Explanatory Memorandum to SEC(2011)1565 - Impact Assessment accompanying the document: Energy Roadmap 2050

Please note

This page contains a limited version of this dossier in the EU Monitor.

dossier SEC(2011)1565 - Impact Assessment accompanying the document: Energy Roadmap 2050.
source SEC(2011)1565 EN
date 15-12-2011
52011SC1565

COMMISSION STAFF WORKING PAPER Impact Assessment /* SEC/2011/1565 final - */


Annex 1 Scenarios – assumptions and results

Contents

1.

Part B: Decarbonisation scenarios. 2


2.

1. Assumptions. 2


3.

1.1 Macroeconomic and demographic assumptions. 2


4.

1.2 Energy import prices. 2


5.

1.3 Policy assumptions. 3


6.

1.4 Assumptions about energy infrastructure development 7


7.

1.5 Technology assumptions. 7


8.

1.6 Drivers. 9


9.

2. Results. 9


10.

2.1 Overview: outcome for the four main strategic directions to decarbonisation. 9


11.

2.2 Energy consumption and supply structure. 13


12.

2.3 Power generation. 20


13.

2.4 Other sectors. 31


14.

2.5 Security of supply. 34


15.

2.6 Policy related indicators. 36


16.

2.7 Overall system costs, competitiveness and other socio-economic impacts. 39


17.

2.8 Conclusions. 53


Attachment 1: Numerical results. 56

Attachment 2: Assumptions about interconnections and modelling of electricity trade. 78

Attachment 3: Short description of the models used. 85

Part B: Decarbonisation scenarios 1. Assumptions

18.

1.1 Macroeconomic and demographic assumptions


On the basis of the European Council's objective for EU decarbonisation of at least 80% below 1990 by 2050 in the context of necessary reductions by developed countries as a group[1] it is assumed that competitiveness effects throughout decarbonisation would be rather limited. Therefore, the decarbonisation scenarios are based on the same demographic and macroeconomic assumptions as the Reference scenario and Current Policy Initiatives scenario. Such an assumption also facilitates comparison of the energy results across scenarios. These macro-economic (sectoral production) assumptions also hold for energy intensive industries. However, under fragmented action, measures against carbon leakage may be necessary. The analysis of this particular case (see below) deals with energy and emission effects of such measures, but does not address potential changes in sectoral production levels under fragmented action. The aim of measures against carbon leakage is indeed to avoid such relocation of energy intensive production.

19.

1.2 Energy import prices


The decarbonisation scenarios are based on 'global climate action' price trajectories for oil, gas and coal i reflecting that global action on decarbonisation will reduce fossil fuel demand worldwide and will therefore have a downward effect on fossil fuel prices. Oil, gas and coal prices are therefore lower than in the Reference scenario and Current Policy Initiative scenario. Their trajectories are an outcome of the global analysis in the Low carbon Economy Roadmap, which is similar to recent IEA projections that assessed the impacts of ambitious climate policies i.

20.

Figure 18: Fossil fuel prices in the decarbonisation scenarios



21.

1.3 Policy assumptions


In addition to policy assumptions in the Current Policy Initiatives scenario, the following policies and measures were added to scenarios:

22.

Table 15 Measures included in all decarbonisation scenarios


Climate policies for respecting carbon constraints to reach 85% energy related CO2 reductions by 2050 (40% by 2030), consistent with 80% reduction of total GHG emissions according to the 'Roadmap for moving to a competitive low carbon economy in 2050' (including achievement of cumulative carbon cap) in a cost effective way| Supplementary to specific energy policies in the scenario, ETS prices and carbon values for non ETS sectors are determined in such a way as to reach the 2050 reduction goal; ETS and non ETS sectors use equal carbon prices/values (from 2025 onwards); cumulative emissions are similar across scenarios

Stronger RES facilitation policies| Represented by higher RES-values in the model. These facilitating RES policies include for example the availability of more sites for RES, easier licensing of RES installations, greater acceptance and support deriving from the improvement of local economies and industrial development; operational aids remain at the same level as in the REF/CPI scenarios.

Transport measures| Energy efficiency standards, internal market, infrastructure, pricing and transport planning measures leading to more fuel-efficient transport means and some modal shift Encourage the deployment of clean energy carriers by establishing the necessary supporting infrastructures[4]

Guarantee funds for all low carbon generation technologies| The model reflects support to early demonstration and first of a kind commercial plants for all innovative low-carbon technologies in the energy sector (nuclear, RES and their infrastructure needs, CCS, etc.).

Storage and interconnections| Higher penetration of variable generation leading occasionally to excess electricity is dealt with by increased pump storage and more interconnection capacity. Moreover, large parts of such excess electricity generation from variable sources is transformed into hydrogen, which is fed, up to a certain degree, into the natural gas grid, thereby providing a means for (indirect) storage of electricity and reducing the carbon content of gas delivered to final consumers enabling deeper emission cuts. Where for technical or economic reasons, simulated in the model, feeding into the natural gas grid is not feasible, excess electricity (mainly from RES) is stored in form of hydrogen at times of excess supply and transformed back into electricity when demand exceeds supply. (Hydrogen storage is used to a different degree in various decarbonisation scenarios, see also measures under Scenario 4).

23.

Scenario 2: High energy efficiency


This scenario is driven by a political commitment of very high primary energy savings by 2050. It includes a very stringent implementation of the Energy Efficiency Plan and aims at reaching close to 20% energy savings by 2020. Strong energy efficiency policies are also pursued thereafter.

Table 16 Policies/measures included (in addition to measures in table 15):

|| Measure| How it is reflected in the model

Additional strong minimum requirements for appliances| Progressive adaptation of modelling parameters for different product groups. As requirements concern only new products, the effect will be gradual.

High renovation rates for existing buildings due to better/more financing and planned obligations for public buildings (more than 2% refurbishment per year)| Change of drivers (ESCOs, energy utilities obligation, energy audits) influence stock – flow parameters in the model reflecting higher renovation rates (higher than 2% pa), with account being taken of tougher requirements for public sector through specific treatment of the non-market services sector

24.

Passive houses standards after All new houses after 2020 comply with passive house standards - around 20-50 KWh/m2 (depending on the country) which might to a large extent be of renewable origin


Marked penetration of ESCOs and higher financing availability| Enabling role of ESCOs is reflected in lower discount rates for household consumers (from 17.5% to 16% in 2015, 14% in 2020, 13% in 2025 and 12% from 2030 onwards) and for industry, agriculture and services (from 12% to 11% by 2015 and to 10% from 2020 onwards)

Obligation of utilities to achieve energy savings in their customers' energy use over 1.5% per year (up to 2020)| Induce more energy efficiency mainly in residential and tertiary sectors by imposing an efficiency value for grid bound energy sources (electricity, gas, heat)

Strong minimum requirements for energy generation, transmission and distribution including obligation that existing energy generation installations are upgraded to the BAT every time their permit needs to be updated | Higher efficiency of power plants through removing less efficient items from the generation portfolio, allowing however for efficiency losses where CCS is deployed Less transmission and distribution losses

Full roll-out of smart grids, smart metering| Enabling more efficiency and decentralised RES; Reflected as costs in the distribution grid costs, electricity prices and overall costs of the energy system

Significant RES highly decentralised generation| More advanced power dispatching and ancillary services to support reliability of power supply Higher penetration of small wind, solar and hydro

25.

Scenario 3: Diversified supply technologies scenario


This option is mainly driven by carbon prices and carbon values (equal for ETS and non ETS sectors). Carbon values are a still undefined proxy for policy measures that bring about emission reduction. They do not represent a cost to economic actors outside EU ETS (where they coincide with the EU ETS price), but are economic drivers that change decision making of the modelled agents. Yet, the changes triggered by carbon values may entail costs (e.g. for investment in energy savings or for fuel switching), which are accounted for in the modelling framework. They are applied to all sectors and greenhouse gas emissions, covering ETS and Non ETS sectors. As economic drivers, they influence technology choices and demand behaviour. Their respective level is not an assumption but a result of the modelling depending among other things on the level of ambition in GHG reduction. The modelling applies equal carbon values across sectors and ensures thereby efficient reductions across sectors.

This option assumes acceptance of nuclear and CCS and development of RES facilitation policies. It reproduces the 'Effective and widely accepted technologies' scenario used in the Low Carbon Economy Roadmap and Roadmap on Transport on the basis of scenario 1bis.

Table 17 Policies/measures included (in addition to measures in table 15):

|| Measure| How it is reflected in the model

26.

MS and investors have confidence in CCS as a credible and commercially viable technology; acceptance of storage and CO2 networks is high|


MS, investors and society at large have confidence in nuclear as safety is considered adequate and waste issues are solved| Applicable for all countries that have not ruled out the use of nuclear, i.e. Germany and Belgium for the longer term and the currently non-nuclear countries except for Poland


27.

Scenario 4: High RES


This scenario aims at achieving very high overall RES share and very high RES penetration in power generation (around 90% share and close to 100% related to final consumption). Recalling security of supply objectives, this would be based on increasing domestic RES supply including off-shore wind from the North Sea; significant CSP and storage development, increased heat pump penetration for heating and significant micro power generation (PV, small scale wind, etc.). Regarding assumptions for the demand sectors, scenario 4 is similar to scenario 3, with the exception that RES are more intensively facilitated.

Table 18 Policies/measures included (in addition to measures in table 15):

|| Measure| How it is reflected in the model

Facilitation and enabling policies (permitting, preferential access to the grid)| Represented by significantly higher RES-values in the model than in other decarbonisation scenarios; these RES facilitating policies include for example lower lead times in construction, and involve greater progress on learning curves (e.g. small scale PV and wind) based on higher production.

Market integration allowing for more RES trade| Use of cooperation mechanisms or convergent support schemes coupled with declining costs/support result in optimal allocation of RES development, depending also on adequate and timely expansion of interconnection capacity (point 4);

Stronger policy measures in the power generation, heating and transport sectors in order to achieve high share of RES in overall energy consumption in particular in household micro power generation and increased power production at the distribution level.| Higher use of heat pumps, significant penetration of passive houses with integrated RES reflected through faster learning rates (cost reductions), higher penetration rates (e.g. due to RES building/refurbishing requirements)

Infrastructure, back-up, storage and demand side management| Substantial increase in interconnectors and higher net transfer capacities including DC lines from North Sea to the centre of Europe. Back-up functions done by biomass and gas fired plants. Sufficient storage capacity is provided (pumped storage, CSP, hydrogen). Smart metering allows time and supply situation dependent electricity use (peak/off-peak) reducing needs for storing variable RES electricity. All these measures allow for exploiting greater potentials for off-shore wind in the North Sea.

28.

Scenario 5: Delayed CCS


The delayed CCS scenario shows consequences of a delay in the development of CCS, reflecting acceptance difficulties for CCS regarding storage sites and transport; large scale development of CCS is therefore assumed feasible only after 2040.

Table 19 Policies/measures included (in addition to measures in table 15):

|| Measure| How it is reflected in the model

Acceptance difficulties for CCS regarding storage sites and transport, which allow large scale development only after 2040.| Shift of cost-potential curves to the left (higher costs reflecting delays and public opposition). The learning curve for CCS is also delayed accordingly, resulting in higher capital costs for CCS than in scenario 3

MS , investors and society at large have confidence in nuclear as safety is considered adequate and waste issues are solved| Low risk premiums for nuclear Applicable for all countries that have not ruled out the use of nuclear, i.e. Germany and Belgium for the longer term and the currently non-nuclear countries except for Poland

29.

Scenario 6: Low nuclear


This scenario shows consequences of a low public acceptance of nuclear power plants leading to cancellation of investment projects that are currently under consideration and no life time extension after 2030. This leads to higher deployment of the substitute technologies CCS from fossil fuels on economic grounds.

Table 20 Policies/measures included (in addition to measures in table 15):

|| Measure| How it is reflected in the model

Political decisions based on perceived risks associated with waste and safety (especially in the aftermath of the Fukushima accident) leading to no new nuclear plants being build besides the ones presently under construction: 1600 MWe in Finland, 2x1600 MWe in France and 2x505 MWe in Slovakia. Moreover, the recourse to deciding instead on nuclear lifetime extension is available only up to 2030.| No extension of nuclear lifetime on economic grounds after 2030 No new nuclear plants are being built besides reactors under construction : 1600 MW in FIN; 2*1600 MW in FR and 2*505 MW in SK

MS and investors have confidence in CCS as a credible and commercially viable technology; acceptance of storage and CO2 networks is high| Low risk premiums for CCS

30.

1.4 Assumptions about energy infrastructure development


Infrastructure modelling for decarbonisation scenarios was done similarly to the approach described in Part A, section 1.4 for the Reference and Current Policy initiatives scenarios.

For decarbonisation scenarios the analysis done showed that except for very high RES penetration, the 2020 interconnection capacity would allow for most intra-EU electricity trade provided that some bottlenecks would be dealt with. The identified bottlenecks concerns interconnections around Germany, in Austria-Italy-Slovenia, Balkans and Denmark-Sweden. Greater investment and capacity for these specific links were assumed.

For very high RES penetration, which involves much more RES based electricity trade, stronger growth of interconnection capacity will be required. Under the assumptions of this scenario, full exploitation of off-shore wind potential at North Sea is foreseen. It is assumed that a dense DC interconnection system will develop mainly offshore but also partly onshore, to facilitate power flows from the North Sea offshore wind parks to consumption centres. In this scenario, the links of Sweden with Poland, Sweden with Lithuania, Austria with Italy, France with Italy and links in the Balkan region appear to be congested and need to be reinforced mainly with DC lines.

31.

For more details on the modelling approach and results see Attachment 2.


32.

1.5 Technology assumptions


Many technology assumptions are the same as in the Reference scenario and Current Policy Initiatives scenario (with revised assumptions about nuclear). In the decarbonisation scenarios, however, there are additional features and mechanisms that produce high decarbonisation and technology penetration.

Whereas all decarbonisation scenarios rely on technologies that exist today, they might become commercially mature only over time supported also by decarbonisation requirements. The uptake of the technologies is endogenous in the scenarios with their large-scale deployment leading to lower cost and higher performance, which correspond to a fully mature commercial stage.

All scenarios simulate merit order dispatching for power generation with contribution of variable generation from renewables. Electricity balancing and reliability is ensured endogenously by various means such as import and export flows (in case of high RES it is facilitated by expanding interconnections), investment in flexible thermal units, pumped storage and if required hydrogen based storage. In this latter case, excess variable generation from RES at times of lower demand may be used to produce hydrogen via electrolysis which is then used to produce electricity in turbine based power plants when electricity demand exceeds production from RES and other available sources (e.g. in situations of high demand).

The modelling approach also considers the possibility to mix hydrogen produced through electrolysis in the low and medium pressure natural gas distribution system (up to 30%) in order to reduce the average emission factor of the supplied blend, thereby contributing to the decarbonisation of final energy consumption.

Photovoltaic in High RES Scenario evolves along more optimistic trajectories than in the Reference scenario, as it is presumed that the higher penetration of the technology leads to stronger learning by doing. The higher uptake of RES technologies is driven mainly by the lower cost potentials for RES power, which are due to policies facilitating access to resources and sites.

A further change is in the Delayed CCS scenario where the development of CCS is delayed, and does not reach the same levels of development as in the other scenarios.

There is also faster progress in energy efficiency related technologies due to bigger scale and carbon prices effects. The energy technologies on the demand side follow a different development from the Reference scenario variants. In any situation there are different choices to consumers regarding the energy performance of appliances, buildings and equipment (evident from e.g. energy labelling where such transparency is provided by legislation). In decarbonisation scenarios, there are stronger shifts towards the more efficient technology vintages, which improve the average energy efficiency of a given energy use (e.g. of the average lighting appliance) compared to the Reference scenario variants. Energy efficiency progress is therefore supported by consumer choice effects similar to increased learning by doing driven by consumers opting for the more efficient available technologies.

The assumptions on the battery costs for the transport sector were developed along the lines of the White Paper on a Roadmap to a Single Transport Area. Efficiency improvements of ICE vehicles also occur in response to carbon values, making the overall vehicle fleet more efficient than in the Reference scenario and its variants. However, the following decarbonisation scenarios do not produce the same energy related transport outcome due to the fact that these scenarios do not handle the same transport details and that the overall framework conditions are different according to the scenario. In particular, the penetration of some alternative propulsion technologies (electric vehicles, hydrogen, etc.) might be somewhat different.

33.

1.6 Drivers


An internal greenhouse gas emission reduction contribution of around 80% in 2050 is taken as the key constraint for exploring different scenarios. To ensure that decarbonisation efforts are comparable across options and scenarios, the equalisation of cumulative emissions across scenarios is used as an additional constraint, underlining the importance of the climate impacts of cumulative emissions over the whole period until 2050 (and beyond). The corresponding decarbonisation effort from energy related CO2 emissions is 85% CO2 reductions compared to 1990, as demonstrated by the modelling underlying the Low Carbon Economy Roadmap of March 2011.

Common carbon values applied to all sectors and greenhouse gas emissions, covering ETS and Non ETS sectors, are used as key driver to reach the emission reductions and to ensure cost efficient reductions across sectors. As economic drivers, they influence technology choices and demand behaviour, in addition to the energy policies mirrored in the various scenarios for the Energy Roadmap. The respective level of carbon values is not an assumption but a result of the modelling.

Another important driver concerns international energy prices. Given that these scenarios assume global action, significantly lower fossil fuel prices are assumed than those in the reference and Current Policy Initiatives scenarios. Their order of magnitude has been set at a similar level as the results of the global analysis done for the Low Carbon economy Roadmap and recent IEA projections which assessed the impacts of ambitious climate policies.

To increase the penetration of renewable energy sources the RES-value was increased compared to the Reference scenario. In 2050, the RES-value in the decarbonisation scenarios is twice as high as in the current trend cases: instead of 35 €/MWh in Reference and CPI it amounts to 71 €/MWh in all decarbonisation scenarios, except for the high RES scenario, in which RES support is much more pronounced (RES-value of 382 €/MWh). The RES-value is a modelling tool used to reflect the marginal value of not explicitly modelled facilitation RES policies. These facilitating RES policies include for example the availability of more sites for RES, easier licensing of RES installations, benefits deriving from the improvement of local economies and industrial development. In High RES scenario the RES-value is the shadow value associated with the additional target of maximisation of the RES share in power generation and in the overall energy mix.

34.

2. Results 2.1 Overview: outcome for the four main strategic directions to decarbonisation


Decarbonisation can be achieved through energy efficiency, renewables, nuclear or CCS. Pursuing each of these main directions can bring the energy system a long way towards the decarbonisation objective of reducing energy related CO2 emissions by 85% below 1990 by 2050. The policy options (scenarios) proposed explore 5 different combinations of the four decarbonisation options. Decarbonisation options are never explored in isolation as interaction of different elements will necessarily be included in any scenario that evaluates the entire energy system. Moreover, the climate change issue is about atmospheric concentrations of GHG, i.e. with the long lifetimes of gases involved it is essentially about cumulative emissions. All scenarios achieve also the same level of cumulative GHG emissions. This makes energy, environmental and economic impacts comparable across the scenarios.

35.

Energy Efficiency


Energy Efficiency is a key ingredient in all the decarbonisation pathways examined. Its contribution is most important in the Energy Efficiency scenario (Scenario 2). Energy savings in 2050 from 2005 (virtually the peak energy consumption year) amount to 41%, while GDP more than doubles over the same period of time (+104%). The lowest contribution from energy efficiency towards decarbonisation comes in the Delayed CCS scenario, having a high nuclear contribution, in which primary energy consumption declines 32% between 2005 and 2050. As GDP does not change between scenarios, these energy savings from 2005 are entirely due to energy efficiency gains in a broad sense (including structural change), but not involving income losses.

In the Energy Efficiency scenario, one unit of GDP in 2050 requires 71% less energy input than in 2005. The average annual improvement in energy intensity (primary energy consumption / GDP) amounts to 2.7% pa, which is almost a doubling from historical trends (1.4% pa in 1990 to 2005 including the major efficiency raising restructuring in former centrally planned economies). All the decarbonisation scenarios have energy intensity improvements around 2.5% pa given e.g. synergies between energy efficiency and RES.

Energy savings in the High RES scenario are almost as high as in the Energy Efficiency case (minus 38% for energy consumption in 2050 compared to 2005 instead of minus 41%), this is however achieved by different means: the energy efficiency scenario focuses on direct impacts on final demand, whereas energy savings in the high RES case come largely through highly efficient RES technologies replacing less efficient nuclear and fossil fuel technologies.

A clear result concerning the strategic energy efficiency direction is that a substantial speeding up of energy efficiency improvements from historical trends is crucial for achieving the decarbonisation objective.

36.

RES


RES, too, are a key ingredient in any decarbonisation strategy. The RES share in gross final energy consumption (i.e. the definition for the existing 20% target) rises to at least 55% in 2050.

In the High RES scenario the RES share in gross final energy consumption reaches 75%, up 65 percentage points from current levels. The RES share in transport increases to 73%. The RES share in power generation reaches 86%. RES in electricity consumption account for even 97% given that electricity consumption calculated in line with the procedure for the calculation of the overall RES share excludes losses related to pump storage and hydrogen storage of electricity, the latter being necessary to accommodate all the available RES electricity in particular at times when electricity demand is lower than RES generation.

The second highest RES contribution (58%) materialises in the Low nuclear scenario. The RES share is also rather high under strong energy efficiency policies (57%).

The High RES scenario is the most challenging scenario regarding the restructuring of the energy system including major investments in power generation with RES capacity in 2050 reaching over 1900 GW, which is more than 8 times the current RES capacity and also more than twice today's total generation capacity (including nuclear, all fossil fuels and RES) (for more details see under power generation)

Nuclear

There is also a rather wide range with regard to the contribution of nuclear towards decarbonisation. The nuclear share is highest in the scenario that models the delayed availability of CCS (Scenario 4), given in particular issues arising with transport and storage of CO2 and has no additional policies on renewables and energy efficiency giving rise to an 18% share for nuclear in primary energy demand in 2050, which is 4 percentage points more than is projected under Current Policy Initiatives.

Least use of the nuclear option is made in the Low Nuclear Scenario (Scenario 6), which mirrors a hypothetical Europe-wide sceptical approach to nuclear deployment and investment. This scenario has still a nuclear share in primary energy of 3% in 2050 for reaching 85% CO2 reduction in 2050 similar to all the other decarbonisation scenarios.

The Diversified supply technology scenario (the other scenario, in which technologies compete on their economic merits alone) for reaching decarbonisation has a nuclear share in 2050 of 16% despite of nuclear phase-out in some Member States, which is still slightly higher than the current share. The High RES scenario would leave only little room for nuclear, bringing its share down to 4% in primary energy supply.

37.

CCS


The energy contribution of CCS towards decarbonisation is contingent upon the level of fossil fuel consumption[5] in sufficiently large units to justify economically the deployment of this technology. Hence the CCS share in e.g. gross electricity generation is limited by the degree of energy efficiency and decentralisation of energy supply as well as by the level of RES and nuclear penetration.

The highest share of CCS materialises in the Low Nuclear scenario (scenario 6). This case gives rise to a 32% share of CCS in gross electricity generation in 2050. CCS can substitute for nuclear in the case that this option was available only to a very limited extent. The CCS share would be particularly small in a scenario, in which almost all power generation stems from RES, i.e. Scenario 3, in which the CCS share drops to a mere 7%. The other scenarios have around 19-24% CCS share in gross electricity generation in 2050, with the lower end of the range stemming from delays in CCS technology introduction (mainly linked to storage issues).

