The Energy Collective Group

This group brings together the best thinkers on energy and climate. Join us for smart, insightful posts and conversations about where the energy industry is and where it is going.

9,731 Subscribers

Article Post

Tracking Energiewende Performance


  • The Energiewende is proceeding ahead of schedule with high grid reliability. 
  • However, costs are much higher than originally planned, while CO2 emissions are stagnating. 
  • Net value of new wind/solar is approaching zero as market value declines and integration costs increase.
  • Wind/solar market shares have now reached the point where large grid expansion projects become critical. 
  • It will be very interesting to see how Germany performs in this complex next phase of the Energiewende.  


The German Energiewende has long been the world’s leading experiment in large-scale deployment of variable and non-dispatchable renewables. Data from this €200 billion (and counting) experiment is very important to quantify the practical and economic challenges of large-scale wind/solar power deployment.

The primary experiences from the Energiewende are well documented. On the positive side, Germany has been able to continue expanding its renewable electricity output at a very impressive rate. On the negative side, the costs of this transition have been higher than expected and CO2 emissions have barely budged since the 2008/2009 financial crisis.

Wind/solar shares and CO2 emissions under the Energiewende (source).

This article will delve a little deeper into these trends with the help of the Energiewende tracker from McKinsey & Co. But first, let’s take a brief look at perhaps the most important datapoint from this experiment: wind/solar market value.

Current wind/solar share and market value

Last year was a good year for German renewable electricity production, especially wind power. In total, 104 TWh of wind and 38 TWh of solar electricity was generated. Total German electricity production amounts to about 650 TWh, resulting in wind and solar shares of 16% and 6% respectively.

When considering practical integration of variable renewables, however, the effective market shares are about 25% lower given that Germany relies heavily on imports/exports to balance its wind/solar output (quantified earlier). A sample week in 2017 is shown below as an example of the clear correlation between wind/solar output and electricity exports.

This brings the effective wind and solar shares to 12% and 4% respectively. At these shares, the market values of wind and solar power are already down to 82% for wind and 92% for solar PV (below) – in line with data used in a previous article illustrating the perpetual subsidy dependence caused by this self-cannibalization effect of variable and non-dispachable renewables.

Market value of different electricity generating technologies relative to the system average in Germany for the year 2017 (source).

When we use the marginal value of new wind and solar capacity (see aforementioned article), correctly accounting for the fact that new capacity will lower the value of all existing capacity, the value factors of new wind and solar drop to 54% and 62% respectively. This is a critical insight from the Energiewende experiment: even at current modest market shares, new wind and solar energy is worth just over half the current wholesale price, about €18/MWh and €20/MWh respectively.

Energiewende performance measures

The McKinsey & Co report contains 13 performance trackers that will be briefly discussed below. In all the graphs, the target trajectory is indicated by the light blue line and the actual trajectory by the dark blue line.

Firstly, the well-known stagnation of CO2 emissions is shown. This is primarily the effect of displacing nuclear with renewables while coal remains relatively constant.

Annual equivalent CO2 emissions.

As mentioned earlier though, the rate at which Germany has expanded renewables is very impressive. The Energiewende has therefore successfully proven that renewable energy can be rapidly built out if subsidies are large enough. It should be mentioned though, that the German renewable energy buildout is about 2x slower than the French nuclear buildout of the 1980s.

Percentage of renewables in annual electricity consumption.

Next, we see that primary energy consumption and electricity consumption are remaining quite flat. Given that very high electricity prices provide a large incentive for increased efficiency, this result therefore appears to illustrate the limits of what can be achieved by energy efficiency measures.

Annual primary energy consumption.

Annual electricity consumption.

Next, we see that Germany has managed to maintain impressively high grid reliability, primarily by maintaining a large reserve margin.

Minutes of power supply failure per year.

Percent reserve margin maintained.

The costs of maintaining this good grid reliability are rising though. As shown below, network costs are increasing rapidly and are currently up to €13/MWh of wind/solar electricity. This is the cost associated with grid stabilization and reserve power plants. It is important to note that this cost is approaching the marginal value of new wind and solar, implying that the net value of new wind & solar capacity is already approaching zero.

Added grid costs from wind/solar power.

Part of the reason behind this rapid cost increase is that electricity networks are not being expanded rapidly enough. The two following graphs show that Germany is falling behind in terms of general grid expansion and interconnections with the broader European grid. The need for vast transmission system expansion, often across international borders, is one of the primary challenges of accommodating high wind/solar market shares.

Transmission expansion.

Interconnection capacity as percentage of generating capacity.

Next, we take a look at electricity prices. Firstly, the famous Energiewende surcharge on consumers is shown below. It has flattened out at about double the original target.

Renewable energy surcharge.

As a natural result, German households pay almost 50% more for electricity than the European average.

Percentage by which household electricity prices exceed the European average.

Industries are protected from this cost increase to a certain degree, although prices remain higher than the target.

