What We Can Learn from Germany’s Windy, Sunny Electric Grid
Today’s release of the final Clean Power Plan by President Obama ushers in an exciting period of change on US power grids. Wind and solar energy will get a big boost through the plan. The plan recognizes that to sustain public support for this strategy, the costs of integrating the new technologies need to be kept down and the grid needs to remain reliable. Several revisions to the draft plan were made specifically to provide time and flexibility to ensure the lights stay on while more clean energy sources join the grid. An important accommodation is carefully phasing in targets.
Much of the US and many countries rely on wholesale markets to help manage electric grids. It’s a good strategy. Well-designed markets can maintain reliability while also keeping costs down. However, policymakers and regulators are faced with a bewildering number of choices to ensure these markets function well.
Fortunately, the US can look to Germany. In Germany, wind and solar already represent 43% of installed generating capacity. While the US is at a modest 7%, some regions are approaching 20% (see graph). The Energy Information Administration projects that under the Clean Power Plan, US-wide wind and solar penetration could reach an overall average of 22% by 2030. Again, though, this masks significant regional variation.
As Max discussed in a recent blog, Germany is also phasing out nuclear energy and accelerating coal’s exit.
German policymakers need to know that with these changes its market will work harmoniously into the future — maintaining reliability and keeping costs reasonable. Fossil fuel power plant owners see problems on the horizon. German energy policies are pushing down wholesale energy prices and could potentially cause fossil fuel power plants to go out of business. As a result, the country could experience shortages and blackouts, say the fossil fuel plant owners.
Concerns about the future have prompted the Germans to review their wholesale markets. American and German policymakers shared their perspectives on this important topic at a recent joint US-Germany Electricity Market Design Workshop held in Berlin, Germany. With support from the US Department of State, I attended the event, which was co-hosted by the German Ministry of Economic Affairs and Energy and the U.S. Departments of State and Energy.
Germany’s electricity market is currently structured as what’s referred to as an “energy-only “ market. In an energy-only market, power plants earn revenues primarily by selling energy. The German government is evaluating whether they should create an additional market, referred to as a “capacity market”.
Power market concepts can be a bit peculiar, so let me try to describe the capacity market concept with an analogy.
Think about milk before the era of affordable refrigeration. Buying energy is like buying milk. You buy it when you need it, perhaps by getting daily deliveries from the milkman. Buying capacity is like buying a cow. When you want milk, your cow can provide it. Except the cow’s milk production doesn’t exactly line up with your needs. You need to buy enough cows to make sure everyone in your family has enough milk for their breakfast cereal. When you have extra milk, hopefully you can find someone to sell it to. Otherwise it spoils. If all your neighbors are selling milk at the same time, you won’t get much for it.
A power market that compensates power plants for energy and capacity is like having consumers pay for milk and also pay an extra amount to sustain the herd of cows. They aren’t your cows, but you happily pay knowing the cows are out there somewhere. When you or anyone else needs milk, these herds of cows are supposed to make it.
On July 3rd the German Ministry of Economic Affairs and Energy completed its deliberations with the release of its White Paper. So far the paper has only been released in German, but an English-language summary is here and the predecessor Green Paper is here.
The Ministry has concluded that a slightly reformed energy-only market—what they are calling electricity market 2.0—will serve the country into the future. The new market reforms will enter legislation in the Fall and reforms will be implemented in 2016.
The Ministry does not buy into the narrative that fossil fuel plants need to be compensated through a capacity market to ensure grid reliability. The Ministry sees other ways to ensure reliability, such as by increasing links to neighboring countries. This would allow Germany and its neighbors to benefit from differences in demand and supply between countries.
The Ministry is also recommitting to not impose explicit or implicit price caps in the energy markets. Flexible power plants can earn revenues by selling energy if prices spike as wind and solar ramp up and down.
How will we know if the Germans made the “right” choice? Ideally, we could run an experiment with multiple Germanys, implement a capacity market in one Germany, but not the other. Maybe we should designate the recently discovered earth-like Kepler-452b for economics experiments like this!
Until NASA takes up my proposal, we’re stuck in the world of observation, theory, and conjecture.
As the German experience unfolds, policymakers in markets such as the US should take a hard look at their own markets. Should capacity market structures be introduced, as was recently debated in Texas? Or should price caps continue to be raised, as the US Federal Energy Regulatory Commission has been urging? Should markets be kept local as in Texas or better integrated with their neighbors, as is beginning to happen in California?
We should continue to keep an eye on Germany.
Categories
Andrew G Campbell View All
Andrew Campbell is the Executive Director of the Energy Institute at Haas at the University of California, Berkeley. At the Energy Institute, Campbell serves as a bridge between the research community, and business and policy leaders on energy economics and policy.