38.

Decarbonisation requires substantial progress on both energy intensity and carbon intensity


The 4 decarbonisation dimensions, explored in 5 decarbonisation scenarios, can also be expressed in terms of energy and carbon intensity. Energy efficiency reduces energy intensity (energy consumption divided by GDP) while the other three options (RES, nuclear and RES) impact overwhelmingly on carbon intensity (CO2 divided by energy consumption). Substantial progress needs to be made on both indicators- energy and carbon intensity – which are to some degree substitutes for each other. The more successful policies to reduce energy consumption are the less needs to be done on fuel switching towards zero/low carbon energy sources, and vice versa[6] (see Figure 19). The five decarbonisation scenarios show substantial improvements in energy intensity which sinks 67%-71% compared with 2005 and 73%-76% compared with the higher 1990 level in terms of energy intensity (1990 had lower energy consumption, but also much lower GDP). Fuel switching continues in the decarbonisation scenarios up to 2050 and carbon intensity would improve substantially falling 76%-78% from 1990 (73%-75% from 2005).

39.

Figure 19: Decarbonisation scenarios: Improvements in carbon and energy intensities (reductions from 1990)


With ongoing economic growth, decarbonisation poses a formidable challenge given that meeting higher demand for energy services (heating and cooling, lighting, cooking, process energy, mobility, communication, etc) is part of increasing welfare. Upward pressure on energy consumption and CO2 emissions from economic growth is substantial given that GDP might increase almost threefold between 1990 and 2050 (see figure 20). The 80% GHG reductions objective by 2050 will however require deep cuts into energy related CO2 emissions, which in turn require energy consumption to decrease substantially as well.

Figure 20: Decarbonisation scenarios: development of GDP, primary energy consumption and energy related CO2 emissions: 1990 = 100

40.

2.2 Energy consumption and supply structure


Primary energy consumption is significantly lower in all decarbonisation scenarios as compared to the Reference scenario. This is also true for the Current Policy Initiatives scenario that shows 6 and 8% lower demand in 2030 and 2050, respectively than in the Reference scenario reflecting effects of energy efficiency measures in the Energy Efficiency Plan. The biggest decline of primary energy consumption comes in Energy Efficiency scenario (-16% in 2030 and -38% in 2050) showing effects of stringent energy efficiency policies and smart grid deployment. Compared with the actual outcome for 2005, primary energy consumption shrinks by 41%. The decrease in energy consumption compared with Reference for the decarbonisation scenarios spans a range from 11% - 16% in 2030 and from 30% to 38% in 2050. Energy efficiency is therefore an essential building block in all decarbonisation scenarios.

41.

Table 21: Total Primary energy consumption, changes compared to the Reference scenario


(Mtoe) 2050

Reference| 1763

Current policy Initiatives| 1615

% difference to Reference| -5.0%| -5.8%| -8.4%

Energy efficiency| 1084

% difference to Reference| -8.1%| -16.0%| -38.5%

Diversified supply technologies| 1217

% difference to Reference| -6.1%| -11.3%| -31.0%

High RES| 1134

% difference to Reference| -6.2%| -12.7%| -35.7%

Delayed CCS| 1238

% difference to Reference| -6.1%| -11.4%| -29.8%

Low nuclear| 1137

% difference to Reference| -5.8%| -13.9%| -35.5%

It is important to note that these levels of reduced primary energy demand do not come from reduced activity levels (which remains the same across all scenarios). Instead they are mainly the result of technological changes on the demand and also supply side: from more efficient buildings, appliances, heating systems and vehicles and from electrification in transport and heating, which combines very efficient demand side technologies (plug-in hybrids, electric vehicles and heat pumps) with a largely decarbonised power sector. Some changes related especially to fuel switching also contribute to reducing primary energy demand, such as switching from lignite or nuclear power generation to gas or wind based electricity production, which is associated with higher conversion efficiencies. In addition, behavioural change, triggered by e.g. changes in prices, information, energy saving obligations, etc, contributes to better energy efficiency.

Energy intensity of GDP (primary energy divided by GDP) reduces by 53% between 2005 and 2050 in the Reference scenario; the CPI scenario scores significantly better by improving energy intensity 57%. Energy intensity diminishes further in all decarbonisation scenarios: by at least 67% in the delayed CCS scenario. It improves 70% in the high RES and the low nuclear scenarios and even 71% in the energy efficiency scenario. Under decarbonisation, a unit of GDP in 2050 requires only one third of the energy needed today (or slightly less under e.g. a strong energy efficiency focus). By 2030, energy intensity would improve around 45% from current levels under decarbonisation, while this improvement would amount to some 40% under current policies.

Absolute energy savings, not considering the doubling of GDP between now and 2050, show still impressing numbers. Compared with the recent peak in energy consumption in 2005/6, the energy efficiency scenario depicts 41% less energy consumption, which means a substantial energy saving with respect to the levels reached just before the economic crisis.

42.

Figure 21: Primary energy savings in 2050 compared to 2005


It is important to note that these levels of reduced primary energy demand do not come from reduced GDP or sectoral production levels (which remain the same in all scenarios). Instead they are mainly the result of technological changes on the demand and supply side, coming from more efficient buildings, appliances, heating systems and vehicles and from electrification in transport and heating. All decarbonisation scenarios over-achieve the 20% energy saving objective in the decade 2020-2030.

The scenarios are based on model assumptions, which are consistent with the input for the 2050 Low Carbon Economy Roadmap. Recognising the magnitude of the decarbonisation challenge, which implies a reversal of a secular trend towards ever increasing energy consumption, this Energy Roadmap has adopted a rather conservative approach as regards the effectiveness of policy instruments in terms of behavioural change. However, the Roadmap results should not be read as implying that the 20% energy efficiency target for 2020 cannot be reached effectively. Greater effects of the Energy Efficiency Plan are possible if the Energy Efficiency Directive is adopted swiftly and completely, followed up by vigorous implementation and marked change in the energy consumption decision making of individuals and companies.[7]

Not only the amount, but also the composition of energy mix would differ significantly in a decarbonised energy system. Figure 22 shows total energy consumption as well as its composition in terms of fuels in 2050 for the various scenarios.

43.

Figure 22: Total Primary Energy in 2050, by fuel


Low and zero carbon content energy sources are strongly encouraged by going the various decarbonisation routes, each of them focusing on different aspects. This has different repercussions on the fuel mix. Energy efficiency encourages primary sources that can be used with small losses (e.g. many renewables or gas) and electricity at the level of final demand. CCS strategies affect the fuel mix by largely neutralising the high carbon content of fossil fuels, notably coal and lignite, through removal of the associated emissions. RES and nuclear routes are directly targeting the fuel mix. The modelling leads to rather wide ranges for primary energy sources with these fuel mixes in the decarbonisation cases all satisfying the decarbonisation requirement by 2050. Moreover, the development of all the fuel mixes under decarbonisation give rise to the same cumulative GHG emissions from 2011 to 2050.

Table 22: Share of fuels in primary energy consumption in %

||| Reference scenario| Current Policy Initiatives| Decarbonisation scenarios

| 2050

Solids| 17.| 12.| 11.| 12.| 9.| 7.2-9.| 2.1-10.2

Oil| 37.| 32.| 31.| 34.| 32.| 33.4-34.| 14.1-15.5

Gas| 24.| 22.| 20.| 22.| 21.| 23.4-25.| 18.6-25.9

Nuclear| 14.| 14.| 16.| 12.| 13.| 8.4-13.| 2.6-17.5

Renewables| 6.| 18.| 19.| 19.| 23.| 21.9-25.| 40.8-59.6

Renewables increase their share significantly under adopted policies and would substantially rise in all decarbonisation scenarios to reach at least 22% of primary energy consumption by 2030 and 41% by 2050. The RES share is comparably low in those scenarios, in which nuclear plays a rather strong role (scenarios 4 and 5). The RES share is highest in High RES scenario reaching 60% in primary energy by 2050. It is also pretty high (44% and 46% in primary energy in 2050) in the Energy Efficiency and Low nuclear scenarios, respectively.

The RES share is higher when calculated in gross final energy consumption (indicator used for the 20% RES target). It represents at least 28% (2030) and 55% (2050) in all decarbonisation scenarios and rises up to 75% in 2050 in the High RES scenario.

44.

Figure 23: Range of Fuel Shares in Primary Energy in 2050 compared with 2009 outcome


Nuclear developments have been significantly affected by the policy reaction in Member States after the nuclear accident in Fukushima (abandoning substantial nuclear plans in Italy, revision of nuclear policy in Germany). These reactions and the forthcoming nuclear stress tests have been reflected in the modelling assumptions for the Current Policy Initiatives scenario (1bis). The downward effects for nuclear penetration in CPI are also present in the decarbonisation scenarios, since the modelling of these cases also included the recent policy adjustments on nuclear.

The share of nuclear varies depending on assumptions taken. In the scenario without new nuclear investment (except for plants under construction) and extension of lifetime only in this and the next decade, the nuclear share declines gradually to 3% by 2050. In the most ambitious nuclear scenario - Delayed CCS, the share rises to 18%[8].

The share of gas under Current Policy Initiatives is higher than in the Reference scenario reflecting abandon of the nuclear programme in Italy, no new nuclear power plants in Belgium and higher costs for new plants and retrofitting. The gas share increases slightly to 26% in 2050 in the Low nuclear scenario where the CCS share in power generation is around 32%.

The oil share declines only slightly until 2030 (and even 2040) due to high dependency of transport on oil based fuels. However, the decline is significant in the last decade 2040-2050 where oil in transport is replaced by biofuels and electric vehicles. The oil share drops to around 15% in 2050 when following any of the examined main directions towards decarbonisation.

The share of solid fuels continues its long standing downward trend already under Reference and CPI developments. Under substantial decarbonisation the solids share shrinks further to reach levels as low as 2% in the High RES scenario in 2050 and only 4% and 5% under Energy efficiency and Delayed CCS, respectively. The solids share would remain rather high only in the Low nuclear scenario (10% in 2050) with a high CCS contribution which allows a continued use of solids in a decarbonisation context.

Final energy demand declines similarly to primary energy demand. Current Policy Scenario shows around 5% decrease (in 2020-2050) compared to the Reference scenario. In the Energy Efficiency scenario the reduction on Reference in final energy demand is -14% in 2030 and -40% in 2050. The decrease in the decarbonisation scenarios is at least -8% in 2030 and -34% in 2050. Compared with actual 2005 outcome, final energy consumption decreases in 2050 by 37% in the High Energy Efficiency scenario and by around 32% in all the other decarbonisation scenarios.

Sectors showing higher reductions than the average are residential, tertiary and generally also transport.

45.

Table 23: Final energy demand, changes compared to the Reference scenario


|| Reference scenario| Current Policy Initiatives| Energy efficiency| Diversified supply technologies

||

Final Energy Demand (Mtoe)| -6%| -4%| -5%| -9%| -14%| -40%| -7%| -9%| -34%|

Industry| -4%| -5%| -5%| -4%| -5%| -30%| -4%| -5%| -26%|

Residential| -9%| -6%| -4%| -13%| -20%| -43%| -9%| -12%| -35%|

Tertiary| -8%| -5%| -7%| -13%| -25%| -53%| -8%| -15%| -42%|

Transport| -4%| -2%| -6%| -7%| -12%| -40%| -7%| -9%| -38%|

|| High RES| Delayed CCS| Low nuclear|

||

Final Energy Demand (Mtoe)| -7%| -8%| -34%| -7%| -10%| -35%| -6%| -10%| -35%|

Industry| -4%| -4%| -25%| -4%| -5%| -26%| -3%| -6%| -26%|

Residential| -9%| -9%| -34%| -9%| -12%| -35%| -9%| -13%| -36%|

Tertiary| -8%| -13%| -44%| -8%| -16%| -42%| -7%| -17%| -43%|

Transport| -7%| -8%| -38%| -7%| -9%| -39%| -7%| -9%| -39%|

There is a lot of structural change in the fuel composition of final energy demand. Given its high efficiency and emission free nature at use, electricity makes major inroads already under current policies (increase by 9 pp between 2005 and 2050 in CPI).

The electricity share soars further in decarbonisation scenarios reaching 36% - 39% in 2050, reflecting also its important role in decarbonising further final demand sectors such as heating and services and in particular transport. The electricity share would almost double by 2050. The crucial issue for any decarbonisation strategy is therefore the full decarbonisation of power generation (see below).

46.

Table 24: Final energy consumption by fuel in various scenarios


Also RES make major inroads under current policies including the 2009 RES Directive. The direct use of RES in final demand (i.e. not counting here the RES input to power and distributed heat generation) rises strongly to 2030 coming close to a doubling of the share. However, without additional policy push beyond the current RES/climate change measures, this RES share could be stagnant. On the contrary, in decarbonisation scenarios the share of directly used RES (e.g. biomass, solar thermal) would go up to 24% in 2050 in almost all decarbonisation cases, except for the high RES scenario, where this share reaches even 30%.

Oil has been dominating final energy for many years and might continue doing so until 2030 even in the decarbonisation scenarios, when is would still account for one third of energy deliveries to final consumers. The big changes come after 2030 when more and more parts of final energy consumption based on oil, especially in transport, are replaced by electricity (e.g. electric and plug in hybrid vehicles, heat pumps). The oil share in 2050 would drop to around 15%.

The gas share has been declining in recent years and would be lower than today under both current policies and decarbonisation in 2030, when gas would account for not more than a fifth in final demand. The gas share after 2030 would be decreasing further in particular in decarbonisation scenarios, which is due to the greater role of electricity in both heating and for providing energy in productive sectors.

Distributed heat would deliver 7-8% of final energy demand in 2030 under both current policies and decarbonisation, raising its share substantially from current levels. The heat share in 2050 would be highest (10%) in the Low nuclear scenario where high electricity production is ensured by CCS equipped generation from gas and solids, often in a CHP mode.

Solid fuels become rather obsolete in final energy demand under current policies (falling to around 3% in 2030-2050). The decline of the solids share reflects higher use of electricity and gas in heating and industry. Solid fuels become marginal under decarbonisation, especially by 2050, when most solids base processes have been replaced by electricity or other fuels. The solid fuel share in 2050 would shrink to 0.3-0.4%.

47.

Figure 24: Shares of Electricity in Current Trend and Decarbonisation Scenarios


48.

2.3 Power generation


Electricity demand increases in all scenarios compared to 2005 levels, following greater penetration of electricity using appliances, heating and propulsion systems. The increased use of electric devices is partly compensated by the increased energy efficiency of electric appliances as well as increased thermal integrity in the residential and service sectors and more rational use of energy in all sectors, but overall the effect from emerging new electricity uses at large scale for heating and transport is decisive. The development of electricity consumption varies between sectors.

Transport electricity demand increases strongest. The increase of electricity use in transport is due to the electrification of road transport, in particular private cars, which can either be plug-in hybrid or pure electric vehicle; almost 80% of private passenger transport activity is carried out with these kinds of vehicles by 2050. Despite substantial progress regarding energy efficiency of appliances and for efficient heating systems, such as heat pumps, household electricity demand in 2050 under decarbonisation exceeds the current level given the additional deployment of electricity in heating and cooling.

Electricity demand in the other sectors decreases or remains flat under decarbonisation. Electricity demand in services/agriculture diminishes in all decarbonisation scenarios as a result of strong energy efficiency policies, although there is a substitution from other energy carriers to more efficient electric devices e.g. heat pumps. . Industrial electricity demand remains broadly at the current level by 2050 under decarbonisation.

49.

Table 25: Electricity final energy demand


|| 2050

|| Reference| Scenario 1bis| Scenario 2

Final energy demand (in TWH)| 3203

Industry| 1109

Households| 913

Tertiary| 518

Transport| 663

|| 2050

|| Scenario Scenario Scenario Scenario 6

Final energy demand (in TWh)| 3552

Industry| 1191

Households| 1013

Tertiary| 677

Transport| 671

50.

Power generation: level and structure by fuel


Given the assumed limited electricity import possibilities from third countries, the increased electricity demand will have to be generated nearly exclusively within the EU. Moreover, electricity production has to cover also power plant own consumption (e.g. for desulphurisation), the consumption of the other energy producing sectors (energy branch) as well as transmission and distribution losses. Furthermore, additional electricity generation is appropriate under strong decarbonisation objectives to produce hydrogen mixed in low and medium pressure gas networks (bringing down emission factors in final demand) and for producing hydrogen, which is used for balancing in the case of high RES scenarios. Therefore, similar to electricity demand there is a strong increase from current levels for power generation in all scenarios. Under decarbonisation, power generation will be lower in 2050 compared with Reference and CPI scenarios. The highest electricity generation level in 2050 among the decarbonisation cases comes about in case of CO2 reduction focussing particularly strongly on RES.

The structure of power generation changes substantially between the scenarios. The Reference scenario and the Current Policy Initiatives scenario show renewable shares in 2050 reaching 40 and 49% respectively and fossil fuels still having a share of 33 and 31% respectively. Among the decarbonisation scenarios, only the Low nuclear scenario has a share of fossil fuels above 30%, as it makes substantial use of CCS. In the other scenarios the fossil fuel share lies below 25% and is particularly low in High RES scenario, where fossil fuels account for under 10% of electricity generation.

Under decarbonisation, power generation in 2050 is based on renewables for at around 60%-65%, except for the high RES case, in which this share is much higher. Wind alone accounts for about one third of power generation in most decarbonisation scenarios. In the high RES case, the wind share reaches even close to 50% in 2050. The nuclear share falls from the present level in all decarbonisation scenarios. This share is highest in 2050 under delayed CCS, in which case it is around 20%. On the contrary, in the low nuclear scenario, nuclear would account for just 2.5% of power generation.

51.

Table 26: Power generation


||| 2050

||| Reference| Scenario 1bis| Scenario 2

Electricity generation| TWh| 4281

Nuclear energy| Shares (%)| 30.| 26.| 20.| 14.2

Renewables| 14.| 40.| 48.| 64.2

Hydro| 9.| 7.| 8.| 9.2

Wind| 2.| 20.| 24.| 33.2

Solar, tidal etc.| 0.| 5.| 7.| 10.6

Biomass & waste| 2.| 7.| 8.| 10.9

Geothermal heat| 0.| 0.| 0.| 0.3

Fossil fuels| 55.| 33.| 30.| 21.6

Coal and lignite| 30.| 15.| 11.| 4.8

Petroleum products| 4.| 2.| 2.| 0.0

Natural gas| 20.| 15.| 16.| 16.7

Coke & blast-furnace gasses| 0.| 0.| 0.| 0.0

Other fuels (hydrogen, methanol)| 0.| 0.| 0.| 0.0

||| 2050

||| Scenario Scenario Scenario Scenario 6

Electricity generation| TWh| 4853

Nuclear energy| Shares (%)| 16.| 3.| 19.| 2.5

Renewables| 59.| 83.| 60.| 64.8

Hydro| 8.| 7.| 8.| 8.1

Wind| 31.| 48.| 32.| 35.6

Solar, tidal etc.| 9.| 16.| 9.| 10.8

Biomass & waste| 9.| 9.| 9.| 9.8

Geothermal heat| 0.| 0.| 0.| 0.4

Fossil fuels| 24.| 9.| 20.| 32.7

Coal and lignite| 8.| 2.| 5.| 13.1

Petroleum products| 0.| 0.| 0.| 0.1

Natural gas| 16.| 7.| 14.| 19.5

Coke & blast-furnace gasses| 0.| 0.| 0.| 0.0

Other fuels (hydrogen, methanol)| 0.| 3.| 0.| 0.0

NB: power generation is presented in the most comprehensive way in this table involving in a sense some 'double counting' in the denominator of shares for the high RES scenario: first electricity generation from RES is counted including those parts of RES based generation that, in case supply exceeds demand, are transformed into hydrogen for later use by producing electricity for a second time from these original renewables sources. This specific representation for showing also the magnitude of hydrogen based RES electricity storage (4% in 2050) leads to total electricity generation numbers that are in a sense inflated, which in turn gives rise to lower RES share numbers in this specific representation that counts production from RES once as such and secondly under hydrogen based generation (shown separately) for the part that is not lost in transformations into hydrogen and back from hydrogen to electricity.

52.

Power plant investments by fuel type (e.g. RES, nuclear, fossils with CCS, fossil without CCS)


The installed capacity increases in all scenarios compared to the Reference scenario due to the additional balancing and power reserve capacities needed for the variable RES which increase in all scenarios. The scenario with the least increase is Energy Efficiency scenario which requires the least amount of electricity and therefore also the least amount of installed capacity. All scenarios still have fossil fuel fired power plants as installed capacity, which are used mainly as back-up.

The share of CCS capacity in thermal power plants for the decarbonisation scenarios ranges from 48% in Low nuclear scenario to 12% in High RES scenario. The share in the other scenarios is between 35 and 44%.

53.

Table 27: Installed power capacity


||| 2050

||| Reference| Scenario 1bis| Scenario 2

Net Installed Power Capacity| GWe| 1473

Nuclear energy| 79

Renewables (without biomass/geothermal) | 1012

Hydro (pumping excluded)| 125

Wind power| 548

Wind on-shore| 370

Wind off-shore| 177

Solar| 330

Other renewables (tidal etc.)| 9

Thermal power| 382

Solids fired| 70

Oil fired| 15

Gas fired| 187

Biomass-waste fired| 108

Hydrogen plants| 0

Geothermal heat| 2

||| 2050

||| Scenario Scenario Scenario Scenario 6

Net Installed Power Capacity| GWe| 1721

Nuclear energy| 16

Renewable (without biomass/geothermal)| 1193

Hydro (pumping excluded)| 127

Wind power| 674

Wind on-shore| 452

Wind off-shore| 222

Solar| 381

Other renewables (tidal etc.)| 11

Thermal power| 513

Solids fired| 125

Oil fired| 18

Gas fired| 255

Biomass-waste fired| 112

Hydrogen plants *| 0

Geothermal heat| 2

||| 2050

||| Reference| Scenario 1bis| Scenario 2

Total CCS capacity| GWe| 149

Solids| 28

Oil| 0

Gas| 121

||| 2050

||| Scenario Scenario Scenario Scenario 6

Total CCS capacity| GWe| 248

Solids| 79

Oil| 0

Gas| 169

* Hydrogen capacity in the above table refers only to plant technologies dedicated to specific hydrogen use, such as fuel cells. Capacity for generating electricity from hydrogen, serving only the purpose of storing RES based electricity that was previously produced at times when electricity supply exceeded demand, is accounted for under gas fired capacity, given that hydrogen would be burnt is such types of plants, including as a mixture with natural gas.

