How expensive is an energy transition? A lesson from the German Energiewende
Unnerstall Energy, Sustainability and Society
How expensive is an energy transition? A lesson from the German Energiewende
Thomas Unnerstall 0
0 Stockstadt , Germany
The Paris climate agreement of December 2015 is generally considered a breakthrough on the way to a sustainable future for mankind. In particular, the agreement calls for fundamental transitions in the energy systems worldwide, since more than 80% of CO2 emissions stem from the use of fossil fuels in the energy supply. Considering such energy transitions, in any country there will certainly be technical issues, there will be debates as to which political instruments are most suitable, and others; but there is no doubt that the question of cost is one of the most crucial issues in the course of such a long-term project. After all, the expected financial burden on the national economy and its stakeholders is the most convincing argument for putting in energy transition off or for slowing it down. There is also no doubt that the German Energiewende, at first sight, does not serve as an encouraging example in this respect: The cost bill - in the sense of the direct, perceptible financial effects - has already run up to almost € 500 billion, and the German private households as well as many businesses pay significantly more for electricity than in most other OECD-countries. As a consequence, in Germany there is a growing opposition against going forward with the Energiewende as planned, and also in the international media, the initially positive image of the German project has suffered. A closer look at the costs of the German energy transition, however, reveals that around 75% of them are due to two particularities of the Energiewende that do not hold true for other energy transitions: the politically enforced nuclear phase-out and the fact that Germany massively expanded renewable energies at a time when they were still very expensive. Therefore, the real lesson of the German example is the opposite of what it may seem: The transition to renewable energies in the electricity sector in a highly industrialized country can be quite affordable.
Ever since its official launch in 2010/2011, the German
Energiewende has received worldwide attention. At first,
international comments were mostly positive, sometimes
even full of admiration, due to its—at least at the
time—very ambitious goals for reducing CO2 and for
expanding renewable energies, notably in electricity
generation. Nowadays, many observers from abroad have
become more skeptical; and the reason is not that
unforeseen technical difficulties have emerged in the
course of the project so far; the reason is that the costs
are perceived to have gotten out of hand. “Other
countries simply cannot afford an energy transition” is a
conclusion that suggests itself, and thus the German
example—even though a frontrunner and meant to
inspire others—actually risks to have an adverse effect.
In the debate within Germany, too, the question of the
cost of the Energiewende and the ensuing burden on
private households and especially on businesses is the most
controversial and the most crucial issue [
]. Indeed, the
initial broad consensus between all major political parties
on the main Energiewende goals and principles seems to
Just recently, several studies of the estimated costs of
the Energiewende until 2025/2030 have been published
], and despite a number of methodological
differences they arrive at comparable figures. The
Energiewende in the electricity sector alone—i.e., not yet taking
into account the necessary transformation of the heating
and transportation sectors—up to 2030 is expected to
require financial support for renewable energies, costs
for grid expansion, etc. on the order of €600–700 billion
]. It is true that these costs are spread out over a
period of 50 years (2000–2050); but the cost after 2030
on the route to an almost completely decarbonized
electricity system envisaged for 2050 or 2060 cannot even be
In any case, such figures could certainly have a
deterring effect in view of other energy transitions—energy
transitions that are or at least should be on the political
agenda of many countries in the wake of the Paris
climate agreement of 2015.
The aim of this article is to show that this need not be
the case: correctly analyzed, the German example can
actually inspire optimism with respect to the
affordability of energy transitions in comparable countries, at least
with respect to the electricity sector.
The argument will proceed in three steps:
We define two stages of the Energiewende up to
2030 and estimate the costs incurred during these
We identify nuclear phase-out and early action as
distinctive features of the German Energiewende
We roughly estimate the cost of a (fictitious)
transition of Germany’s electricity sector without
nuclear phase-out and without early action.
Before getting started, however, it is important to
briefly outline the methodology of this article, to
delineate its scope and to define the principal terms used.
We will use the term “cost” to mean the direct financial
effects of the Energiewende: in particular, the so-called
“differential costs” due to the expansion of renewable
energies (RE) explained below, costs for grid expansion,
government funding for energy efficiency investments
such as CHP plants, government funding for R&D, and
the so-called “merit order effect.”1
This is certainly quite a narrow view on the cost
issue—we do neglect secondary cost effects (effects on
jobs, taxes etc.), and we do neglect the so-called external
costs of power production (such as cost of
Taking such a narrow view, however, seems justified
for our purposes since the direct costs are the ones
immediately noticeable and quantifiable. Thus, the public
and political debates often focus solely on these costs.