Energy Institute Blog 
Look at: https://www.energy-charts.de/power_de.htm
Choose: “Woche 30”, “Alle Quellen” “Gestapeld”, and select also “Import Saldo”. What do we see?
On 26.07.2015 12.00 Germany was able/obliged to export 13 GW, otherwise p.e. all of it’s Braunkohle (lignite) plants would have been put on hold. However on 26.07.2015 20.00 16 GW Braunkohle were needed to supply Germany without import.
What can we learn? Germany is able to accommodate such a hugh amount of renewables only as long as other European countries dont have to do that.
A couple of points based on Andy’s article and some of the comments.
First, Germany’s renewable energy production is causing all sorts of reliability- and operating-related problems for its neighbors, including unscheduled flows, even as it’s own citizens dig in against new transmission to carry power from the windy north. It’s a bit misleading to trumpet Germany’s success at building renewables while ignoring the adverse impacts.
Second, the lowest estimate I’ve seen for the amount of storage required to support ultra-high renewable penetration levels in California (roughly the size of Germany’s grid) is about 210 GWh. I’m guessing there isn’t that much storage in all of Europe. PG&E’s Helms plant, which is quite large, can store about 6 GWh, As more and more countries increase renewable penetration, there will be a need for more and more storage in the range of hundreds of GWh, but there are not enough suitable sites in Europe to build pumped storage with its own environmental problems and the amount of battery capacity is staggering.
Third, absent a stochastic analysis that looks at the probability distributions around load and renewable supply, any suggestion that geographic and technology diversity plus some amount of storage will provide the level of supply reliability customers have come to expect is speculation. I like back-of-the-envelope solutions more than I like the output of often opaque computer models, but assessing the reliability impacts of high renewable penetration levels doesn’t lend itself to a back-of-the-envelope answer.
My personal view based on examining some of the California ISO’s studies is that penetration levels in the 30% range are feasible at costs that won’t wreck the economy if regulators, politicians and renewable advocates stop trying to micromanage the problem or promote their own narrow agendas. Above that level, the costs go up pretty quickly or the level of supply reliability plummets very quickly.
What politicians should do is put a price on carbon and then go home for a decade or two. Instead they seem to be throwing everything they can at the proverbial wall and hoping what sticks won;t make too big a mess.
There are a few things this article didn’t consider:
The reason to use renewable energy is to produce less global warming gases, without a significant decrease in GDP or living standards. What is the reduction in Germany’s global warming gas production per capita per GDP? There is some data here:
https://en.wikipedia.org/wiki/List_of_countries_by_ratio_of_GDP_to_carbon_dioxide_emissions
but it does not show the change as Germany has adapted more renewable energy. What is this change? Does it include imported nuclear electricity from France?
If sustainable CO2 production equivalent is 1 to 2 metric tonnes per capita per year
http://www.manicore.com/anglais/documentation_a/greenhouse/quota_GHG.html then this can
only be done at low cost of reduction of CO2 per tonne. What is the cost of reduction in Germany? They have high electricity prices, and some chemical firms have threatened to leave the country because of high energy or feedstock cost. If the cost of energy becomes to high in Germany and firms leave and produce CO2 elsewhere, how is this accounted for?
Germany is an utter failure and will keep on being an utter failure for another decade while it is decommissioning its nuclear power. Likely, the US will go the same way. It’ll add 20% renewables and decommission 20% nuclear, and any improvement will come from replacing coal with gas. Just as with Germany, the nuclear decommissioning will be a bit back-loaded, so there will be initial progress which will vanish later.
“Germany and its neighbors is sufficient to supply Germany’s non-renewable needs, sounds more like Germany is exporting its fossil energy use to its neighbors so it can appear to be more green.”
This, possibly, and also possibly exporting its capacity needs to other countries, as well. If indeed, the German plan is to free-ride reliability, then it’s not exactly a thrilling exercise.
I am not enough of an economist to opine if capacity markets are required for a stable system in the long term, but I am enough of an engineer to know that reliable capacity is required for a stable system over any time period. It could very well be that creating strong interconnections over a very wide area frees up excess capacity that would otherwise be trapped in its local pockets, thus providing a cushion to allow retirement of unfavored types of capacity from the mix. But that seems like a one-shot solution (admittedly, could work for a long time). But eventually, if demand grows (for example, from electrification of transport) you’ll need to build something to provide capacity. Perhaps by then the something (battery storage, directed EV charging, etc) will be cheap and easy.
Best,
Dave J
PS — Regarding Kepler 452-b, I think the experiment might run into IRB problems (Intergalactic Review Board) 🙂
Note that renewable power also has capacity value, just not the metal and cement in the ground that engineers want to rely on. Renewables offer portfolio reliability by distributing resources with probabilities of availability. This is much like demand response, which engineers also too often are uncomfortable with. If we are to move forward in this arena, our understanding of “capacity” and “reliability” will have to change to reflect new technologies and better understandings of how uncertainty and risk can be managed.