The high RES scenario is a particularly challenging scenario regarding the restructuring of the energy system involved; RES policy related challenges in this scenario include the following:

Huge investments in RES power capacity need to be ensured with wind capacity alone reaching over 980 GW in 2050, this is 20% more than today's (2010) total power generation capacity (including nuclear, fossil fuels and all RES); similarly, solar capacity would need to soar to 600 GW, which amounts to almost three quarters of our present total generation capacity; all RES power generation capacity (Renewables + biomass/waste + geothermal in table 27) would need to increase to over 1900 GW, which is more than 8 times the current RES capacity and also more than twice today's total generation capacity. It might be a challenge to ensure the raw material needed for RES technologies and there may be upward pressure on e.g. steel prices, which could be a challenge to such a development (not modelled with the energy model); other logistic challenges would relate to ensuring the maritime equipment to install and maintain the off-shore wind capacity that rises from just close to 5 GW today to over 370 GW in 2050; In order to accommodate RES production from remote sites with respect to consumption centres and to take advantage of the cost differences across Member States for cost-effectiveness reasons, the grid needs to be extended substantially and also smartened to deal with variable feed in from many dispersed sources (e.g. solar PV); the scenario analysis identified needs for grid extension beyond 2020 under a decarbonisation agenda and in addition a set of additional DC links (electricity highways) needed to accommodate a very high RES contribution to electricity supply (see attachment 2 to this Annex); Another challenge relates to the skilled workforce required, the lack of which can lead to a stalled development unless a major RES related education and training strategy is pursued taking account of ageing EU population over the next decades, which is even shrinking after 2035. Skilled workforce will also be needed for the construction of expanded, smart grids, which will also be necessary for the penetration of other low carbon technologies. In addition to economic, logistical, resource security and manpower challenges, there is the acceptance issue for new transmission lines and perhaps also regarding the substantial expansion of (on-shore) RES installations;

It will also be challenging in the other decarbonisation scenarios to ensure the required RES capacity in 2050 and to accommodate it by the grid. The Energy Efficiency scenario poses the least challenge given the lowest electricity demand, but nevertheless, RES power generation capacity would need to soar to 5 times the current level, exceeding today's total electricity generation capacity (nuclear, fossil fuel and RES combined) by more than a third. On the other hand, increased energy efficiency and decentralised RES might require more sophisticated solutions for distribution level.

Other scenarios pose also substantial challenges throughout the transition. For example, higher nuclear deployment in the delayed CCS scenario leads to more requirements for nuclear fuel and more nuclear waste that needs to be safely transported and stored. Electrification of passenger transport involves many changes in car production and infrastructure provision. A smooth transition from a petrol/diesel to an electricity based system for mainly urban transport requires a lot of logistical changes.

Widespread penetration of CCS will require dedicated CO2 transport grids that need to be financed, constructed and accepted. Acceptance challenges could be particularly pronounced for nuclear and CO2 storage. As carbon capture, transport and storage require significant quantities of electricity that need to be generated in addition to electricity for final use, there would be higher input demand also for fossil fuels. This effect would be particularly pronounced if global decarbonisation includes an important contribution from CCS for energy consumption and also for abatement of industrial process emissions. This could exert upward pressure on the level of world fossil fuel prices.

All scenarios involve substantial changes in production, transformation, smart transmission/distribution and consumption patters for energy, requiring a skilled workforce against the background of ageing population. Enhancement of the European capacity for innovation, appropriate RTD as well as education and training will be instrumental for a cost-effective transition to a low carbon economy that fosters competitiveness and security of supply.

Decarbonisation requires also considerable capacity for CCS, except for the high RES scenario. The other scenarios involve around 150 GW – 250 GW CCS capacity in 2050, with the upper end materialising in Low nuclear scenario, which is the case with the greatest use of CCS for power generation (32% share

In Table 28 the capacity investment per decade for the scenarios can be seen; as can be observed the highest investments take place in RES in all scenarios. As can be seen no new investment is undertaken in nuclear in Low nuclear scenario after 2030; only Delayed CCS sees higher nuclear investment than in the Reference scenario for the last two decades of the projection period. Investment continues in thermal power plants throughout the projection period in all scenarios; it is lowest in High RES and Energy efficiency scenarios. These investment numbers include lifetime extensions of existing plants, refurbishments and replacement investments on existing sites, which is particularly relevant for nuclear. These investment numbers must not be confused with additional new plants of e.g. nuclear.

54.

Table 28: Net Power Capacity Investment in GWe per decade


||| 2011-| 2021-| 2031-| 2041-2050

Reference| Nuclear energy| 62

Renewable energy| 259

Thermal power fossil fuels| 183

of which: CCS| 41

Thermal power RES| 24

Scenario 1 bis| Nuclear energy| 49

Renewable energy| 309

Thermal power fossil fuels| 198

of which: CCS| 17

Thermal power RES| 29

Scenario Nuclear energy| 22

Renewable energy| 436

Thermal power fossil fuels| 92

of which: CCS| 90

Thermal power RES| 29

Scenario Nuclear energy| 35

Renewable energy| 463

Thermal power fossil fuels| 101

of which: CCS| 98

Thermal power RES| 25

Scenario Nuclear energy| 0

Renewable energy| 817

Thermal power fossil fuels| 91

of which: CCS| 30

Thermal power RES| 53

Scenario Nuclear energy| 39

Renewable energy| 464

Thermal power fossil fuels| 115

of which: CCS| 110

Thermal power RES| 23

Scenario Nuclear energy| 0

Renewable energy| 515

Thermal power fossil fuels| 121

of which: CCS| 118

Thermal power RES| 27

Investment in generation capacity entails substantial cumulative investment expenditure in all scenarios over the period 2011-2050. Cumulative investment expenditure for power generation is most pronounced in the high RES scenario amounting to over 3 trillion € in real terms up to 2050. Among the decarbonisation scenarios cumulative investment expenditure for power generation is lowest in the Energy Efficiency scenario given the marked savings in electricity consumption.

Figure 25: Cumulative investment expenditure in 2011-2050 for power generation (in € of 2008)

55.

These investment expenditure results impact on electricity generation costs in the different scenarios (see below)


Impacts on infrastructure

Infrastructure requirements differ in scenarios. Decarbonisation scenarios require more and more sophisticated infrastructures (mainly electricity lines, smart grids and storage) than Reference and CPI scenarios. High RES scenario necessitates additional DC lines mainly to transport wind electricity from the North Sea to the centre of Europe and more storage. The biggest share of costs relate to the upgrade and improvement of distribution networks including smartening of the grid. Investments needed in transmission lines are much lower and new interconnectors represent only a fraction of these transmission costs.

Table 29: Grid investment costs

(Bn Euro'05)| Grid investment costs

2011-| 2021-| 2031-| 2011-2050

Reference| 1269

CPI| 1357

Energy Efficiency| 1518

Diversified supply technologies| 1712

High RES| 2195

Delayed CCS| 1717

Low nuclear| 1793

Euro'| Transmission Grid investment (bEUR)

2011-| 2021-| 2031-| 2041-| 2011-2050

Reference| 47.| 52.| 53.| 52.| 205.7

CPI| 47.| 49.| 64.| 66.| 228.2

Energy Efficiency| 49.| 63.| 80.| 80.| 272.5

Diversified supply technologies| 52.| 70.| 88.| 86.| 297.8

High RES| 52.| 95.| 137.| 134.| 420.4

Delayed CCS| 52.| 71.| 88.| 87.| 299.9

Low nuclear| 52.| 73.| 95.| 94.| 316.6

Euro'| Distribution Grid investment (bEUR)

2011-| 2021-| 2031-| 2041-| 2011-2050

Reference| 243.| 263.| 280.| 276.| 1063.7

CPI| 245.| 239.| 317.| 325.| 1127.8

Energy Efficiency| 256.| 289.| 408.| 291.| 1245.5

Diversified supply technologies| 284.| 345.| 454.| 329.| 1414.1

High RES| 283.| 440.| 619.| 431.| 1774.8

Delayed CCS| 283.| 349.| 445.| 339.| 1417.5

Low nuclear| 286.| 350.| 472.| 366.| 1476.3

Euro'| Investments in new electricity interconnectors

2006-| 2021-| 2031-2050

Reference| 13.| 0.| 0.0

CPI| 21.| 9.| 0.6

High energy efficiency| 21.| 9.| 0.6

Diversified supply technologies| 21.| 9.| 0.6

High RES| 21.| 21.| 50.8

Delayed CCS| 21.| 9.| 0.6

Low nuclear| 21.| 9.| 0.6

The model assumes that grid investments, that are prerequisites to the decarbonisation scenarios in this analysis, are undertaken and that costs are fully recovered in electricity prices. The reality might differ from this model situation in a sense that current regulatory regime might be more short to medium term cost minimisation oriented and might not provide sufficient incentives for long-term and innovative investments. There might also be less perfect foresight and lower coordination of investments in generation, transmission and distribution as the model predicts.

56.

Power generation costs


Fixed operational and capital costs for power generation increase over time in all scenarios. The increase in capital costs is more pronounced in decarbonisation scenarios, notably in the High RES case. A substantial RES contribution (high RES scenario) leads to an increase of fixed and capital costs of 155% in 2050 compared with 2005 (81% rise by 2030) due to the additional investment needs in generation, grid, storage and back-up capacities. On the contrary, the increase in variable and fuel costs over time under Reference and CPI developments would be more or less cancelled in the decarbonisation cases. This effect of shifting variable and fuel costs towards capital costs is most pronounced in the High RES scenario. In this decarbonisation case, the substantial RES contribution leads to a decline of variable and fuel costs by 45% below Reference in 2050 and also a decrease by 21% on the 2005 level.

Unit costs of transmission and distribution increase substantially in all decarbonisation scenarios. The High RES case has the greatest increase. Due to the decarbonisation of the power sector in all scenarios in the last two decades of the projection period, the costs related to ETS auction payments decrease substantially.

These effects on cost components allow for a decrease in electricity prices between 2030 and 2050 in all decarbonisation scenarios, except for the High RES scenario. This is in stark contrast to the period up to 2030, in which electricity prices increase due notably to increases in capital cost, grid costs and auctioning payments. The High RES case is an exception from other cases because of the very high investment requirements combined with stronger requirements on the electricity grid extension, which is not fully compensated by savings in fuel and other variable costs.

Therefore the High RES case features the highest electricity prices among the decarbonisation scenarios, as it would not allow for the flattening out of the strong price increase up to 2030 (observed in all scenarios) but continues with major capital intensive changes to the power system.

Table 31: Electricity prices and cost structure [9]

It is important to note that, as explained in the assumptions part, the PRIMES model makes sure that the full costs of electricity production and distribution are recovered through electricity prices. Both marginal costs and the appropriate portion of fixed capital and operation costs are allocated to the various sectors according to the Ramsey Boiteux methodology taking into account price elasticities in the allocation of fixed costs. This procedure is necessary to ensure a sustainable modelling solution because in internally consistent scenarios electricity sector investments need to be financed by the revenues from selling electricity.

However, power exchanges in wholesale markets work on the basis of marginal costs for determining spot prices with suppliers having lower marginal costs that the equilibrium price being able to cover (parts of) fixed costs. In a situation with a very high contribution of capital intensive low carbon technologies with marginal costs close to zero, such as RES, all suppliers succeeding to place bids might be bidders with such RES power plants and competition at power exchanges would drive this electricity price down close to zero. Obviously, close to zero prices over very long time segments every year would not be a sustainable solution in such a scenario, as the necessary capital expenditure and investment under such market structure could not be financed from selling revenues and such a scenario would not materialise. While PRIMES, presenting functioning scenarios, presents economically sustainable electricity prices, this issue appears to be an institutional challenge for the transition to a low carbon electricity system, especially for one that is nearly entirely based on RES.

57.

2.4 Other sectors


Heating and cooling: distributed heat/steam and RES

Demand for distributed heat in the decarbonisation scenarios rises compared to current level but is 2%-10% lower by 2030 as compared to the Reference scenario, with the greatest decline occurring in the high RES scenario. The decrease is more pronounced towards 2050 with 46% decrease as compared to Reference scenario in the High RES scenario; 26% decrease in the Energy Efficiency scenario and at least -20% decrease in other decarbonisation scenarios. The High RES scenario shows lowest distributed heat demand after 2025 due to the highest penetration of RES in power generation which leads to decrease of CHP[10] and due to the shift towards electricity use for heating reducing especially district heating from fossil fuels.

When comparing results for distributed heat between Reference and decarbonisation scenarios, it is important to note that final energy demand in the decarbonisation scenarios is 34% - 40% lower in 2050 than under reference developments (around 10% lower in 2030).

The biggest decrease as compared to the Reference scenario in 2050 occurs in the residential sector (-63% in High RES scenario and -32-42% in all other decarbonisation scenarios) reflecting stringent energy efficiency policies in buildings. Demand stays at current levels of around 240 TWh until 2015 and then gradually declines to 69 TWh in the High RES scenario and 126 TWh in the Low nuclear by 2050, showing the higher distributed heat demand among the decarbonisation cases.

The decrease in the tertiary sector is important as well with -43% in the Energy Efficiency scenario and at least -31% in all other scenarios. The demand peaks in 2015 at 120 TWh and goes down to 52 TWh in Energy Efficiency scenario and around 60 TWh in other decarbonisation scenarios.

Contrarily to residential and tertiary, industrial demand for heat increases massively from 160 TWh in 2005 to reach 503 TWh in High RES scenario and up to 733 TWh in Low nuclear/High CCS scenario by 2050. Industrial demand is still lower as compared to Reference scenario by at least 17% in all decarbonisation scenarios and by -43% in High RES scenario. However, industry needs steam for some processes that can hardly be substituted by other fuels.

Heat consumption is also rising in the energy branch from 54 TWh in 2005 to 71-77 TWh in 2050, with the Energy Efficiency and delayed CCS scenarios at the lower end of the range and the Low Nuclear scenario at the upper one.

Following the diverging trends in different sectors the shares of sectors in total distributed heat changes significantly up to 2050.

58.

Table 32: Heat/steam final consumption


|| 2050

Reference scenario| Decarbonisation scenarios

Industry| 161TWh| 31%| 880 TWh| 76%| 503 - 733 TWh| 81- 80%

Households| 240 TWh| 46%| 186 TWh| 16%| 69 - 126 TWh| 11 - 13%

Tertiary| 116 TWh| 22%| 92 TWh| 8%| 52 - 64 TWh| 8 - 7%

Final demand| 517 TWh| 100%| 1.159 TWh| 100%| 627 – 923 TWh| 100%

With lower final energy demand under decarbonisation, the share of distributed heat in total heating in the residential, services and agriculture sectors rises somewhat from current level of slightly over 11% in most scenarios, except for the High RES scenario. This decrease in the share of distributed heat is compensated by the increased direct use of biomass for heating, which soars from approx. 13.5% in 2010 to approx. 33% in 2050 in the High RES scenario.


59.

Table 33: Share of distributed heat in total heating for residential and tertiary


||

CPI| 11.6%| 12.0%| 12.0%|

Energy Efficiency| 12.0%| 12.8%| 13.3%|

Div. Supply Technology| 11.6%| 12.4%| 13.4%|

High RES| 11.6%| 11.4%| 8.5%|

Delayed CCS| 11.6%| 12.4%| 12.4%|

Low Nuclear| 11.6%| 12.5%| 13.7%|

60.

Heat and steam generation



Heat from CHP rises from 473 TWh in 2005 to 1030 TWh by 2025 in the High RES scenario and then declines to 682 TWh by 2050. In other scenarios, including Energy Efficiency, the rise continues until 2035 with the highest CHP generation in the Low nuclear scenario at 1113, exhibiting a slight decline thereafter. CHP heat production in 2050 covers a range from 682 TWh in the high RES scenario to 1019 TWh in the low nuclear case. As in the Reference scenario the growth is driven by support policies resulting from the application of the CHP directive and ETS carbon prices.

CHP share in power generation is the highest in the Low nuclear scenario reaching 22% in 2030. This share in 2030 is the lowest in the High RES scenario at 19%. By 2050, CHP share decline in all scenarios to 18% in Low nuclear and to 11% in High RES scenarios reflecting higher penetration of wind and solar in power generation (no combined production of heat possible) and electrification of heating in combination with energy efficiency policies to reduce demand for heat.

District heating is already declining from its 2000 levels of almost 190 TWh and this decline continues in the Reference scenario as well as in decarbonisation scenarios to 109 TWh in the Reference scenario and 29 -52 TWh in decarbonisation scenarios. The development of district heating is due to its benefits in reducing emissions in the short and medium term but in the long run, similarly to CHP plants, if district heating boilers do not use biomass, they emit GHG.

RES in heating and cooling

The modelling of energy demand formation by sector includes heating and cooling requirements as well as a detailed coverage of various ways of satisfying these needs including distributed heating and cooling from co-generation and district heating. As can be seen from table 34, there is very significant progress in all decarbonisation cases regarding the share of RES in heating and cooling. The RES share in heating and cooling doubles between 2005 and 2020 in all scenarios, reaching at least 44% by 2050 under decarbonisation. The highest share of well over 50% in 2050 is achieved in the High RES scenario.

61.

Table 34: Percentage share of RES in gross final consumption of heating and cooling


% share| 2050

CPI| 20.| 22.| 23.8

Energy Efficiency| 21.| 23.| 44.9

Div. Supply Technology| 20.| 23.| 44.0

High RES| 20.| 26.| 53.5

Delayed CCS| 20.| 24.| 44.9

Low Nuclear| 20.| 24.| 44.6

62.

Transport


In the decarbonisation scenarios, transport energy demand is projected to decline by close to 40% below Reference in 2050 due to active policies for tightening CO2 standards (essentially impacting on fuel efficiency), taxation, internal market and infrastructure measures[11]. The highest energy savings, in order of 155 Mtoe, are achieved in the Energy Efficiency scenario but all decarbonisation scenarios deliver savings in the same order of magnitude (around 150 Mtoe). Over 60% of these energy savings originate from passengers transport.

Energy intensity in passenger transport improves by slightly over 60% between 2005 and 2050 in the decarbonisation scenarios, mainly due to the enforcement of such efficiency standards. For freight transport, the efficiency standards together with measures encouraging a shift in modal choices lead to around 40% decrease in the energy intensity.

The EU transport system would remain extremely dependent on the use of fossil fuels in the Reference scenario. Oil products would still represent 88% of the EU transport sector final demand in 2030 and 2050 in the Reference scenario. Consumption of oil would decrease by 11% by 2050, relative to the Reference scenario, in the Current Policy Initiatives scenario mainly driven by the revision of the Energy Taxation Directive.

In the decarbonisation scenarios, final consumption of oil by transport is expected to decrease by almost 70% in 2050, relative to the Reference scenario; the oil share in final demand would amount to around 45%. This decline is compensated to a certain extent by the rise in the electricity demand by road and rail transport and the increased demand for biofuels, especially in aviation, inland navigation and long distance road freight, where electrification is not or less an option. Biofuels would represent slightly below 40% of energy consumption in aviation and inland navigation and 41% in long distance road freight by 2050. The role of biofuels in energy demand by passenger cars and light duty vehicles would be more limited, ranging between 13% and 15%. Electricity would provide around 65% of energy demand by passenger cars and light duty vehicles in all decarbonisation scenarios. Electro-mobility would need to be supported by the upgrade of Europe’s networks towards a European super grid and decarbonisation of electricity sector.

As a result of the higher demand for electricity and sustainable biofuels, the share of renewables in transport would increase by 2050, ranging between 62% and 73%. This difference between the decarbonisation scenarios can be explained by the different power generation mix, despite similar shares of biofuels and electricity demand in energy consumption by transport mean. Therefore, the highest share of renewables in transport is achieved in the High RES scenario.

63.

2.5 Security of supply


Import dependency in 2030 does not change substantially in decarbonisation scenarios as compared to Reference scenario and Current Policy Initiatives scenario due to decline in both gross inland consumption and imports. There is however a substantial decrease in 2050, driven by increased use of domestic resources, mainly renewables. Import dependency is only 35% in High RES scenario (compared to 58% in the Reference scenario and Current Policy Initiatives scenario) and 39-40% in all other decarbonisation scenarios besides Low nuclear scenario (45%) where it is higher due to significant use of fossil fuels with CCS. Decarbonisation will significantly reduce fossil fuel security risks.

64.

Table 35: Import dependency


%| 2050

1.Reference| 53.| 56.| 57.6

1 bis Current Policy Initiatives| 57.| 58.0

2. Energy efficiency| 56.| 39.7

3. Diversified supply technologies| 55.| 39.7

4. High RES| 55.| 35.1

5. Delayed CCS| 54.| 38.8

6. Low nuclear| 57.| 45.1

Large scale electrification combined with more decentralised power generation from variable sources brings other challenges to high quality energy service at any time. An adequate stability of the grid is a precondition for the consistent modelling of all scenarios; that is why differences in indicators such as reserve margin are rather small.

Utilisation rates of electric capacities decrease from 49% in 2005 to 36% in 2050 in the Reference scenario and to a range of 25% (High RES) to 33% (Diversified supply technologies scenario) in decarbonisation scenarios. This reflects higher requirements for reserve power and balancing services in order to keep supply of electricity reliable and secure in all scenarios.

All scenarios see a high increase in the share of variable RES in the electricity supply; this naturally leads to higher balancing requirements in the system. In the long term the balancing is met to the greatest extent by increased pumped storage (to the extent there is still increased potential available), the development of flexible gas-based units, higher import-exports and in the case of very high RES penetration with hydrogen based balancing. Thermal power plants, mainly gas fired ones, remain available as reserve power and provide ancillary services. The reduction in utilisation rates of thermal power plants is driven by economic considerations, not by predetermined exogenous inputs.

Utilisation rates for steam stay stable in the Reference scenario at around 43% but decrease to a range of 26% (High RES) and 36% (Diversified supply technologies scenario) in decarbonisation scenarios. Energy savings and electrification in heating which takes place in the decarbonisation scenarios limits the scope for further expansion of distributed heat/steam and CHP, except in cases of production with carbon free (or very low carbon content) resources (e.g. biomass, gas mixed with hydrogen).

Import-export flows of electricity are also driven by economic considerations in the internal market, for which simulations were carried out separately for every scenario. This allows for trade between countries and therefore for optimal use of the interconnections and generation capacities across countries, taking into consideration the limits of the interconnector capacities, which have been adapted according to the challenges posed by the different scenarios. . The simulation thus allows for a better cost optimisation of the power generation system across the EU Member States in the context of stable grid operations at European level at any time.

It emerges clearly from table 36 that decarbonisation would involve greater electricity trade among Member States, which is most pronounced in the case that decarbonisation focuses overwhelmingly on RES.

65.

Table 36: Grid stability related indicators


Power Reserve Margin (%)|| Volume of electricity trade (TWh)

Ratio of dispatchable nominal capacities with RES contributing with (small) capacity credits divided by total peak demand (EU net imports not included)|| Sum of all export and import flows of electricity as simulated by the model (lower than in reality)

|||| 2050

Reference| 24,| 16,| 17,|| Reference| 212,| 217,| 222,3

Scenario 1bis| 26,| 19,| 22,|| Scenario 1bis| 255,| 307,| 322,8

Scenario 29,| 24,| 27,|| Scenario 303,| 450,| 618,9

Scenario 25,| 21,| 23,|| Scenario 326,| 476,| 623,6

Scenario 25,| 21,| 32,|| Scenario 304,| 602,| 1040,9

Scenario 25,| 21,| 25,|| Scenario 322,| 489,| 648,6

Scenario 25,| 20,| 26,|| Scenario 317,| 482,| 599,1

|||||||||

Contribution of electricity storage (%)|| Volume of electricity trade as % of gross final electricity demand

Extraction of electricity from storage systems as percentage of gross final demand of electricity|| Sum of all export and import flows of electricity as simulated by the model (lower than in reality) as percentage of gross final electricity demand

|||| 2050

Reference| 1,| 1,| 1,|| Reference| 6,| 5,| 4,9

Scenario 1bis| 1,| 1,| 1,|| Scenario 1bis| 7,| 8,| 7,4

Scenario 1,| 1,| 1,|| Scenario 9,| 13,| 15,4

Scenario 1,| 1,| 1,|| Scenario 9,| 13,| 13,6

Scenario 1,| 1,| 6,|| Scenario 8,| 17,| 24,3

Scenario 1,| 1,| 1,|| Scenario 9,| 13,| 14,3

Scenario 1,| 1,| 1,|| Scenario 9,| 13,| 13,4

|||||||||

Share of decentralised power generation (%)|| Investment in electricity grids (bn EUR'08)

Share of generation by small scale power plants which are connected to low voltage and medium voltage grid over total net power generation|| Investment expenditure on electricity networks over the indicated time period

|||| 2006-| 2021-| 2031-2050

Reference| 6,| 9,| 10,|| Reference| 389,| 308,| 649,0

Scenario 1bis| 6,| 10,| 13,|| Scenario 1bis| 387,| 291,| 773,6

Scenario 7,| 13,| 21,|| Scenario 405,| 352,| 860,5

Scenario 7,| 13,| 20,|| Scenario 436,| 416,| 958,9

Scenario 7,| 17,| 31,|| Scenario 434,| 535,| 1323,5

Scenario 7,| 13,| 21,|| Scenario 436,| 420,| 960,9

Scenario 7,| 14,| 24,|| Scenario 438,| 424,| 1029,0

66.

