Implications of a study by Hans Werner Sinn, ifo Institute Munich

For a long time Germany’s attempt to grapple with atomic power, climate change and energy issues through its so called “Energiewende” (Energy Transformation) has been inspirational to many green activists and seen as a process to learn from. The priority given to “clean energy”, to wind and solar in its electrical grid, incentivised by feed in tariffs and favourable prices has taken wind and solar added together to 3.5 % of its energy supply and 16 % of its electrical power generation.

However, there is a long way to go to 100% green energy. 58% of power generation is still by fossil fuels and fossil fuels are still predominant in 78% of energy consumption that is not electrical, for example for transport fuels and non electrical space heating.

No problem, just a matter of time? A lot of activists probably think this but sadly it is not likely to be true. Yes, there are things to learn from Germany’s attempt to make an Energy Transformation. Unfortunately these things are that it will not be easy and it will probably not be possible at all without a considerable reduction in overall energy consumption and/or major new technological breakthroughs in energy storage. Such breakthroughs currently do not look very likely and/or would involve very high costs. Such costs would cripple the German economy in its current form.

This anyway is the conclusion that I draw from a study by one of Germany’s leading economists, Hans Werner Sinn, that appeared in the European Economics Journal, in the summer of 2017. I was alerted to his article, published in English, by a weblink which connected to a lecture that Sinn gave at Munich University just before Xmas. The lecture, in German, contains much the same material as the article with one or two small differences.

Before I go further I think it important to say that Sinn is not a climate denier. He acknowledges climate change as real and in need of addressing. It is important to be clear that the issue of whether climate change is real is completely independent from how easy or difficult or costly it will be to develop a renewable energy system. There are no guarantees that just because humanity has a serious problem there are easy and cheap engineering solutions. In any case Sinn does not address these issues – he is addressing the practicalities and limits of the Energiewende.

Whether in German or English the data he presents is bad news because it is about the difficulty of storing electricity for the German economy at its current scale of energy and electricity use – and storing energy is going to be necessary to further expand renewable generation without having fossil fuel based generation to back it up.

This is because under current conditions the coal and gas generators in Germany are necessary complements to balance the volatility of wind and solar and the variable nature of electricity demand. When the wind is not blowing and the sun not shining – the coal and/or gas generation must step in to provide the power. Or perhaps there is wind and solar power but not enough as the demand for power rises. It is the fossil fuel generators that must step in and provide the buffer between them and if fossil fuel generated electricity is going to be driven out then some other means must be found to buffer between fluctuating supply and demand. There is a missing technology needed to make this possible – electrical storage.

What gives Sinn’s article and lecture credibility is that they are based on real world intermittent data for wind and solar power generation in Germany in 2014 as well as data from an EU research project called ESTORAGE. ESTORAGE set out to find Western Europe’s potential for pumped hydro power – by finding all the locations where it could conceivably be developed along with how much electricity could be stored altogether.

The use of real world data from Germany in 2014 completes the picture because it enables Sinn to show how much storage is needed over a year to balance the grid at different levels of penetration by renewables. This volume is then compared to what is available in potential pumped hydro sites.

Pumped hydro is a way of storing electric power by using surplus electricity to pump water uphill into a storage lake, that can then be released through turbines downhill later, when electric power is wanted. Its significance is that it is by far the cheapest and easiest way of storing electric power on a grid scale. The findings of the ESTORAGE project therefore enables Sinn to explore if there is enough pumped storage capacity in Germany, in Germany and Norway and in an energy union between Germany, Norway, Denmark, Austria and Switzerland. The figures are sobering – firstly there is no way that Germany has enough undeveloped new sites where it could develop sufficient pumped hydro storage on its own territory to balance its grid without fossil fuel generation doing buffering. The furthest it can get in the direction of an entirely green electricity supply is 49% of power generation by renewables, if it is in an alliance with 4 other countries which have the best pumped storage options – assuming they are prepared to develop these options.

Sinn does consider other storage methods in his lecture but considers them too expensive and impractical for storing electric power – for example lithium ion batteries are practical up to a point for powering electrical cars but it would require the batteries of 524 million BMW electric vehicles to balance the German grid and the cost of storing a kilo watt hour in a lithium battery is 50 times the cost of storing a kilo watt hour using pumped hydro. Sinn also considers storing energy by using surplus electricity to generate hydrogen or methane but again considers them too expensive particularly because of the “round trip” power conversion losses from power to methane and back to power (only a quarter of the power left) and with hydrogen only a half of the power left. (Added to which hydrogen is a very corrosive stuff to work with.) This is a thermodynamic problem first studied by Carnot for which there is no pat solution.

There is also the option of shifting demand. The problem with wind and solar is that what is generated must be made to match what is demanded – but can this done by shifting demand around so that, for example, the washing machine is switched on when the wind is blowing? To explore the magnitude of what is possible Sinn again uses real world data. He calculates how much buffering storage could be reduced by shifting demand around during the course of each day. He also calculates how much storage could be reduced by shifting demand during the course of a week and shifting demand during each month. His results are disappointing. Shifting demand during a month it is only possible to reduce the need for energy storage by 11%. This is because energy storage is mostly needed between seasons and the amount of storage required would be astronomically expensive to achieve without pumped hydro. Switching the washing machine on when the wind is blowing is one thing – you cannot wait till summer to switch a heater on in winter when there is no wind and it’s the middle of a cold night.