By far the most important direct cost factor are the
so-called differential costs for the expansion of RE: the
remuneration/feed-in-tariff to the investors of the
REplants for the electricity produced (granted by the German
Renewable Energies Act (GREA) for the first 20 years of
operation) minus the market value of this electricity.
These differential costs have to be paid via the
GREAsurcharge by the electricity customers. For each RE
plant, the differential costs can be calculated as
20 years × feed-in-tariff × average electricity production
per year—average market value of the electricity
produced over these 20 years.
We will assume here that the average market value
will turn out to be 3 ct/kWh over the time periods in
question, which roughly corresponds to the average
market value over the past few years. Of course, this is a
somewhat bold assumption—but it is relatively easy to
see that the arguments to be developed in the following
sections do not depend on this figure. In other words,
assuming, e.g., 2.5 or 4 ct/kWh or a certain development
over time does not alter our main conclusions (at least
in all reasonable scenarios).
It can be shown [
] that the other direct cost factors
enumerated above—grid expansion, funding for CHP
and for R&D—are much lower than the differential cost
of the RE expansion itself, and they are being at least
partly offset by the positive merit order effect. Taken
together, these factors amount to no more than 10% of the
Finally, we must assume here that the original targets
for the Energiewende in the electricity sector up to
2030 remain unchanged: no nuclear power, 50%
renewables in the electricity mix, but no further heavy
political intervention in the power market. (There is
certainly a debate in Germany about significantly
accelerating the transition speed in order to better meet the
overall CO2-reduction goals: phasing out coal until
2030, 60% renewables or more in 2030, massive
subsidies for power-to-gas and storage technologies, etc. If
such measures should become reality, the cost
estimates given here are no longer valid).
Summing up, the differential costs—the immediate
financial burden on households and businesses due to the
expansion of RE power plants—give to a certain extent
an estimate of the (direct) costs of the Energiewende in
the electricity sector which is admittedly very limited in
scope and quite rough, but which is sufficient for our
purposes. Differential costs do not reflect the real costs
of the RE expansion in a strictly scientific sense; but our
aim here is not to give a comprehensive scientific
account of the cost issue, but to address in an easily
understandable way important trends in the public and
political discussion on the cost issue.
Two stages of the Energiewende up to 2030
With respect to the costs as just defined, the German
Energiewende in the electricity sector until 2030 can be
divided up into two stages:
In stage 1, renewable energy (RE) plants of around
100 GW were built which produce on average approx.
170 TWh per year of electricity. The average cost of this
electricity—the remunerations granted by the GREA—is
approximately 16 ct/kWh for 20 years [
], while the
current market value is only around 3 ct/kWh.2
Assuming this to be the average market value for the
time periods in question, the net cost to the German
national economy—paid via the GREA
apportionment (EEG-Umlage; 6.9 ct/kWh in 2017) by the
electricity consumers—can thus be estimated to be
approx. 13 ct/kWh. It follows that the total costs of
this previous RE expansion are probably on the order
of €450 billion (13 ct/kWh × 170 TWh × 20 years).
In stage 2, the task is—with respect to the
electricity sector alone, i.e., not taking into account the
socalled sector coupling—to achieve the milestone in
2030 of an approx. 50% share of RE in electricity
generation, which translates to an RE electricity
production of 280–300 TWh. Since it can be estimated that
50–70 TWh/a of RE electricity from plants built in
stage 1 will go out of the system by 2030 (due to end
of technical lifetime or to insufficient economic viability
without the GREA remunerations), it will be necessary to
build RE plants with an average electrical production of
The latest auction results in Germany for PV, for
offshore wind farms and for onshore wind parks suggest
that the necessary remunerations to the investors for
this electricity will not exceed 6–7 ct/kWh—far below
anything that was expected only 2 or 3 years ago. And
they might still be considerably lower due to further cost
degressions to be expected on the way to 2030. But even
not taking this into account, and again assuming an
average market value of only 3 ct/kWh (many experts
expect rising prices on the electricity exchange EEX in
the next decade), the net cost can be estimated to be
3–4 ct/kWh. The total cost, then, of the expansion of
RE in the second stage can be estimated to be in the
range of €90–130 billion (3–4 ct/kWh × 140–
160 TWh × 20 years).
Taken together and considering smaller cost effects (see
“Methodology” section), stages 1 and 2 will probably
amount to a cost balance of between €600 and 700 billion.