Of course, renewable resources of course have capacity value, some more, some less. In any given energy market, capacity value has a specific meaning according to the rules in effect in that market. But the big, giant, electric machine doesn’t know the market definition. Solar in CA peaks hours before demand. On the margin, it’s capacity value is low, as in, feel free to install infinite solar, it’s not going to do a damn thing for a 7pm demand peak.
Engineers, by temperament, tend to be “small ‘c'” conservative people. this serves them, because they are accustomed to being held accountable when things do not work. I wish policy advocates felt the same way about their work product.
If folks want to sell a new “understanding” of the word “reliability,” then have at it, but let’s not pretend we’re offering the same product as before. You want to build a power system that cannot function properly without a big chunk of real-time DR? OK, sure. But please do not act surprised when ratepayers are less than thrilled, especially if they “get to” participate in DR/RTP _and_ end up with Germany-like overall rates.
I don’t even mean to harsh on solar and DR. I like them and want to see lots of both. Moreover, engineers, system operators, utilities, all need to be pushed out of their comfort zones, otherwise there will be no progress. But caution is justified.
I’m pessimistic about getting Intergalactic Review Board approval.
There may be a better analogy than cows. Here’s one we understand from California: Dams. Let’s say I’m a farmer near an aqueduct, downstream from a dam. I usually get enough water from rain, but when it’s dry, I can buy water and take it from the aqueduct. Should I pay something in wet years, to be sure the dam is kept full and well-maintained? If I pay only for water delivered, then I expect to pay a lot in dry years, and nothing in wet. If I also pay for dam maintenance (i.e. capacity), then I expect to pay less in dry years, since I’ve cut the costs and risks for the dam operator.
And the water markets to a certain extent have converted that into a single-dimension per acre-foot market price. Water can be bought in the market for a single year on the “spot” market, e.g., Metropolitan Water District of Southern California is paying $660/AF to Sacramento Valley districts. Or water rights can be bought in perpetuity. The cities of Davis and Woodland are buying rights from Conway Ranch for about $4,000 per acre-foot. Water rights have varying seniority which leads to variations in reliability.
Trying to create separate short-run capacity markets is trying to solve a problem is better addressed through more transparent long-term markets. The capacity value will be rolled into the long-term prices.
Britain, which is much less connected and has a lower capacity margin, recently instituted a capacity market.
For a US example look at PJM with both energy and capacity markets with some price caps (although price and applicability are currently a bit fluid). PJM is BIG, from the shores of the Hudson River to shores of Lake Michigan (over 175 GW of capacity), complicated and works reasonably well, but PJM is also wrestling with everything mentioned in the post. Also, I have a hard time believing that differences in load patterns between Germany and its neighbors is sufficient to supply Germany’s non-renewable needs, sounds more like Germany is exporting its fossil energy use to its neighbors so it can appear to be more green. If this is true it is not a solution, just deferring the day of reckoning.
Your mention of PJM is very helpful. Different US utilities and operators will resemble different European countries, so comparisons make sense at a sub-national level.
About energy trading across markets: It’s not just about differences in load patterns (although those would be significant for southern neighbors like those on the mediterranean). It also refers to renewable supply patterns. The sun sets later over your western neighbors; wind farms often have peak periods that vary due to geography; and hydro (and nuclear) supplies have their own patterns. It would be interesting to try to quantify these variations to understand how much they may help.
Germany is net-exporting in the moment, because wholesale prices are cheaper, and wholesale prices are cheaper because of increased renewable capacity. Which hints at Germany exporting at least also renewable electricity. Having said that, this scheme of course would not work any more if the neighbours increased their renewable capacity, too, as long as (as you say) load patterns are more or less similar. However, the key rationale of better integrating with neighbours is to make use of the large hydro/pumped storage capacities in Austria, Switzerland and Norway (which is to be an “electrical neighbour” by underwater cable). And those should be flexible enough to balance out renewables.
And there is Slovenia, which has significant hydro capacity which is not used for pumped storage; then there are several countries in Western Balkans where there is untapped hydro potential.
Can laboratory experiments with students and/or electricity market players teach us anything about the strengths and weaknesses of the different market designs?
No
Frank Wolak, Mark Thurber, and Trevor Davis, all of Stanford, have run some simulated games looking at wholesale market outcomes in the face of renewable policies and carbon prices. I’m not sure if electricty market designs could be studies through such an approach. See: http://pesd.fsi.stanford.edu/publication/simulating-interaction-renewable-portfolio-standard-electricity-and-carbon-markets