Agora 12 Insights On Germanys Energiewende Web

12 Insights on
Germanys
Energiewende
February 2013
12 Insights on
Germanys
Energiewende
Imprint
12 Insights on Germanys Energiewende.
A Discussion Paper Exploring Key Challenges
for the Power Sector
This document is the slightly updated English version
of 12 Thesen zur Energiewende. Ein Diskussionsbeitrag
zu den Herausforderungen im Strommarkt,
Agora Energiewende 2012 (original: German language).
Agora Energiewende
Rosenstrae 2 | 10178 Berlin | Germany
T +49. (0) 30. 284 49 01-00
F +49. (0) 30. 284 49 01-29
www.agora-energiewende.de
info@agora-energiewende.de
Typesetting:
UKEX GRAPHIC
www.ukex.de
Print: Oktoberdruck, Berlin
Cover: iStockphoto.com/visdia
010/03-I-2013/EN
Printed on 100% recycling nature paper
FSC Circleoffset Premium White
Introduction
In June 2011 the German Parliament ended a long debate

of several decades over the future of the countrys energy
system with a historical decision. It was historical
because of its ambitious goal:
Germany wants to transform its power sector from
nuclear and coal to renewables within the next four
decades. The decision was also historical because it was
an almost unanimous vote, a consensus of ruling and
opposition parties.
This transition of Germanys energy system is called

Energiewende. Germany will phase out nuclear energy;
the remaining nuclear power plants will be gradually
closed down by the end of 2022. The countrys targets to
transform its energy system are summarized below.
Since the long-term targets are in place the big question

remaining to be answered is how to accomplish these
goals. Namely, how to manage the transition from a fossilnuclear system to a mainly renewable energy system
at the lowest possible cost and without compromising
the high standard in system reliability? Our attention is
focused on the next 10 to 20 years, or the period up to
2030.
We know from numerous studies and scenarios that

the power sector will be crucial in this transition. This
sector is where key decisions have to be made in coming
years from reforming the energy market design to infrastructure planning. Knowing that the time horizon for
implementing such decisions in the energy sector should
be contemplated in decades rather than years, the need
Status Quo and Main Targets of the Energiewende

2011
2020
2030
2040
2050
-26.4 %
-40%
-55%
-70%
-80 to -95%
primary energy use (against 2008)
-6%
-20%
-50%
electricity demand (against 2008)
-2.1%
-10%
-25%
n.a.
-20%
-0.5%
-10%
-40%
share in electricity consumption
20.3%
35%
50%
65%
80%
share in final energy use
12.1%
18%
30%
45%
60%
Greenhouse Gas Emissions

GHG (against 1990)
Efficiency
heat in residential sector

energy use in transport sector (against 2005)
Renewable Energy
Introduction
for urgent action is becoming more and more obvious. For

these reasons this paper focuses on the power sector.
are abundantly available in most regions of the world and

production costs are going down rapidly.
The 12 Insights on Germanys Energiewende specifically describe the German context. However, we strongly
believe that many of the developments we currently see
in Germany will be highly relevant for other countries and
regions as well certainly, but not only, in Europe where
Germany is a key part of the integrated energy system.
In many countries, wind power and photovoltaics will

form the basis of a future, low-carbon energy system and
bring with them similar challenges that Germany is facing
today. We leave it to the reader to make up his or her mind
on this assertion and we look forward to any comments
and ideas in this regard.
Our main insight entitled Its all about Wind and Solar
certainly applies beyond Germany, as wind and solar
Rainer Baake
and the team of Agora Energiewende
GW
1. Its all about Wind and Solar!

features
70
Wind and PV are the

cheapest renewable
energy sources
The potential of other
renewable energy
sources is limited
> weather dependent

> variable
> only capital expenditure
60
50
40
30
20
10
0

Mo Tu We Th Fr Sa Su
How to balance demand and supply?

How to minimise costs?
How to realise the Energiewende in the European context?
technical system
2. Base-Load power plants disappear altogether,
and natural gas and coal operate only part-time
3. There is plenty of flexibility but so far it has no
value
4. Grids are cheaper than storage facilities
5. Securing supply in times of peak load does not
cost much
6. Integration of the heat sector makes sense
market design and regulation

7. Todays electricity market is about trading kilowatt
hours it does not guarantee system reliability
8. Wind and PV cannot be principally refinanced via
marginal-cost based markets
9. A new Energiewende Marked is required
10. The Energiewende Market must actively engage the
demand-side
1 1. It must be considered in the European context
12. A saved kilowatt hour is the most cost-effective kilowatt hour
Content
Insight 1
Its all about wind and solar!
Insight 2
Base-load power plants disappear altogether, and natural gas and coal operate only part-time
Insight 3
Theres plenty of flexibility but so far it has no value
11
Insight 4
Grids are cheaper than storage facilities
14
Insight 5
Securing supply in times of peak load does not cost much
16
Insight 6
Integration of the heat sector makes sense
18
Insight 7
Todays electricity market is about trading kilowatt hours
it does not guarantee system reliability
20
Insight 8
Wind and PV cannot be principally refinanced via marginal-cost based markets
22
Insight 9
A new Energiewende market is required
24
Insight 10
The Energiewende market must actively engage the demand-side
27
Insight 11
The Energiewende market must be considered in the European context
29
Insight 12
Efficiency: A saved kilowatt hour is the most cost-effective kilowatt hour
31
References33
Agora Energiewende | 12 Insights on Germanys Energiewende
12 Insights on Germanys Energiewende
Insight 1:
Its all about wind and solar!
Two winners have emerged from the technology competition initiated by the German
Renewable Energy Act wind power and
photovoltaics, the most cost-effective technologies with the greatest potential in the
foreseeable future
For the time being, the winners of the technological competition initiated by the German Renewable Energy Act of the
year 2000 (Erneuerbare-Energien-Gesetz, EEG) have been
determined: the lowest-cost ways to generate electricity
from renewable energy are by wind power and photovoltaics (PV). Given the present state of the art, no other
renewable energy technology can generate electricity in
sufficient quantity at such a low cost. As a result, Germanys
Energiewende will be based upon these two technologies.
Underlying this fact is the enormous decline in cost for the
key technologies in wind power and solar energy that has
occurred during the past 20 years. For wind, the costs for
generated power despite rising raw material costs for steel
have fallen by about 50% since 1990. For photovoltaics,
the change has been even more remarkable. In this sector,
systems costs have fallen by 80 to 90% over the same time
period. Furthermore, there is no end in sight to this trend
toward falling costs for either technology.1
All other renewable technologies are either

significantly more expensive or have limited potential for further expansion (water,
biomass/biogas, geothermal energy) and/or
are still in the research stage (wave power,
energy from osmosis processes, etc.)
Bioenergys share of German power generation will continue to be limited to less than 10% for the long term (2012:
1 See IPCC (2011) and IRENA (2012a); (2012b). The cost of

installing a photovoltaic plant in Germany has since fallen
significantly more since the figures cited in these studies.
approx.6%). The reason is that agricultural and forest

acreage is limited in Germany and other countries, and
the use of wood and energy crops in the energy system
competes directly with many other potential land uses, such
as raising food crops, producing raw materials for industry
(for example, the paper or chemical industries) and nature
conservation. In addition, biomass is a relatively expensive
energy source for generating electricity, the costs of which
have risen rather than fallen in recent years.2 In addition,
the quantity of low-cost, sustainably produced wood
imports is limited, among other reasons due to the growing
demand for food and biomass in developing and emerging
countries.3
In addition to wind, PV and biomass, hydropower and
geothermal both currently contribute to total power production. However, all forecasts suggest that these sources will
not contribute substantially more to power production in
the future. Even if some limited potential still exists for
expanding hydropower in Germany, it will not play a central
role in the electricity supply.4 Given the current state of the
technology, the cost of power produced from geothermal
resources will remain very high over the long term, and
for this reason, its current contribution of less than one
per thousand is not expected to increase very much in the
future.5 Other technologies, such as tidal power, wave power
and energy from osmosis are still in the research stage and
are not nearly ready for large-scale application.6
2 In 2002, basic remuneration for small biomass plants was
10.1 ct/kWh, and in 2012 it is 14.3 ct/kWh. On top of this, there
are additional bonuses of up to 18 ct/kWh which have been
established in recent years. As a result, the average remuneration for electricity from biomass plants today is 19.6 ct/kWh.
3 See, for example DLR/FhG IWES/IfNE (2012); Prognos/EWI/GWS
(2010); Prognos/ko-Institut (2009); SRU (2011); UBA (2010).
4 See Ingenieurbro Floecksmhle et al. (2010).
5 Thus, for example, the EEG remuneration rate of 9 ct/kWh in
2000 was increased to the current rate of 25 ct/kWh, without
bringing about any significant increase in production.
6 More research support for all renewable energy technologies makes
By 2015, the cost for power produced by

wind and photovoltaic plants will be 710
euro cents per kilowatt-hour the cost of
electricity from a system consisting of wind,
PV and backup storage capacity will thus be
about the same as electricity generated by
new gas and coal-fueled plants
The EEG remunerates onshore wind farms at a current rate
of about 710 (euro) ct/kWh and PV plants at about 1218ct/
kWh; both rates dependent on plant size and location, and
remuneration is guaranteed for 20 years.7 As a result of the
decreases in the feed-in tariff established by the EEG and
further decreases in the costs of these technologies, it will be
possible by 2015 to generate electricity by newly built wind
and PV power plants in the range of 710 ct/kWh.8
Since neither the wind nor the sun is always available, any
power system that depends upon these sources must also
have supplementary power plants, and for the time being,
sense overall as a way of further reducing costs and, if possible,
creating renewable technologies for generating electricity with
even more favorable power production costs than wind and PV.
7 See Federal Ministry for the Environment (2012a) and (2012b).
Offshore wind power is currently significantly more expensive:
The feed-in tariff in the EEG is 15 ct/kWh for at least 12 years,
or, in the acceleration model, 19 ct/kWh for at least eight years.
After 8 or 12 years, offshore wind facilities receive 3.5 ct/
kWh. However, the time for receiving 15 ct/kWh is extended
according to the distance of the plant from the coast and the
depth of the sea. In addition, the network connection costs
are covered through network fees as well as support from the
KfW Development Banks Offshore Guarantee Program.
8 This is generally valid for wind-power plants on land (onshore)
as well as for larger PV plants. For offshore wind facilities,
future cost trends remains to be seen, as the decline in current
remuneration rates is only scheduled to begin in 2018 according
to the EEG. Depending on their size, PV plants will be receiving a remuneration of approx. 914 ct/kWh as of the end of
2014 if they expand in line with the expansion corridor set in
the EEG; if they expand very rapidly, depending on the size of
the plant, the remuneration rate will be 69 ct/kWh. After this
time, even smaller PV rooftop systems in sunnier locations in
Germany will be able to produce electricity for 10 ct/kWh, for
example, under the following quite realistic conditions: installation costs: 1000 per kWp; electricity yield: 1000 h per year;
25 years working life; operating costs: 1% of installation costs
per year; interests costs for capital investment: 5% annually.
these will continue to be primarily dependent on fossil fuels.