It is obvious here that roughly 75% of these costs accrued
in stage 1. The main reasons are clear: up until 2010,
especially PV electricity was extremely expensive (Table 1),
aIncluding auction result from February 2017; own calculations
and in general, the massive expansion of RE first in
Germany and then in many other countries has led to
drastic cost degressions of PV and Wind in the last years.
Nuclear phase-out and early action as distinctive features
No matter the future of energy policies around the
globe, there is little doubt that the German
Energiewende is and will remain unique in two respects:
Despite ambitious climate goals, Germany
decided to phase-out the CO2-free nuclear
energy until 2022. Obviously, this decision—which
is still unparalleled in any other country with
nuclear power plants—renders it much harder and
more expensive to achieve CO2 reduction goals3
Germany was the first country to massively expand
RE for electricity generation. This “early action”
certainly has had its merits, but—as just shown—it
is also responsible for much of the substantial
Energiewende bill up to now.
Put another way: because of these two unique features
heavily influencing the financial aspects, no inference
whatsoever can be made from the cost of the
Energiewende to the cost of energy transitions in other
countries. Consequently, the German figures should have no
deterring effect at all—they simply do not give a clue as
to what an energy transition might cost in a comparable
Cost of a transition of Germany’s electricity sector
without (rapid) nuclear phase-out and without
We can take the same point further and consider a
transition scenario for Germany’s electricity system without
these two distinct characteristics of the Energiewende: a
scenario where Germany—with the same CO2 reduction
goals in electricity generation—continues to operate its
newer nuclear power plants (i.e., those built after 1980)
beyond 2030, and where the expansion of RE is launched
only in 2017.
What would such a transition cost (until 2030)?
The starting point for this fictitious energy transition
would be the electricity generation mix, had the
Energiewende between 2000 and 2016 not happened4;
presumably, it would look something like this5:
Nuclear energy 160 TWh.
Renewable energies 30 TWh.
Fossil fuels 425 TWh.
Total 615 TWh
The target state in 2030 would be defined by a 50%
share of CO2-free energies as well as by a substantial
advance in energy efficiency. Taking the same rate of
decrease in electricity consumption in the years 2017 to
2030 as has actually happened in the real Energiewende
between 2010 and 2016, we assume a domestic demand
of approximately 570 TWh in 2030. The target
electricity mix in 2030 then looks like this:
Nuclear energy 100 TWh
Renewable energies 190 TWh
Fossil fuels 280 TWh
Total 570 TWh
Consequently, the transformation in this scenario
consists of expanding RE by around 160 TWh between 2017
and 2030. We further assume that the new GREA to be
introduced in 2017 is basically similar to the actual
GREA in place—it works with the tender model, it
grants fixed remunerations for 20 years, and it provides
around 15 GW of offshore wind, 40 GW of onshore
wind, and 30 GW of PV (only open space systems).
As seen in the “Two stages of the Energiewende up to
2030” section, the upper limit for the cost of this RE
expansion can be estimated to be only 3–4 ct/kWh (net); the
total cost would thus probably not exceed (3–4 ct/kWh ×
160 TWh × 20 years =) €100–120 billion, to be paid in the
years 2018 to 2050.
In this scenario, until 2030 only moderate grid
expansion will be needed6—the amount of RE electricity in
2030 (190 TWh) equals the actual RE production in
Germany of 2016, which has been integrated into the
existing electricity system with only limited extra
measures (switching off RE-plants, redispatching
conventional power plants); these costs would in all probability
be offset by the positive effect of lower EEX prices.
Taking into account costs to enhance electricity
efficiency, the total cost of such a transition of Germany’s
electricity system up to 2030 can be estimated not to
exceed €150 billion,7 or on average less than €5 billion per
In terms of GDP—assuming moderate real growth of
1% and not even taking into account inflation—, this
corresponds to an annual average of less than 0.15% of GDP.
Let us discuss this result with a few comparative figures:
– The GREA apportionment would not exceed
2 ct/kWh (compared to a maximum of approx.
8 ct/kWh in the actual Energiewende, expected
for around 2023).
– The burden on private households would thus not
exceed approximately €5 per month (2017 prices);
on average, it would amount to only roughly 0.1%
of future consumption spending (whereas, e.g.,
alcoholic beverages and tobacco products each
amount to more than 1% of consumption spending).
– The average “electricity transition bill” to German
businesses would equally be less than 0.1% of sales
volumes in the future.