There are in fact three ways of balancing a grid rendered unstable by intermittent renewables. One is a double structure where fossil fuel generation balances the grid but we want to go beyond that. Another is storage which we have seen is expensive with not enough options – but what about just continuing to expand wind and solar capacity – more installations at each place and over a wider area. This is the strategy of “over extension”. If its not windy or sunny everywhere it will be somewhere so one just has to have enough kit there to capture enough of the wind and/or the sun.

In fact Sinn considers this option too. He has a “thought experiment” in which a greater and greater percentage of the German grid is supplied by renewables and a smaller and smaller % of electricity is balanced by fossil fuel generation. At 89% wind and solar generation the German grid would in fact be 100% green energy since 11% would be electricity from hydro power and through burning biomass. (He ignores those who question whether biomass is really “renewable”). But at this point of 89% wind and solar the average efficiency of wind and solar generation would be 39% and the marginal efficiency would be 6%. Put in another way 61% of all electricity would on average have to be dumped or curtailed because there would be too much power for the demand. To say the marginal efficiency is 6% means that to extend renewable energy by 1% of the overall capacity at this point you would need to dump or curtail 94% of the extra generated electricity.

I hope this is clear – you can extend wind and solar more and more but in order to have power all the time, including those times when there is not a lot you need to develop a capacity that, in the face of intermittent wind and solar, is most of the time oversupplying.

Any way you look at it you have a lot of cost.

Now to my own comments. What Sinn does not explore is if the German energy demand were only half its current size or even smaller. His figures suggests that renewables can maximally supply a balanced grid for only half the current power supply in the 5 country association. But what if only half the energy were needed?

I do not think that Hans Werner Sinn is an exponent of degrowth…far from it….but that is what we should be looking at.

The aim is not unreal or unrealisable if we start thinking about “energy sufficiency” (rather than energy efficiency). In a recent article titled “How Much Energy do we Need” in Low Technology Magazine Kris de Decker explores the many opportunities for reducing energy consumption once we adopt a sufficiency approach. He writes

“In principle, public service delivery could bring economies of scale and thus reduce the energy involved in providing many household services: public transport, public bathing houses, community kitchens, laundrettes, libraries, internet cafés, public telephone boxes, and home delivery services are just some examples.

Combining sufficiency with efficiency measures, German researchers calculated that the typical electricity use of a two-person household could be lowered by 75%, without reverting to drastic lifestyle changes such as washing clothes by hand or generating power with exercise machines. Although this only concerns a part of total energy demand, reducing electricity use in the household also leads to reductions in energy use for manufacturing and transportation.

If we assume that similar reductions are possible in other domains, then the German households considered here could do with roughly 800 kgoe per capita per year, four times below the average energy use per head in Europe. This suggests that a modern life is compatible with much lower energy demand, at least when we assume that a reduction of 75% in energy use would be enough to stay within the carrying capacity of the planet.”

Suddenly we are back in the realms of practicality IF, that is, it is politically practical to adopt a sufficiency agenda – but perhaps that is what will have to happen anyway as the decline of the oil and gas industry accelerates.

In conclusion. It looks very as much as if before “over developed” countries like Germany can hope to develop an all-renewables power system, let alone an all-renewables based energy system including non-electric energy uses, it will have to dramatically reduce its power consumption. Even though studies based on energy sufficiency show that most people could probably live a comfortable enough life the changes in economic organisation and thinking would or will have to be massive for that to happen. I therefore doubt that this is going to happen as a result of well-meaning policy intiatives any time soon. The inertia will in all probability be too great.

That said countries like Germany are not just under pressure to change their energy system because of climate change – Germany and other countries too must respond to the global trend to depletion of fossil energy sources and the rising cost of extracting them. While it is true that renewable energy together with energy storage would be expensive if attempted above a limited scale, it will be expensive in the future to extract fossil fuels too. As we reach the limits to growth we are probably looking at economic contraction anyway- and no doubt a good deal of political turmoil because politicians and the German (and world) public will be disorientated and not really understand that is happening.

There is an irony here. The best chance of developing grids adapted to renewables will probably be in countries where electricity demand and energy use is currently very low and where it can develop “organically” without having first to go backwards in a retreat from “overdevelopment” before it can again “go forward” in conditions of much depleted resource availability.

If humanity survives the next few decades of turmoil – and it is a big IF given the collective psychosis likely in heavily armed countries thrown into economic contraction – IF… then the best chance for technologies to evolve into 100% renewables-based systems are in what are today regarded as poor countries. Then the last would be first and the first last. That at least is something to hope for.


Hans Werner Sinn in European Economic Review “Buffering Volatility. A study on the limits of Germany’s energy revolution” – on his website at
Hans Werner Sinn “Wie viel Zappelstrom verträgt das Netz? Bemerkungen zur deutschen Energiewende” Lecture in German for the IFO institute at the University of Munich 18.12.2017
Kris de Decker in Low technology magazine – “How Much Energy do we Need?”

Featured image: A. Source: