The market has spoken – clean energy just became the cheapest source of electricity

Spreadsheets, analysis, and predictions are all great. But nothing compares to the free market if you want to really understand the trajectory of clean energy pricing.

A recent bidding process for utility giant, Xcel Energy, just rocked the industry. Bids for clean energy projects including solar, wind, and batteries, came in lower than anything ever seen before.

This wasn’t just a few crazy, low bids, it was hundreds. These rock bottom prices weren’t the lowest bidders, they were the average of all bids.

Xcel told state regulators that “the response to this solicitation is unprecedented.”

From Xcel’s report on their bid response: https://www.documentcloud.org/documents/4340162-Xcel-Solicitation-Report.htm

Who is Xcel and why does their bidding process matter?

Xcel is a US$11 billion gas and electric utility giant that serves 3.3 million people across 8 states.

Last year, Xcel reached an agreement to shut down two Colorado coal plants 10 years earlier than planned. But there was a hitch. The plants would only be shut down if Xcel could replace their 660 megawatts of power with lower-cost clean energy alternatives.

On August 30, 2017, Xcel put out a solicitation for bids to see what the market could offer.
The last time Xcel did an open solicitation, they received 55 bids in totalThis time 430 bids were received. 251 included solar, 119 included wind, and most surprising of all, 105 included storage.
The sheer number of responses was stunning but it was the price of the bids that made international news headlines.

How do these bid prices compare to the current market?

Source: Freeing Energy Project

According to the US Energy Information Administration (EIA), a new coal plant would produce electricity at a cost of 12-14 cents per kilowatt hour.  But, no one is building new coal plants.
So analysts fall back to looking at the marginal cost, or just the cost of running an existing coal plant. The EIA says that’s about 3.6 cents per kilowatt hour – 2.5 cents of coal and 1.1 cents for operations and maintenance.

The world’s collective jaws hit the collective floor when bids for wind came back to Xcel at 1.8 cents per kilowatt hour and solar came back at 3.0 cents.

Of course, it would be fair to point out that solar and wind are not a true replacement for coal because they only run intermittently. So, the really big news from the Xcel solicitation was the more than 100 bids that also included storage. Storage can be expensive but in the Xcel responses, the average bid for wind+storage came in at 2.1 cents and solar+storage was 3.6 cents per kilowatt hour.

Yes, 3.6 cents per kilowatt hour for solar + storage is the same as the marginal cost of operating a coal plant. Xcel did not provide details on how much storage was included in the bids so it’s likely that true 24/7 electricity would require more storage than was bid.

The Freeing Energy perspective…

In the world of energy, this is a tectonic event. Years earlier than even the most optimistic projections, clean energy is less expensive than fossil fuels. Even with storage, these bids remain cost competitive with every form of traditional energy.

The Xcel bidding process highlights the emerging story on clean energy. It started out as a way to help the planet by cleaning up old, dirty power plants. But, it ended up being a story about saving money. They say you can’t have your cake and eat it, too. But, when it comes to wind, sun, and batteries, it looks like maybe you can.

15 thoughts on “The market has spoken – clean energy just became the cheapest source of electricity

  1. Looks so good until the line near the end “Xcel did not provide details on how much storage was included in the bids so it’s likely that true 24/7 electricity would require more storage than was bid.” So, no, you aren’t price competitive with Combustion Turbines or Coal.

    FURTHER, the price of combustion turbine bids is $4.80/kW*month or $0.007 per kWh so your graph is misleading because you don’t include relevant competing bids in the graph.

    I look forward to your reply and I hope you correct the misleading article.

    1. In the data that Xcel published, they used capacity numbers (power or KW max power over a month) for the fossil fuel data and LCOE number data for renewables (energy). My understanding is that these are apples and oranges and can’t be compared directly. The gold standard for LCOE cost comparisons is from investment bank, Lazard: https://www.lazard.com/perspective/levelized-cost-of-energy-2017/. The Xcel numbers are way under this, probably because they are very new and the Lazard data covers the entire last year.

      I agree with your point intermittency. But as you probably know, wholesale markets are measured hour to hour (or even more often) so the actual LCOE cost really does drive the economics of the marketplace. Also, the cost of electricity for commercial and industrial customers varies throughout the day (as opposed to residential which is usually a fixed rate in most states). The most expensive electricity is almost always peak summertime rates because the power company has to generate or purchase wholesale electricity from the most expensive sources. At times like this, when preventing blackouts is most important, the cost of fossil fuels are at their most expensive just when the cost of solar is usually the least expensive.

      Bottom line: I agree that intermittency is a real challenge and the numbers in the article are not all apples to apples. That being said, the low cost of solar and wind are extremely valuable to the marketplace, particularly at those times when electricity is most important (hot summer days when air conditioners are running full tilt).