– The subsidies for hard-coal mining in Germany
between 1970 and 2010 amounted to around 280
billion, equating to an annual average of more
than 0.3% of GDP.
In other words, this fictitious transition of Germany’s
electricity system—up to a 50% decarbonization by
2030—could be considered as a feasible challenge to the
German national economy and as quite affordable for its
Contrary to first sight, the experience of Germany with
its Energiewende does not show that any energy
transition in an industrialized country is an expensive
undertaking and is bound to place a heavy financial burden on
private households and impair the competitiveness of its
businesses. Indeed, the cost figures often cited in this
context are decisively influenced by unique features of
the German energy transition that do not hold true for
Actually, the true message is the opposite one: at least
in Germany, the transformation of the electricity system
with a goal of 50% decarbonization by 2030, launched
today and without nuclear phase-out, would not be
expensive at all. While this result, of course, cannot be
translated to other countries as such, it should
nevertheless be encouraging to all those around the globe who
work for energy transitions in their countries.
1With “merit order effect” we denote the fact that the
prices on the German energy exchange are lower (by
around € 10–15/MWh) due to the Energiewende since
the most expensive power plants are being ousted by the
RE (with variable costs close to 0). This effect lowers the
electricity bill for the customers.
2All prices will be given in Euro-cent per kilowatt
hour = ct/kWh.
3This is true for the direct costs which are the focus of
this article; we do not take into account external costs
here (“Methodology” section).
4We will assume in the following that even without
the Energiewende, the cost degression of RE would have
taken place, knowing that, in fact, it has played an
important role in this.
5We do not take into account here possible electricity
6New power lines would be necessary mostly to
connect the offshore windfarms to the grid and to transport
part of the wind electricity down south.
7This holds all the more true since without the
transition envisaged here, a few conventional power plants
might have to be replaced until 2030.
I am grateful to many colleagues for discussions on the German Energiewende
which have helped me to enhance my knowledge and to balance my views, in
particular to Prof. Peter Birkner, Dr. Christoph Müller, and Jörg Stäglich.
Consent for publication
The author declares that he has no competing interests.
1. Handelsblatt 2017. Strompreise geraten ausser Kontrolle http ://www. handelsblatt.com/unternehmen/energie/kosten-fuer -energiewendestrompreise-geraten-ausser-kontrolle/14667630.html . Accessed 13 Sept 2017
2. Welt 2016. Die Energiewende droht zum ökonomischen Desaster zu werden https://www .welt.de/wirtschaft/article162600538/ Die-Energiewende-drohtzum-oekonomischen-Desaster-zu-werden . html. Accessed 13 Sept 2017
3. DICE Consult 2016 . Kosten der Energiewende-Untersuchung der Energiewendekosten im Bereich der Stromerzeugung in den Jahren 2000- 2015 in Deutschland http://www.insm.de/insm/dms/insm/text/soziale-marktwirtschaft/ EEG/INSM_Gutachten_Energiewende.pdf. Accessed 13 Sept 2017
4. Fraunhofer ISE 2015 . Was kostet die Energiewende? https://www .ise. fraunhofer.de/de/veroeffentlichungen/studien/was-kostet -die-energiewende . html. Accessed 13 Sept 2017
5. Unnerstall , T. 2017 . The German energy transition-concept, implementation, cost and lessons . Springer May 2017
6. Netztransparenz 2016. Prognose der EEG-Umlage 2017 nach AusglMechV https ://www.netztransparenz.de/portals/1/Content/EEG-Umlage/EEGUmlage%202017/20161014_Veroeffentlichung_EEG-Umlage_ 2017 .pdf. Accessed 13 Sept 2017
7. BDEW 2015 . Erneuerbare Energien und das EEG ( 2015 ) https://www.bdew. de/internet.nsf/id/20150511-o -energie-info-erneuerbare-energien-und-daseeg-zahlen-fakten- grafiken- 2015-de/$file/Energie-Info_ Erneuerbare_ Energien_und_das_EEG_ 2015 _ 11 . 05 . 2015 _final. pdf. Accessed 13 Sept 2017
8. Bundesnetzagentur 2017 https://www.bundesnetzagentur.de/DE/Sachgebiete/ ElektrizitaetundGas/Unternehmen_Institutionen/Ausschreibungen/ Solaranlagen/BeendeteAusschreibungen/Ausschreibungen2017/ Ausschreibungen2017_node. html. Accessed 13 Sept 2017