In the short term, existing power plants will take on this
backup function (current electricity prices on the exchange:
approx. 5 ct/kWh). In the medium term, investments will
be needed for new fossil-fueled power plants to be able to
cover the demand even at those times when no electricity
can be produced from renewable energies. Given that the
costs of generating electricity from new gas or coal power
plants is similarly about 710 ct/kWh,9 and that providing
peak power needs is also relatively inexpensive (see Insight
5), then by 2015, it is anticipated that the cost of generating
electricity from a system based upon new wind, photovoltaic and flexible fossil fuel power plants will be comparable
to that of alternative investment in traditional coal or
gas-based power systems.
Wind and PV power are the two essential

pillars of the Energiewende
Clearly, Germanys Energiewende will be based on wind
power and PV. There is no realistic alternative path. By
the time renewable energy covers one-half of total power
demand, the share of wind (onshore and offshore) and PV
will already be 35%. The greater the share of renewable
energy, the more important will be the role of wind and PV
when compared to other renewable technologies, whose
potential for expansion is limited. According to the Federal
Network Agency (Bundesnetzagentur) by 2022 wind and PV
power will be generating about 70% of the power obtained
from renewable energy.10 Subsequently, their share will
continue to rise to as high as 8090%.
Figure 1 illustrates the importance of power generation from
wind and PV for Germanys Energiewende over the medium
term. It presents the findings of model calculations concerning electricity demand and generation in Germany for three
different weeks in 2022. The upper red line represents
demand in GW, and the different colors represent generation from renewable energy, while the gray area depicts
9 See. (EWI) 2011, p. 2729 and 40 or DLR/IWES/IfnE (2012), p. 217.
10 According to Lead Scenario B for the 2012
Network Development Plan.
Electricity Demand and Generation by Renewable

Energy in Three Examplary Weeks in 2022
Early February (calendar week 6)
GW
80
60
40
20
Mo
Tu
We
Th
Fr
Sa
Su
Wind and solar energy plants have three key

characteristics:
Mid-August (calendar week 33)
GW
residual demand that will still need to be covered by existing

fossil fuel power plants. The results show that as early as
2022, there will be about 200 hours during which power
production from wind, sun, water and biomass will exceed
Germanys total demand for electricity. However, there will
also be many hours when renewable energy will only be able
to produce a small quantity of electricity. From this analysis,
we can identify important challenges, which are described
further in this paper.11
At the heart of the challenge for the Energiewende will be
to continuously balance fluctuating electricity production
from wind and solar plants with consumer demand. This
will require flexibility on both the supply side and the
demand side.
80
60
40
20
Mo
Tu
We
Th
Fr
Sa
Su
They have high capital costs but (virtually)

no operating costs
Running costs (maintenance, operation) for wind and photovoltaic facilities are very low, and represent about 13% of
capital costs per year.12 Because there is no need for fuel, the
Late November (calendar week 47)
GW
80
60
40
20
Mo
Tu
We
Th
Fr
Sa
Su
Electricity demand
Fossil fuel power plants
Photovoltaik
Wind onshore /offshore
Hydropower
Biomass
Illustration based on Agora Energiewende (2012a)
They dependent upon energy source availability; that

is, power production depends upon the weather
Unlike fossil energy sources, the output from wind and PV
power plants cannot be controlled in response to either
demand for electricity or the price signals on the exchange.
These power plants produce energy when the sun shines or
when the wind blows.
11 This graph was prepared by Fraunhofer IWES at the request of

Agora Energiewende (Agora 2012a). The calculations were based
upon what is known as Lead Scenario B from the framework of
scenarios for the network development plan authorized by the
Federal Network Agency at the end of 2011. Complete data for all 52
weeks can be found at www.agora-energiewende.de/download.
12 Fixed operating costs of a PV plant are about 1-1.5% of capital
costs per year, where the short-term marginal production costs
of a functioning facility are zero. For wind power plants, operating costs are about 2 4% of capital costs per year, where the
operating costs also include wear and tear of materials and as a
result, the short-term marginal production costs are a bit greater
than zero (see McKinsey (2010), p. 63; DLR/IWES/IfnE (2012),
marginal costs are virtually zero. This means that with wind
and solar power, the initial capital investment pays almost
completely for the cost of producing electricity over the next
2030 years.
Their production of electricity fluctuates rapidly
As the result of weather variations, such as wind blasts or
periods of calm as well as the passage of cloud banks, the
power fed into to the system from solar and wind plants
sometimes fluctuates widely. This means that the rest of
the power system fossil fuel plants, electricity demand,
electrical storage facilities must be very flexible in order
to adjust to the fluctuating input behavior of wind and PV
power plants.
These characteristics are fundamentally

different from those of coal and natural gas;
they profoundly alter the energy system and
energy market
Natural gas and coal power plants are controlled according
to the market price of electricity on the exchange, and their
operating costs (fuel, CO2 emissions rights) vary greatly.
In addition, to date power plants have not been built with
the technology to enable them to make rapid adjustments
in production. The Energiewende, with a growing share of
wind and solar power, will fundamentally alter the power
system and the power market.
Wind and PV power should be expanded in

tandem since they have mutually complementary features; generally speaking, the
wind blows when the sun is not shining and
vice versa.
is shining at its strongest.13 Since electricity is relatively

expensive to store because of efficiency losses, it makes
sense to use less costly options for continuously balancing
supply and demand from the perspective of the system as
a whole. This includes raising the overall flexibility of the
system (see Insight 3).
However, we should also take advantage of the differences
in the input behavior of wind and solar power even if
electricity production costs are somewhat lower for wind
power than for solar power.
This logic using asynchronous input behavior as a way
to minimize total system costs also applies to the regional
distribution of renewables. Since the wind blows at different times in different regions in Germany, wind power
should not be generated exclusively in Northern Germany,
and solar power should not be generated only in Southern
Germany.
It is true that production costs for wind power are somewhat
lower on the coast than in the rest of Germany, and that for
solar power they are lower in Bavaria and Baden-Wrttemberg, but if generation only took place in the most advantageous regions, one could only secure wind or solar power
at times when weather conditions in that region permitted.
From the perspective of optimizing the system as a whole, it
makes sense to take advantage of the differences in weather
conditions across Germany in such a way that electricity production from wind and PV is distributed across
the greatest possible number of hours in the year, with
electricity transmitted between regions using an expanded
transmission grid. Little research has been conducted as of
yet about the precise times that PV and wind power plants
typically produce electricity and which temporal feed-in
patterns new facilities should have.
Wind power plants are especially good at generating electricity in the winter, and solar plants in the summer. Most
sunshine occurs around the middle of the day, whereas
wind can happen at any time throughout the day and
often wind is blowing the least at those times when the sun
p. 1 of the data appendix; IRENA (2012a); IRENA (2012b).
13 See for example, the analyses by Gerlach/Breyer (2012) for

central Germany and E.on Bayern (2011) for Lower Bavaria.
Insight 2
Base-load power plants disappear altogether,
Wind and PV will form the basis of the power
supply, with the rest of the power system
being optimized around them; most fossilfueled power plants will be needed only at
those times when there is little sun and wind,
they will run less hours, and thus their total
production will fall: Base load power plants
will be a thing of the past.
Base load is a category of demand for electricity and, in
Germany, refers to the 35 to 40 GW that is the minimum
consumed at all times over the year. Previously, the base
load was supplied through fossil-fueled power plants that
ran 24/7, which led to the misleading term, base-load
power plant. By guaranteeing priority access to the grid, the
Demand for Fossil Fuel Power Plants in 2022:
Example of a Week in August
GW
Almost no fossil
fuel power plants
20-30 GW from
fossil fuel power plants
80
EEG has turned renewables into base load power by law.

Renewable energy is effectively taking the place of conventional base load power plants. In the future, wind and solar
power will cover an ever-growing share of the electricity
demand. Already in 2022, total load (base, mid- and peakload) will be completely covered by renewable energy alone
during many hours throughout the year, as shown in Figure
2 below. In the week that is illustrated, a large number of the
coal and gas fired power plants remain out of service during
the first half of the week, but are needed during the second
half. Use of the remaining fossil-fueled power plants must
be oriented to demand and the production from renewable
energy sources.
Wind and PV will reduce the total quantity of power
produced from fossil fuels, and thus reduce the load factor
and number of operating hours. With a 40% share of power
coming from renewable energy, only 10 to 25 GW conventional capacity operating between 6000 to 8000 hours per
year will still be needed.14 In ensuing years, this need will
decrease even further.
Rapid changes in feed-in from renewables as

well as forecasting uncertainties will create
new requirements for both short- and longterm flexibility
60
40
20
Mo
Tu
We
Th
Fr
Sa
Su
From Monday to Thursday, wind and solar power cover most

of the demand for electricity, fossil fuel power plants are
scarcely needed
B
etween Thursday afternoon and Sunday morning, 20-30 GW
of additional power plant capacity are continuously required
The fluctuating patterns of generation from wind and solar

power will set completely new requirements for the power
plants of the future: electricity generation from controllable
14 See VDE (2012a), p. 43; Consentec/r2b (2010a), p. 51; IWES
(2009). For 40% renewables, the VDE (2012a) calculates the
need for 10-15 GW of capacity which would be used almost
constantly (more than 8000 load hours per year). For 50%
renewables, Consentec/r2b (2010a) calculates that 18 GW would
be used for more than 6000 load hours, and IWES (2009) calculates a need for 21-26 GW for more than 7000 load hours.
power plants must be rapidly ramped up and down over

short periods of time in order to compensate for these fluctuations. As the share of wind and solar power increases,
this will also apply to the few remaining base-load power
plants. In the future, all remaining fossil fuel power plants
will need to operate on a flexible basis.
The quantities of wind and PV can never be accurately
predicted in advance. They always entail forecasting
uncertainties, which become ever greater the further one
looks into the future. The daily and hourly forecasting of
wind15 and PV may improve with more accurate weather
forecasting data. With the right regulatory framework,
producers can take account of these short-term forecasts
when scheduling the operation of their power plants,
thereby optimizing generation.
Long-term forecasting for one or more years in advance is
even more difficult, and fraught with greater uncertainty.
Major fluctuations occur between one year and the next,
especially with regard to wind. This was particularly
evident in 2010 when power production from wind farms
in interior locations in Germany was 25% lower than the
average for the previous ten years, and in coastal areas, it
lagged by 15%.16 Since the total installed capacity of nonrenewable power plants must be sufficient to accommodate
for wind-poor years, they will experience lower load factors
during years with ample wind.
Over the medium term, combined heat-andpower as well as biomass plants need to be
operated according to the demand for electricity
Today, the operation of combined heat-and-power (CHP)
and biomass power plants is not generally dictated by
demand for electricity. Instead, CHP facilities are driven by
the need for heat, which is to say, when there is a need for
15 For offshore wind and PV, many solutions have already been
developed for the challenges that have arisen. Offshore wind is
characterized by further challenges resulting from greater and
more rapid fluctuations, and until now, there have been fewer
measuring points, which may require new technological solutions.
16 See IWR (2012).
10
heat, the facility runs to produce heat, and merely produces

electricity as a by-product. Similarly, most biomass plants
run in continuous operating mode, since they are most
economical this way as a result of remuneration rules set
by the EEG. In the medium to long term, this will have
to change. Electricity generation from CHP plants shall
increase to 25% by 2020 according to official government
targets. Therefore, over the medium term, these plants will
constitute the majority of controllable power generation
in Germany.17 While CHP and biomass power plants only
contributed about one-fifth of controllable power generated in 2010, already in 2020 they will produce more than
one-third of such power.18 Therefore, as wind and PV take
on a greater proportion of total electricity production, the
operation of CHP and biomass plants will have to be geared
to the demand for electricity.
Demand side management and storage

contribute to maintaining system balance
Flexibility options such as demand side management and
pumped storage will help assure the more efficient use of
the power plant fleet in the future. Through demand side
management, the demand for electricity will be shifted to
those times when more wind and sun are available. During periods of high renewable power availability, pumped
storage facilities will take in electricity. When there is little
wind and the sun is not shining, they release it back into the
system. This approach will reduce expensive start-up and
shutdown procedures at fossil-fueled power plants, optimize the use of cheaper power plants, and minimize overall
costs to the system as a whole.
17 See Cogeneration Protection Law, Paragraph 1.