2.6 Policy related indicators


Emissions and ETS prices

All decarbonisation scenarios achieve 80% GHG reduction and close to 85% energy related CO2 reductions (83.4-84.4%) in 2050 compared to 1990 as well as equal cumulative emissions over the projection period. In 2030, energy-related CO2 emissions are between 38-41% lower, and total GHG emissions reductions are lower by 40-42%. In 2040, energy related CO2 emissions are 63-66% below their 1990 level, while total GHG emission fall by 61-63%.

Power generation would be almost completely decarbonised with CO2 emissions in 2050 plummeting 96-99% compared with 1990. CO2 emission reductions by 2050 are particularly high (minus 86-88%) also in the services/agriculture sector as well as in households (minus 85-87%). Energy related CO2 emissions in industry fall 77-79% below their 1990 level. Transport CO2 emission are 60-62% lower in 2050 compared with 1990.

The ETS price rises moderately from current level until 2030 and significantly in the last two decades providing support to all low carbon technologies and energy efficiency. Concrete policy measures such as those pushing energy efficiency and/or those enabling penetration of renewables depress demand for ETS allowances which subsequently lead to lower carbon prices. Carbon prices are the lowest in the Energy Efficiency scenario where energy demand is the lowest followed by High RES scenario (second lowest in 2030 and 2040) and the Diversified supply technology scenario (second lowest in 2050). Delay in penetration of technologies (CCS) or unavailability of one decarbonisation option (nuclear) put an upwards pressure on demand for allowances and ETS prices.

Table 37: ETS prices in €'08/t CO2

|| 2050

Reference scenario| 50

Current Policy Initiatives| 51

Energy Efficiency| 234

Diversifies supply technologies| 265

High RES| 285

Delayed CCS| 270

Low nuclear| 310

The same carbon value as in the ETS applies also to non-ETS sectors after 2020 assuring cost-efficient emissions abatement in the whole economy.

67.

CCS storage needs


Making use of the CCS option will require considerable storage capacities for CO2 over time. The Reference scenario developments, including a more optimistic picture on CCS demonstration and availability of storage sites, would require storage capacity for the cumulative CO2 emissions captured up to 2050 of 8 billion tonnes of CO2.

In the CPI scenario, CCS penetration is more moderate leading to storage requirements of 3 bn t CO2 up to 2050. The lowest storage needs come about under high RES scenario, in which case the additional storage requirements over CPI amount to 0.5 bn t CO2. The highest storage needs comes in the Low nuclear scenario leading to considerable CCS penetration, which requires almost 13 bn t CO2 storage capacity up to 2050. Also the Diversified Supply Technology scenario would require considerable storage capacity.

68.

Table 38: CCS storage needs for power generation and industrial processes up to 2050 (in bn t CO2)


RES targets and biomass

The Reference scenario assumes that the RES target is reached in 2020. The RES share (as % of gross final energy consumption according to the definition of the RES directive) is slightly higher in all decarbonisation scenario in 2020 (21%), rises to at least 28% in 2030 and 55% in 2050. In the High RES scenario this share is at 31% in 2030 and 75% in 2050.

The share of renewables in power generation is even higher and stands at 86% in 2050 in the High RES scenario. The share in consumption is even higher, since with much more variable supply and demand some electricity produced needs to be stored and losses linked to such storage processes lead to lower consumption compared to production, i.e. reducing significantly the denominator of such a share. When calculating the RES-E share in line with the calculation of the overall RES share in gross final energy consumption, i.e. excluding energy losses linked to pump storage and hydrogen storage of electricity, the RES share in electricity consumption amounts to 97% in 2050 in the High RES case.

The share of renewables in transport (target of 10% for 2020 in the RES directive) is 1 percentage point higher in all decarbonisation scenarios in 2020; it rises to 19%-20% in 2030 and to 62%-73% in 2050. The share of renewables in transport in the High RES scenario is 20% in 2030 and even 73% in 2050. The increase between 2030 and 2050 as well as the difference to the Reference scenario and Current Policy Initiatives scenario of almost 50 percentage points in 2050 for the decarbonisation scenarios is remarkable and shows the importance of RES based decarbonisation of transport, either directly via biofuels or indirectly via RES based electricity. Decarbonisation efforts and RES share in transport are rather moderate till 2030 but rise significantly from 2030 to 2050.

The large share of RES in the scenarios is driven by a strong support for RES in the form of an implicit facilitation of RES in the scenarios. These lead to shifts in RES potential curves in the decarbonisation scenarios, allowing for more RES exploitation at a given deployment cost level, compared to the Reference scenario. This includes facilitation policies such as:

- For biomass: agricultural policies stimulating the production of energy crops, increased residue collection, and/or increased yield of crops;

- For wind: regarding on-shore it comprises the availability of more land area and a facilitation of the licensing requirements; for off-shore it also represents a facilitation of licensing and the development of technologies that allow placing off-shore power plants in deeper areas or further offshore; and

- For small scale solar PV and wind: development of smart grids and other facilitation policies.

The total use of biomass in the various scenarios is shown in table 39. Whereas the Reference and CPI scenarios have about 100 Mtoe more biomass use in 2050 compared with today's level, there is around 70-80 Mtoe additional biomass use in most decarbonisation scenarios in 2050, except for the high RES case, in which the additional biomass use amounts to around 120 Mtoe.

69.

Table 39: Use of biomass and biofuels


ktoe| Reference scenario| Current policy Initiatives

| 2050

Total domestic biomass| 188914

of which biofuels| 38912

Biofuels in bunkers| 2325

Total use of biomass| 191239

||| Energy efficiency| Diversified supply technologies

||| 2050

Total domestic biomass|| 253209

of which biofuels|| 71047

Biofuels in bunkers|| 17995

Total use of biomass|| 271204

||| High RES| Delayed CCS

||| 2050

Total domestic biomass|| 252893

of which biofuels|| 69370

Biofuels in bunkers|| 17523

Total use of biomass|| 270415

||| Low nuclear||

||||

Total domestic biomass|||

of which biofuels|||

Biofuels in bunkers|||

Total use of biomass|||

Biofuel consumption rises by a factor of more than ten between 2005 and 2050 under current policies to reach 37-39 Mtoe in 2050. Decarbonisation of transport requires substantially greater biofuels use, which increases to 68-72 Mtoe in 2050, with the highest levels being reached in the High RES and Diversified Supply Technology scenarios.

70.

2.7 Overall system costs, competitiveness and other socio-economic impacts


This section deals with the costs for providing the energy services to the EU economy and society. One key element of such costs is the external fuel bill, i.e. the amount of money that they EU economy has to pay to the outside world for procuring all the net imports of oil, gas and solid fuels from the rest of the world.

The external fuel bill arising from the net imports of fossil fuels decreases below 2005 levels in all decarbonisation scenarios by 2050. This result stems from the pursuit of this major decarbonisation as a part of a global effort with industrial countries as a group reducing GHG emissions by 80%. In such a global setting, fossil fuel import prices will be much lower (see part on assumptions) and actual imports of fossil fuel will be much lower, too. These both effects reduce the expenditure for each of the fossil fuels and thereby the total external fuel bill of the EU. The decrease of the fuel bill in the decarbonisation scenarios is smallest in the Low Nuclear scenario at 31% and highest in the high RES scenario with 43% with RES replacing most fossil fuels.

Compared with current level, all decarbonisation scenarios increase the external fuel bill in 2030, but to much lower levels than the Reference and Current Policy Initiative scenarios. While the external fuel bill would double between 2005 and 2030 under Reference and Current Policy Initiatives developments, this increase would be limited to around 40% under these decarbonisation policies.

Table 40: External fossil fuel bill (in bn € (08))

|| Reference| CPI

|| 2050

Bn. EUR'| 269.| 549.| 752.| 531.| 704.2

Diff. to| 104%| 180%| 98%| 162%

||| Energy Efficiency| Diversified supply technologies

||| 2050

Bn. EUR'|| 364.| 165.| 379.| 180.1

Diff. to| 35%| -38%| 41%| -33%

||| High RES| Delayed CCS

||| 2050

Bn. EUR'|| 374.| 154.| 377.| 180.4

Diff. to| 39%| -43%| 40%| -33%

||| Low nuclear||

||||

Bn. EUR'|| 382.| 186.||

Diff. to| 42%| -31%||

Savings in the external fuel bill are most striking in 2050. Compared with Current Policy Initiatives, the EU economy could save in 2050 between 518 and 550 bn € (08) by going this strong decarbonisation route under global climate action. The largest energy bill savings come about in the high RES scenario. Such fuel bill savings have strong impacts on overall energy system costs.

Total costs for the entire energy system include capital costs (for energy installations such as power plants and energy infrastructure, energy using equipment, appliances and vehicles), fuel and electricity costs and direct efficiency investment costs (house insulation, control systems, energy management, etc), the latter being also expenditures of capital nature. Capital costs are expressed in annuity payments. Total costs exclude disutility and auction payments.

Auction payments are expenditures for individual actors/sectors that are not costs for the economy as a whole, since the auctioning revenues are recycled back to the economy. Disutility costs are a concept that captures losses in utility from adaptations of individuals to policy impulses or other influences through changing behaviour and energy consumption patterns that might bring them on a lower level in their utility function. Such disutility costs correspond to a monetary estimation (income compensating variation) of lower utility from useful energy services (lighting, heating, mobility, etc.) resulting from a more rational use behaviour by consumers who for example adjusts thermostats, switch lighting off or travel less in order to adapt to higher costs of useful energy services. Such costs monetisation captures relevant issues regarding new consumption patterns especially for a short to medium time horizon, but becomes more challenging and uncertain in the long term, given that monetisation requires the comparison with a counterfactual development assuming unchanged tastes, habits and values over up to 40 years.[12]

71.

Table 41: Energy system costs


Average annual energy system costs 2011-|||

Bn. EUR'| Ref| CPI| High Energy effic.| Div. supply techn.| High RES| Delayed CCS| Low nuclear

Capital cost| 1104

Energy purchases| 1311

Direct efficiency inv. costs *| 161

Total system cost excl. all auction payments and disutility **| 2552

||||||||

Absolute Difference to Reference||||||

Bn. EUR'||| High Energy effic.| Div. supply techn.| High RES| Delayed CCS| Low nuclear

Δ Capital cost||| 149

Δ Energy purchases||| -| -| -| -| -312

Δ Direct efficiency inv. costs *||| 133

Δ Total system cost excl. all auction payments and disutility **| -| -| -29

||||||||

Absolute Difference to CPI|||||||

Bn. EUR'||| High Energy effic.| Div. supply techn.| High RES| Delayed CCS| Low nuclear

Δ Capital cost||| 109

Δ Energy purchases||| -| -| -| -| -300

Δ Direct efficiency inv. costs *||| 125

Δ Total system cost excl. all auction payments and disutility **| -| -| -| -| -67

||||||||

Percentage change to Reference||||||

%||| High Energy effic.| Div. supply techn.| High RES| Delayed CCS| Low nuclear

Capital cost||| 16,| 15,| 14,| 14,| 15,6

Energy purchases||| -24,| -20,| -16,| -20,| -19,2

Direct efficiency inv. costs *||| 937,| 462,| 475,| 466,| 465,5

Total system cost excl. all auction payments and disutility **| 1,| -1,| 0,| -2,| -1,1

||||||||

Percentage change to CPI|||||||

%||| High Energy effic.| Div. supply techn.| High RES| Delayed CCS| Low nuclear

Capital cost||| 12,| 10,| 9,| 10,| 10,9

Energy purchases||| -24,| -19,| -15,| -19,| -18,6

Direct efficiency inv. costs *||| 729,| 349,| 359,| 353,| 352,2

Total system cost excl. all auction payments and disutility **| -0,| -3,| -1,| -3,| -2,5

* Include costs for insulation, double/triple glazing and for efficiency enhancing changes in production processes not accounted for under energy capital and fuel/electricity purchase costs;

** These macroeconomic costs do not include ETS auctioning payments that represent a cost from the individual economic actors point of view, but do not present a cost to society given that auctioning revenues are recycled back to the economy (societal perspective); auctioning payments are partly included in energy purchase costs (e.g. in electricity prices) and partly paid directly by actors subject to ETS; total costs in table 41 differ from the sum of the items shown; table 42 on additional information below gives more detail

Table 42: Additional information on auctioning payments, disutility and total costs from the individual economic actor's point of view (bn € (08) per year on average in 2011-2050)

Bn. EUR'| Ref| CPI| High Energy effic.| Div. supply techn.| High RES| Delayed CCS| Low nuclear

Auctioning payments| 30

Total energy system cost (a)| 2583

Disutility costs (b)| 190

Total energy system costs including auction payments and disutility (c)| 2772

(a) From the individual economic actors' point of view, including direct and indirect (via purchase of e.g. electricity) auctioning costs, but excluding disutility costs;

(b) Disutility costs are costs stemming from behavioural change, such as changing lighting quality, lowering thermostat temperature, replacing fuel consuming mobility with other types of mobility (e.g. bikes) or telecommunication that are not accounted for by expenditure flows in the model, but change the level of utility of consumers; such changes are linked to carbon values in non-ETS (which do not represent a cost in cash terms), but are a proxy for policy measures bringing about such behavioural change; direct costs of such change in terms of investment and fuel bills are accounted for in the normal modelling procedure; given the long time horizon and possibly changing preference, the estimation of disutility costs is surrounded with uncertainty.

(c) From the individual economic actors' point of view, including direct and indirect (via purchase of e.g. electricity) auctioning costs as well as disutility costs;

NB: The lower system cost (without auctioning revenue and disutility) in the Delayed CCS scenario compared with the Diversified supply technologies scenario (that is unrestricted regarding technology) is not present when auctioning revenues and disutility costs are included, i.e. the point of view of the economic actors is taken (numbers denoted with (c) above). In this case, the Diversified Supply Technology scenario has the lowest costs. The modelling approach simulates the system from the point of view of economic actors, who perceive auctioning payments and disutility as cost to them that they want to minimise. Disutility costs are however surrounded with uncertainty given the long time horizon and their dependence on preferences and values. Moreover they represent a monetary equivalent in terms of imputed income compensation of changes in utility and are not associated with payments represented in the process of modelling (e.g. energy purchases, investment sums). Given the uncertain and somewhat controversial nature of disutility costs for a 40 year time horizon this long term assessment of economic impacts reports on costs without disutility. Furthermore, taking a macro-economic perspective auctioning revenues can be seen as transfers as they are supposed to be recycled, justifying their exclusion from the macro-economic cost evaluation.

The average additional energy system cost per year from 2011 to 2050 compared with the Reference and Current Policy Initiatives scenario are rather small due to the pursuit of this major decarbonisation as a part of a global effort. Given that the Current Policy Initiatives scenario is the most up to date current trend scenario and that all decarbonisation scenarios base themselves on this updated baseline, the following comparison starts from the CPI scenario (1bis).

The Delayed CCS scenarios and the Diversified Supply Technologies have the lowest level of average annual energy system costs, representing even a cost saving compared with CPI (of 94 bn €(08) and 84 bn €(08), respectively) given the large fossil fuel import cost savings discussed above. These are scenarios, in which there is a rather high nuclear penetration in addition to substantial RES penetration and strong energy efficiency progress. Given these fossil fuel import bill effects, also the Low Nuclear Scenario would produce average annual fuel bill savings of 67 bn € (08) when compared with CPI. The High RES scenario gives rise to a annual energy system cost saving of 29 bn € (08) when compared with CPI, while the annual cost savings for the Energy Efficiency scenario amount to 4 bn €(08).

The cost savings in the Energy Efficiency scenario are smaller (4 bn €) given that very high energy efficiency progress requires strong action on the building stock entailing major expenditure for accelerated building renovation, in addition to costs for other energy efficient equipment including the costly transition to electric and plug in hybrid vehicles. High renovation rates are one of the salient features of the energy efficiency scenario. Electro-mobility also provides for greater energy efficiency in the system. However, this higher cost does not disqualify energy efficiency policies as such, as strong energy efficiency policies leading to substantial improvements and energy savings, are present in all scenarios. The Energy efficiency scenario just shows that there are certain limits from where on other decarbonisation routes are less costly than further reductions of energy consumption.

All scenarios show higher annual costs in the last two decades 2031-2050 reflecting mainly increased investments in transport equipment as the major transition to electric and plug in hybrids vehicles is projected after 2030. In High RES scenario costs are also linked to significant expansion of RES based power generation capacity.

Cumulative auction payments are lowest in Energy efficiency scenario due to the reduced energy consumption, decreasing emissions and therefore the necessity to buy ETS permits. The scenario with the highest auction revenues is Delayed CCS where the delay in the use of CCS leads to high carbon prices in the long-term to ensure the achievement of the decarbonisation target via the uptake of this technology in these later years. The PRIMES model works with perfect foresight in the supply side module, therefore the high carbon prices are expected, influencing choices already in previous years. The auction revenues represent an equivalent of around 1% of total cumulative energy system costs.

When relating the cumulative costs to the GDP (which remains constant in these scenarios) the ratio of costs to GDP is similar across the scenarios (around 14.1% to 14.6%) exhibiting costs at the low end of the range in case of diversified supply technologies and delayed CCS.

72.

Table 43: Energy system costs (without auction payments and disutility) related to GDP


|| Cumulative costs as percentage of GDP (*)

Reference| 14.37%

CPI| 14.58%

Energy Efficiency| 14.56%

Diversified Supply Technology| 14.11%

High RES| 14.42%

Delayed CCS| 14.06%

Low Nuclear| 14.21%

73.

Change in cost structure: fixed costs versus variable costs


The composition of energy costs changes over time and varies across scenarios. The share of fixed cost (capital costs including for e.g. insulation) rises in all scenarios. Following larger capital expenditure for e.g. power generation, grids, energy efficiency investment over time energy, the progress in energy efficiency, greater use of technologies with low operating costs (most RES) and lower world fossil fuel prices in the decarbonisation scenarios bring lower fuel and emission allowances costs. Consequently, the share of capital costs increases over time, especially in the Energy Efficiency and High RES scenarios, which have the highest fixed cost shares (see table 44).

Table 44: Share of fixed costs* in total energy costs** (averages over the time periods indicated)

74.

Energy related costs for companies


Energy related costs in relation to sectoral value added rise from 5.8% in 2005 to 7.8% in 2030 in the Reference/CPI cases and to around 7.5% in the decarbonisation scenarios. In 2050, under current policies, this indicator declines to 7.5% and even more so in the decarbonisation scenarios falling to under 7%. Long term energy costs relative to value added of companies under decarbonisation are lower in the decarbonisation cases than under current policies thanks to substantial global decarbonisation efforts. Whereas relative costs for stationary use (heating, process energy, appliances, lighting, etc) in the decarbonisation scenarios remain at the current level by 2030, there is a strong increase in costs related to value added for transport services. After 2030, both stationary and transport energy costs decline somewhat when related to value added. Overall, energy costs relative to value added in 2050 are only somewhat higher than they were in 2005 under decarbonisation, whereas there would be a much more pronounced increase of such costs in the absence of such decarbonisation under significant global climate action.

75.

Table 15: Energy related costs of companies


%| Reference| Scenario 1 bis

| 2050

Ratio of energy related costs to value added| 5.| 7.| 7.| 7.| 7.5

of which stationary uses| 4.| 4.| 4.| 4.| 4.3

of which transportation uses| 1.| 3.| 2.| 3.| 3.1

||| Scenario Scenario 3

||| 2050

Ratio of energy related costs to value added|| 7.| 6.| 7.| 6.4

of which stationary uses|| 4.| 3.| 4.| 3.8

of which transportation uses|| 3.| 2.| 3.| 2.6

||| Scenario Scenario 5

||| 2050

Ratio of energy related costs to value added|| 7.| 6.| 7.| 6.3

of which stationary uses|| 4.| 4.| 4.| 3.8

of which transportation uses|| 3.| 2.| 3.| 2.5

||| Scenario|

||||

Ratio of energy related costs to value added|| 7.| 6.||

of which stationary uses|| 4.| 3.||

of which transportation uses|| 3.| 2.||

Energy intensive industries face particularly high energy costs for their highly energy consuming production processes. Five industrial sectors (iron and steel, non-ferrous metals, non metallic mineral products, chemicals, paper and pulp industries) have such high energy costs and are therefore particularly concerned by potential changes from decarbonisation in the energy component of their costs. Table 46 shows for these energy intensive industries combined the ratio of energy related costs for production processes and other stationary use, on the one hand, and their value added, on the other.

76.

Table46: Ratio of energy related costs to value added for energy intensive industries


|| 2050

Reference| 33.7%| 40.8%| 40.5%

CPI|| 39.4%| 39.5%

Energy Efficiency|| 35.6%| 30.6%

Diversified Supply Technologies|| 36.4%| 32.4%

High RES|| 36.1%| 34.8%

Delayed CCS|| 36.5%| 33.2%

Low Nuclear|| 37.1%| 33.5%

Energy costs of energy intensive industries relative to value added would increase under Reference and CPI developments. This development stems also from rising world fossil fuel prices under current trends. It is worth noting that under global climate action bringing with it lower energy import prices and due to substantial energy efficiency progress, the ratio of energy costs to value added in energy intensive industries would decline in all decarbonisation scenarios – most markedly in the Energy Efficiency scenario

77.

Effects of fragmented climate action: competitiveness and energy consequences of safeguards for energy intensive industries


This Energy Roadmap has assumed the implementation of the European Council's decarbonisation objective that includes similar efforts by industrialised countries as a group. The analysis presented focuses on energy consequences. A more comprehensive analysis of different global paths to decarbonisation was presented in the Low Carbon Economy Roadmap 2050 i exploring impacts of three global climate situations: a) business as usual; b) global climate action and c) fragmented action. Fragmented action assumes strong EU climate action that is however followed globally only by the low end of Copenhagen pledges up to 2020 and afterwards the ambition level of the pledges is assumed to stay constant. It analyses impacts on energy intensive industries (EII) both in a global macroeconomic modelling framework to address carbon leakage issues and by means of energy system modelling to address effects of fragmented action, including electricity costs for companies. Electricity costs are, in fact, higher in the fragmented action scenarios as compared to global action scenarios due to higher energy import prices. On the other hand, carbon prices are lower under fragmented action.

