Update: I think I buried the takeaway points a bit too far down below the industry background and definitions, so I’m bumping this summary up to the top (so you can understand the problem before reading more details on what leads to the problem and potential solutions):
1. Wholesale electricity is sold to the grid on spot markets in many places. Under such a system, power plant operators have the option to sell electricity from their power plants to the grid on a day-ahead, hourly, or other short-term basis. As long as the power plant operators sell electricity for more than the marginal cost of producing electricity (which almost $0 for solar and wind without subsidies, and below $0 with some subsidies), they make some money.
2. In a system where all power plants are solar or wind, however, that means everyone is bidding each other down to almost $0 (without subsidies). But selling electricity for just barely more than $0 for years is going to result in never earning back the money invested into the power plants to set them up (or company overhead) — non-marginal costs.
3. So, if we project out to a 100% renewables future, or even a future in which renewables are dominating but not capable of providing 100% of electricity, we need to figure out a way to compensate power plant owners for non-marginal as well as marginal costs — something different from the simple spot markets that many grids run today. How best to compensate for non-marginal costs in a competitive way is what is quite unclear at the moment.
[This is quite an “inside baseball” topic for CleanTechnica, but I think it’s a fascinating one and one that is ripe for discussing in this community and more broadly.]
For several years now, there’s been a bit of concern about how the energy industry will proceed as solar and wind take more and more of the market — and I’m not talking about integration or flexibility, but about the insanely low marginal (operational and “fuel”) costs of solar and wind farms (close to nilch).
To understand the issue, you have to have somewhat of an industry insider’s view of how electricity markets work (but not too much, so don’t run away as I explain a few key industry terms and processes).
In certain electricity markets, to get your power plant’s electricity onto the grid, you have to outbid other electricity providers on the wholesale spot market (under what is referred to as a “merit order” system). Given that your power plant is already built, what you have to look at in order to determine how low of a bid you can offer is what it costs your power plant to actually produce the electricity at a certain point in time. If you can produce another kWh of electricity for 3 cents, and you can sell it for 5 cents, you’re good to go. However, if you can produce another kWh of electricity for no fewer than 4 cents, and you can’t sell it for more than 3 cents, you lose out and you can’t sell the kWh of electricity.
Since wind and solar power have nearly no marginal electricity production costs (sunshine and wind are free, and no one has to operate the power plants as the sun shines down or the wind blows), they can outbid practically everything else. Thus, the more that solar and wind electricity becomes available for bidding, the more higher-marginal-cost electricity options get pushed off the table. At some point, of course, if a coal, nuclear, or natural gas power plant can’t sell enough electricity to cover its overhead (and marginal) costs, it is not financially viable to keep the power plant open.
As one study has indicated, wind energy in Germany saves German consumers €5 billion per year due to the “merit order effect.” This is a good system for lowering the cost of electricity, and solar and wind offer a lot of value for being most capable of doing that.
However, at some point, the ~$0 marginal cost of solar and wind could result in these renewable technologies pricing so much electricity generation capacity (power plants) off of the grid (including other solar and wind farms) that there isn’t an adequate amount of generation capacity for high-demand times.
Furthermore, if it were just or primarily solar and wind competing with each other, and they go down to, say, a bid of 0.5¢/kWh (or whatever would cover marginal costs + a slight profit) on a regular basis, that would end up driving the power plant owners into bankruptcy. Over the life of a plant, after all, the initial investment costs need to be recouped from selling electricity to the grid/consumers.
By the way, for some deeper background pieces on the above topics, check out some of our previous articles on the merit-order effect:
Wind Power is Making Electricity Cheaper (Exxon: Wind to be Cheapest Source of Electricity) — That article was from 2011. In many places, wind is certainly now the cheapest source of electricity.
Industry professionals and researchers have been debating for years how to deal with this logical “game over” conclusion to the current system.
A great article titled “Do We Really Need Solar That’s Too Cheap to Meter?” (which built off of this article in this journal Nature Energy) inspired me to jump into this topic. It covered several related topics, but jumping off of it all, the question I want to highlight/explore a little bit is “How should we modify the system so that it doesn’t collapse on itself?” Here are three ideas:
- Capacity markets: To some degree or another, in some way or another, a capacity market compensates power plants for offering latent electricity generation capacity that can be dispatched as needed, and especially that can be dispatched quickly. (Click on this link, this link, this link, or this link to read more about capacity markets — or the link a little further up, which explains the idea much more simply and clearly.)
- Modified spot markets: This is a bit similar to the above, but pricing in the extra security of reliable generation capacity for all times of the year is not done through direct payments for that capacity. Rather, the idea is that either 1) power plants can “economically withhold capacity” while bidding on the spot market, in order to raise prices, or 2) higher scarcity prices, or spot-market-clearing prices, could be determined to preserve economic viability for enough power plants. (This is a new concept to me, and I may be misinterpreting the arguments made for this approach, so please have a look at page 25 of this document as well and check my interpretation/logic.)
- Long-term power purchase agreements (PPAs): To guard against power plants having to shut down and inadequate electricity supply, utilities can also contract electricity under certain prices from individual power plants (over the course of, say, 20 or 30 years). Even enough is contracted in this way, from an adequate variety of sources, in order to guarantee electricity security, the problem is essentially solved. However, getting the calculation precisely right for the lowest potential cost using such long-term contracts is practically impossible. Thus, while PPAs can be a useful part of electricity security while providing viable compensation for power plants, these alone would not offer the most cost-competitive solution. However, this would offer the simplest solution, which cuts costs in other ways. And if the costs get low enough, how much is it actually worth to have a complicated capacity market or modified spot market system and bureaucracy?*
Better demand management: Note that, if the electricity grid could cut back demand 100% when needed, there would be no need for backup generation capacity at all. That’s an absurd example, but the point it is trying to demonstrate is that, as electricity demand management gets better, the grid is more capable of working around periods of relatively high electricity demand and low electricity supply. This isn’t a solution for the concern described above, but it’s a helpful tool to limit the breadth of power plants needed, the capacity/backup needed, and thus the challenge of compensating power plants for not actually producing electricity but simply being capable of it. So, it’s not surprising that the UK’s Capacity Market and other capacity markets and proposals have been “designed to support the development of more active demand management in the electricity market.”
As part of better demand management, better real-time electricity pricing (or nearly real-time electricity pricing) for retail electricity consumers could help a great deal to cut demand at times of limited supply. More accurately pricing electricity would result in less “waste” in the electricity capacity market. Furthermore, better transmission and storage capacity can help electricity generated to “go farther” (i.e., not be wasted or undervalued).
I haven’t done a PhD in this topic, so I’m sure I’m missing some key arguments and ideas. I know we have plenty of industry professionals reading CleanTechnica, so I encourage you to chime in below in order to improve this story and the broader discussion of what to do about the growth of low-cost solar and wind power.
*Note: I think this was essentially James Wimberley’s point underneath the “Do We Really Need Solar That’s Too Cheap to Meter?” article, and I think it’s a very compelling one.
Charts by 1) Richard Caperton for Think Progress, and 2) Lazard, modified by me to highlight today’s low solar power costs, which are now lower than all other electricity generation sources other than wind.
Appreciate CleanTechnica’s originality? Consider becoming a CleanTechnica Member, Supporter, Technician, or Ambassador — or a patron on Patreon.