18 See the Federal Environment Ministrys Lead Study Renewable
Energies 2011 (DLR/IWES/IFnE (2012)), p. 19; Scenario 2011
A describes a capacity of 22 GW of power produced by CHP
and biomass for 2010, and 90 GW from other power plants;
the anticipated capacities for 2020 are 31 GW from CHP and
biomass and 61 GW from other power plants; for 2040: 30
GW from CHP and biomass and 32 GW from other power
sources. The share for biomass in total electricity generation
thus remains limited to about 10%, and the power provided
from non-biomass plants without CHP to about 3 GW.
Insight 3
Theres plenty of flexibility
but so far it has no value
In the future, fluctuations in wind and PV
production will demand significantly greater
flexibility from the power system
As previously shown in Insight 2, the electrical system will
have to respond more flexibly as the share of wind and PV
increases. Figure 3 illustrates this need for flexibility. In the
case presented, the wind dies down in tandem with a drop
in the generation of solar power. As a result, controllable
power plants have to cover a major portion of the demand
within a few hours. In a worst case scenario, demand might
increase at the very same time for example, if a large part
of the population comes home at the time of sunset and
turns on electrical appliances, television sets and lights.
Flexibility Requirements in 2022:

Example of a Week in August
Difference in wind and solar

production: 30 GW in 4 hours
GW
80
60
40
20
Mo
Tu
We
Th
Fr
Sa
Su
Thursday from 10 AM to 1 PM, electricity demand is covered

almost completely by wind and solar power
S
tarting from 1 PM, power generation from both wind and
solar goes down , by 5 PM about 30 GW of supplementary
power plant capacity is required
At a renewable energy share of 50% in the German power

system, one has to anticipate extreme cases in which the
load that must be covered by controllable power plants
increases by about 40 GW within a span of four hours.19
This is the equivalent of more than half of todays load in
Germany. Within a 15-minute-intervall, changes in load of
up to 6 GW can be expected. The need for this level of flexibility will create entirely new challenges for power systems
in the future.
Technical solutions to provide sufficient

flexibility readily exist today
The need for flexibility can be met through various flexibility options on the supply-side, demand-side, through
storage and improved grids. With increasing requirements
for flexibility, these options should be deployed in the order
of their overall economic cost efficiency. In addition to grid
expansion (see Insight 4), from todays perspective the most
important flexibility options are as follows:
Operation of combined heat-and-power and biomass
plants according to electricity demand
To achieve flexibility in an efficient way, we first need to
avoid unnecessary inflexibilities in the electrical system.
Power plants that are readily controllable from a technological perspective should be run to optimally supplement generation from wind and PV plants. As described in Insight2,
this is not the case today for either CHP plants or biomass
facilities. In the future, however, such plants will constitute
a large percentage of the controllable power generation
fleet in Germany. Accordingly, they will need to be operated
above all to respond to electricity demand, along with the
demand for heat. This form of operation poses no technical
19 See IAEW/Consentec (2011), p. 17; considering a
scenario with 50% renewable energy in 2030.
11
problems and is associated with relatively low costs. In the

case of CHP plants, this merely requires that heat is fed into
storage facilities or district heating grids, which can be done
for a few hours without major challenges (see Insight6).20
In the case of new biomass plants, this merely requires an
adapted design of the overall facility for example, with
regard to the relationship between fuel storage capacity and
generator output.
their load by as much as 4 GW within five minutes (today:

1 GW). Similarly, optimization would allow to ramp down
generation output from 10 GW hard-coal fired plants under
continuous operation to 2GW (today: 4 GW).
Greater flexibility from retrofitted existing power plants
and from new plants would also contribute substantially to
reducing the minimum feed-in from thermal power plants
(the so-called must-run level). 21
Improved flexibility of fossil-fueled power plants

(minimum output, start-up times)
Fossil-fueled generators offer very large potentials for
improved flexibility. Coal and gas-fired power plants can
be made more flexible through technical and organizational
modifications. The minimum output rate can be reduced,
load gradients increased, and start-up times shortened. The
differences between the usual degree of flexibility today
and the technical optimization potential are presented in the
following table (Figure 4).
Avoid generation peaks from wind and PV or use

them for heating
With very high shares of electricity generated by wind and
solar it is likely to make economic sense in the future to
limit generation peaks or to use them for producing heat. Yet
designing the grid to transport every generated kilowatthour would be unacceptably expensive, as the grid would
be designed for transport capacity that is only required for
a few hours during the year. Wherever possible, electricity
that cannot be transported should be used purposefully. One
option is to use such power for heating (see Insight 6). Heat
production can take place using electrical heating rods in
warm water accumulators (1 kWh of electricity generates
For example, an optimized gas and steam turbine could be

brought into full service within two hours (today, this takes
four hours). Moreover, optimization would allow 10 GW
gas and steam turbines that are already running to adjust
20 Upgrading to heat accumulators is subsidized by the 2012

amendment to the Combined Heat and Power Law.
21 Today, depending on the particular time point, there is still

a minimum must-run capacity of up to 25 GW in Germany
required for system stability, but it should be possible to
reduce this in the long term, see BMU (2012c), p 22.
Flexibility of Fossil Fuel Power Plants
Optimization potential (first figure) and typical status today (figure in parentheses) per 1000 MW
Hard coal power plant
Lignite coal power plant
Combined cycle power plant
Gas turbines
Minimum load
MW
200 (400)
400 (600)
300 (500)
200 (500)
Maximum change in
load within 5 minutes
MW
300 (75)
200 (50)
400 (100)
750 (400)
4 (10)
6 (10)
2 (4)
<0.1
Start-up time
cold start
Illustration based on VDE (2102a)
12
1kWh of heat) or by using heat pumps (1 kWh of electricity

generates about 4 kWh of heat).
Load shifting and interruptible loads in industry
Load management is another cost-effective flexibility
option with great potential, especially in industry. Industry consumes about 40% of the electricity generated in
Germany,22 and a considerable proportion of this consumption takes place in large plants with centrally controlled
processes. From a technical point of view, it would be readily
possible in many cases to shift demand by several hours
by adapting processes and, if necessary, installing storage
capacity for intermediate products, for heat, cold or compressed air. The flexibility potential achievable in industry
over the medium term is estimated to be about 4.5 GW.23
Further large and cost effective potentials can be achieved
in the area of commerce. Large refrigeration or heating
facilities, for example, can be centrally controlled as well as
upgraded to store heat or cold for a short time.
The challenge is not about technology and

control, but rather about incentives
The flexibility options described here are already technically available today and can be implemented at relatively
low cost. Since they primarily involve large-scale facilities
(CHP plants, biomass facilities, industrial processes, large
heat storage), the problem of controlling them is technically easy to solve in contrast to small-scale household
facilities such as washing machines or refrigerators. The
major challenge in accomplishing this is not about technical
implementation, but about providing effective incentives.
The goal should be that in each case, the most cost-effective
option is used first, and to accomplish this, there needs to
be a level playing field for competing flexibilities. Both the
supply side and the demand side need to be able to participate in this competition (see also Insight 10). The speed at
which flexibility potential can be tapped once the proper
incentives are provided was demonstrated by the response
to negative prices on electricity markets in Germany. For
example, in 2009, as a result of high feed-in from wind

and PV plants in a number of hours with low demand, and
because of the lack of flexibility in conventional power
plants, in nearly 100 hours a situation occurred in which
market participants were paid for accepting electricity. By
2011 the number of hours with negative prices had been
reduced by a factor of 10 despite significant increases in
generation from wind and PV.24 Clearly, market participants
were able to provide additional flexibility.
Leveraging small-scale flexibility options at

the household level by using smart meters is
currently too expensive
Seen from todays perspective, small-scale flexibility
options, which can be leveraged in households by using
smart meters, are simply too expensive. It is only in the long
term that such options will make an efficient contribution
to the overall system. Small-scale options should only be
leveraged after all of the more cost-effective options have
been exhausted. For example, in order to use a washing
machine as a flexibility option it must contain a dedicated
control system. The household must have a meter able to
measure real-time electrical consumption, and a control
signal from the electricity market has to get to the washing
machine. The specific expense for implementing such a
system is very large in comparison to the specific expenses
for the large-scale flexibility options described above. In
the long-term perspective, however, especially heat pumps
and electrical vehicles will create substantial potential for
demand side management at the household level. For this
reason, it is important to make sure that in further development of these technologies, consideration be given to their
potential contribution to increasing the flexibility of the
power system.
22 See DLR/IWES/IfnE (2012), p. 20; EWI/GWS/Prognos (2011), p. 37.