A 'fragmented' action scenario including measures against carbon leakage was not analysed in this IA report as the challenges for the energy sector arising from decarbonisation are the biggest under 'global climate action' assumption, given that fragmented action with measures against carbon leakage will deliver lower GHG reductions by 2050. Decarbonisation scenarios that accommodate action against carbon leakage under fragmented action would either go for lower ambitions in terms of GHG reduction or would have measures included that imply such lower efforts for energy intensive industries and consequently for the total energy system[14]. With action on carbon leakage the challenge for the transition in the energy system would be smaller given lower efforts in parts of the system. Such results are however modified through countervailing effects from lower world fossil fuel prices under global action that encourage somewhat higher energy consumption and emissions. In any case, the implementation of measures will be crucial. The real difference for industrial and thereby climate policy might come from the concrete design of policy instruments that is not discussed in this Energy Roadmap Impact Assessment (e.g. special provisions on ETS for EII).

From the analysis undertaken for the Low Carbon Economy Roadmap it can be concluded that under fragmented action with the EU reducing emissions much more than other regions, certain industries supplying low carbon technologies would benefit from improved competitiveness due to higher internal demand and first mover advantages. However, EII would suffer from higher costs for allowances and/or significant mitigation costs in order to avoid the need to purchase such allowances. Furthermore, under fragmented action they would not benefit from the fuel and electricity price reductions stemming from a global climate deal that lowers world fossil fuel prices.

This situation of fragmented action might require countervailing action to combat carbon leakage, which was investigated in the Low Carbon Economy Roadmap, notably by exploring a scenario, in which energy intensive industries (iron and steel, non-ferrous metals, chemicals, non metallic minerals, paper and pulp industries) would benefit from the same ETS prices that prevail in the reference scenario, whereas other sectors would be exposed to higher carbon costs. These provisions have only a limited impact on the CO2 emission reduction of all sectors, which instead of reaching minus 86% on 1990 under fragmented action (85% under global action) would amount to 78% with these specific provisions for EII. Clearly, the CO2 emission reduction for EII, i.e. their level of effort, would be reduced more markedly, falling from 87% reduction below 1990 under fragmented action (88% under global action) to only 51% by 2050.

These measures keeping the ETS price for energy intensive industries at the reference case levels lead to significant cost savings for purchasing fuel, electricity, steam and energy using equipment. Compared with the reference case situation with no additional climate action, the average costs in 2011-2050 decrease by 6 bn € (08) annually over 40 years. Higher energy, especially electricity prices from decarbonisation action together with the still significant carbon price signal lead to significant energy savings in energy intensive industries (22.7% in 2050 from Reference).

These cost savings take into account that electricity prices rise significantly under fragmented action (7% in 2050 compared with Reference) and this to a higher degree than under global action given that the cost reducing effect through lower fossil fuel input prices (global action reducing world fossil fuel demand) would not materialise. Electricity prices in 2050 would be 6 % lower on average under global climate action compared with fragmented action with specific measures for EII.

Under global action, the energy saving effect of energy intensive industries is reinforced through higher carbon prices, entailing even greater energy savings. Combined with lower fossil fuel import and therefore final consumer prices, there would be additional cost savings, amounting to 21 bn € per year from 2011 to 2050 when comparing global climate action with fragmented action with less effort for EII.[15]

Table 47 compares the energy related results of decarbonisation scenario under fragmented action with specific carbon leakage measures for energy intensive industries with the Reference case. It includes also a comparison between global action and fragmented action with these specific measures for energy intensive industries. The energy results for this analysis are taken from the energy modelling results for the Low Carbon Economy Roadmap, which includes, in addition to the Reference scenario, the Fragmented action, effective technology and less effort for EII scenario and the Effective Technology Global Action scenario.

The effective technology scenarios are driven by carbon prices and assume the absence of significant obstacles for technology penetration, especially CCS and nuclear, as well as the absence of specific strong push for RES and energy efficiency. The rationale of these scenarios is similar to the Diversified Supply Technologies scenario, which includes however recent policy initiatives, especially on energy efficiency and energy taxation as well as recent changes in nuclear policies. The most relevant comparison of energy results when dealing with carbon leakage in the case of Fragmented action, effective technology and less effort for EII is therefore in relation to Reference (no additional climate action), on the one hand, and Effective Technology under global climate action, on the other.

Table 47: Comparison of energy results for 2050* between fragmented action with specific measures for energy intensive industries (FAEII) and Reference as well as between global action and FAEII

|| Less effort for EII compared with Reference| Global action compared with less effort for EII

Final energy consumption EII Other sectors Primary energy consumption Gross electricity generation ** Average electricity prices Energy related CO2 emissions Import dependency RES share in gross final energy demand Cumulative investment expenditure in power generation Average annual fuel, electricity and equipment costs| -22.7% -33.6% -24.1% +10.3% +7.2% -72.4% -26.2 pp +26.6 pp +30.1% -6 bn| -11.2% +1.6% +2.5% +6.6% -5.7% +1.2% +1.3 pp -0.3 pp +1.2% -21 bn

* For investment expenditure and costs this comparison relates to the 40 year period up to 2050

** including new uses, such as hydrogen as a means for electricity storage and for feeding into the gas grid thereby contributing to decarbonisation by lowering the carbon content of the gas supplied

Climate action with specific measures for EII against carbon leakages leads to quite significant energy consequences in 2050 compared to reference regarding energy consumption, fuel and electricity costs, prices and emissions. Import dependency would fall strongly, whereas the RES share would rise to a large extent. Investment in power generation would also need to rise strongly while average costs would fall significantly.

Energy consumption of EII would drop further significantly when undertaking decarbonisation in the context of global action, as EII would face higher carbon prices in this case (the same as other sectors). The small increase of energy consumption and emission levels (outside EII) when moving to global action stem from the markedly lower fossil fuel prices under globally reduced demand. Energy related results are either reinforced, if the policy response to climate change moved from fragmented action with specific carbon leakage measures for EII to global action without such measures, or they are modified reflecting the impacts from lower fossil fuel prices.

78.

Energy related expenditures of households


Affordability of energy services as regards fuel and electricity costs but also equipment (insulation, more efficient appliances, etc) is one of the essential elements of the analysis. The sector that is mostly concerned is households. All decarbonisation scenarios show significant fuel savings compared to the Reference and CPI scenarios but also higher costs for energy appliances, boilers and insulation. Energy related expenditures for heating and cooling of households as well as for lighting and appliances almost double from around 2000 EUR'08/year today to 3800 to 3900 EUR'08 in 2050 in the Reference and CPI scenarios reflecting rising fuel and electricity prices and increasing direct household investments in energy efficiency. Expenditures per household amount to some 4500 EUR'08 in most decarbonisation scenarios in 2050, with expenditure per household reaching some 4800 € (08) and almost 4900 € (08) in the Energy Efficiency and high RES scenarios respectively.

It is important to note that per capita income in 2050 will also almost double from today's level, but also that households will be composed of fewer members reflecting aging and changing lifestyles. Energy costs per household exceed the Reference/CPI case level by 16-17% in 2050 in most decarbonisation scenarios. They are 25-27% higher in the Energy Efficiency and High RES scenarios, as these scenarios are particularly intensive in investment. While these costs might be affordable by an average household, vulnerable consumers might need specific support to cope with increased expenditures due to decarbonisation.

Households spend money on transport services, too. Such costs concern expenses on tickets for rail, bus, metro, air and other travel as well as costs for purchasing privately owned vehicles and paying for other fuel and operational expenses. These transport costs per household would even almost triple by 2050 reaching 3900 € (08) and 4100 € (08) in the Reference and CPI case, respectively. The strong growth of such costs reflects rising oil prices as well as changes in the vehicle fleet towards more efficient cars (hybrids, plug in hybrids, electric cars) that involve higher costs[16]. In the decarbonisation scenarios, transport related energy costs per households are lower in 2050 than under Reference or CPI developments, markedly so (broadly around 10%) under Diversified Supply Technologies and delayed CCS, given substantial improvements in energy efficiency in transport and limited price increases with respect to reference for transport fuel.

Relating the costs of households for stationary energy use (heating, appliances, etc) plus those for transport to household expenditure gives the following picture. The share of energy in household expenditure rises over time in all scenarios from 10% in 2005 to around 16% in 2030, decreasing thereafter to around 15-16% by 2050. Among the decarbonisation scenarios, the Delayed CCS and the Diversified Supply Technology scenarios have costs at the lower end of this range, whereas the High RES and Energy efficiency scenarios show 2050 costs at the upper end of the range.

79.

Table 48: Energy related expenditures of household for stationary use and transport


%| Reference| Scenario 1 bis

| 2050

Share of energy related costs in household expenditure| 9.| 15.| 14.| 16.| 15.1

of which stationary uses| 5.| 8.| 7.| 7.| 7.3

of which transportation uses| 4.| 7.| 7.| 8.| 7.8

||| Scenario Scenario 3

||| 2050

Share of energy related costs in household expenditure|| 16.| 16.| 15.| 15.4

of which stationary uses|| 7.| 9.| 7.| 8.4

of which transportation uses|| 8.| 7.| 8.| 6.9

||| Scenario Scenario 5

||| 2050

Share of energy related costs in household expenditure|| 15.| 16.| 15.| 15.1

of which stationary uses|| 7.| 9.| 7.| 8.5

of which transportation uses|| 8.| 7.| 8.| 6.6

||| Scenario|

||||

Share of energy related costs in household expenditure|| 16.| 15.||

of which stationary uses|| 7.| 8.||

of which transportation uses|| 8.| 7.||

Whereas companies enjoy long term energy costs relative to value added that are lower (or at most as high) as such costs under current policy initiatives, the 2050 energy costs of households relative to household expenditure generally exceed such costs without strong decarbonisation albeit only to a rather small extent, especially under Delayed CCS and Diversified Supply Technologies.

80.

Electricity prices


Another important indicator on costs relates to final consumer prices especially the prices of electricity for industrial, household and services consumers as well as the average price. Electricity prices are calculated in such a way that total costs of power generation, balancing, transmission and distribution are recovered, ensuring that investments can be financed. Table 49 shows the average price for electricity in the EU27 for different sectors; the residential sector has the highest user price and industry the lowest as it is currently the case. In 2050, average electricity costs are highest in High RES scenario reaching 199 €/MWh. The lowest electricity prices are in Diversified supply and Energy efficiency scenario, with prices below the Reference and Current Policy Initiatives scenarios because of cheaper procurement of fossil fuels under global climate action.

Average prices of electricity are rising compared to current levels until 2030 and continue increasing in the High RES scenario. In the Energy Efficiency and Diversified Supply Technology scenarios, electricity prices remain similar to those in the Reference/CPI scenario up to 2030 thanks to lower fossil fuel input costs with lower world market prices. With somewhat higher investment or ETS costs, the other decarbonisation scenarios have slightly higher costs in 2030, exceeding Reference/CPI by around 5%. By 2050, the average price exceeds reference/CPI level markedly in the High RES scenario (around 30%) to recover costs for the high generation capacity needs including for back-up and for greater grid and storage capacities, while it remains almost at that level in the Low nuclear case (+4%). In the Diversified Supply Technologies and Energy Efficiency scenarios, electricity prices in 2050 are even below those in the Reference/CPI cases, whereas beneficial effects from lower import prices are compensated by effects from restricted choices on nuclear or delayed penetration of CCS in the respective scenarios. Electricity prices are already slightly higher than reference in Current policy Initiatives scenario reflecting less nuclear in power generation at somewhat higher costs.

It should also be noted that prices rise strongly up to 2020/30, but that after 2030 prices either fall or show an average annual price increase that is much smaller than in the period 2005-2030, which applies in particular for the High RES scenario.

Table 49: EU27 average electricity prices [17]

81.

Diesel prices


Another pertinent indicator on costs across scenarios is the price of diesel, which is relevant for both passenger transport (in private cars and buses/coaches) and freight transport.

Prices for diesel in transport in CPI and the decarbonisation scenarios reflect the new energy taxation directive as well as different bio-diesel blends. The energy system changes between scenarios cause only limited changes to end-user diesel prices. The strong decline in diesel prices between CPI and decarbonisation scenarios in 2030 reflects oil and product import price savings. This effect is compensated in 2050 by the impact of a significantly higher biofuel penetration in the diesel market.

Table 50: Average EU27 diesel (including blended biodiesel) end –user prices for private transport[18]

||| 2050

Reference| (EUR(08)/toe)| 2250

CPI| % diff. to Reference| 0%| 20%| 16%

Energy Efficiency| 0%| 3%| 17%

Diversified Supply Technology| 0%| 3%| 19%

High RES| 0%| 3%| 21%

Delayed CCS| 0%| 2%| 18%

Low Nuclear| 0%| 3%| 22%

82.

2.8 Conclusions


The Commission services conducted a model-based analysis of decarbonisation scenarios exploring energy consequences of the European Council's objective to reach 80% GHG reductions by 2050 (as compared to 1990), provided that industrialised countries as a group undertake similar efforts. These scenarios explore also the energy security and competitiveness dimension of such energy developments. Businesses as usual projections show only half the GHG emission reductions needed; increased import dependency, in particular for gas; and rising electricity prices and energy costs. Several decarbonisation scenarios highlighting the implications of pursuing each of the four main decarbonisation routes for the energy sector – energy efficiency, renewables, nuclear and CCS - were examined by modelling a high and low end for each of them. The model relies on a series of input assumptions and internal mechanisms to provide the outputs.

83.

The most relevant assumptions and mechanisms of the model


Ø All scenarios were conducted under the hypothesis that the whole world is acting on climate change which leads to lower demand for fossil fuel prices and subsequently lower prices.

Ø The model assumes perfect foresight regarding, policy thrust, energy prices and technology developments which assures a very low level of uncertainty for investors, enabling them to make particular cost-effective investment choices without stranded investments. There is also no problem with uncertainty on whether all the infrastructure and other interrelated investment (e.g. grid connections) needed to make a particular investment work will be in place in time.

Ø Regulatory framework in model allows for investments to be built and costs fully recovered.

Ø The model assumes a 'representative' or average household or consumer while in reality there is a more diversified picture of investors and consumers.

Ø The model assumes continuous improvements of technologies.

The model-based analysis has shown that decarbonisation of the energy sector is feasible; that it can be achieved through various combinations of energy efficiency, renewables, nuclear and CCS contributions; and that the costs are affordable. The aim of the analysis was not to pick preferred options, a choice that would be surrounded with great uncertainty, but to show some prototype of pathways to decarbonise the energy system while improving energy security and competitiveness and identify common features from scenario analysis.

84.

Common elements to scenario analysis


Ø There is a need for an integrated approach, e.g.; decarbonisation of heating and transport relies heavily on the availability of decarbonised electricity supply, which in turn depends on very low carbon investments in generation capacity as well as significant grid expansions and smartening.

Ø Electricity (given its high efficiency and emission free nature at use) makes major inroads in decarbonisation scenarios reaching a 36-39% share in 2050 (almost doubling from current level and becoming the most important final energy source). Decarbonisation in 2050 will require a virtually carbon free electricity sector in the EU, and around 60% CO2 reduction by 2030.

Ø Significant energy efficiency improvements happen in all decarbonisation scenarios. One unit of GDP in 2050 requires around 70% less energy input compared with 2005. The average annual improvement in energy intensity amounts to around 2.5% pa.

Ø The share of renewables rises substantially in all scenarios, achieving at least 55% in gross final energy consumption in 2050, up 45 percentage points from the current level (a high RES case explores the consequences of raising this share to 75%).

85.

Ø The increased use of renewable energy as well as energy efficiency improvements require modern, reliable and smart infrastructure including electrical storage.


Ø Nuclear has a significant role in decarbonisation in Member States where it is accepted, especially if CCS deployment were delayed.

Ø CCS contributes significantly towards decarbonisation in most scenarios with a particularly strong role in case there were problems with nuclear investment and deployment. Developing CCS can be also seen as an insurance against energy efficiency, RES and nuclear (in some Member States) delivering less or not that quickly.

Ø All scenarios show a transition from high fuel/operational expenditures to high capital expenditure.

Ø Substantial changes in the period up to 2030 will be crucial for a cost-efficient long term transition to a decarbonised world[19]. Economic costs are manageable if action starts early so that the restructuring of the energy system goes in parallel with investment cycles thereby avoiding stranded investment as well as costly lock-ins of medium carbon intensive technology.

Ø The costs of such deep decarbonisation are low in all scenarios given lower fuel procurement costs with cost savings shown mainly in scenarios relying on all four main decarbonisation options.

Ø Costs are unequally distributed across sectors, with households shouldering the greatest cost increase due to higher costs of direct energy efficiency expenditures in appliances, vehicles and insulation.

Ø The external EU energy bill for importing oil, gas and coal will be substantially lower under decarbonisation due to substantial reduction in import quantities and prices dependent on global climate action lowering world fossil fuel demand substantially.

When considering these scenario results it might be useful to consider as well that energy supply structures are being transformed. Today we have, for the most part, concentrated rather invisible items, such as mines, import terminals, large power plants outside towns, and underground pipelines for energy dense fossil fuels and nuclear energy. Under decarbonisation we would increasingly have well visible land consuming configurations, such as very large numbers of wind turbines, solar devices, biomass plantation, and additional transmission lines. This might raise issues with public acceptance and local opposition.

Deployment of nuclear technologies is fraught with acceptance problems in a large number of Member States. CCS is already now experiencing local opposition in some Member States. Temporary delays in CCS were modelled but not the complete unavailability of this option. Permanent unavailability of CCS could mean that decarbonisation would almost entirely hinge upon very strong progress with RES penetration (and energy efficiency) given the existing limitations to nuclear with many Member States having opted out. In the high RES scenario with much energy efficiency (discussed above), the CCS role is very small, given the predominance of RES, requiring in turn large efforts in terms of financing and finding accepted sites for very substantial investments in production and transmission.

Some policy relevant conclusions can be drawn based both on the results of the scenario analysis as well as on a comparison of the hypothetical situation of ideal market and technological conditions needed for modelling purposes and what is found in the much more complex reality.

86.

Implications for future policy making


Ø Successful decarbonisation while preserving competitiveness of the EU economy is possible. Without global climate action, carbon leakage might be an issue and appropriate instruments could be needed to preserve the competitiveness of energy intensive industries.

Ø Predictability and stability of policy and regulatory framework creates a favourable environment for low carbon investments. While the regulatory framework to 2020 is mainly given, discussions about policies for 2020-2030 should start now leading to firm decisions that provide certainty for long-term low-carbon investments. Uncertainty can lead to a sub-optimal situation where only investment with low initial capital costs is realised.

Ø A well functioning internal market is necessary to encourage investment where it is most cost-effective. However, the process of decarbonisation brings new challenges in the context, for example, of. electricity price determination in power exchanges: deep decarbonisation increases substantially the bids based on zero marginal costs leading in many instances to prices rather close to zero, not allowing cost recovery in power generation. Similarly, the necessary expansion and innovation of grids for decarbonisation may be hampered if regulated transmission and distribution focuses on costs minimisation alone. Building of adequate infrastructure needs to be assured and supported either by adequate regulation and/or public funding (e.g. financed by auctioning revenues).

Ø Energy efficiency tends to show better results in a model than in reality. Energy efficiency improvements are often hampered by split incentives, cash problems of some group of customers; imperfect knowledge and foresight leading to lock-in of some outdated technologies, etc. There is thus a strong need for targeted support policies and public funding supporting more energy efficient consumer choices.

Ø Strong support should be given to R&D in order to bring down costs of low-carbon technologies.

Ø Due attention should be given to public acceptance of all low carbon technologies and infrastructure as well willingness of consumers to undertake implied changes and bear higher costs. This will require the engagement of both the public and private sectors early in the process.

Ø Social policies might need to be considered early in the process given that households shoulder large parts of the costs. While these costs might be affordable by an average household, vulnerable consumers might need specific support to cope with increased expenditures. In addition, transition to a decarbonised economy may involve shifts to more highly skilled jobs, with a possibly difficult adaptation period.

Ø Flexibility. The future is uncertain and nobody can predict it. That is why preserving flexibility is important for a cost efficient approach, but certain decisions are needed already at this stage in order to start the process that needs innovation and investment, for which investors require a reasonable degree of certainty from reduced policy and regulatory risk.

Ø External dimension, in particular relations with energy suppliers, should be dealt with pro-actively and at an early stage given the implications of global decarbonisation on fossil fuel export revenues and the necessary production and energy transport investments during the transition phase to decarbonisation; new areas for co-operation could include renewable energy supplies and technology development.

87.

Attachments


Attachment 1: Numerical results

88.

1. Reference scenario


Reference scenario with Low energy import prices

Reference scenario with High energy import prices

Reference scenario with High GDP

Reference scenario with Low GDP

89.

1bis. Current Policy Initiatives scenario


90.

2. Energy Efficiency scenario


91.

3. Diversified supply technologies scenario


92.

4. High RES scenario


93.

5. Delayed CCS scenario


94.

6. Low nuclear scenario


Attachment 2: Assumptions about interconnections and modelling of electricity trade

Short description of the model

The electricity trade model of PRIMES covers all countries in the European continent except countries of the CIS and Turkey. Interconnector capacities at the various country borders are determined exogenously.

The model performs unit commitment, endogenous use of interconnectors (with given capacities and Net Transfer Capacities (NTC)) and also optimal power generation capacity expansion planning in a perfect foresight manner until 2050. Simulations of different electricity demand levels with the model allow identification of bottlenecks and of the amount of investment in interconnectors necessary to remove such bottlenecks.

The model covers demand both for electricity and CHP steam/heat, as given from results of the entire PRIMES model. Demand for electricity and for steam/heat is supposed to be given and is represented through two typical days (for winter and summer).


Investment in new power plants is endogenous. The rate of use of power capacities and interconnectors is endogenous. Regarding the use of interconnectors the model performs a linear Direct Current optimal power flow under oriented NTC constraints defined per each couple of countries. The model makes distinction between AC lines and DC lines, the use of the latter being controlled by operators. All interconnectors existing today or planned to be constructed in the future are represented (one by one) in the model.


Among the inputs, the model considers non linear cost-supply curves for fuels used in power generation and non linear investment cost curves for nuclear and renewable energy power plants, which are a function of total installed capacity (unit investment costs increase as approaching the potential).

The electricity model, used in stage 1, is identical to the model used in the entire PRIMES model, but could be used with endogenous electricity trade only for the work during stage 1 because of very long computing times for each model run when iterations are performed between demand and supply and for meeting carbon targets.

95.

Assumptions for the modelling exercise


All data about NTCs and interconnection capacities were taken from ENTSOe databases. Information on new constructions was taken from the latest “Ten-year network development plan 2010-2020”, complemented, where necessary, with information from the Nordic Pool TSOs and the Energy Community (for South East Europe). Some of the planned new constructions would justify increase of NTCs values until 2020, as mentioned in the ENTSOe’s TYNDP document. Other mentioned new constructions regard directly the building of new interconnection lines which are introduced as such in the model database.

According to assumptions agreed with the Energy DG of the European Commission, the following three cases were formulated regarding the NTC values:

a) NTC-0: keeping the NTC values of 2020, which are much higher than today, unchanged until 2050; the projection of NTC values to the year 2020 from today levels follows a study by KEMA, except few cases either because the links were not included in that study or because ENTSOe’s NTC values announced for 2010-2011 were exceeding the KEMA’s values. This assumption does not use the TYNDP information about new constructions aiming at increasing the NTC values in the future, except indirectly if in some cases the KEMA values for 2020 increase from today’s levels.

b) NTC-2: apply a doubling of 2020 NTC values between 2020 and 2050 and interpolate linearly between 2020 and 2050; increase capacities of interconnectors where necessary so as to keep NTC values lower than total interconnection capacity by individual couples of countries. Some additional DC lines were added (linking Italy with western Balkans).

c) NTC-4: apply a quadrupling of 2020 NTC values by 2050 and interpolate with extension of interconnection capacities where needed.

