $0.02. If you need 400 miles, it would be double. If you only had 350MW, it would be double again, etc. In Solano’s case, they are in a great location, near PG&E’s transmission lines, so maybe their costs are around $0.005, so we’re up to around $0.105 per KWH total costs, whereas a remote location in Texas which needs a 200 mile line, would be around $0.12, just about the same as the average power grid costs.
But there is one more very important and significant cost left – Distribution. Transmission lines go from the power source to a town or city for example. Distribution are all the wires you see connecting up to homes and businesses plus the local utility’s operations in maintenance, meter reading / smart meter installations, billing, legal, operations, profit margins, etc. This number is not easy to come by, but it appears to be around $0.06 per KWH; the single biggest component of cost, more than traditional power generation which runs around $0.018 – $0.05. So for Solano, we’re at around $0.165 / KWH, whereas a remote location with dedicated transmission lines of 200 miles would be around $0.18. Numbers from the department of energy indicate around $0.17 per KWH for wind onshore (and $0.19 offshore), so we’re pretty close. It turns out there are other costs like dealing with planning, avian/critter studies, and zoning at a County, State, and Federal level over a 5 – 7 year period in California (about 2 years in Texas) which is required to build a facility, but it turns out if you have a large enough facility and company which operates many wind facilities, this does not contribute significantly to total costs. There is one element of cost which I did not quantize which are the roads that need to be built at the facility, but let’s assume for the moment these costs are already built in, or are small relative to everything else. In the case of Solano, the facility is located on farms and the farm owners love having the roads, so perhaps the wind facility got a deal on their lease that offset the cost of the roads.
Best case, wind appears to be $0.165 / KWH and the US average is $0.12 / KWH. If you compare an optimal wind facility (a lot of wind, and near transmission lines) with traditional, dirty power, it is significantly more expensive. Cheap power is most often the dirtiest, you can’t get around that. Let’s compare just power generation (not transmission and distribution), costs:
Increasing order of costs (% indicate percentage of total US power generation):
- Hydro: $0.01 (6%) – Ideal, but limited unless you live in Switzerland
- Nuclear $0.018 (20%) – How dirty this is depends on your view of nuclear waste issues
- Coal: $0.023 (50%) – The most popular fuel throughout the world, plenty of cheap supply for many decades
- Natural Gas: $0.05 (20%) – About half the CO2 emissions relative to coal
- Wind: $0.05 – $0.09 (2%), the $0.05 seems low, but these may be older wind facilities in ideal locations
- Fuel Cells: $0.10 – $0.20?? – Very experimental
- Large scale Solar Thermal: $0.15 – $0.20 – Experimental, but very promising. Key issue is transmission costs.
- Solar Photovoltaic (Solar panels): $0.20 – $0.40 (< 1%?)
So wind is not quite cost competitive, but perhaps with tax benefits, subsidies, etc we can make it competitive? Well, there is one important factor we need to consider before we ponder our conclusions. Wind is not getting cheaper, it’s getting more expensive; the cheapest wind has already been deployed. The best wind locations have already been taken – lots of wind, large acreage with minimal zoning issues, near transmission lines. There are a lot of great wind locations left with 1 major issue, most of the ones that are left are nowhere near transmission lines – think remote areas of Texas, the Midwest etc. Well according to our calculations, there is a premium of somewhere in the $0.015 – $0.04 range or even greater if your facility is more than 400 miles from existing available distribution available. The costs would be even greater still if we are talking about transmitting a great deal of power from the Midwest to the coasts. How much more? Hard to say because I can’t find enough data about very long haul distribution. There are similar issues with generating solar power in areas like Arizona and shipping it throughout the country. In addition, the cost of wind turbines is going up, as my contacts have told me, it’s a sellers market. Perhaps this will change over time, but for the moment we must assume that wind has a premium of approximately 40%, best case.
The economics of clean power generation (e.g. solar, wind tides, etc.) vary significantly by location, the best locations being less expensive. So there are several ways to look at these economics:
- How much of a premium is clean power in volume?
- How much subsidies or tax benefits do we need from the government to reach grid parity (the same price as the power grid) and must these subsidies be sustained, or will the economics eventually drive down the actual costs to grid party?
- How much clean power can we generate at grid parity? Wind seems to have already reached past its peak at 2%, but I’ve heard figures indicating that this number could be as high as 10%, although I’m not clear what in particular will drive down the costs. Wind turbines have been around for a long time and they represent approximately 90% of the total power generation costs and around 56% of total costs – and their costs are going up.
One more thing to consider – who currently purchases wind power and why? Are they doing it because it’s clean, because it’s economical, or both? At Solono, they are privately financed (i.e. no government subsidies or tax benefits) and they sell 100% of their capacity, and their customers are roughly equally split 3 ways:
- Utilities – long term contracts
- To wholesalers through long term contracts who deliver the power to municipalities. These often have the name “renewables” or “clean” in their name, but it’s not clear if small municipalities are purchasing this power because it’s clean, or simply because they get a good deal on the price. I’m told, it’s mostly the price, and that it being clean or renwable is a minor benefit.
- Selling to the spot market (customers who need to purchase power on demand when they run out of capacity).