Percentage by which industrial electricity prices exceed the European average.

Finally, a measure of the number of jobs in renewable energy is provided. It is clear that significant contraction has taken place in recent years following the great solar PV boom of 2010-2013.

Number of renewable energy jobs.

Discussion and conclusion

To date, the German Energiewende has clearly proven that renewables can be expanded rapidly if the population is willing to pay. Renewable energy expansion is proceeding ahead of schedule and grid reliability remains high, but this comes at a cost that is more than double original targets. Cost will continue to pose a significant challenge as the net value of wind and solar power approach zero due to falling market value and rising integration costs.

The Energiewende now enters the next stage of wind/solar integration where substantial grid expansion is required to balance variable renewables. Currently, Germany is falling behind with this task, leading to rapidly increasing grid stabilization costs.

This will be an interesting test for the Energiewende given the complexity and scale. Up to this point, the modular nature of wind/solar power made their expansion attractively simple. From this point onward, however, continued expansion will require large and complex national and international grid expansion projects.

Aside from this transmission buildout challenge, it will not be long before Germany will require significant expansions of energy storage. The current electricity mix already achieves occasional scenarios where wind/solar supply approach total demand, resulting in negative electricity prices. This will either require curtailment or energy storage, both of which are very costly.

The Energiewende therefore remains a fascinating large-scale energy experiment. As far as I’m concerned, the jury is still out on whether this will work or not. I will certainly continue to follow closely as Germany embarks on this next, much more complex, phase of wind/solar power expansion.

Content Discussion

Robert Hargraves's picture
Robert Hargraves

Schalk, I like your articles. Can you update the links in this one to reflect the new hosting?

Audra Drazga's picture
Audra Drazga

Robert - the links to this are broke because of the site transition.  They should be back live shortly.  Sorry about the inconvience! 

Rick Engebretson's picture
Rick Engebretson

Nice to read your articles again, Schalk. Instead of allowing "the German experiments" to define "renewable energy," I would again like to share a different direction we have discussed before, biofuels.

A very interesting link to the U. of MN. Chem. Engineering (and beyond) is here;

Further, recently visiting the lovely St.Paul, MN. district heating plant area, biomass advocacy is now emphatic. Nested between Excel Energy Center, the Science Museum of Minnesota, etc., is a working prototype with abundant information.

So I am confidant the people that provide abundant "dispatchable" food are ready to ramp up abundant "dispatchable" renewable energy resources. I am greatly relieved the weather will not control sustainable energy resources.

Schalk Cloete's picture
Schalk Cloete

Thanks Robert. My previous TEC article linked in this article is now here:, but itseems like the images for this article are not ported over yet. 

Schalk Cloete's picture
Schalk Cloete

Hi Rick, it is good to be back on the new TEC.

I also think that biomass will be an important contributor in the future. Of course the availability of sustainably produced biomass is limited, but there is enough to take over many of the applications where we have no viable alternatives to fossil fuels. Balancing intermittent renewables could be another important application. 

Rick Engebretson's picture
Rick Engebretson

Schalk, this became more timely with President Trump asking Germany how we reconcile NATO and new gas piped from Russia. One might also question Iran energy. After 30 years we have the data, windmills and solar panels won't power Europe.

Not "of course the availability of sustainably produced biomass is limited." Instead, the availability of wastefully consumed biomass is limited. I was absolutely delighted to discover serious chemical engineers and serious industry now know this.

Matt Robinson's picture
Matt Robinson

A good article. Thanks.

I wonder if there'll ever be an article comparing the current costs of Germany's Energiewende with the strategy of keeping their nuclear plants and replacing coal plants with nuclear?

I would be very interested to see all those graphs with another line showing the nuclear strategy and how it might have developed.

Schalk Cloete's picture
Schalk Cloete

Hi Matt. This is a difficult question.

First off, it is important to keep in mind that Germany is not phasing out nuclear because of economics, but rather due to (perceived) safety concerns. 

If we only consider cost-effectiveness, there can be no doubt that a nuclear strategy would have been much cheaper and much more effective at curbing emissions. However, we should be mindful that most of the Energiewende costs were accrued in the earlier years when wind and solar were still expensive (particularly the solar PV boom of 2010-2012). If another country wants to replicate the Energiewende today, they would be able to do it at a cost below nuclear.

Thanks to their impressive cost reductions, wind and solar are great at moderate market shares (up to the 10-20%  range). Their modular nature makes them easy and fast to deploy and intermittency is not such a big issue yet. The challenge comes when we start pushing beyond this level. As stated in the article, new wind and solar are already approaching zero net value in Germany. This means that, no matter how cheap they get, deployment will always need to be subsidized.

Thus, if deep decarbonization is the goal (as recommended by climate science) a nuclear pathway will probably be cheaper. But there can be little doubt that wind and solar will be cheaper if the goal is only moderate decarbonization and improvements in energy independence. 