Two energy demand and pricing contexts were considered to analyze the implications from the above mentioned NTC assumptions, which are as follows:

1. Reference scenario: demand, prices, taxes and ETS carbon prices are taken as identical to the DG ENER Reference scenario. Some adjustments on electricity demand figures were made only for year 2010, based on monthly statistics for 2010, in order to be able to simulate the true NTC values for this year.

2. Decarbonisation scenario: demand, prices and ETS carbon prices, as well as the parameters mirroring RES facilitation and other policies, are taken from the DG CLIMA “Decarbonisation scenario under effective technologies and global climate action” scenario.

96.

Discussion of model results for the Reference scenario with three NTC value cases


The model results show that the NTC values retained for the year 2020 do not lead to substantial changes compared to results for the standard Reference scenario, i.e. the Reference scenario referred to in the Low Carbon Economy Roadmap). The countries projected to be net exporters in the standard reference scenario remain so in the model results presented here; the same applies to countries projected to be net importers in the standard reference scenario. There are differences in the magnitude of exports or imports for the year 2020, as for example for Belgium, Portugal, Lithuania and Latvia (higher net imports), for Hungary and Denmark (lower net imports) and for Slovenia, Slovakia, Sweden and Bulgaria (more net exports). It is reminded that for the standard Reference scenario import-exports of electricity were derived following a different methodology, which applied common balancing by region, contrasting the pan-European balancing applied for the model runs presented here.

NTC-0 case.

Regarding the scenario with NTC values remaining unchanged at the year 2020, the model results provide information about congestion by considering whether the NTC constraints are binding or close to be binding for couples of countries. The findings from this analysis regarding the projected NTC values for 2020 are summarized below

· Link Switzerland-Germany: appears congested and NTC is 32% of capacity

· Link Germany-Poland: appears congested and NTC is 17% of capacity

· Link Denmark-Sweden: appears congested and NTC is 54% of capacity

· Link Austria-Italy: appears very congested and NTC is 16% of capacity

· Link Italy-Slovenia: appears congested and NTC is 15% of capacity

· Link Austria-Hungary: appears congested and NTC is 31% of capacity

· Link Slovenia-Croatia: appears congested and NTC is 18% of capacity

· Links in the Balkans (FYROM-Greece, Albania-Greece, Bulgaria-Greece, Serbia-FYROM, Romania-Serbia, Serbia-Albania, Bulgaria-FYROM) appear very congested and NTC are below 30% of capacity, except Greece-Bulgaria NTC which is 68% of capacity

Congestion is detected in the model runs due to NTCs that are only a small part of existing capacities. One option for dealing with congestion would be to increase NTC without necessarily construct new lines. From the above overview it can be seen that the congestions after 2020 remain between Germany and neighbours to the east and south, between Austria, Italy, Slovenia, and Hungary, and finally in the Balkans, both within the Balkans and the linkages with northern neighbours.

97.

NTC-2 and NTC-4 cases


The NTC-2 and NTC-4 cases assume doubling and quadrupling of NTC values, respectively from 2020 to2050, with linear interpolation applied between 2020 and 2050. The model results show that this way of uniformly increasing the NTC values does not really solve the problem of systematic congestions mentioned above for the case NTC-0. These congestions are removed only in the case NTC-4 and after 2030, with the exception of the Austria-Italy and Germany-Poland links, which remain congested until 2050 despite the quadrupled NTCs. The congestion problems in the Balkans are removed only in the NTC-4 case after 2030, but the area remains strongly congested under the NTC-2 assumptions. The congestions in links with Germany (Switzerland, Poland, Czech Rep. and Austria) are not removed in the NTC-2 case.

The doubling and quadrupling of NTCs values do not provide any advantages concerning the large list of links, which are not found congested under the NTC-0 assumptions.

The doubling of NTCs under the assumptions of NTC-2 case lead to lower rates of use of interconnection capacities (reported as percentage of NTCs), compared with NTC-0 results in the following cases:

· UK-Ireland: 17 percentage points less use

· France, Belgium, Netherlands, Luxembourg, Germany: between 15 and 30 percentage points less use

· Nordic area: around 15 percentage points less use

· Czech Rep., Slovakia, Poland, Hungary, Romania, Croatia: between 20 and 30 percentage points less use

· Latvia-Estonia: 20 percentage points less use

Passing from the doubling to the quadrupling implies even lower rates of use of interconnection possibilities.

Both cases NTC-2 and NTC-4 have adverse implications on the rate of use of DC lines leading to lower rates of use compared to case NTC-0, which under NTC-4 are close to zero in some cases. The NTC constraints help using the DC links for which the NTC values are usually equal to the interconnection capacities. Excessively high NTC constraints, which also mean more AC links, imply much less use of DC lines, which of course is unrealistic, as the DC lines correspond to today known constructions and are furthermore expensive. So the companies would not build so many new AC lines as the ones corresponding to NTC-4 on economic grounds including the adverse effects on DC lines.

A major issue with NTC-2 and NTC-4 cases regards the investment cost implicitly associated with the increase of interconnection capacities stemming from the doubling and quadrupling of NTC values. Total interconnection capacity is projected to increase by 43% in 2020 compared to 2010 levels, as a result of implementing the construction program of the TYNDP. In NTC-0 the capacity remains roughly unchanged until 2050. But in NTC-2 the capacity has to increase by 95% in 2050 compared to 2020 levels and in NTC-4 this increase is 277%. Such a construction program exceeds by far capacity requirements and would unnecessarily penalize costs and electricity prices in the scenarios.

According to the model results, we obtain the following changes in energy terms from NTC-2 and NTC-4 assumptions compared to NTC-0 results:

· Total volume of electricity traded increases by 5% in NTC-2 and by 8% in NTC-4 compared to NTC-0 in cumulative terms for the period 2015-2050. It is evident that the additional cost of interconnectors cannot be justified by such small increases in total traded volumes (i.e. adding absolute values of flows between countries).

· Total electricity production costs decrease by 0.13% in NTC-2 and by 0.23% in NTC-4 compared to NTC-0 in cumulative terms 2015-2050

· CO2 emissions from electricity production decrease by 0.8% in NTC-2 and by 0.9% in NTC-4 compared to NTC-0 in cumulative terms 2015-2050

· Nuclear and RES cumulative production are found slightly higher in NTC-2 and NTC-4 compared to NTC-0, but the changes are less than 1% in cumulative terms.

It can therefore be concluded that the NTC expansion according to the NTC-2 and NTC-4 assumptions are not needed for the functioning of the electricity system and would entail high unnecessary cost without providing any noticeable benefit. These assumptions do not solve the serious congestion issues, do not provide gains for the non congested areas and have adverse effects on the economics of DC lines.

The conclusion for a Reference or Current Policy Initiatives framework is therefore to follow an approach that focuses on identified bottlenecks. For stage 2 of the modelling it is appropriate to increase NTC values and interconnection capacities after 2020 in a selective way, with priority to areas that would be congested in the future according to the reference scenario results. Such areas are the southern and eastern connections of Germany, the area linking Italy, Austria and Slovenia, the linkages of Balkans with northern neighbours and the linkages within Balkans. Some NTC additions should be also made for the linkages Denmark-Sweden and Latvia-Estonia.

With lower electricity demand due to the assumed strong energy efficiency policies, these results also hold for the Current Policy Initiatives scenario.

98.

Discussion of model results for a Decarbonisation scenario with three NTC value cases


Under the assumptions of the decarbonisation scenario, total demand for electricity (in the 32 countries included in the model) increases by 15% in 2050 compared to the Reference scenario for year 2050. ,

It is assumed that the renewable facilitation policies develop in all countries in favour of domestic renewable potential. The scenario does not assume inflows of RES electricity from outside EU countries (e.g. North Africa) and does not include the possibility of exploitation of offshore wind located at long distance from the coasts.

The results from the model show that the NTC values retained for the period until 2020 do not alter the electricity trade pattern projected in previous decarbonisation exercises and compared to the Reference scenario.

The congestions identified in the context of the decarbonisation scenario for the year 2020 are the same as in the context of the reference scenario (see previous section).

Under the assumptions of the NTC-0 case the results show congestions similar to those found for the reference scenario, i.e. in south and east of Germany, in the Balkans, in the northern connections of the Balkans, in the linkages between Italy, Austria and Slovenia. Some additional congestion cases, found in the context of decarbonisation, relate to the link Germany-Sweden, Norway-UK and Germany-UK which are based on DC-links and do not concern the NTC values.

The doubling of NTC values under the assumptions of the NTC-2 case does not help removing the congestions. The quadrupling of NTC values (NTC-4 case) helps removing the congestions only in the long term, after 2040. So the linear interpolation method seems not to be useful as it brings little benefits and entails high costs for building new interconnectors. Increase of NTC values in a selective way and at an early stage after 2020 seems more suitable.

In the context of the decarbonisation scenario, the NTC-2 case allows increase of total volumes traded by 12% when compared to NTC-0. The increase obtained for the NTC-4 case is 14% (up from NTC-0). NTC-2 reduces total power generation costs roughly by 0.85% in cumulative terms compared to NTC-0. In NTC-4, the additional effect on power generation costs is smaller, NTC-4 power generation costs are 0.2%) lower compared with NTC-2. It is important to note that these statements related to power generation costs, and that the move from NTC-0 to NTC-2 and even more NTC-4 involves large costs for grid investment. NTC-2 has small impacts favouring slightly more nuclear and RES generation, whereas NTC-4 add very little to NTC-2 effects.

99.

Overall conclusions on decarbonisation scenarios (except for those with very strong reliance on RES)


Following these economic modelling results, the approach for further modelling was chosen to start fromNTC-0 assumptions and to increase in selective way NTC values immediately after 2020 for the linkages found to be congested. This concerns interconnections around Germany, in Austria-Italy-Slovenia, Balkans and Denmark-Sweden.

For very high RES penetration, such linkages may not be sufficient. Therefore, this case has been examined separately. The results of this analysis are reported in the following chapter.

100.

Assumptions about interconnections in the Decarbonisation scenario with High RES deployment both domestically and in the North Sea


Under the assumptions of this decarbonisation scenario, full exploitation of off-shore wind potential at North Sea is foreseen. In this modelling, exploiting the highest possible offshore wind potential is envisaged for Denmark, the UK, France, Germany, Netherlands, Sweden, Norway, Belgium and Ireland, according to the division of the sea in economic zones. Data on potentials come from published reports (e.g. EEA); the additional potentials, compared to standard RES scenario, are remarkably high for Norway, UK and Netherlands. It is assumed that a dense DC interconnection system will develop mainly offshore but also partly onshore, to facilitate power flows from the North Sea offshore wind parks.

After several model runs with different DC topology configurations and after considering elimination of congestions arising from wind offshore power flows, we have concluded to the following assumptions about the additional DC interconnections:

In MW| Investment in additional new interconnectors in the 4.1 scenario – North Sea

||| Total

Ireland| UK| 1000

Spain| France| 2000

France| Germany| 2000

France| Belgium| 1000

Belgium| Netherlands| 2000

Netherlands| Germany| 2500

UK| France| 2500

UK| Belgium| 1500

UK| Netherlands| 1000

UK| Germany| 3000

Norway| Belgium| 5000

Norway| Netherlands| 3000

Norway| Germany| 5000

Germany| Denmark| 4500

Norway| Denmark| 1000

UK| Norway| 2000

Norway| Sweden| 1000

Sweden| Poland| 10000

Netherlands| Denmark| 2500

Denmark| Sweden| 4000

Germany| Poland| 5000

Denmark| Poland| 6000

|| Total| 67500

The NTC values are identical to the DC capacities, as assumed for all DC lines.

The congestions in this scenario are related to the wheeling of electricity from the North Sea region to consumption centres. The links of Sweden with Poland, Sweden with Lithuania, Austria with Italy, France with Italy and links in the Balkan region appear to be congested.

In this scenario, the electricity trade changes drastically. The United Kingdom, Netherlands, Denmark, Sweden, Norway export large amount of electricity while France, Belgium Germany, Italy, Czech Republic, Slovakia, Poland become or remain importing countries. This changes the results for the decarbonisation scenario as regards several countries.

101.

Attachment 3: Short description of the models used


The scenarios were derived with the PRIMES model by a consortium led by the National Technical University of Athens (E3MLab), supported by some more specialised models (e.g. GEM-E3 model that has been used for projections for the value added by branch of activity and PROMETHEUS model that has been deployed for projections of world energy prices).

102.

GEM-E3


The GEM-E3 (World and Europe) model is an applied general equilibrium model, simultaneously representing World regions and European countries, linked through endogenous bilateral trade flows and environmental flows. The European model is including the EU countries, the Accession Countries and Switzerland. The world model version includes 18 regions among which a grouping of European Union states. GEM-E3 aims at covering the interactions between the economy, the energy system and the environment. It is a comprehensive model of the economy, the productive sectors, consumption, price formation of commodities, labour and capital, investment and dynamic growth. The model is dynamic, recursive over time, driven by accumulation of capital and equipment. Technology progress is explicitly represented in the production function, either exogenous or endogenous, depending on R&D expenditure by private and public sector and taking into account spillovers effects. The current GEM-E3 version has been updated to the GTAP7 database (base year 2004) and has been updated with the latest Eurostat statistics for the EU Member States. PRIMES model

The PRIMES model simulates the response of energy consumers and the energy supply systems to different pathways of economic development and exogenous constraints and drivers. It is a modelling system that simulates a market equilibrium solution in the European Union and its member states. The model determines the equilibrium by finding the prices of each energy form such that the quantity producers find best to supply match the quantity consumers wish to use. The equilibrium is forwarding looking and includes dynamic relationships for capital accumulation and technology vintages. The model is behavioural formulating agents’ decisions according to microeconomic theory, but it also represents in an explicit and detailed way the available energy demand and supply technologies and pollution abatement technologies. The system reflects considerations about market competition economics, industry structure, energy /environmental policies and regulation. These are conceived so as to influence market behaviour of energy system agents. The modular structure of PRIMES reflects a distribution of decision making among agents that decide individually about their supply, demand, combined supply and demand, and prices. Then the market integrating part of PRIMES simulates market clearing.

PRIMES is a partial equilibrium model simulating the entire energy system both in demand and in supply; it contains a mixed representations of bottom-up and top-down elements. The PRIMES model covers the 27 EU Member States as well as candidate and neighbour states (Norway, Switzerland, Turkey, South East Europe). The timeframe of the model is 2000 to 2050 by five-year periods; the years up to 2005 are calibrated to Eurostat data. The level of detail of the model is large as it contains:

· 12 industrial sectors, subdivided into 26 sub-sectors using energy in 12 generic processes (e.g. air compression, furnaces)

· 5 tertiary sectors, using energy in 6 processes (e.g. air conditioning, office equipment)

· 4 dwelling types using energy in 5 processes (e.g. water heating, cooking) and 12 types of electrical durable goods (e.g. refrigerator, washing machine, television)

· 4 transport modes, 10 transport means (e.g. cars, buses, motorcycles, trucks, airplanes) and 10 vehicle technologies (e.g. internal combustion engine, hybrid cars)

· 14 fossil fuel types, new fuel carriers (hydrogen, biofuels) 10 renewable energy types

· Main Supply System: power and steam generation with 150 power and steam technologies and 240 grid interconnections

· Other sub-systems: refineries, gas supply, biomass supply, hydrogen supply, primary energy production

· 7 types of emissions from energy processing (e.g. SO2, NOx, PM)

· CO2 emissions from industrial processes

· GHG emissions and abatement (using IIASA’s marginal abatement cost curves for non CO2 GHGs).

103.

For further information see


www.e3mlab.ntua.gr/e3mlab%3Amanual-for-primes-model&catid=35%3Aprimes&Itemid=80&lang=en Prometheus model

A fully stochastic World energy model used for assessing uncertainties and risks associated with the main energy aggregates including uncertainties associated with economic growth and resource endowment as well as the impact of policy actions (R&D on specific technologies, taxes, standards, subsidies and other supports). The model projects endogenously to the future the world energy prices, supply, demand and emissions for 10 World regions. World fossil fuel price trajectories are used for the EU modelling as import price assumptions for PRIMES.

Annex 2 - Energy Roadmap 2050 – Selected Stakeholders' Scenarios

104.

1. Introduction


105.

2. Scanning of Stakeholder Scenarios


106.

3. Comparative Analysis of Scenario Studies


107.

3.1 Policy Assumptions and Targets


108.

3.2 Economic Assumptions


109.

3.3 Assumptions on Social Issues


110.

3.4 Further Technology Assumptions


111.

3.5 Key Results of Scenarios


112.

3.6 Models Used and Interdependencies Between Studies


113.

4. Summary of comparison


References

114.

1. Introduction


Stakeholders are continuing their work on scenarios for long term transformation of energy systems. These analyses, using a variety of models and assumptions and exploring a variety of constraints, all help in assessing the robustness of conclusions on policy actions needed in the coming years.

The bulk of this report, chapters 2-4, is a systematic presentation of a representative sample of European long term energy scenarios. Their policy targets, assumptions on various economic, social and technological factors, and resulting outcomes of model-based analyses are compared. The purpose is not to judge the outcomes of the scenarios but to try to understand and clearly describe the similarities and differences in the scenarios[20]. This work was completed in April 2011.


Since then, several scenarios this year explore consequences of the Fukushima accident and unconventional gas. In the IEA[21]'s global scenario to 2035 entitled The Golden Age of Gas, ample availability of gas, much of it unconventional, keeps average gas prices well below levels assumed in WEO-2010. Especially in growing economies in China and other non-OECD countries, gas consumption increases throughout the energy system, driven by price, improved access to supplies, efficiency improvements in technologies, also emissions benefits. Its flexibility is a distinct benefit in a perspective of much change in energy systems and much uncertainty about how drivers will play out. In Europe, scenario analyses by the European Gas Advocacy Forum[22] and Eurogas[23] underline this flexibility and how it can be used. EGAF argues that with greater use of gas in the short to medium term, to 2030 or so, implementation risks in the early years of a long-term strategy focused on renewables[24] can be reduced as well as overall costs. Eurogas similarly argues that the balance which will emerge between renewables and CCS/fossil fuels cannot be known today and that investing in gas keeps these long-term options open. The importance of CCS in these strategies in the long term is evident in the IEA scenario which does not assume availability of CCS by 2035. In this scenario, the long-term trajectory for CO2 emissions is towards 650ppm, thus a probable temperature rise well above the 2 degrees C target.

The European Climate Foundation in this year’s phase of its Roadmap 2050 work, concentrates on trade-offs in the period till 2030, exploring coherent policy actions needed to keep the European energy system on track to 2050. With further analyses of its 60% renewables and high renewables scenarios for the power sector[25], trade-offs among additional grid infrastructures, generation capacities and their location, storage and demand side management are examined. Additional grid investments beyond 2020, although substantial, are low compared to generation investments. If these grid investments are not made, the result is an increase in back-up and operational costs amounting to far more than the grid investments saved. Demand response, within day, reduces the need for additional transmission infrastructure. The deployment of renewables in order of productiveness across Europe reduces cumulative generation capital costs by over a fifth by 2030 compared to a Member State by Member State approach. ECF also examines price setting in regional markets and utilisation rates of additional back-up plant, crucial for understanding market design issues.

Greenpeace concentrated on grids in its 2011 scenario analyses[26], building on its earlier Renewables 24/7 study. Looking beyond 2030, a High Grid scenario encompassing much trade and North African solar resources and a Low Grid scenario with more local solutions within Europe are explored. With adequate transmission, both would imply shrinking utilisation rates for coal and nuclear plants and later for gas fired plants, which could then be converted to biogas.

Scenarios for sustained transformation of the energy system are now being developed by a whole range of organisations, at local, Member State and European level[27]. Many look explicitly at the European market and policy context[28].

The conclusions of these scenario analyses and the analyses by the Commission are consistent on many but not all issues. All agree on the importance of energy efficiency in any strategy. The increased reliance on capital investments in the transformation of the energy system and in energy efficiency improvements is evident in all scenarios, raising financing, risk management and cost of capital issues to the top of the agenda. All see a much stronger reliance on renewables than currently, which raises issues notably for the power system. Flexibility from all sources is increasingly important. Grid investments and the market developments that go with them look like a no-regrets policy, at least in the period to 2030. Areas of difference among scenarios often concern timing. They include the degree of early reliance on electrification as opposed to direct use of, notably, gas, in heating, transport and industry. Estimates of total system costs in scenarios are still very different. They are not easy to compare.

115.

2. Scanning of Stakeholder Scenarios


A variety of international organisations, industry associations, individual companies, NGOs and research/academic institutions have put forward mid- and long-term energy scenarios. In order to make a representative sample, 28 studies were identified by screening contributions and publications from stakeholders.

A representative set of 7 studies was selected (see Table 1). The criteria used were time horizon until at least 2030, geographical coverage of EU-27 (or Europe[29]), public availability of main results in a quantitative form, coverage of at least the electricity sector, level of detail, and the scenarios being well known and discussed internationally. For example, studies covering only the world as a whole without defining Europe as a region were not selected. The time horizon, geographical and sectorial coverages, as well as the level of detail, vary greatly among the scanned studies.

116.

Table 1: Scanning of Energy Scenario Studies


The 7 studies selected to be compared in detail are, as follows (see full references at the end of this report):

· European Commission Reference Scenario to 2050, published in 2011, [1]:

o "The 2050 Reference scenario depicts energy and greenhouse gas (GHG) emission developments on the basis of policies implemented up to March 2010, mirroring as well the achievement of the legally binding 2020 targets on renewables (RES) and GHG and the implementation of the ETS Directive. It shows the magnitude of the additional effort needed for EU policies to achieve the European Council's GHG mitigation objective."

· European Climate Foundation (ECF) – Roadmap 2050, 2010, i:

o "The objectives of the Roadmap 2050 are: a) to investigate the technical and economic feasibility of achieving at least an 80% reduction in greenhouse gas emissions below 1990 levels by 2050, while maintaining or improving today's levels of electricity supply reliability, energy security, economic growth and prosperity; and b) to derive the implications for the European energy system over the next 5 to 10 years."

· Greenpeace/EREC – Energy [R]evolution (+EREC (2010), Re-thinking 2050), 2010, i:

o "The report demonstrates how the world can get from where we are now, to where we need to be in terms of phasing out fossil fuels, cutting CO2 while ensuring energy security. This includes illustrating how the world's carbon emissions from the energy and transport sectors alone can peak by 2015 and be cut by over 80% by 2050."

· International Energy Agency (IEA) – Energy Technologies Perspectives (ETP), 2010, [4]:

o 'The goal of the analysis in this book is to provide an IEA perspective on the potential for energy technologies to contribute to deep emission reduction targets and the associated costs and benefits. It uses a techno-economic approach to identify the role of both current and new technologies in reducing CO2 emissions and improving energy security.'

· IEA – World Energy Outlook (WEO), 2009, [5]:

o 'The results of the analysis presented here aim to provide policy makers, investors and energy consumers alike with a rigorous, quantitative framework for assessing likely future trends in energy markets and the cost-effectiveness of new policies to tackle climate change, energy insecurity and other pressing energy-related policy challenges.' (Reference scenario);

o 'More specifically, this report is intended to inform the climate negotiations by providing an analytical basis for the adoption and implementation of commitments and plans to reduce greenhouse-gas emissions.' (450 Scenario).