So that’s the data that I have. Let’s zoom out, and think about what does this mean in term of addressing our big question: Can clean electricity only be generated at a premium cost over fossil fuel (coal and natural gas) and nuclear energy and thus be a difficult sell to the public, and if so, what can be done to make it competitive.
There is NO evidence that people will pay more for green energy coming from the power grid in any non-trivial volume, and more importantly that they can. On the demand side, the smart power grid theoretically will allow customers to choose green energy through interfaces with their smart meter and the new smart grid infrastructure. For example, they may choose to have 20% of their power come from green sources at a maximum price of a $0.05 premium, or $20 a month maximum premium, or they may want the charger for their new electric car to only charge from clean energy sources. But this is problematic from many perspectives. Most people don’t understand that electric cars actually pollute more than a hybrid, as 70% of power in the US comes from fossil fuels. The focus is mostly on the belief that electricity is clean, and the cost. On the supply side, there is no evidence that individual municipalities will choose green power unless the cost is at or pretty close to grid parity. There are no plans to allow customers to choose where their electricity comes from in the same way for example, that you can choose your long distance carrier, even though the smart grid allows for that option through accounting mechanisms. The power grid infrastructure is just too complicated, and the way that bulk deals are made is the primary driver of consumption, and that is about cost, supply and demand of a commodity; electric power. In order to understand the primary driver of demand for clean power, you need to understand how electricity is bought and sold, regardless of whether it is coal, nuclear, wind, solar, etc. And this is complicated, but let’s take a shot at summarizing all of this.
The average cost of power is $0.12 per KWH. But it varies significantly by the time of day, the location/state, the time of year, etc. Also many locations throughout the US are at capacity, so they will pay a premium for excess power on the spot market (on demand), and they may choose a long term contract for extra capacity so they don’t have to purchase it on the spot market. The task of ensuring that you don’t run out of capacity and the costs are manageable, eclipses any concern about green energy. Green energy will be chosen over dirty energy only if it’s a tie breaker – i.e. it’s at the same cost as other power, and it’s available. So wind has to compete with coal, nuclear, natural gas, etc – on the same terms as dirty energy. And wind is less predictable, thus harder to use as base power. So in summary, since the demand for power is growing, and the costs are going up for traditional power, and we are running out of capacity, wind is simply another alternative power source and it must be competitive. That it’s green is mostly irrelevant. Perhaps a large corporation will step up and be willing to make a wholesale deal for wind power and pay a premium for it to be a good citizen, but there is no evidence that this is happening and that this will not significantly hurt a company’s competitive position.
I went into this wanting to believe that a case could be made for large scale wind power without government involvement – i.e. that the markets could deliver clean wind power in substantial volume. In the process of working with utilities, regulators at the State and Federal level, electric vehicle and infrastructure vendors, etc, I have been fighting for technology (for example in the smart grid and electric vehicle infrastructure) to allow this, and I have chosen to believe all the data I could to support this position. However, as I dug in further, I saw more and more evidence that indicated that this was not the case. Power is simply a commodity, and the trend towards green technologies like electric vehicles is missing an essential component – the appreciation of the importance of the impact of our electric power infrastructure.
So the final missing piece of the puzzle is why do companies invest in and purchase wind power when it would be cheaper to just build a coal plant. I can think of several possibilities, and perhaps the truth is a combination of all of them:
- Coal plants require approvals, zoning issues, etc. Since power demand is constantly rising, wind facilities might be more readily built in some locations
- Many facilities were built years ago when the belief was that certain segments of the market would pay more for green energy. Maybe a small portion are, or maybe this capacity is being sold below cost.
- Maybe some facilities were built with the expectation that regulations, subsidies, tax benefits will eventually favor wind or mandate a certain minimum amount of alternative, clean energy. Maybe in some locations this is a key driver.
- Maybe some were subsidized by the government – For example the Altamont Pass Wind Farm outside San Francisco was deployed in response to the 1970’s energy crisis via tax incentives for investors.
- Maybe some are subsidized by the wind industry
- Maybe in some markets (e.g. California), there is a higher propensity to pay a premium for green energy and/or electric power rates are higher. California has the highest rates in the nation (around $0.014) so the premium for wind is less than average.
- Wind is somewhat scalable – you have to build the entire coal plant to generate a single Megawatt, whereas you can deploy the wind turbines over time. This only makes sense if you have an ideal facility near transmission lines (e.g. a pre-existing wind farm).
Maybe things will change in the future. Peak power demand will continue to stress the supply. Maybe turbine costs will come down (but it would have to be at least by at least 45% to compete with natural gas which is still more expensive than coal). Maybe we will finally have a significant carbon tax initiative. Maybe the advent of electric vehicles will raise the awareness of this issue. But without an energy policy at the Federal level, it appears that this will be unlikely in the foreseeable future. Additional wind facilities require a lot of capital, especially if you have to build a 200 mile transmission line. There are the economic costs, then the risk considering that there is much uncertainly about the status of green energy regulations, subsidies, etc. How this will play out is up to the way the wind blows. However, if we are serious about green energy, for example if we are serious about electric vehicles, we ought to be smart and be just as serious about energy policy at the Federal Level to continue this kind of conversation. Why isn’t this happening? Well that’s another fascinating story, for another time.