23 See VDE (2012b), p. 55.
24 See EnBW (2012), p. 5.
13
Insight 4
Grids are cheaper than storage facilities
Grids decrease the need for flexibility: fluctuations in generation (wind and PV) and demand
are equilibrated across large distances
The larger the area connected by the grid, the greater the
extent to which fluctuations in generation and demand can
be pooled: while generation by a single wind farm (for example, on the North Sea coast) may vary greatly, the sum of the
generation from all wind turbines in Germany (for example,
on the North Sea coast, in Thuringia and in Bavaria) is much
more balanced. The same is true for demand, where regional
fluctuations offset each other as well. More extensive
geographical linkage reduces the need for flexibility.
Grids enable access to cost-effective flexibility

options in Germany and Europe
Grids also make it possible to take advantage of the cheapest
flexibility options over larger distances in Germany and
throughout Europe. For example, at times when there is a large
amount of sun and wind, surpluses can be sold to European
neighbours instead of having to be stored or cut back.
By 2020, surpluses of up to 22 GW are to be expected during
certain hours in Germany, and by 2030, such surpluses
could be as great as 41 GW.25 If this demand for flexibility
was to be met exclusively through storage facilities, it would
be very expensive, since most of these facilities would
be used only rarely. Instead, using grid connections to
European neighbours, a large portion of the surplus could be
sold internationally. During times of little wind and sun in
Germany, electrical power could be bought back from neighbouring countries. In this way, a grid connection functions
as an indirect storage facility.26 Expansion of transport
capacities to nations with especially cost-effective flexibility options would be most advantageous for example,
25 IAEW/Consentec (2011), p. 20.
26 See Prognos (2012), p. 17.
14
to alpine nations and Scandinavia, with large amounts of

hydroelectric and pumped storage facilities. The production of electricity by hydroelectric plants in those nations
can be throttled back when cheap electricity is available
in Germany. For Germany, it would be more advantageous
to sell electricity than to store it. For the opposite case of
times with low wind and PV generation in Germany, it may
be cheaper to buy electricity abroad than to provide power
plants in Germany exclusively for periods of peak demand.
Transmission grids reduce overall system

costs with relatively small investment costs
Transmission grids reduce overall system costs since they
reduce the need for flexibility and at the same time permit
the use of the most cost-effective flexibility options.
Significance of Grids and Pumped Hydro Storage
in 2022: Example of a Week in February
GW
~16 GW grids
~9
GW pumped
hydro storage
80
60
40
20
Mo
Tu
We
Th
Fr
Sa
Su
In a week in early February there is so much wind that a

significant surplus is produced for almost two days
During this period, about 16 GW can be exported using grids
An additional 9 GW can be taken up over about five hours in
existing pumped storage facilities
New storage technologies would not be used due to their
higher costs (assuming sufficient grid expansion)
Illustration based on Agora Energiewende (2012a) and TAB (2012)
Furthermore, the costs for increased transmission capacity

are relative low. Overall, the expansion of transmission grids
across Europe will only constitute about 6% of overall costs
for the power system in the long term.27
Expanding and upgrading distribution grids is

less expensive than local storage facilities as
well
The same logic that applies to transmission grids is also true
for distribution grids. Local surpluses can be transferred
to neighbouring regions or to the transmission grid by
expanding distribution grids and upgrading transformers
for transferring electricity to high-voltage grids. From
todays perspective, new construction and the upgrading of
distribution grids is many times more cost effective than
local storage using new storage technologies.28
New storage technologies will only become

necessary as the share of renewable energy
exceeds 70 percent
For quite some time to come, grid expansion will continue to
be the cheaper option for integrating renewable energy in the
power system in comparison to new storage technologies.
From todays perspective, new storage technologies such
as batteries, adiabatic compressed air storage, and powerto-gas systems should only be deployed in the long term.29
Currently, the costs for these technologies are prohibitively
large; and will likely remain high in the medium term as well.
To keep overall system costs low, new storage technologies
should only be used after the flexibility potentials of other,
more cost-effective options have been fully exhausted. Since
they are relatively expensive flexibility options, new storage
technologies will only contribute to limiting overall system
27 McKinsey (2010), p. 45.
28 Consentec/r2b (2010b), p. 36.
29 This assessment is essentially in agreement with the expert opinion
presented in BMU (2012b), but it needs to be qualified by two factors.
For one, in the face of long term delayed or limited grid expansion,
storage facilities could represent a second-best efficient option, and
for another, breakthroughs in the manufacturing costs for new storage
technologies could make them more cost-effective in the future.
costs in Germany at the point when renewable energy

accounts for about 70% of power generation.30
Local PV- battery systems may provide a

business case for individual investors sooner
because of savings in taxes and fees
Battery storage systems combined with PV power units
can enable households and firms to use a greater proportion
of the decentralized power they generate and to purchase
less electricity from the system. The savings in taxes and
fees (grid charges, taxes, EEG levies, etc.) incurred can make
investments in facilities of this kind attractive from an
individual economic perspective in the medium term. This
does not by any means imply that decentralized storage
facilities will reduce the overall cost of electricity generation
in Germany. The main reason for this is that most taxes and
fees are calculated per kWh, but overall costs for the system
are not reduced proportionately with the reduction of kWh
purchased from the grid. For example, increasing private
consumption does not reduce the costs for transmission
and distribution grids or for assuring system reliability, but
merely leads to a different distribution of these costs across
the kWh still purchased from the grid. Presumably no private
homeowner or firm will chose to be disconnected from the
grid. For the design of the grid, however, the most salient
factor is not the annual quantity of electricity transported,
but instead, the maximum required capacity. Should a
breakthrough occur in the costs of battery storage technology (for example, related to the accelerated deployment of
electric vehicles and economies of scale for lithium-ion batteries) an accelerated expansion of decentralized storage can
be expected if and as long as the infrastructure costs for
the electrical system are apportioned according to kilowatt
hours and not according to connected load.
30 70% is a rough estimate in the case of the German power system.

The specific share of renewables, at which new storage technologies will be able to contribute to reducing total system cost depends
mainly on the type of renewables (fluctuating vs. controllable), the
residual power plant fleet (amount of CHP, flexible generation), the
level of interconnection and the future cost of storage technologies.
It is likely to be a similar percentage in most other countries which
are able to connect to a transmission grid.
15
Insight 5
Securing supply in times of peak load does not cost much
At certain times (e.g. during windless days in
the winter), wind and PV cannot contribute to
coverage of peak loads, and, for this reason,
controllable resources will be required in the
same order of magnitude as today
Power supply must be secured during peak-hours even
when there is no wind blowing and the sun does not shine
even if a major power plant goes out of service unexpectedly. Wind and PV, as naturally fluctuating power sources,
can only make a small contribution in such moments to
assuring system reliability. Figure 6 illustrates such a
potential supply and demand situation.
Securing Supply in Times of Peak Load in 2022:

Example of a Week in November
0 GW from PV
4 GW from wind
80 GW
demand
GW
80
Accordingly, to assure the same level of system reliability in

the future, peak demand levels experienced today will need
to be covered by resources other than wind and PV. This
type of controllable resources can be provided both on the
supply-side by power plants, as well as on the demand-side
through controllable loads.
Peak-load can be met reliably by firm generation capacity, or be reduced through demandside measures; almost a quarter of the demand
(approx. 15 to 25 GW in Germany) occurs during
only very few hours in the year (<200).
A new feature of the Energiewende world is that major
quantities of controllable resources will be required that
will only be used during a few hours of the year or only as
a reserve. As can be seen in Figure 7, it is predicted that
by 2020 Germany will need about 20 GW of controllable
capacity, which will almost never be used or used only for a
few hours each year. The graph presents calculations by the
VDE [German Electrotechnology Association] based on the
assumption of a 40% share of renewable energy in 2020.
60
40
20
Mo
Tu
We
Th
Fr
Sa
Su
At the moment of the maximum load of 80 GW (for example,

on a Thursday in November at 7:00 PM) there is no sunshine,
and because of a lull in the wind, only approx. 4 GW generated
by wind turbines
F
or the possibility that at this moment, zero wind power is
generated, sufficient controllable ressources must be retained
A
bout a fourth of total controllable capacity will be retained
solely for this eventuality occurring only a few hours per year
16
On the lower axis, load ranges are presented in GW, and the
continuous light blue (2010) and violet colored (2020) lines
indicate the number of hours per year that the respective
power plant capacity will be required in Germany. According to these calculations, the load ranges between 65 and
85 GW will be used for less than 100 hours per year. These
20 GW make up one fourth of the total required capacity
of approximately 80 GW. Many other studies have come
to a similar conclusion and determined, depending on the
underlying assumptions, that between 14 and 27 GW of gas
turbines or other options to provide controllable resources
will be needed, and that these resource would be used for
only a few hours during the year.31
31 See Consentec/r2b (2010a), p. 78f; IWES (2010), p. 94; TAB
(2012), S. 103; ECF (2010), Appendix Generation, p. 16.
Demand for Controllable Capacity to Cover Maximum Peak Loads

9000
8000
In 2020 approx. 20 GW of
controllable capacity will be
needed, which will be employed
for less than 200 hours
Full load hours
7000
6000
5000
4000
3000
Full load hours 2010
2000
Full load hours 2020
1000
80 - 85
75 - 80
70 - 75
65 - 70
60 - 65
55 - 60
50 - 55
45 - 50
40 - 45
35 - 40
30 - 35
25 - 30
20 - 25
15 - 20
10 - 15
7,5 - 10
5 - 7,5
2-5
0-2
0
Residual load in GW
Illustration based upon VDE (2012a)
Gas turbines can meet this demand quite

cheaply (3570 million EUR per year per GW),
controllable loads or retired power plants
might be even cheaper
The need to cover peak-load in times when high load
intersects with minimal feed-in from wind and PV does not
need to be met by using expensive, regular power plants.
Instead, it is possible to use more cost-efficient open cycle
gas turbines (OCGTs) for this purpose. OCGTs have been
used for generating electricity during peak load periods for
many years and are capable of reaching their full capacity
in less than ten minutes. Because of their relatively low
efficiency an efficiency factor of about 30% compared to
about 60% for combined-cycle gas turbines (CCGTs) and
their associated high fuel costs, OCGTs are not well suited
for continuous use in electricity production. However, for
the new Energiewende world, as a way to cover peak load
during only very few hours per year, OCGTs represent a
cost-effective option. Their expected costs per GW per year
are between 35 and 70 million EUR.32
In principle, other options can perform this function just as
well as long as they provide the same level of reliability.
On the supply-side, such options include the use of older,
retired power plants or units as well as smaller diesel and
gas generators, which are often employed for emergency

power generation. Experiences in other countries show that
the demand side can provide a promising and inexpensive
contribution as well for example, through controllable
loads in industry. In the US, this form of controllable capacity has been shown to be a very cost-efficient option (see
Insight 10).
European cooperation reduces the cost and

simplifies securing supply in times of peak
loads
Through the pooling of controllable capacity with other
European countries, the cost for securing supply in times
of peak-load is reduced even further. For one thing, the
combined peak-load across several countries that must be
covered without wind or PV is smaller than the sum of the
individual peak loads, since these peaks never take place at
exactly the same time. The load curves differ significantly
and fluctuating production does not reach minimum levels
in all countries simultaneously. In addition, neighbouring
countries can jointly use the most cost-effective options
at any particular moment regardless whether these are
controllable loads, retired power plants or new OCGTs. In
this way, costs can be reduced for all participating countries.
32 TAB (2012), p. 114; BMU (2012b), p. 21.
17
Insight 6
Integration of the heat sector makes sense
The heat sector offers enormous potential
for increasing system flexibility
Share of Heat and Electricity in Total

Energy Consumption in Germany for 2020 (%)
As the power system is adapted to the fluctuating output

from wind and PV power, it is important to keep in mind its
interaction with other energy sectors. In the future, the heat
sector will play a pivotal role in the transition of the power
system. There are three reasons for this:
Heat is easy to store contrary to electricity