· Eurelectric – Power Choices, 2009, [6]:

o 'The Eurelectric Power Choices study was set up to examine how the vision, of cutting Greenhouse Gas (GHG) emissions by 75% in 2050, can be made reality. Power Choices looks into the technological developments that will be needed in the coming decades and examines some of the policy options that will have to be put in place within the EU to attain a deep cut in carbon emissions by mid-century.'

· FEEM i et al., EU-RTD Project PLANETS: Probabilistic Long-term Assessment of New Energy Technology Scenarios, 2010, [7]:

o 'PLANETS is a research project funded by the EC under the 7th Framework Programme with the scope of devising robust scenarios for the evolution of energy technologies in the next 50 years. The project aims to assess the impact of technology development and deployment at world and European levels, by means of an ensemble of analytical tools designed to foresee the best technological hedging policy in response to future environmental and energy policies.'

117.

3. Comparative Analysis of Scenario Studies


118.

3.1 Policy Assumptions and Targets


All scenario studies analysed use a 'baseline scenario' to show the impact of presently implemented policies (e.g. until 2009). These baseline scenarios are used as a basis for assessing impacts of alternative scenarios.

The 'alternative scenarios' all aim at reducing GHG or CO2 emissions (and are generally in line with the EU 2020 target of -20% and to the long term target of -80% to -95% by 2050).

Most models concentrate on the electricity sector and are much less detailed (or provide no details) on developments in the heating and transport sectors (except insofar as they may assume major electrification in these sectors).

Table 2 gives an overview of main pre-defined policy assumptions and targets across the scenarios (for EU-27 or OECD-Europe, depending on study) for:

· GHG or CO2 emissions reduction (economy-wide),

· Share of renewables (RES),

· Role of nuclear,

· Efficiency,

· Emission Trading System (ETS) and remarks on status of policies taken into account.

119.

Table 2: Overview of Main Policy Assumptions and Pre-Defined Targets in the Scenarios


Short name scenario| GHG or CO2-emissions reduction, economy-wide| Share of renewables in gross final energy consumption| Share of nuclear in power generation| Reduction in primary energy by improved energy efficiency| Carbon policy

WEO Ref| § GHG: -20% below 1990 levels by 2020 for EU| § 20% by 2020 for EU| §| § 20% by 2020 for EU| § Policies until mid 2009 § ETS

WEO 450 ppm| § GHG: -20% below 1990 levels by 2020 and -80% by § 20% by §| § 20% by § Policies until mid-2009 § ETS (OECD+, OME)

ETP BL OECD Europe| § GHG: -20% below 1990 levels by 2020 for EU| § 20% by 2020 for EU| §| § 20% by 2020 for EU| § Policies until mid-2009 § ETS

ETP Blue Map OECD Europe| § CO2eq: -74% below 2007 levels by 2050 § GHG: -20% below 1990 levels by 2020 for EU| § 20% by 2020 for EU| §| § 20% by 2020 for EU| § Policies until mid-2009 § ETS (OECD+, OME)

EC Reference Scenario to § GHG: -20% below 1990 levels by 2020 (in Reference scenario)| § 20% by 2020 (in the Reference scenario)| § Economic modelling with currently non nuclear MS remaining non nuclear except Poland and Italy; phase-out in 2 MS| §| § Implemented Policies until March 2010 & achievement of legally binding targets § Revised ETS Directive applied until 2050

ECF BL| § GHG: -20% below 1990 levels by 2020 for EU| § 20% by 2020 for EU| §| § 20% by 2020 for EU| § Policies until mid-2009 § ETS

ECF 80% RES| § GHG: -80% below 1990 levels by § 80% RES of power generation by § 10% nuclear of power generation by § 20% by 2020 for EU| § ETS (OECD+OME)

ECF 60% RES| § GHG: -80% below 1990 levels by § 60% RES of power generation by § 20% nuclear of power generation by § 20% by 2020 for EU| § ETS (OECD+OME)

ECF 40% RES| § GHG: -80% below 1990 levels by § 40% RES of power generation by § 30% nuclear of power generation by § 20% by 2020 for EU| § ETS (OECD+OME)

E[R] Ref| §| §| §| §| § No specific targets or policies mentioned

E[R]| § CO2: -80% below 1990 levels by §| § Phasing out| §| § No specific targets or policies mentioned

E[R] Adv| § CO2: -95% below 1990 levels by § High RES share: 'Close to fully renewable energy system' by § Phasing out| §| § No specific targets or policies mentioned

Eurelectric BL| §| §| § Germany and Belgium phased out| §| § Policies until mid-2009 § ETS

Eurelectric Power Choices| § GHG: -40% below 1990 levels by 2030 and -75% by § 20% by § Germany and Belgium phased out| § 20% by 2020 for EU| § Policies until mid-2009 § ETS (all sectors)

FEEM-WITCH| §| §| § No exogenous constraint| §| §

120.

Prognos 2011


Abbreviations used: ETS: Emissions Trading System, GHG: Greenhouse Gas, OME: Other Major Economies (Brazil, Russia, South Africa and the countries of the Middle East), MS: EU Member States.

From Table 2 it can be seen that:

In relation to reduction of GHG emissions,

· Most of the studies do not take into account negative or positive effects of climate change on the economy in the models used. One exception found are the FEEM-scenarios where the WITCH-model incorporates an integrated assessment module which is able to take into account a dynamic linkage of climate change and economic activity.

· In general, some form of European Emissions Trading (ETS) is considered in most studies (exceptions are Greenpeace/EREC and FEEM), some models used for scenarios development even have specific modules which simulate a market for emission allowances i (e.g. PRIMES used by both Commission services and Eurelectric).

· The scenarios differ in their assumptions about future emissions trading markets. There is a large consensus about the sectors included, but not about the geographical coverage. Some studies assume no extension of the current EU emissions trading, others assume an expansion of the market from OECD+ up to a global dimension. With the Clean Development Mechanism (CDM), another possibility to enlarge the geographic coverage of the allowances market exists. The EU DG ENER scenarios focus on this issue, other scenarios give little information. Finally, some scenarios envisage small deviations from the current status of the allocation process, assuming full auctioning in the power sector and grandfathering in the other sectors. Other scenarios tend towards a general full auctioning of allowances.

Carbon pricing in the different scenarios is shown in Table 3:

121.

Table 3: Comparison of Carbon Pricing in the Scenarios


Scenario| Sectoral coverage| Geographical coverage| Auctioning or grandfathering| CDM

IEA – WEO| Existing EU-ETS, including aviation| n/a| n/a| CDM taken into account

IEA – WEO 450 ppm| n/a| OECD+ in 2013, major economies as of n/a| CDM taken into account

IEA – ETP Reference| n/a| n/a| n/a| n/a

IEA – ETP Blueline| n/a| n/a| n/a| n/a

EU DG ENER Reference| Existing EU-ETS including aviation| n/a| Auctioning in power sector, grandfather­ing for other-sectors| CDM taken into account

EU DG ENER Baseline| n/a| n/a| Auctioning in power sector, grandfather­ing for other-sectors| Limited use of CDM-credits

ECF Roadmap Reference| Industry, power sector, aviation| n/a| n/a| n/a

ECF Roadmap Pathways| Industry, power sector, aviation| Until 2020 OECD-countries, from 2020 including developing countries| n/a| n/a

Energy [R]evolution| n/a| Global CO2 trading system in the long term| n/a| n/a

Energy [R]evolution Advanced| n/a| Global CO2 trading system in the long term| All allowances should be auctioned| n/a

Eurelectric Baseline| n/a| n/a| Full auctioning as of 2015 (except some new Member States)| n/a

Eurelectric Power Choices| ETS extended to all major economic sectors after International carbon market after Full auctioning as of 2015 (except some new Member States)| n/a

FEEM et al. - Planets| n/a| n/a| n/a| n/a

In relation to future energy mixes,

A few studies use pre-defined future energy mix targets, by preferring or excluding certain technologies from the beginning (in a 'back-casting approach'): Predetermined Role of Renewables (RES): Only Greenpeace/EREC and ECF make specifications on the desired shares of RES energies in 2050:

o Greenpeace/EREC sets in its advanced Energy [R]evolution scenario the 2050 RES target share at 100% (all sectors).

o ECF sets in its alternative scenarios the 2050 power sector RES target share at 40%, 60% and 80%, respectively.

Predetermined Role of Nuclear Energy (NUC) and Carbon Capture & Storage (CCS): Greenpeace/EREC and ECF specify pre-defined shares of NUC and CCS in 2050:

o Greenpeace/EREC sets in its advanced Energy [R]evolution scenario the 2050 NUC as well as CCS target shares to 0%.

o ECF focuses on the RES share. For the purposes of the analysis, particularly of infrastructure needs, it divides the remaining share equally between NUC and CCS, thus 30%, 20% and 10% for each, in the three alternative scenarios i.

The other scenarios determine the contribution of NUC and CCS on the basis of cost assumptions and optimisation rather than pre-defined policy targets.

In relation to sustainability aspects other than GHG reduction,

Economic constraints or the issue of maintaining high levels of grid stability and overall system reliability are in most cases either not considered or at least not fully quantified:

· Economic constraints, e.g.:

o Minimisation of private financial costs (investment in new generation capacities and infrastructure (an exception is e.g. ECF)),

o Minimisation of social costs, such as environmental externalities (costs of GHG avoided, other environmental pollution, land use, etc.) i.

Maintaining high levels of grid stability and overall system reliability i, e.g.:

The high relevance of this issue is due to the fact that scenarios with high shares of RES energy sources, particularly wind and solar energy, increase the need for backup capacity or other means of ensuring grid stability. Substitution of electricity for FOS fuels in buildings and transportation, results in higher electricity demand but also expanded possibilities for demand management. These challenges are addressed by all of the studies in one way or another. Several approaches can be identified in the scenarios:

o Flexible thermal power plants (NUC, FOS) for load-following operation and back-up capacity,

o Greater use of non variable RES energy (biomass, solar with storage, geothermal, hydro with pumped storage facilities),

o Transmission expansion. This approach is constrained in some of the scenarios by model limitations. In PRIMES, interconnections are exogenous. The model used in ECF's scenarios derives transmission needs,

122.

o Large-scale storage


o Smart grids and demand side management developments.

Maintaining high levels of system reliability and thus high levels of power supply security is qualitatively mentioned across most studies as a key objective and in some studies also as a key challenge.

Regarding realisation, particularly studies with ambitious GHG reduction targets implicitly assume significant progress in grid technology (ECF maintains that they use existing technologies) and social acceptance related to transmission expansion to be able to achieve their targets. However, analysis is typically not taken further i from such largely qualitative statements and it is usually concluded that financing needs to be found for the large increases in pan-European transmission and storage capacities to be able to cope with the expected large future shares of intermittent generation.

Implications for distribution networks are not addressed by most of the studies. This is particularly concerning as almost all studies emphasize at the same time the relevance of technologically advanced smart grids and smart metering, especially those confronted with ambitious emission reductions (Energy [R]evolution, ECF Pathways, ETP Blueline, Eurelectric Power Choices).

In relation to security of supply of energy resources,

· All scenarios expect reserves of natural gas to be sufficient to meet future demands. Unconventional oil reserves are expected to be deployed in some scenarios without ambitious emission reductions (e.g. ETP Reference). No indictors of security of supply are developed. Possible indicators (diversity of imports, stability of exports, reliability of supply, diversity of supply, etc.) are not developed.

123.

3.2 Economic Assumptions


Regarding general economic assumptions,

· The scenarios assume a steady increase of GDP of ~1-2% per year until 2030/2050. The recent financial crisis is taken into account in the projections of GDP.

Regarding fossil fuel prices,

· Fossil fuel prices are often exogenously determined (in PRIMES scenarios by using a separate modelling framework). ECF and Greenpeace/EREC use price developments from WEO 2009. In WEO 2009, international fossil fuel prices are based on a top-down assessment of prices which would create enough investment to meet energy demand over the projection period (global balance of supply and demand). Therefore, fossil fuel prices in WEO are endogenously determined and sensitive to scenario assumptions. ETP takes prices up to 2030 from WEO 2009 and calculates prices for the period beyond 2030 by taking into account the long-term oil supply cost curve.

· Recent studies suggest a range of ~90-120 USD/barrel until 2030 and 2050 for the oil price. Only Greenpeace/EREC considers an oil price that increases to 150 USD/barrel in 2030. Oil prices in Greenpeace/EREC and ECF are assumed to stay constant after 2030.

· Until 2030 most scenarios presume an increasing gas price. In the IEA alternative scenarios the prices of gas, as for oil, stabilise or decrease after 2030 due to weaker energy demand, while in the reference case gas prices increase in respond to increasing demand (e.g. from additional gas-fired power plants).

· Most studies agree on the idea that gas prices will keep their linkage with oil prices, i.e. the ratio of gas and oil prices remains quite constant i. Main exceptions are the Greenpeace/EREC Energy [R]evolution and – to some extent – the alternative scenario of the PLANETS-WITCH project (see Figure 1). The PLANETS alternative scenario assumes a higher increase of gas prices than oil prices, motivated by the high gas demand and relatively low oil demand.

Figure 1: Development of Gas to Oil Prices, in %

· A moderate increase of coal prices is assumed in most of the scenarios. Some differences exist in expectations of future gas-to-coal price ratios. Most of the studies (e.g. Eurelectric) expect coal prices to increase at far lower rates than gas prices. A slight decoupling can be observed in most of the scenario studies. In contrast to the other studies, the Energy [R]evolution of Greenpeace/EREC and the alternative scenario of the PLANETS-WITCH project show a stronger increase of coal than oil prices in the long run.

Regarding incentives for RES,

· Some studies (e.g. Eurelectric Power Choices) explicitly assume decreasing direct incentives for RES in the future due to assumed increasing cost-competitiveness.

Regarding CO2-certificate prices,

· Different developments for the (typically assumed) CO2-certificate prices are to some part also determined by targets set and the resulting CO2-emissions development. As shown in Figure 2, emissions in the ECF Pathways and the WEO 450 ppm show a faster decline than emissions in the Eurelectric Power Choices scenario, which seems to allow a higher degree of flexibility to reach the targets set for 2050. Furthermore, the sharp increase of CO2-certificate prices in the Eurelectric Power Choices scenario from 2030 onwards partly results from the assumed removal of mandatory RES-targets after 2020. Therefore, carbon prices gain high importance to deliver required emission reductions by 2050.

124.

Figure 2: Development of CO2-certificate prices, in EUR2008/t CO2


On the other hand, assumed geographical extension of emission trading systems (e.g. in the ECF Pathways, WEO 450 ppm and the Power Choices scenario international carbon markets are assumed not later than 2020) can be interpreted to prevent carbon prices from rising unlimited. This effect is due to more abundant and cheap opportunities for emission reduction outside the EU/OECD.

Relatively low prices for emission certificates in the Greenpeace/EREC study may be partly determined by the idea that the process of emission trading remains unclear and is not able to help RES energy expansion (and is thus not considered adequate to become an important parameter for Greenpeace/EREC in their model).

In summary, the pre-defined importance of carbon prices as an instrument in different scenario studies may also explain their different resulting price levels (e.g. in the Power Choices scenario, carbon prices are assumed to be important to reach emission targets).

Regarding investment costs,

· All scenarios confronted with high emission reduction requirements estimate a considerable increase in capital expenditure. Even baseline scenarios suggest an increase in capital expenditure in the coming years. The somewhat higher estimation in the emissions reduction scenarios is generally based on several effects: higher capital intensity of RES technologies in terms of costs per power produced and the need for higher power transmission capacity due to intermittency of most of the expected new RES (investment in power transmission capacity is roughly estimated to be 20-50 % higher in most of the alternative scenarios, compared to Baseline or Reference scenarios); higher capital intensity of new NUC and CCS investments. The scenarios also agree in the estimation of lower expenses for FOS fuels in due course due to large substitutions of RES for FOS fuels and energy efficiency improvements.

· Overall, these effects lead to somewhat different total cost results across scenarios with large methodological uncertainties, strongly influenced by different modelling mechanisms (e.g. cost-optimization vs. accounting frameworks), framework parameters (e.g. price developments; see above) and conventions for cost-estimations. Furthermore, results are often not available for the same timeframes and geographical boundaries.

· Results on future investment costs are also strongly influenced by the chosen assumptions on technological developments in energy transformation and end-user applications. A lot depends on learning rates For example, in ECF, learning rates are 5% for wind offshore/onshore, 15% for solar PV and 12% for CCS and yearly reductions in investments costs per capacity are estimated at 1% for biomass and geothermal plants, compared to 0,5% for FOS-fired plants.

· Table 4 shows compliance costs available in alternative scenarios, differentiated into total costs and grid costs or investment:

125.

Table 4: Comparison of Compliance and Grid Costs/Investment in Alternative Scenarios


Scenario| Estimated compliance costs/investment| Estimated grid costs/investment| Comments

IEA – WEO 450 ppm| § EU-27: +1600 bn USD (vs. Ref.) cumulative investment in the energy sector (incl. grid costs) till § Global: 5100 bn USD (20% lower vs. Ref.) cumulative investment till § External costs not included (except GHG) § Grid investment (Ref.): 25% transmission, 75% distribution

IEA – ETP Blue Map| § EU-27: additional cumulative invest­ment (energy sector) compensated by cumulative fuel savings: 7100 bn USD vs. 13100 bn USD till 2050 (vs. Bas.)| § Global: 12300 bn USD (incl. smart grids) cumulative grid-investment till 2050 (+50% vs. Bas.)| § Grid investment (Ref.): 30% transmission, 70% distribution § Back-up costs may be considered implicitly

EU DG ENER Ref.| § ~175 bn € p.a. (2030) capital and O&M costs in power generation (i.e. 51,0 €/MWh)| § EU-27: grid costs of 10,8 €/MWh (2030) vs. 7,4 (2010), i.e. ~165 bn € cumulative grid costs| § Distribution grid not included § Back-up costs considered implicitly

§ ECF Roadmap 80% RES| § Lower fuel costs dominate capital cost expenses: overall -80 bn € in 2020 (-205 bn € in 2030) vs. Ref.| § Cumulative additional transmission capex: 95-129 bn €, additional back-up capex: 63-99 bn € (vs. Ref.) § Cumulative additional distribution capex: 200-300 bn €| § Amount by which distribution costs are incremental to the Ref. is unclear

Energy [R]evolution Advanced| § Global: 292 bn USD add. invest­ment p.a. 2007-2030 (vs. Ref.) § 42 bn € additional investment p.a., fuel savings of 62 bn € p.a. (2007-2050, vs. Ref.)| § Costs of 209 bn € p.a. for the assumed new European 'Supergrid'| § Grid costs estimated externally, cost structure of grid costs not further specified

Eurelectric Power Choices| § Capital and O&M costs of 53,3 €/MWh in § Grid-costs rise from 7,3 to 12,6 €/MWh (2050) § Cumulative grid investment: 1.500 bn € (+35% vs. Baseline)| § No external costs besides CO2-costs § Not clear if back-up costs are considered implicitly

FEEM et al. - Planets| § Global: ~800 (2030) and 2500 (2050) bn € yearly costs (i.e. 1-2,5 % of GDP)| § n/a| § Costs are measured as consumption losses vs. the Reference scenario

· Table 4 shows that:

o A comparison of total cost results from the different scenario studies is hardly possible as the underlying assumptions on methods and data used are in most cases not presented sufficiently transparently to give a clear picture on the dependability of figures presented (see also above discussion about grid costs).

o Macroeconomic costs or benefits are not provided, so the net economic cost or benefit (e.g. including the gains or losses from competitiveness factors) are not available.

o Distribution costs are hardly ever estimated although they seem to represent the majority of necessary grid investments. This makes it doubtful that costs for infrastructure changes are realistically included in most scenarios.

Regarding electricity prices,

· Electricity prices increase in most of the studies at least in the medium term (up to 2030). Some studies with high emission reduction targets expect a decrease of electricity prices in the long term (up to 2050), mainly driven by lower consumption of FOS fuels in the power sector in combination with assumed technological improvements for RES power plants. Not all studies actually calculate electricity prices for a market environment with supply of and demand for electricity. Therefore Table 5, providing an overview on electricity prices and their main drivers, displays electricity generation costs as a proxy for electricity prices in these cases.

126.

Table 5: Comparison of Properties of Electricity Prices in the Different Scenarios


Study and scenario| Electricity price/cost developments| Main drivers

IEA – WEO Ref and 450 ppm| § No data for Europe| § No data for Europe

IEA – ETP Ref and Blue Map| § No data for Europe| § No data for Europe

EU DG ENER Reference| § 1.4% average annual rise 2000-2030, declining after § Increasing fuel prices, higher capital costs of RES, NUC and CCS, auctioning of CO2-allowances

EU DG ENER Baseline| § 1.5% average annual rise 2000-2030, declining after § Increasing fuel prices, higher capital costs of RES, NUC and CCS, auctioning of CO2-allowances

ECF Roadmap Reference| § n/a| § Carbon prices, fossil-fuel prices, technology learning rates

ECF Roadmap Pathways| § Higher levelised costs of electricity (LCOE) than in the Ref. (short term), slightly higher LCOE by § Carbon prices, fossil-fuel prices, technology learning rates

Energy [R]evolution| § Generation costs increase up to 2020, upward tendency until § Fossil fuel prices, technology improvements of RES-technologies, costs for CO2-allowances

Energy [R]evolution Advanced| § Generation costs increase up to 2030 and decrease afterwards (-34-43 % 2050 compared to the Baseline)| § Fossil fuel prices, technology improvements of RES-technologies, costs for CO2-allowances

Eurelectric Baseline| § Strong increase up to 2025, stabili­zation afterwards| § Fossil fuel prices, restructuring of the power plant fleet

Eurelectric Power Choices| § Strong increase up to 2025, slight decrease afterwards| § Fossil fuel prices, restructuring of the power plant fleet, lower fossil fuel consumption and lower demand for CO2-allowances)

FEEM et al. – Planets| § Electricity prices stay almost constant| § Restructuring of power generation

FEEM et al. – Planets Fb 3.| § Increase until 2015, stagnation 2015 to 2035, sharp increase after § Restructuring of power generation, increasing electricity demand

Key points:

· The economic performances of all energy technologies – FOS, NUC and RES – are reflected by their specific generation costs which are heavily influenced by assumed future fuel and carbon prices, and assumed technology learning rates.

· Technology-neutral studies, such as from IEA, DG ENER or Eurelectric, give high importance to the carbon price as a key driver to deploy the most competitive low-carbon technologies and leave it then to the market to develop the future energy mix.

· Comparison of total costs for developing a more sustainable EU energy system by 2050 is hardly possible due to lack of transparency in most scenarios on methodological and data assumptions.

· Most scenarios seem to lack a realistic consideration of the costs for necessary infrastructure changes. For example, although investments in the distribution grid represents the majority of necessary grid investments, in almost all scenarios only transmission costs (if at all) are considered.

· Electricity prices increase in most of the studies at least in the medium term (2030).

127.

3.3 Assumptions on Social Issues


The most important effect in the EU social structure considered in the scenarios is change in size of population. Throughout the studies, a slight increase of the EU population is expected in the medium term (immigration), with the tendency to a stabilised population in the long term. Some studies also assume a significant decrease in the size of households.

128.

However, in none of the scenarios analyzed evidence on fundamental changes in the behavioural patterns of the economic agents was found.