Charles  Forsberg's picture
Charles Forsberg

Germany has what is called a faith-based energy policy--wind and solar at 22% of electricity production with price collapse limiting value of added wind and solar. One more experimental point on where wind and solar limits are today. The question is how many times does this experiment have to be done before people believe the answer?

The solution to enable larger-scale use of low-carbon wind and solar is cheap storage with assured electric generating capacity. The only cheap storage is thermal storage. The DOE capital-cost battery goal for battery storage is $150/kWh and double that after add the electronics. The DOE capital storage goal for heat storage is $15/kWh. Some big concentrated solar thermal plants appear to be below that goal but concentrated solar only works in some locations.

Nuclear reactors produce heat and thus couple to heat storage. When couple nuclear to heat storage and electric prices are low: (1) operate reactor at full power, (2) operate turbine at minimum load to enable rapid return to full power  (3) send excess steam to heat storage (steam accumulators, hot oil, concrete, etc.) and (4) buy low-price electricity to provide more stored heat. When electric prices are high: (1) all reactor steam to turbine, (2) added steam from storage to turbine and (3) peak power greater than base-load power with oversized turbine or separate peaking turbine.

The limitation of all storage technologies is that storage can be depleted. If heat storage, can assure peak power capacity by adding a steam boiler to provide steam when storage is depleted. Because have heat storage system that provides peak power capability most of the time, this boiler may be used less than 100 hours per year. If bought storage system with oversized turbine, the incremental cost of that boiler will be $100-300/kWe or half or a third of a gas turbine—the cheapest method to assure generating capacity.

Nuclear with heat storage and steam boiler can be (1) the replacement for fossil as a dispatchable electricity source and (2) the enabling technology for large-scale use of wind and solar—which requires really cheap storage and assured backup generating capacity to be viable. Open source paper on these options available Forsberg, Electricity Journal April 2018:

Edmund Kelly's picture
Edmund Kelly

The latest BNEF clean energy investment report for Q2 2018 has a chart that illustrates the decline of clean energy investment in Germany (and others that show a similar overall decline for Europe) The last three quarters have been unbelievably low, way less than $1B/Q. This on top of a steady decline since 2011 would indicate that the Energiewende experiment may well be over. Its clear from Germany and California that penetration levels significantly less than utilization factor rapidly raise the costs of intermittent clean energy. All additions to mitigate these costs add new costs and increase overall cost. Overall clean energy investment wordwide has stalled since 2011 and now with China's PV pullback, overall investment seems set to decline. With Europe's decline and Trump in the US, its hard to see who is leading the clean energy charge. Following the money is a reliable indicator that cuts through hype and wishful thinking.

Schalk Cloete's picture
Schalk Cloete

Hi Charles,

It is true that heat storage is cheap, but it is important to keep in mind that heat storage costs should be multiplied by a factor of at least 3 when storing the heat for electricity production via steam cycle. This brings the DOE target you mention up to about $50/kWh. Oversizing a turbine for 100 hours per year of peak production also does not sound like a very economic proposition because efficiency drops in part-load operation.

Personally, I’m more in favour of chemical energy storage. Conversion efficiencies are still fairly low, but, depending on the produced synfuel, storage costs can be almost negligible. More importantly, deployment of these synfuels in a future economy with volatile electricity prices and an increasingly complex interconnected energy system will be attractively simple and versatile. 

Schalk Cloete's picture
Schalk Cloete

Hi Edmund,

I share your skeptisism about wind and solar market share expansion beyond their average utilization factor. From a climate change point of view, the biggest risk with the current expansion of wind and solar power is that the world will eventually reach a point where it is forced to admit that the required deep decarbonization is simply impossible with wind and solar. Shifting back to a baseload dominated system will then be incredibly expensive and time consuming. 

As long as investment remains flat, deployment rates of wind and solar are still increasing due to their impressive cost declines. But several regions are now arriving at the point where the practical and economic challenges of wind/solar integration are becoming undeniable. It will be very interesting to see how this pans out over the next couple of years...

John Oneill's picture
John Oneill

Hi Schalk - as regards heat storage, this presentation by Moltex Energy claims (pages 25-26) that one of their high temperature reactors using hot salt storage, as at some solar thermal plants, and running at a thirty percent capacity factor, would have lower capital cost per kw installed than combined cycle gas turbines. They use a tertiary salt loop anyway, to separate the nuclear island from the steam turbine hall, so adding extra insulated tanks for hot and cold salt, plus extra turbines for peak demand periods, would be cheaper than having a whole redundant reactor. Other high temperature reactors could do likewise. Moltex has recently signed an agreement to begin development at Point Lepreau, New Brunswick. Another company will be working there on a development of the Experimental Breeder Reactor, which ran for thirty years in Idaho. I understand that sodium cooled reactors are limited to 510 C outlet coolant temperatures, though - better than the 315 C of a light water reactor, but perhaps a bit low, without reheat, for an off-the shelf turbine, as commonly used in the coal or gas  industry, once salt storage losses are factored in.