      One other point: the cost of batteries is dropping quickly as well. I am hearing about contracts closing today for $90khw. If this scales, solar+battery will be cost competitive with dispatchable resources like coal and eventually natural gas.

  2. Bill, it’s good to see true quantitative analysis like this, so thank you. There’s not enough of this analysis in the usual news and we need more of it.

    Biodiesel is closer in price to fossil fuels than many old-school O&G folks like to admit, and even with plentiful renewable electricity, there will be a continued need for traditional fuels for heating, including water heating to support dishwashers, dryers, washing machines, showers, etc. There are renewable fuel alternatives that are likely already viable as replacements to motor fuels and propane/LP gas. Just throwing that out there as a possibility for a future article.

    1. Brian – great point on biofuels. What kinds of fuel sources are most interesting to you? What’s your view on the water requirements of crop-based biofuels like corn?

  3. Bill, one point I’m missing in the article is a distinction or clarification on the renewable bids. Coal is bid @ 13 for a new build, but opex is only 3.6. Are all the renewable builds including the cost of the build, or are they operational expenses only?

    I guess worded differently, are all the prices in the charts for new builds, with the exception of the coal operational and fuel line?

    Thanks!

    1. Great point. We could have been clearer and will try to do so in the future.

      To your question… I am fairly certain that the Xcel data on solar and wind includes ALL costs: initial building and long term operational costs. The Lazard and US Energy Information Administration data, which is mentioned elsewhere in the comments here and on reddit, is definitely ALL costs for all power plant types – initial construction, ongoing operations, and fuel costs for nuclear, natural gas and coal plants (solar and wind have no fuel costs). The industry refers to these kind of metric as LCOE – levelized cost of electricity.

  4. @Bill: does your cost also calculate the added costs for environmental damage and health concerns from generating classic fuel sources which everybody pays for in hidden costs but which are not included in the kind of numbers that contracts talk about?

    1. Great question.

      No. We do not include health or environmental impact for most of our research.

      This was a hard decision because the impacts on traditional energy on health and the environment are already substantial. Worse yet, our children will be stuck with even more massive costs incurred by the decisions our generation is making today.

      The two primary reasons these broader costs are not used are:

      1. The entire reason that the Freeing Energy Project came into being was the realization that clean energy is now economically competitive purely on a current dollar to current dollar basis. Even if you ignore health and environment costs, and even if you level the playing field of subsidies, energy from the sun and wind are quickly becoming the least expensive source of electricity.

      2. The health and environment costs can be calculated many different ways so even experts projections vary widely. These differences are often cited by opponents of clean energy as a way to question the validity of all clean energy economic analysis. Per point 1, we are attempting to focus our analysis largely on data that is widely accepted and more uniformly calculated.

    2. The environmental impacts of battery manufacturing vary tremendously on the types of batteries and the regulations around them. I’m encouraged as I see new, grid-scale battery chemistries emerging that use fewer (or no) rare earth materials or mining-intensive materials. The Freeing Energy Project is working on a comprehensive health and environmental impact on all energy related technologies and batteries will be included in this. The EV industry seems locked into Lithium Ion and Cobalt and mining of these materials will likely get a lot more scrutiny (which is a good thing) as the volumes increase.

  5. The first 5 lines are the prices for “standby capacity”, ready to use when needed, and paid regardless if it is used or not. The remaining lines are for actual energy delivered on an “as available” basis (i.e. the sun is shining or the wind is blowing) and not assured to be available at all times. A utility has a variable demand across time of day and seasons. What they want to do is fill that demand at the lowest cost. Today that means filling the demand with lower cost sources first, and then using the standby capacity when necessary. The standby capacity get paid for the actual energy delivered *in addition* to the standby charge.

  6. Apples and Oranges…you refer to Lazzard. Lazzard, nor Xcel, include the cost of the transmission and ancillary services to get the power to the load. Wind is rarely located anywhere near load centers and requires large investments in transmission to deliver the power. Ancillary services are what keep the grid stable; VAR support, frequency response, voltage support and reserves. Renewables need to include T&D costs plus ancillary services to come up with the true cost to serve the load. Wind in Kansas cannot support the grid in Minneapolis. Solar with batteries is DC power…does not supply frequency response and only gives VAR support if run uneconomically. You also are playing games with your chart…you need to include natural gas because it is the cheapest by far when you include all the costs. Xcel’s lowest cost option would be to replace the coal unit with a combined cycle unit in the same footprint…then you have the transmission to support it already there. But, they can make more money by installing wind/solar plus spend a fortune on T&D plus install a gas turbine in Minneapolis to support the grid. All that goes into ratebase and is why retail electric rates have gone up 50% across the country in the past 10 years. And they are going to continue to climb as utilities have figured out they can make a ton of money off wind and solar and commissions won’t fight them.

Leave a Reply