Unlike electricity, heat is easy to store. Heat for homes
(warm water and interior heating) can be stored easily in
insulated water tanks at private residences, in city-wide
district heating systems, or in decentralized local heating
networks. Thermal storage systems like these can supply
heat for several hours or days at very low cost. Energy losses
are far lower than those associated with power storage. The
properties of heat also extend to cold, which can be stored
for short periods at relatively low cost with relatively low
levels of energy loss, for example in commercial refrigeration.
33 See ECN (2011).
18
Electricity
47%
Heat
x2
Energy consumption for heating is twice as large as

for electricity; gas and oil must (almost) entirely be
replaced to achieve emission targets
Total energy consumption in the heat sector in Germany
is twice as large as that in the electricity sector, as the
2020 forecast in Figure 8 exemplifies.33 To meet Germanys
official targets for greenhouse gas reduction 80% to 95%
by 2050 energy consumption for heating must decrease
and renewables must replace oil and gas in heat generation almost entirely. Due to the limitations of biomass (see
Insight1), biogenic fuels are likely to assume only a small
share of that burden. In the long run, both the heat sector
and the transportation sector must increase their utilization
of electricity from wind and PV.
25%
Illustration based on ECN (2011)
Most heat is needed in winter, when winds are

strongest
Most of Germanys heating needs arise from October to
April. These are also the months in which winds and wind
power are strongest (see Figure 9). This is an advantageous
correlation, as electricity from wind is likely to be the least
expensive source of energy for Northern Europe.
CHP plants already provide a link between the

electricity and heat sectors; in the medium
term, dual-mode heating systems, capable
of using either fuel or electricity will be
deployed; over the longer term, integration
will occur by using a common fuel natural
gas, biogas, or power-to-gas
Combined heat and power plants (CHP) produce both
heat and power and already provide a link between the
electricity and heat sectors. While operation today mostly
follows the demand for heat, these facilities can be easily
upgraded to be responsive both to demand for electricity
as well as demand for heat (Insight 3). This upgrade is done
by the addition of a thermal storage system - if there is
little demand for heat, but only for power (e.g. in times of
Heating requirement and Wind Power Generation

in Germany per Month
Power Generation
(TWh)
Difference in Temperature
(Celsius)
20
5.0
4.0
15
3.0
10
2.0
1.0
0.0
0
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Wind power generation
(TWh per month based on
20032008 averages)
Heating requirement (difference of average monthly

temperature to 20 Celsius)
Illustration based on ISI/ISE (2009) and DWD (2012)
little wind) the plant produces power and stores the heat.
This approach is already being used by municipal utility
companies (Stadtwerke) and other electricity producers.
Facilities in Flensburg, Lemgo, and Hamburg are already
considering installation of, or have installed such power-toheat systems.

In the medium term, dual-mode heat systems will be used.
Such systems link the power and heat sectors and provide
a high degree of flexibility by producing heat either with
fossil fuels or with electricity. At times of high power
production from wind and sun these systems use electricity
to generate heat, taking advantage of the low power prices.
When wind and sunlight levels are low, and power prices are
high, they use natural gas or oil instead.
Longer-term, both sectors will increasingly be linked by a
common and interchangeable gaseous fuel natural gas,
biogas, or power-to-gas. This fuel can be used for centralized or decentralized electricity generation, for electricity
and heat generation in CHP plants, or in systems used only
for generating heat. The advantage of such a gaseous fuel
is that it can be stored over long periods, and an extensive
storage infrastructure caverns, pipeline networks is
already in place.
19
Insight 7
Todays electricity market is about trading kilowatt
Todays electricity market handles energy
quantities (Energy-Only)
Todays electricity market is an energy-only market: suppliers and consumers trade in kilowatt hours, i.e. a specific
amount of energy at a specific point in time. The contractual
relations between participants do not address system reliability. This is the responsibility of the grid operator, who
must maintain the requisite balance between demand and
supply in every moment, in order to maintain a stable and
reliable system.
The price of electricity is determined in each

hour by the operating costs of the most
expensive plant running on the system
(marginal costs); this mechanism ensures that
power plants with the lowest operating costs
are run first, followed by those with higher
operating costs.
Prices on todays electricity market are determined by what
is known as the merit order. Different energy suppliers
offer electricity from available power plants to the market
at a certain price. Electricity generated from wind and solar
energy is cheapest. Next, by order of increasing price, comes
electricity from water, electricity from nuclear power, electricity from lignite, and (depending on CO2 and fuel prices)
electricity from hard coal and gas-fired plants (see Figure
10). The cheapest electricity is used first. As consumption
increases, more expensive power plants kick in to cover
demand. These power stations supply electricity only until
demand is met. The most expensive power plant needed to
cover demand the marginal power plant determines the
electricity price on the spot market.34
34 Operators of marginal power plants try not to base their
prices on their own variable operating costs. Instead, to
20
The more electricity generated from wind and solar energy,

the lower the operational costs of the plant that serves as the
marginal power station, thus prices on the exchange vary in
accordance to the levels of wind and sun.
The energy-only market may not provide

sufficient incentives for new and existing
resources to continuously ensure system
reliability.
There is heated debate among economists whether todays
energy-only markets can guarantee system reliability.
Arguments against are, inter alia, the lack of elasticity in
electricity demand, the missing money problem for power
plants with few operating hours, and regulatory uncertainty.
On the other side it is argued that every demand produces a
supply.35 All in all, it is fair to conclude that economic theory
cannot answer the question once and for all.
Regulatory agencies in a number of countries with competitive wholesale electricity markets (e.g. United States, Brazil,
Spain, United Kingdom, South Korea) have therefore decided
to introduce additional measures to ensure the availability
of sufficient firm generation capacity. The reason behind
this is that system reliability is considered a public good,
where there is a high risk that the energy-only markets
alone do not deliver at an adequate level. 36
generate additional revenue, they set them just under
the variable operating costs of the next power station in
the merit order. In markets where the power stations all
have similar cost structures, the additional revenues are
low, so that stations electricity prices are almost identical to the marginal costs of the last power station in use.
35 For an overview of the pro and contra arguments, see
Cramton/Ockenfels (2012) and Msgens/Peek (2011).
36 The capacity markets have a variety of configurations with different degrees of efficiency and effective-
Electricity Prices and the Merit-Order Curve
10
Low Level of Wind and Solar power
High Level of Wind and Solar power

oil
Marginal costs
EUR/MWh
natural gas
hard coal
wind, solar,
hydro
oil
Marginal costs
EUR/MWh
natural gas
hard coal
wind, solar,
hydro
lignite
High elec
tricity price
when levels
of wind and
solar power
are low
lignite
nuclear
Low elec
tricity price
when levels
of wind and
sun are high
nuclear
Capacity
GW
Demand for electricity
Capacity
GW
Demand for electricity
Own illustration
In Germany, the issue of back-up capacity needed to ensure

system reliability has special relevance: the countrys
decision to phase out nuclear power will eliminate 4 GW in
energy capacity between 2015 and 2019 and another 8 GW
between 2020 and 2022.
guarantees system reliability at any one moment in time.

Because new OCGTs need two to three years for approval
and construction, new CCGTs three to five years, and
coal-fired plants longer still, politicians will have to find a
regulatory answer to the question of system reliability in
the coming legislative period 2013-2017.37
The Energiewende brings this issue to the

forefront, because power production from
wind and PV will reduce the average market
price of electricity and with it the operating
times of fossil-fueled power stations
The issue of system reliability in energy-only markets is
aggravated by the fact that the Energiewende will continue
to shrink the market for electricity from fossil-fueled
power stations. As the Energiewende increases the share
of renewables, the operating hours of fossil power plants
will decrease, especially those of gas-fired and hard coal
power stations that usually find themselves behind the
lignite power stations in the merit order. And as the share
of renewables with marginal costs close to zero grows, the
market price for electricity will fall.
It is doubtful, therefore, whether investors will build new
power stations or operate existing ones to an extent that
ness. For an overview, see Senbacher et al. (2011).
37 For an overview of models currently in discussion on strategic reserves, comprehensive capacity markets, and focused
capacity markets, see Agora Energiewende (2012b).
21
Insight 8
Wind and PV cannot be principally refinanced via
Wind and solar power have operating costs
close to zero
Wind and PV are characterized by high investment costs,
zero fuel costs, and low operating costs (see Insight 1). Once
operational, wind or PV facilities exhibit marginal costs
close to zero, that is, the production of an additional kilowatt
hour creates essentially no incremental costs - in contrast
to coal- or gas-fired plants.
Wind and PV produce electricity when the

wind blows and the sun shines, regardless of
electricity price
Wind and PV power production is variable depending upon
the weather and the time of day. Unlike coal- and gas-fired
plants, the electricity output from wind and PV cannot
be controlled by plant operators (curtailment is the only
exception). Since their marginal operating costs are close to
zero, wind and PV will always produce electricity when the
wind blows and the sun shines, independent of whether the
exchange price for electricity is high or low.38
In times when wind and/or sun is plentiful,

wind and PV facilities produce so much
electricity that prices decrease on the spot
market, thus destroying their own market
price
When the wind blows or the sun shines, all wind (or PV)
power stations in the same weather zone produce electricity
simultaneously. And when there is a significant number of
wind and PV power stations in the system, this will have a
38 One exception is when electricity prices are negative.

Operators of wind and PV can turn off their facilities to
avoid having to pay out money for their production.
22
price effect in the market: With so much electricity being

offered into the market at close to zero price, the market
exchange price will decrease. Power plants with more
expensive marginal costs will not come on-line and power
plants with lower marginal costs determine the exchange
price (Merit-Order Effect). In Germany, the installed generation capacity of PV and wind is on the order of 30GW each.
In the sunny month of May 2012, the exchange price during
the day was often only 30/MWh; the former midday peak
in electricity prices due to peak demand does not occur
anymore at sunny days. Further evidence for this argument is provided by the forecasts of the grid operators: To
estimate the revenues from the sale of EEG electricity, grid
operators use so-called market value factors for electricity
generated from specific renewable energy sources. These
factors describe whether the wholesale spot market price at
feed-in is above or below the yearly average. The forecasted
market value factors for electricity generated by wind and
PV are both below 1, i.e. prices are always below average in
sunny and/or windy times. Furthermore, these factors are
continuously going down.
As a consequence, wind and PV destroy their own prices at
the marginal-cost based wholesale spot market. This effect
worsens as more wind and PV power plants are built, which
then produce electricity concurrently (see Figure 11).
Impact of High Levels of Wind and Solar on Electricity

Pricesin 2022: Example of Week in August
11
High CO2 prices do not fundamentally change

this effect
Marginal costs: ~0 EUR/MWh

GW
70
50
30
10
Mo
Tu
We
Th
Fr
electricity market models have confirmed this effect using

Germanys current power mix.40
Sa
Su
Between Monday night and Tuesday evening electricity

from wind and solar energy (plus small quantities from other
renewables) cover total demand for electricity in Germany
The market price determined on the basis of marginal
costs would tend to zero EUR/MWh over a period of
18-hours
Therefore by principle, wind and PV cannot be

refinanced in a marginal-cost based market,
even when their total costs are below those
of coal and gas.
Even if the EU Emissions Trading Scheme applies stricter

regulations leading to increased CO2 prices, this problem will
not be solved, as high CO2 prices do not change the fact that
wind and PV have marginal costs close to zero. High CO2
prices cause the costs of electricity generation in coal and
gas-fired plants to rise. In times of low wind and sunlight,
wind and PV, running at partial capacity, profit from the
higher marginal costs of coal and gas-fired plants and
receive higher revenues. But these higher revenues cannot
compensate for the many hours in which conditions are
favorable and wind and PV produce high levels of electricity,
lowering market exchange prices.
Hence, the aim to make renewables market competitive, so
that their further development is driven by the electricity
market, faces a fundamental problem: Wind and PV cannot
be integrated into the existing marginal-cost based market,
since these weather-dependent energy sources cannot be
re-financed. Thus, without a regulatory instrument governing electricity from renewables, Germanys Energiewende
will fail.
The fundamental problem is this: Wind and PV cannot earn

enough revenues to cover the average cost of their initial
investment in the market, because the price will always be
lower than the market price average whenever the wind
is blowing or the sun shining, which is precisely when
electricity can be produced from these weather-dependent
technologies. Hence, if the share of wind and PV is high
when there are strong winds and/or intensive sunlight,
electricity market prices are expected to decrease to
minimum levels from 1 to 2 ct/kWh.39 Simulations with
39 If the electricity price on the market is close to zero or
below, new consumers are likely to appear in the electricity market. For instance, by making a small investment, a
heating consumer can convert electricity into heat using
the immersion heater principle, making the lowest price
level on the electricity market 1-2 Cent/kWh depending
on gas prices as opportunity costs in the heating market.

40 See Kopp et al. (2012).
23
Insight 9
A new Energiewende market is required
The future Energiewende market
must fulfill two functions:
Steer the installation of capacity, in order to achieve
an efficient balance between demand and supply
Send investment signals for renewable energy, as
well as for conventional facilities and making energy
demand and storage (longer term) more flexible.
From Insights 7 and 8 above, we conclude that in order to
succeed the Energiewende requires a new market design
(see Figure 12). This market will need to fulfill two basic
functions. First, within the realm of energy-only markets,
supply and demand will need to be balanced as efficiently
as possible based on marginal cost principles just as in
the past. Renewable energy, conventional power plants,
and storage will be dispatched in order of increasing
marginal costs to meet total demand. Second, the new
market must attract the required investment in renewables
on the one hand, as well as in conventional power plants,
demand-side flexibility and storage technologies. This
involves not only ensuring that existing power plants are
optimally deployed, but also that the necessary expansion in generation from renewables is assured, while
continuing to maintain system reliability. Fossil-fueled
power plants, demand-side, and storage systems must
compete with one another, in order to deliver the flexibility
required to balance the fluctuating output from wind and
PV at least cost.
The new market will create two

sources of revenue:
Revenue (as before) from the sale of electricity
quantity (MWh) in the marginal-cost based EnergyOnly Market
Revenue from a new Investment Market (MW)
The new Energiewende market will preserve the energyonly market while adding a new market for capacity investment. Operators of renewable energy facilities, fossil-fueled
24
power plants, and storage systems receive revenues from

both markets in order to recover the investment and operating costs of their facilities.
In addition, fossil-fueled power plants,

renewable energy, demand-side resources,
and storage systems will compete to provide
ancillary services (e.g. balancing energy)
The market for balancing energy as part of the ancillary
services market will continue to be necessary for maintaining system stability. But this market also requires modification that allows a greater variety of options for renewables
and demand-side resources to participate in the tender
process. In addition, these costs should no longer be offset
partly by grid tariffs. Instead of socializing up to two-thirds
of the system balancing costs as is the case currently the
entities responsible for balancing should bear them in full.
This allows clear assignment of responsibility for next-day
electricity production forecasts and demand estimates.
The new investment market will reward

reliable, flexible resources (both supply and
demand-side) to guarantee system reliability
CO2-free electricity, to ensure the transition to
renewables
A new market for investments must take into account supply as well as demand if system reliability is to be secured.
The products on this market must be defined in a way that
enables controllable power stations, flexible demand, and
energy storage systems to participate. Due to the increasing
share of wind and PV, power stations and load displacement
must become very flexible, with quick ramp-up and rampdown times.
For the renewables sector, a market must be designed that
attracts new investments so that the Energiewende-targets
can be reached and total system costs kept to a minimum.
A possible Market Design for the Energiewende
Synchronisation of
supply and demand
12
Energy-only market
Fossil fuel power plants, renewable energy systems, demand side
Investment market
Ensuring system reliability and climate
protection
Safeguarding system
stability
Firm capacity
Fossil fuel power plants, demandside, storage systems
carbon-neutral electricity
renewable energy systems
Market for ancillary services (e.g. reserve and balancing energy)

Fossil fuel power plants, renewable energy systems , demand side, storage systems
Own Illustration
The exact configuration of the new market

requires further study; many options are
possible (premiums/bonuses, tenders,
c ertificates)
There are many options for incentivizing investment:
bonuses on top of the prices on the energy-only market
(bonus or premium model)
a central agency (e.g. the transmission system operator
or Germanys Federal Network Agency) that organizes
tenders and contracts the most cost-efficient provider
(tender or auction model)
obligations for electricity distributors to hold a certain
amount of renewable electricity respectively supply
guarantee certificates (certificate or quota model)
The advantages and disadvantages of these models for the
expansion of renewables have been discussed many times.41
For ensuring system reliability, the main candidates under
consideration have been the second and third options, in
addition to the creation of a strategic reserve.42
41 For a current overall, see Verbruggen/Lauber
(2012) and Fraunhofer ISI et al. (2012).
42 For a current overall, see Agora Energiewende (2012b) and Dena
(2012). On the recommendations now being discussed in Germany,
It is important to analyze these options anew in the specific

context of the Energiewende. Crucially, the costs for the
technologies to be rewarded be they wind or PV, capacities for system reliability, or in the longer term energy
storage systems arise almost entirely with the initial
investment, while operating costs are negligible or in the
case of fossil fuel capacities for system reliability such as
gas turbines yield insufficient return on investment due to
the limited extent of their operating hours.
Installing a new mechanism instead of the

current feed-in tariffs for renewables is only
justified if it brings increased efficiency
The creation of a market for renewables has no inherent
value in itself. Rather, the new market design must allow us
to expect greater efficiency for consumers and society than
that offered by the existing feed-in tariff system.
The problem with feed-in tariffs for renewables is that
legislators can err when setting the level of remuneration, or
they can be prone to influence from interest groups. When
see Consentec (2012), EWI (2012), and ko-Institut et al. (2012).
25
transitioning to a new mechanism three considerations

should be kept in mind:
Once a new wind or PV facility is built, there are few
options for modifying it. The crucial step, therefore, is the
initial investment. Given the various, location-dependent
feed-in profiles of wind and PV, it is important that a new
market for investments provides the correct incentives
for building facilities in a way that is efficient for the total
system.
Wind and PV facilities are costly investments, and hence
seem risky to investors if future revenues are uncertain.
Currently, the system of feed-in tariffs is neutralizing this
price risk through fixed 20-year guarantees. A market
model that gives too little attention to the risks of highcapital investments may drive up costs for consumers.
In the past, the expansion of renewables has been
financed through citizens as well as small and medium
enterprises (SMEs). This has created widespread support
in the population for the Energiewende. A market design
that no longer permits the participation of citizens and
SMEs would endanger the future of the Energiewende.
For these reasons, it is vital that the alternatives to the
existing feed-in tariff system be studied carefully. How,
exactly, will they affect a power system dominated by wind
and PV? A hasty transition to a new mechanism may give
rise to a system that is both less efficient than the current
one and unnecessarily driving up the cost of the Energie
wende.
26
Insight 10
The Energiewende market must actively
engage the demand-side
Greater demand-side flexibility is fundamental to increasing the use of wind and PV
Shifting demand from periods of low levels of sun and wind
to those with high levels increases the ability of the power
system to integrate a growing level of wind and PV, without
relying on expensive storage (see Figure 13). Historically,
electricity demand has been very inflexible to changes in
electricity prices over the course of a day. Currently, about
2 to 3% of demand is flexible (around 1.5 GW in total load
of 50to 80 GW). The technological potential is markedly
higher, however. In the long term, over 50% of todays
demand could be able to respond to the available supply from
Contribution of Demand-Side Measures to Flexibility

in 2022: Example of a Week in February
13
Shift of Demand
GW
70
50
30
10
Mo
Tu
We
Th
Fr
Sa
Su
From Friday night to Saturday morning, excess electricity is

available;, during Friday noon and afternoon, around 20 GW
of fossil power are needed
wind and PV.43 There are two reasons for the lack of demand
responsiveness in Germany: Firstly, the price fluctuations
during the day are not very large, so that load shifts generate
little additional value. Secondly, Germanys system contains
counterproductive incentives especially with respect to
grid tariffs that further inhibit demand-side responsiveness to price changes.
Demand response is usually cheaper than

electricity storage or supply-side options
In most cases, demand-response is the cheapest flexibility
option. For instance, with little investment refrigeration
facilities can be cooled to below minus 18 Celsius when
winds are strong and then do without some electricity when
winds are weak. In the industrial sector, electricity demand
can be temporarily shifted without additional investment,
and in many cases affordable thermal storage systems or
storage for intermediate products such as chemicals can
be introduced. Experience in the United States shows that
the costs of increased demand flexibility are often substantially below those of supply-side solutions. Following
the introduction of a capacity market in the eastern United
States, demand-side offers have won an impressive share
of auctions. 11 GW of demand-side resources (primarily
demand response) was added in delivery year 201213, with
consumer savings of more than one billion U.S. dollars.44
Current regulations for grid tariffs and ancillary services often work at cross purposes
with demand response, and should therefore
be reformed
Industries and other electricity users can shift their demand

from Friday afternoon to Friday night or Saturday.
43 See Bundesumweltministerium (2012b), p. 24-28.

44 See Gottstein (2012).
27
Regulations governing grid tariffs are often at odds with the

flexibility requirements of the energy transition. With many
industrial customers, grid tariffs are calculated at least in
part using the energy prices that result from individual
maximum load. As a consequence, industrial customers
are at a disadvantage when their yearly maximum loads
increase due to demand response measures even if, seen
from the whole, this would be beneficial in times when e.g.
wind power is plentiful.
that the demand-side can offer to the entire system, and not
just focus on the rare instances when load reductions are
required for grid stability.
Current grid tariffs exemptions for facilities with utilization

hours of 7,000 or more are also counterproductive: For as
long as these exemptions offer greater economic incentive
than demand response, the energy-intensive industries
have no reason to modify their demand, for doing so could
lower their utilization hours below the 7,000-hour mark
and disqualify them from exemption. Moreover, companies
that operate close to the threshold have an incentive to use
more electricity to ensure they reap the exemption benefits.
The fee exemption program, therefore, provides perverse
incentives in terms of making the system as a whole more
efficient and flexible.
The new market for investments in firm

capacity must be designed such that demandside resources able to shift loads can actively
participate
The regulations governing interruptible demand currently

under consideration should therefore address the flexibility
28
The demand side can also take part in ancillary services

markets, such as the market for balancing energy, by
bidding negative load. Current prequalification requirements, however, are in some respects so onerous, that many
potential industrial customers are not able to participate.
Products for the capacity market (securing load change for

a specific duration at a certain speed) should be specified
so that they can be delivered, not only by generators, but
also by demand-side changes in load. For example, most
industrial companies as a rule cannot sign multi-year
load management contracts, since their load management
potential depends on their core business orders. Accordingly, at least some of the products for tender must be short
term (e.g., one year) and take into consideration the unique
demand-side conditions.
Insight 11
The Energiewende market must be considered in the
European context
The ongoing integration of the German power
system into the European system makes the
Energiewende simpler and more affordable
because:
Germanys electricity market is already strongly interconnected with neighbouring markets. With some 17 GW of
interconnector capacity, the coupled markets with western
neighbours as well as the regulatory framework for the EU
internal market, it is absurd to take on a purely national
perspective. And all indications point to further market
integration in the coming years.
the fluctuations of wind and PV energy production

becomes less pronounced over a larger geographic
region
firm capacity can be collectively shared
low-cost flexibility options in Europe can be more
fully utilized (e.g. energy storage resources in
Scandinavia and Alpine countries)
This is advantageous for Germanys Energiewende.

Improved interconnection of power systems across Europe,
or at least across sub-regions of Europe, are projected to
yield considerable savings in comparison with an isolated
Fluctuation of Electricity Demand across Europe
30
14
Regional demand variation from average per hour during one day (percent)
20
10
1
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
0
-10
-20
-30
%
40
Regional demand variation from average over the year (percent)
20
-20
-40
Individual regions
Total EU 27
Adopted from ECF (2010)
29
national solution.45 The reason is that differences on the

supply- as well as the demand-side between regions
become advantageous with broader geographic diversification of the system. On the demand side, the marked (daily
and seasonal) load curve differences between individual
member states and regions can result in a lower total
average peak load (see Figure 14). On the supply side,
regional differences in levels of wind and sun can balance
out fluctuating energy production. Furthermore, geological
differences can be partly equalized, for example, by expanding the limited potential for low-cost pumped storage in
Germany with reservoir power stations in Scandinavia or in
Alpine countries, provided the necessary grids are in place.
All in all, less peaking power plants and balancing energy
would be needed.
European electricity trading

stabilizes market prices
Although other European member states are pursuing
somewhat ambitious expansions of renewables, Germany
clearly stands out when it comes to the absolute level of
production from variable renewables. By 2020, Germany
will produce about twice the amount of electricity from
wind and sun as the rest of the EU.46 Since wind and PV
have near-zero marginal costs, the market price for electricity will decrease an effect that we see today mostly
at noontime (see Insight 8). When this happens, demand
from surrounding countries can act to stabilize prices.
Conversely, in times when wind or PV production levels in
Germany are low, energy from neighbouring states can flow
into Germany and dampen high market prices in Germany.
Naturally the same effects can be observed in reverse in
these neighbouring countries.
45 See ECF (2010), p. 59ff. Calculations from KEMA

und Oxford Economics predict savings of up to 40%
should the electric grid be made Europe-wide.
46 According to the recent numbers of National Renewable
Energy Action Plan, Germany will produce around 24% of
its energy from variable sources (with 35% electricity from
renewables in total) by 2020. The rest of Europe is projected to
produce 13%. See European Environmental Agency (2011).
30
In the long run, there will be

total market integration in Europe
Electricity market integration in Europe began two decades
ago. Many steps have already been taken, including the
unbundling of production and grid operations. In coming
years, energy-only markets and ancillary services will
merge more and more. According to the current schedule,
the EU Third Energy Package, which contains regulations
on market coupling and network codes, will be fully implemented by 2015.
In contrast, current discussions related to system reliability
such as capacity markets or strategic reserves of various
designs are exclusively being debated on the national level.
As discussed above, however, separate systems for ensuring
system reliability in an interconnected European market are
not efficient. The debate should therefore be raised to European level discussions very soon. How national capacity
markets can and should be harmonized is still an open issue.
Regional markets would be a conceivable intermediate step
before long-term expansion of interconnector capacity and
a uniform European solution can be achieved.
Insight 12
Efficiency: A saved kilowatt hour is the
most cost-effective kilowatt hour
Energy efficiency decreases total costs;
increased energy productivity enables the
decoupling of economic growth from energy
consumption
Realising the Energiewende requires not only meeting
targets for renewable energy production, but also targets
for improved energy efficiency, among others. Accordingly,
the German federal government aims at reducing electricity
consumption below 2008 levels by 10% by 2020 and by
25% by 2050. In particular, improved electricity efficiency
will play a critical role in meeting Germanys emission
reduction goals, for without greater efficiency measures, a
significantly greater expansion of renewable energy installations will be needed to achieve an equivalent reduction of
greenhouse gases.
Energy efficiency means that the same level of goods and
services can be produced with less energy input. This, in
turn, means that economic growth and energy consumption
can be decoupled, and greater prosperity no longer relies on
a higher consumption of resources.
also beneficial to the economy, as the less electricity consumed, the less gas and coal must be imported. Moreover,
fewer investments in renewable and fossil-fueled power
stations must be made. The net result from improved energy
efficiency is increased domestic product and employment.48
The challenge lies less in technology and

more in creating the right incentives
Market-ready technologies for efficiency exist on both the
supply side and the demand side. Nevertheless, relatively
few measures have been implemented primarily due to
insufficient incentives. In many companies, managements
attention is not focused on energy efficiency because it is
not the most pressing investment decision to make and,
despite the savings that efficiency offers, energy may
only comprise a small fraction of total costs. The situation is similar for households. For most people, electricity
consumption is a side issue: the share of electricity in total
consumer expenditures is, on average, only 2.3%. As a result,
quite profitable efficiency potential from, for example,
replacing heat pumps or refrigerators is not realized.
Every kilowatt saved means

less burning of natural gas and coal
fewer investments in new power plants,
fossil and renewable
Numerous studies have demonstrated that improved electricity efficiency is not only advantageous for the economy,
but also benefits individual businesses and private households. This is because many efficiency measures pay off:
With todays available technology, internal rates of return of
over 30% (which corresponds to a 3 year pay-back period)
can be achieved, especially in the industrial sector.47 This is
47 See, for instance, McKinsey (2007); Roland Berger
(2011); Prognos et al (2009); ZVEI (2008).
Implementation of the EU energy efficiency directive

gives the opportunity to create a new incentive system for
electricity efficiency. It is critical that policy makers take
advantage of this opportunity, most importantly, by clearly
organizing the demand for energy efficiency and activating
market forces for delivering efficiency services. Energy
efficiency must become a viable business model. To this
end, the markets for energy efficiency in other countries
can serve as role models for accessing Germanys large and
unrealized energy efficiency potential.
48 See, for instance, Prognos et al (2009).
31
Comprehensive consideration of electricity,

heat, and transport sectors is required: heat
pumps and electric vehicles may increase
electricity demand but they are not at odds
with efficiency
transport sectors, thereby reducing consumption of these

resources. Moreover, the building and transportation sectors
offer the largest economic potential for energy efficiency,
specifically, in building insulation improvements and better
propulsion technologies.
Especially where energy efficiency is concerned, we need

to look beyond the boundaries of the electricity sector. As
more electricity is used for heating and electric mobility,
demand for electricity may increase. But this does not pose
a contradiction with respect to energy efficiency. The use
of wind and PV power for heating or electric vehicles will
displace demand for oil and natural gas in the heating and
Over the medium term, increased use of electricity in the

heating and transport sectors, together with emerging
hydrogen and power-to-gas technologies, will lead to
stronger integration of the electricity, heating, and transport
sectors, and is thus intensifying the need for coordination
among these different sectors in the future.
Effects of efficiency in the year 2022 in an examplary

week in November
In one week, more than

1.000 GWh less power
production from coal and gas
Load reduction
of up to 8 GW
GW
80
60
40
20
Mo
Tu
We
Th
Fr
Sa
The aim of the German government is to reduce power

demand in 2020 by 10 per cent (against 2008); this equals
a 60 TWh reduction of gross demand
L ower electricity demand reduces power production
from coal and gas fueled plants
In addition, the need of firm capacity is reduced
in this November week, up to 8 GW less firm capacity
or load shifting is needed to ensure system reliability
32
Su
References
Agora Energiewende (2012a): Erneuerbare Energien und
Stromnachfrage im Jahr 2022. Illustration der anstehenden
Herausforderungen der Energiewende in Deutschland
Cramton, P., Ockenfels, A. (2011): Economics and design

of capacity markets for the power sector. Zeitschrift fr
Energiewirtschaft (36) 2, p. 113134
Agora Energiewende (2012b): Brauchen wir einen Kapa

zittsmarkt? Proceedings of the panel discussion held on
24August 2012 at Stiftung Mercators ProjektZentrum
Berlin
Dsseldorfer Institut fr Wettbewerbskonomie (DICE)

(2011): Vor- und Nachteile alternativer Kapazittsmechanismen in Deutschland, assessment undertaken on behalf of
RWE.
Bundesumweltministerium (2012a): Die wichtigsten

nderungen der EEG-Novelle zur Photovoltaik 2012, white
paper available at www.erneuerbare-energien.de
Deutsche Energie-Agentur (dena) (2012): Presentation at the

Conference on Capacity Mechanisms: Experiences in Various European Countries, available at http://www.dena.de/
veranstaltungen/archiv/ conference-on-capacity-mechanisms-experiences-in-various-european-countries.html
Bundesumweltministerium (2012b): Vergtungsstze,

Degression und Berechnungsbeispiele nach dem neuen
Erneuerbare-Energien-Gesetz (EEG) vom 04. August 2011
(EEG 2012), white paper available at www.erneuerbareenergien.de
Bundesumweltministerium (2012c): Bericht der Arbeitsgruppe Interaktion der Plattform Erneuerbare Energien; appendix: Flexibilittsoptionen: Potenziale und
Hemmnisse
Consentec/r2b (2010a): Voraussetzung einer optimalen
Integration erneuerbarer Energien in das Stromversorgungssystem, study undertaken on behalf of the German
Federal Ministry of Economics
Consentec/r2b (2010b): Frderung der Direktvermarktung
und der bedarfsgerechten Einspeisung von Strom aus Erneuerbaren Energien; study undertaken on behalf of the German
Federal Ministry of Economics
Consentec (2012): Praktikabel umsetzbare Ausgestaltung
einer Strategischen Reserve, assessment undertaken on
behalf of the BDEW
DIW/Fraunhofer ISI/CPI (2011): Untersuchung des Energie

sparpotentials fr das Nachfolgemodell ab dem Jahr 2013ff
zu den Steuerbegnstigungen fr Unternehmen des Produ
zierenden Gewerbes sowie der Land- und Forstwirtschaft bei
der Energie- und Stromsteuer, assessment undertaken on
behalf of the German Federal Ministry of Finance
DLR/IWES/IfnE (2012): Langfristszenarien und Strategien
fr den Ausbau der erneuerbaren Energien in Deutschland
bei Bercksichtigung der Entwicklung in Europa und global;
study undertaken on behalf of the German Federal Environmental Industry
DWD (2012): Data on monthly temperatures in Germany,
available at www.dwd.de (under Klimadaten online frei >
Klimadaten Deutschland >Messstationen >Monatswerte)
EnBW (2012): Entwicklung der Mrkte fr Flexibilitt:
Inwieweit wird eine flexible Fahrweise konventioneller
Kraftwerke angereizt?; presentation at the DENA Dialogforum Retrofit und Flexibilisierung konventioneller Kraftwerke
ECF (2010): Roadmap 2050 a practical guide to a prosperous, low carbon Europe
33
ECN (2011): Renewable Energy Projections as Published in

the National Renewable Energy Action Plans of the European
Member States: Summary Report; study undertaken on
behalf of the European Environment Agency
E.ON Bayern (2011): Netz der Zukunft. Research project
conducted by E.ON Bayern AG
EWI (2012): Untersuchungen zu einem zukunftsfhigen
Strommarktdesign. Report for the German Federal Environmental Industry
IRENA (2012a): Renewable Energy Technology Cost Analysis

Solar Photovoltaics
IRENA (2012b): Renewable Energy Technology Cost Analysis
Wind Power.
IPCC (2011): IPCC Special Report on Renewable Energy
Sources and Climate Change Mitigation.
ISE (2009): Integration von Windenergie in ein zuknftiges
Energiesystem untersttzt durch Lastmanagement; study for
the German Federal Environmental Industry
EWI/Energynautics (2011): Roadmap 2050 a closer look.

Forschungsverbund Erneuerbare Energien (2010): Eine
Vision fr ein nachhaltiges Energiekonzept auf Basis von
Energieeffizienz und 100% erneuerbaren Energien
Fraunhofer ISI et al. (2012): RE-Shaping Shaping an
effective and efficient European renewable energy market,
assessment conducted for the Executive Agency for
Competitiveness and Innovation Innovation (EACI) of the
European Commission
IWES (2009): Dynamische Simulation der Stromversorgung

in Deutschland nach dem Ausbauszenario der EE-Branche;
study for the German Association for Renewable Energy
IWES (2010): Energieziel 2050: 100% Strom aus Erneuerbaren Quellen, study for the German Federal Environmental
Industry
IWR (2012): Der IWR-Windertragsindex fr Regionen;
www.iwr.de/wind/wind/windindex
Gerlach/Breyer (2012): PV und Windkraft: Sich hervorragend ergnzende Energietechnologien am Beispiel Mitteldeutschlands, Paper fr das 27. Symposium Photovoltaische
Solarenergie Bad Staffelstein
Kopp/Frey/Engelhorn (2012): Knnen sich erneuerbare

Energien langfristig auf wettbewerblich organiserten
Strommrkten finanzieren? Zeitschrift fr Energie
wirtschaft Online
Gottstein (2012): Mssen wir beyond capacity markets

denken? Erfahrungen mit US-Kapazittsmkten anhand
einer Fallstudie. Available (in German only) at
http://www.raponline.org/document/download/id/4926
McKinsey (2007): Kosten und Potenziale der Vermeidung

von Treibhausgasemissionen in Deutschland
IAEW/Consentec (2011); Bewertung der Flexibilitten von

Stromerzeugungs- und KWK-Anlagen; study on behalf of
the German Association of Energy and Water Industries
McKinsey (2010): Transformation of Europes power system

until 2050 including specific considerations for Germany
Ingenieurbro Floecksmhle et al. (2010): Potentialermittlung fr den Ausbau der Wasserkraftnutzung in Deutschland
als Grundlage fr die Entwicklung einer geeigneten Ausbaustrategie, assessment for the German Federal Environmental Industry
34
McKinsey (2009): Wettbewerbsfaktor Energie
Msgens/Peek (2011): Sind Kapazittsmrkte in Deutschland

erforderlich? Eine kritische Analyse vor dem Hintergrund
der konomischen Theorie, Zeitschrift fr neues Energie
recht 15 (6), p. 576583
Prognos (2012): Bedeutung der internationalen WasserkraftSpeicherung fr die Energiewende, study undertaken on
behalf of the German chapter of the World Energy Council
Prognos/EWI/GWS (2011): Energieszenarien 2011, study
undertaken on behalf of the German Federal Ministry of
Economics
bertragungsnetzbetreiber (2012): Prognose der EEGUmlage 2013 nach AusglMechV. Forecasts and calculations
by NB
VDE (2012a): Erneuerbare Energie braucht flexible Kraftwerke Szenarien bis 2020; ETG Task Force for the Flexibilization of Power Plants
Prognos/IFEU/Fraunhofer ISI/GWS (2009): Klimaschutz,

Energieeffizienz und Beschftigung. Potenziale und volks
wirtschaftliche Effekte einer ambitionierten Energie
effizienzstrategie fr Deutschland, project supported by the
German Federal Environmental Industry
VDE (2012b) Demand Side Integration Lastverschiebungspotentiale in Deutschland; ETG Task Force for Demand Side
Management
ko-Institut/Prognos (2009): Modell Deutschland. Klima

schutz bis 2050: Vom Ziel her denken, Report undertaken on
behalf of WWF Germany
p. 635644
Verbruggen/Lauber (2012): Assessing the performance of

renewable electricity support instruments, Energy Policy 45,
ZVEI (2010): Weibuch Energie-Intelligenz Energie intelligent erzeugen, verteilen und nutzen, 2nd edition
ko-Institut/LBD/Raue LLP (2012): Fokussierte Kapazi

ttsmrkte. Ein neues Marktdesign fr den bergang zu
einem neuen Energiesystem, study undertaken on behalf of
WWF Germany
Roland Berger (2011): Effizienzsteigerung in stromintensiven
Industrien
SRU (2011): Wege zur 100 % erneuerbaren Stromversorgung
Senbacher/Schwaiger/Stigler (2011): Kapazittsmrkte
und -mechanismen im internationalen Kontext. Study for
the 7th International Energy Economy Conference at Vienna
Technical University
TAB (2012): Regenerative Energietrger zur Sicherung der
Grundlast in der Stromversorgung; Working report no. 147
of the Office for Technology Consequences Assessment,
conducted on behalf of the Committee for Education,
Research, and Technology Consequences Assessment of the
German Bundestag
bertragungsnetzbetreiber (2011): Prognose der EEGUmlage 2012 nach AusglMechV. Forecasts and calculations
by NB
35
Publications by Agora Energiewende

12 Thesen zur Energiewende
Ein Diskussionsbeitrag zu den wichtigsten Herausforderungen im Strommarkt
(Lang- und Kurzfassung)
Brauchen wir einen Kapazittsmarkt?

Dokumentation der Stellungnahmen der Referenten der Diskussionsveranstaltung am
24. August 2012 in Berlin
Die Zukunft des EEG Evolution oder Systemwechsel?

Dokumentation der Stellungnahmen der Referenten der Diskussionsveranstaltung am
13. Februar 2013 in Berlin
Erneuerbare Energien und Stromnachfrage im Jahr 2022

Illustration der anstehenden Herausforderungen der Energiewende in Deutschland.
Analyse auf Basis von Berechnungen von Fraunhofer IWES
Kapazittsmarkt oder strategische Reserve?

Ein berblick ber die in der Diskussion befindlichen Modelle zur Gewhrleistung der
Versorgungssicherheit in Deutschland
Steigende EEG-Umlage: Unerwnschte Verteilungseffekte knnen vermindert werden

Analyse des Deutschen Instituts fr Wirtschaftsforschung (DIW)
All Publications are available on our website: www.agora-energiewende.de
36
10
20
30
40
50
60
> weather dependent

> variable
> only capital expenditure
features
12. A saved kilowatt hour is the most cost-effective kilowatt hour
7. Todays electricity market is about trading kilowatt

8. Wind and PV cannot be principally refinanced via
9. A new Energiewende Marked is required
10. The Energiewende Market must actively engage
the demand-side
1 1. It must be considered in the European context
market design and regulation
How to balance demand and supply?

How to minimise costs?
How to realise the Energiewende in the European context?

Mo Tu We Th Fr Sa Su
Its all about Wind and Solar!
technical system
2. Base-Load power plants disappear altogether,
3. There is plenty of flexibility but so far it has no
value
4. Grids are cheaper than storage facilities
5. Securing supply in times of peak load does not
cost much
6. Integration of the heat sector makes sense
The potential of other

renewable energy
sources is limited
Wind and PV are the

cheapest renewable
energy sources
70
GW
010/03-I-2013/EN
How can Germany realise its Energie

wende? What concrete legislation, initiatives or measures are required to make
the Energiewende a success? Agora Energiewende lays the groundwork to ensure
that Germany makes the decisions required to set the course towards full decarbonisation of the power sector. We are
a think-and-do tank, working to enhance
a fuller understanding of the Energie
wende and facilitate a convergence of
views among relevant stakeholders.
Agora Energiewende
Rosenstrae 2 | 10178 Berlin | Germany
T +49. (0)30. 284 49 01-00
F +49. (0)30. 284 49 01-29
www.agora-energiewende.de
info@agora-energiewende.de
Agora Energiewende is a joint initiative of the Mercator Foundation and the European Climate Foundation.

Agora 12 Insights On Germanys Energiewende Web

Uploaded by

Copyright:

Available Formats

Agora 12 Insights On Germanys Energiewende Web

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Agora 12 Insights On Germanys Energiewende Web

Uploaded by

Copyright:

Available Formats

12 Insights on

In June 2011 the German Parliament ended a long debate

This transition of Germanys energy system is called

Since the long-term targets are in place the big question

We know from numerous studies and scenarios that

Status Quo and Main Targets of the Energiewende

primary energy use (against 2008)

electricity demand (against 2008)

share in electricity consumption

share in final energy use

Greenhouse Gas Emissions

heat in residential sector

for urgent action is becoming more and more obvious. For

are abundantly available in most regions of the world and

In many countries, wind power and photovoltaics will

1. Its all about Wind and Solar!

Wind and PV are the

> weather dependent

How to balance demand and supply?

market design and regulation

12. A saved kilowatt hour is the most cost-effective kilowatt hour

Agora Energiewende | 12 Insights on Germanys Energiewende

12 Insights on Germanys Energiewende

All other renewable technologies are either

1 See IPCC (2011) and IRENA (2012a); (2012b). The cost of

approx.6%). The reason is that agricultural and forest

Agora Energiewende | 12 Insights on Germanys Energiewende

By 2015, the cost for power produced by

these will continue to be primarily dependent on fossil fuels.

Wind and PV power are the two essential

12 Insights on Germanys Energiewende

Electricity Demand and Generation by Renewable

Early February (calendar week 6)

Wind and solar energy plants have three key

Mid-August (calendar week 33)

residual demand that will still need to be covered by existing

They have high capital costs but (virtually)

Late November (calendar week 47)

Fossil fuel power plants

Wind onshore /offshore

Illustration based on Agora Energiewende (2012a)

They dependent upon energy source availability; that

11 This graph was prepared by Fraunhofer IWES at the request of

Agora Energiewende | 12 Insights on Germanys Energiewende

These characteristics are fundamentally

Wind and PV power should be expanded in

is shining at its strongest.13 Since electricity is relatively

p. 1 of the data appendix; IRENA (2012a); IRENA (2012b).

13 See for example, the analyses by Gerlach/Breyer (2012) for

12 Insights on Germanys Energiewende

EEG has turned renewables into base load power by law.

Rapid changes in feed-in from renewables as

From Monday to Thursday, wind and solar power cover most

Illustration based on Agora Energiewende (2012a)

The fluctuating patterns of generation from wind and solar

Agora Energiewende | 12 Insights on Germanys Energiewende

power plants must be rapidly ramped up and down over

heat, the facility runs to produce heat, and merely produces

Demand side management and storage

From Monday to Thursday, wind and solar power cover most

Thursday from 10 AM to 1 PM, electricity demand is covered

Flexibility of Fossil Fuel Power Plants

In a week in early February there is so much wind that a

At the moment of the maximum load of 80 GW (for example,