Some studies (e.g. PRIMES-based Eurelectric, DG TREN, FEEM) apply fixed microeconomic decisions of economic agents concerning demand for energy related products and investment in energy supply equipment. These scenarios partly take into account different levels of risk-awareness of agents (higher levels for individuals than for enterprises, reflected by high discount rates for individuals), lack of information, market barriers for new technologies and rebound-effects in energy-efficiency investments. Investment decisions are modelled under full information and perfect foresight assumptions.

Only very little information concerning trends and effects on the labour market was found in the studies. However, in some scenarios (ECF pathways, Energy [R]evolution) sectorial shifts on the labour market from traditional energy sectors (e.g. FOS fuels) to sectors linked to RES installations are expected. Magnitudes of these effects are very differently estimated, usually ignoring the related loss of employment and market leadership in more traditional sectors.

The risk of loss of global competitiveness of energy intensive industries and related deindustrialisation in Europe is usually not considered explicitly.

Issues of public acceptance regarding deployment of new power plants (large-scale RES, new NUC, low-carbon FOS), new RES-support infrastructure (pan-European grid, large storage) or new enforced consumer behaviour (smart metering) are nowhere explicitly modelled (implicitly only for NUC by assuming e.g. growth rates being much more limited than economic optimisation would suggest).

Key points:

· Only few studies explicitly model changes in the behaviour of economic agents with regard to changes in consumer behaviour or public acceptance of deployment of new power generation plants and RES-support infrastructures,

· Effects on the EU labour market and the economy as a whole (e.g. risk of deindustrialisation) as a consequence of visions of a future EU energy mix are not consistently modelled in any scenario study and are usually limited to presenting short-term positive effects of preferred technological solutions.

129.

3.4 Further Technology Assumptions


The following conclusions on technology assumptions in the different scenarios are in addition to the technology-related assumptions already evaluated and compared under Sections 3.1 (Policy Assumptions and Targets) and 3.2 (Economic Assumptions):

· In all of the studies, a Baseline or Reference scenario is compared with scenarios which are more ambitious in reducing GHG-emissions. These 'GHG-ambitious scenarios' mostly assume significant growth rates in RES energy sources for power generation (up to shares of e.g. 50% in the ETP Blueline and 97% in the Energy [R]evolution scenario by 2050) and agree on the main RES electricity generation technologies: onshore/offshore wind, biomass and solar-PV.

· The studies are more diverse regarding the estimations for the shares of thermal and hydro-RES: Of course, higher shares of FOS fuels are estimated in the absence of additional policies promoting RES-deployment. In the scenarios with more ambitious emission-reduction policies, gas-fired power plants often have a high relevance in serving peak-loads and load-following, due to the high shares of variable RES sources. Nuclear power plants, without t restrictions on development, are often considered as a vital option to help significantly reducing GHG-emissions from power generation in a cost-effective way (e.g. ETP Blueline).

· Innovative solutions for road transport (electric vehicles, biofuels) and other new power and energy technologies are identified as crucial for future energy systems throughout the studies. Most of the studies focus on electric vehicles and biofuels besides power sector restructuring.

· The scope for biomass technology improvements to 2050 is not explored in most cases, nor is the prospect of productivity increases driven by rising demand for biomass.

· Most of the studies emphasize the importance of policies concerning end-user efficiency (residential and industrial energy demand) and some studies describe measures in this field as crucial factors in the short run (2010 to 2030) to reach the emission targets set for the long run (e.g. ETP Blueline). The proposed measures comprise the thermal integrity of buildings, heat pumps, technological development in the processes of energy-intensive industries and more energy-efficient vehicles.

· Efficiency considerations on the one hand affect end-user efficiency and on the other hand the energy transformation sector (mainly power generation). There is little information on the latter and if, the studies estimate improvements in the efficiency of traditional power generation technologies, but only small ones compared to current state of the art (e.g. in the ECF Reference efficiencies of 60 % are assumed for CCGT-plants and 50 % for coal-fired plants in 2050).

· In most scenarios except those of ECF, grid development is not modelled or optimised for the given energy mix; it is pre-determined. Given the expected burst in electrification, the role of 'smart grid' technology developments, increased balancing needs and distributed generation, the assumptions about grid development equate to assumptions regarding costs and energy mixes.

Key points:

In addition to the technology-related 'Key points' already presented at the end of Sections 3.1 (Policy Assumptions and Targets) and 3.2 (Economic Assumptions), the following key conclusions on technology assumptions in the different scenarios can be made:

· Most scenarios assume significant growth rates in the use of RES energy sources for power generation.

· NUC, without restrictions on development, is often considered as a vital option to help significantly reducing GHG-emissions from power generation in a cost-effective way.

· Competitiveness of CCS depends strongly on the carbon price.

· Problems in extended use of biomass needed to counter-balance future shares of intermittent wind and solar are nowhere analysed in detail.

· Estimated future investment costs are strongly influenced by chosen assumptions on technological developments in energy technologies..

· Innovative solutions for road transport (electric vehicles, biofuels) are identified as crucial for future energy systems throughout the studies.

· Most studies emphasize the importance of policies concerning end-user efficiency (residential and industrial energy demand) and some studies describe measures in this field as crucial factors in the short run to reach emission targets set for the long run.

130.

3.5 Key Results of Scenarios


From the above modelling assumptions taken by different stakeholders, scenario models result in often different, sometimes similar projections regarding specific future trends:

· Future primary energy demand has to be seen in relation with the final energy demand and the technologies used. As shown in Figure 3, whereas the baseline scenarios show generally slightly increasing primary energy demands, the alternative scenarios aiming at reducing GHG-emissions show generally declining demands:

131.

Figure 3: Development of Economy-Wide Primary Energy Demand, in PJ


· As can be seen from Figure 4, without new energy policies to reduce energy demand or GHG-emissions, final energy demand will increase, similar to GDP-development. With new stringent policy measures, final energy demand can be reduced by 20-25% until 2050.

132.

Figure 4: Development of Economy-Wide Final Energy Demand, in PJ


· Compared to primary energy demand, long term developments in final energy demand are also influenced by the structure of the power generation sector (see Figure 4).

· It has to be noted, that in some cases, differences in efficiency targets may lead to major differences in projected energy demand. For example, ambitious energy efficiency measures are implemented in the Greenpeace/EREC Energy [R]evolution scenarios and in the Eurelectric Power Choices scenario, even in the medium term up to 2020. This results in significant declines of final energy demand and also primary energy demand, if measures aim at reducing energy demand of end-consumers.

· Looking at future electricity demand, a steady increase can be seen in all scenarios. Compared to the picture of the final energy demand, in general a substitution towards electricity can be observed. This tendency is especially relevant for scenarios with high GHG-reduction targets as these scenarios focus on decarbonisation of power generation and substitution for FOS fuels in transportation (e.g. electric vehicles) and buildings (e.g. heat pumps). Generally, this substitution process is induced through cost-optimization, either for individuals (DG ENER, Eurelectric), or for the whole region (e.g. ECF, FEEM), with the exception of the Greenpeace/EREC Energy [R]evolution scenarios. Beneath this substitution effect, electricity demand also increases due to higher income and economic activity. Figure 5 clearly shows that reductions in electricity demand due to energy-efficiency policies are outweighed by additional demand caused by the mentioned factors.

133.

Figure 5: Development of Economy-Wide Electricity Demand, in TWh


· The changes in electricity generation (development as well as structure), which are shown in Figure 6 for 2050 depend on:

· GHG and RES targets set in the scenarios,

· competitiveness of power plants assumed differently in different scenarios (capital costs, fixed and variable O&M costs, fuel and CO2-prices),

· pre-defined RES-targets set in 'back-casting' scenarios (Greenpeace/EREC, ECF),

· bounds set for deployment of NUC/CCS in some scenarios (Greenpeace/EREC, ECF).

134.

Figure 6: Electricity Generation in 2050, in TWh


In the medium term, up to 2030 and especially up to 2020, differences between the alternative scenarios are relatively small. Of course, even in the medium term, differences between scenarios with emission reduction targets and reference scenarios are considerable: Ambitious scenarios generally show higher shares for RES and NUC, with diverse views on CCS, except when NUC and CCS are excluded from the beginning.

In the long term, even differences between alternative scenarios are considerable. In the Eurelectric Power Choices and the ETP Blueline scenarios, nuclear power plants are estimated to obtain a high relevance in reaching the emission reduction targets. Nuclear power plants are assumed to be the most economic option to serve baseload in these scenarios, whereas FOS-fuelled plants are mainly used for load following (gas-fired plants), with the exemption of coal-fired plants with CCS. Differences between the two scenarios could be due to slightly different geographical coverage (ETP focusing on OECD-Europe, including Norway and Switzerland, both with high RES-shares) and differences in the estimated competitiveness of CCS / FOS fuels vis-à-vis NUC and RES power generation.

Deployment of CCS is of importance for all alternative scenarios (except Energy [R]evolution where it is excluded), but significantly higher in the Power Choices scenario and the ECF-pathways. Deployment for this form of emission abatement starts typically in the period from 2020 to 2030, but is assumed to gain importance only after 2030 (ETP, ECF, Eurelectric, PLANETS). The outcomes in the basic and advanced Greenpeace/EREC Energy [R]evolution scenarios are significantly different, due to exclusion of NUC and CCS in these scenarios.

In the low carbon scenarios examined, the quantity of electricity from RES produced by 2050 ranges from 1862 TWh to 4110 TWh. Fossil fuel generated electricity deploying CCS ranges from 490 TWh to 1470. Nuclear powered electricity production ranges from 490 TWh to 2607 TWh.

Key points:

· Without new energy policies to reduce energy demand or GHG-emissions, final energy demand will increase, similar to GDP-development.

· The significant differences across scenarios on assumptions on feasibility of efficiency improvements lead to major differences in projected energy demand.

· Compared to primary energy demand, long term developments in final energy demand are also influenced by the structure of the power generation sector. Higher decreases of final energy demand in relation to primary energy demand can be achieved by a technology-neutral approach in developing future power generation mixes (i.e. resulting in higher shares for CCS and NUC).

· Electricity demand increases across all scenarios due to higher income and economic activity. Reductions due to energy-efficiency policies are outweighed by additional demand.

· If a technology-neutral approach is chosen, high prices of CO2-certificates are the main driver for deployment of both RES and NUC, but also for development of CCS. Therefore, GHG-ambitious technology-neutral scenarios generally show higher shares for RES and NUC, except when NUC and CCS are excluded from the beginning.

135.

3.6 Models Used and Interdependencies Between Studies


In all scenario studies analysed bottom-up models are used, some of them in combination with top-down models, as summarised (for the main models) in Table 6:

136.

Table 6: Characteristics of Models Used


Study| Models used| Type of model| Characteristic

IEA - WEO| § World Energy Model| § Bottom-up-model (with additive top-down model)| § Simulation

IEA - ETP| § ETP MARKAL/TIMES| § Bottom-up-model| § Optimization (lead costs)

EU DG TREN| § PRIMES| § Mixed representation: Bottom-up and top-down model| § Partial market equilibrium

ECF Roadmap| § a.o. McKinsey Power Generation Model| § Bottom-Up-Model (with additive top-down model)| § Simulation

Greenpeace/EREC Energy [R]evolution| § MESAP/PlaNet| § Bottom-up model| § Simulation

Eurelectric Power Choices| § PRIMES| § Mixed representation: Bottom-up and top-down model| § Partial market equilibrium

Not least because of the use of the same models by different scenario studies, a variety of studies uses the input and output of other studies.

Two main studies can be indentified: IEA World Energy Outlook and DG ENER / PRIMES.

· The IEA Energy Technology Perspectives, the ECF Roadmap 2050, Eurelectric's Power Choices and Greenpeace/EREC's Energy [R]evolution use the WEO baseline.

· Input and output of the DG ENER PRIMES study are used for the Eurelectric study.

137.

4. Summary of comparison


From the comparison of stakeholder scenarios presented in this report, the following conclusions can be drawn:

· Overall Goal:

o Scenarios are marked by GHG and/or RES targets and development of future energy mixes is primarily based on optimising this parameter.

o Security of supply indicators are not created (or optimised), except for the grid-oriented modelling of ECF.

o Competitiveness indicators are limited, partial and not optimised.

· Basic Modelling Approaches used by Stakeholders:

o Models used for scenario studies can broadly be grouped into market-based optimisation models ("fore-casts") and models which use exogenously defined market shares ("back-casts").

o If market-based optimisation is applied (i.e. a technology-neutral approach chosen), deployment of the different energy technologies (FOS, NUC, RES) mainly depends on their relative total costs.

138.

o Grid modelling (and its major implications) are modelled by ECF; in most other scenario analyses, they are pre-determined.


· Energy/Electricity Demand:

o Without new policy measures demand will increase due to GDP growth. Final energy consumption in 2030 in low carbon scenarios range from 41000 PJ to 61000 PJ; in 2050 from 34000 PJ to 49000 PJ.

o Electrification is assumed in (almost) all scenarios. Electricity is estimated to gain higher shares in final energy demand, especially in scenarios confronted with ambitious GHG-targets (mainly as a substitute for fossil fuels).

· Development of More Sustainable Future Energy Systems:

o Most scenarios, such as those generated by the PRIMES model, optimise to determine the final energy mix ("technology neutrality"), based on cost input and technology learning assumptions. Greenpeace/EREC and the ECF scenarios backcast from several targeted generation shares, the former excluding NUC and CCS...

o Estimated future investment costs are also strongly influenced by chosen assumptions on technological developments in energy technologies whose dependability is often difficult or impossible to verify.

o Most scenarios seem to lack a clear consideration of the costs for necessary infrastructure changes to enable further deployment of variable RES. For example, although investments in the distribution grid seem to represent the majority of necessary grid investments and although all studies stress the importance of smart grids, in almost all scenarios merely transmission (if at all) is considered.

o Few studies explicitly model changes in the behaviour of economic agents with regard to changes in consumer behaviour.

o Effects on the EU labour market and the economy as a whole (e.g. risk of deindustrialisation) are not consistently modelled in any scenario study.

o Electricity prices increase in most studies at least in the medium term (2030). Some studies with high emission reduction targets expect a decrease of electricity prices in the long term (up to 2050), due to lower fossil-fuel consumption.

· Renewables:

o Absolute and relative increases of RES in the power sector across all scenarios.

o Investment costs for RES decrease across all scenarios, especially in a long term perspective.

· Nuclear Power:

o When optimised purely on costs, nuclear power tends to expand and gain increasing shares.

· Fossil fuel plants:

o CCS plays an increasing role in scenarios with a focus on a strong future role of the carbon price.

139.

References


[1] European Commission Reference Scenario to 2050 (2011), European Commission, Brussels, Belgium.

[2] ECF (2010): Roadmap 2050 – A practical guide to a prosperous, low-carbon Europe, European Climate Foundation, The Hague, The Netherlands.

[3] Greenpeace/EREC (2010): Energy [R]evolution – Towards a fully renewable energy supply in the EU 27, Greenpeace International, Amsterdam, The Netherlands, and European Renewable Energy Council (EREC), Brussels, Belgium.

[4] IEA (2010): Energy Technology Perspectives (ETP) 2010 – Scenarios & Strategies to 2050, International Energy Agency, Paris, France.

[5] IEA (2009): World Energy Outlook 2009, International Energy Agency, Paris, France.

[6] Eurelectric (2009): Power Choices – Pathways to Carbon-Neutral Electricity in Europe by 2050, Union of the Electricity Industry (Eurelectric), Brussels, Belgium.

[7] FEEM et al. (2010): PLANETS – Probabilistic Long-term Assessment of New Energy Technology Scenarios, Fondazione Eni Enrico Mattei (FEEM) et al., Milan, Italy, RTD Project Sponsored by the European Commission under the Seventh EU Research Framework Programme (Project No. 211859).

[8] Analysis and Comparison of Relevant Mid- and Long-term Energy Scenarios for EU and their Key Underlying Assumptions, Study performed by PROGNOS for EC – DG ENER, Basel, 2011.

[9] Key Factors Affecting the Deployment of Electricity Generation Technologies in Energy Technology Scenarios, Paul Scherrer Institut, Villigen, Switzerland, 2009.

[1] European Council, 29/30 October 2009.

[4] The decarbonisation scenarios reflect the transport policy measures included in the White Paper "Roadmap to a Single Transport Area – Towards a competitive and resource efficient transport system" (COM (2011) 144) with highest impact on energy demand in transport.

[5] The discussion here does not deal with CCS used for mitigation of industrial process emissions that do not stem from fossil fuel burning. These considerations exclude also potential removal of CO2 from the atmosphere through fitting CCS to biomass power plants, in which case the atmospheric removal of CO2 during plant growth is not undone by later emissions of CO2 from burning the biomass, with the CO2 from biomass burning being stored instead.

[6] In this respect, carbon intensity is a summary indicator for the fuel mix, while energy intensity captures the efficiency of energy consumption and the composition of economic activity (e.g. share of services versus (heavy) industry).

[7] In modelling terms this means a significant lowering of the discount rate used in energy consumption decision making of hundreds of millions of consumers.

[8] This share is considerably lower than in decarbonisation scenarios of DG CLIMA. There are three main explanations:

1. Decarbonisation scenarios and Current Policy Initiatives scenario are based on revised assumptions on nuclear (abandon of nuclear programme in Italy, no new nuclear plants in Belgium and upwards revision of costs for nuclear power plants).

2. Electricity demand is lower than in the Low Carbon Economy Roadmap Scenarios due to stringent energy efficiency measures.

3. Revised assumptions on the potential of electricity in transport compared to the DG CLIMA decarbonisation scenarios, following more closely the scenarios developed in the White Paper on Transport leading to lower utilisation rate of nuclear power plants than in the Low Carbon Economy Roadmap Scenarios. Electric vehicles flatten electricity demand and thus incentivise base load power generation.

[9] Average electricity prices in this table relate to a somewhat different customer base compared with electricity prices shown in Part A by including also energy branch customers in addition to those in final demand sectors; this explains the slight differences in average prices (e.g. for 2005 between 109.3 €/MWh when including the energy branch and 110.1 €/MWh when excluding it).

[10] CHP leads to emission reductions compared to conventional systems, but is only decarbonised when fired with biomass. The use of biomass in PRIMES is optimally allocated endogenously and might therefore not be used for CHP.

[11] The decarbonisation scenarios reflect the transport policy measures included in the White Paper "Roadmap to a Single Transport Area – Towards a competitive and resource efficient transport system" (COM (2011) 144) with highest impact on energy demand in transport.

[12] The PRIMES model having a micro-economic foundation, deals with utility maximisation and can calculate such perceived utility losses via the concept of compensating variations. However, this concept has to assume that preferences and values remain the same, even over 40 years, and has to compare utility with a hypothetical state of no policy or no change in framework conditions. Examples of such decreases are lowering thermostat in space heating, reducing cooling services in offices, switching light off, staying home instead of travelling, using a bicycle instead of a car, etc.

[15] The energy modelling did not include possible changes in value added of energy intensive industries as a reaction to climate policy measures. However, the low carbon economy roadmap includes a complementary analysis of macroeconomic and industrial competitiveness effects of a fragmented action scenario (SEC (2011)650, section 5.1.3) which provides further insights on these issues.

[16] It should be noted that costs of engines and propulsion cannot be separated from the rest of vehicle costs and that these numbers include therefore the costs for owning the entire vehicle.

[17] Average electricity prices in this table relate to a somewhat different customer base compared with electricity prices shown in Part A by including also energy branch customers in addition to those in final demand sectors; this explains the slight differences in average prices (e.g. for 2005 between 109.3 € when including the energy branch and 110.1 € when excluding it).

[18] The average EU price of diesel is calculated with the weighted average of country prices; differences between scenarios are therefore also due to different amounts of diesel used in the countries per scenario; in addition there are different blending ratios; the different taxation regime between the Reference scenario and the other scenarios including CPI reflecting the new proposal for the energy taxation directive.

[19] Scenarios for the Low Carbon Economy Roadmap of March 2011 show the additional costs of delayed action.

[20] Key references for this work are: (1) 'Analysis and Comparison of Relevant Mid- and Long-term Energy Scenarios for EU and their Key Underlying Assumptions' (PROGNOS, 2011) [8], and i 'Key Factors Affecting the Deployment of Electricity Generation Technologies in Energy Technology Scenarios' (Paul Scherrer Institut, 2009) [9].

[21] World Energy Outlook 2011 - special early insights: 'Are We Entering A Golden Age Of Gas?' International Energy Agency, June 2011 (complete WEO 2011 due 9 November)

[22] 'Making the Green Journey Work', European Gas Advocacy Forum, February 2011

[23] Eurogas Roadmap 2050, 13 October 2011

[24] EGAF refers to European Climate Foundation's 60% RES scenario for the power sector (Roadmap 2050, 2010)

[25] 'Power Perspectives 2030', European Climate Foundation, 7 November 2011; scenarios from Roadmap 2050, ECF, 2010.

[26] “Battle of the Grids”, Greenpeace supported by Energynautics, 2011

[27] For example, members of European Environment and Sustainable Development Advisory Councils

[28] Eg. DIW work for review of German energy concept

[29] 'Europe' is sometimes defined as OECD-Europe, EU-25 (for older scenario studies) or EU-27.

140.

TABLE OF CONTENTS


141.

1........... Section 1: Procedural issues and consultation of interested parties.................................... 3


142.

1.1........ Organisation and timing................................................................................................... 3


143.

1.2........ Consultation and expertise.............................................................................................. 3


144.

1.3........ Opinion of the IAB......................................................................................................... 4


145.

2........... Section 2: Problem definition........................................................................................... 5


146.

2.1........ Context.......................................................................................................................... 5


147.

2.2........ What is the problem?...................................................................................................... 6


148.

2.3........ Underlying drivers of the problem.................................................................................... 7


149.

2.3.1..... General barriers.............................................................................................................. 7


150.

2.3.2..... Sector specific barriers................................................................................................... 9


151.

2.4........ Business as usual developments..................................................................................... 13


152.

2.4.1..... Modelling approach...................................................................................................... 13


153.

2.4.2..... Assumptions................................................................................................................. 13


154.

2.4.3..... Energy developments.................................................................................................... 14


155.

2.4.4..... Sensitivity analysis......................................................................................................... 19


156.

2.4.5..... Conclusion................................................................................................................... 20


157.

2.5........ EU's right to act and EU added-value............................................................................ 20


158.

2.6........ Who is affected?........................................................................................................... 20


159.

3........... Section 3: Objectives.................................................................................................... 21


160.

3.1........ General objective.......................................................................................................... 21


161.

3.2........ Specific objectives........................................................................................................ 21


162.

3.3........ Consistency with other European policies...................................................................... 22


163.

4........... Section 4: Policy options............................................................................................... 22


164.

4.1........ Methodology................................................................................................................ 22


165.

4.2........ Policy options............................................................................................................... 24


166.

5........... Section 5: Analysis of impacts....................................................................................... 26


167.

5.1........ Environmental impacts.................................................................................................. 26


168.

5.2........ Economic impacts......................................................................................................... 28


169.

5.3........ Social impacts.............................................................................................................. 35


170.

5.4........ Sensitivity analysis......................................................................................................... 39


171.

6........... Section 6: Comparing the options.................................................................................. 40


172.

7........... Monitoring and evaluation............................................................................................. 44


173.

8........... Annexes....................................................................................................................... 45


174.

1. Section 1: Procedural issues and consultation of interested parties


Identification: Lead DG: DG ENER. Agenda planning/WP reference: 2011/ENER/002

175.

1.1. Organisation and timing


The IA work started early 2009 with the Reference scenario that is being used for all long-term initiatives of the Commission. An Interservice Steering Group was established early 2009 together with DG CLIMA and MOVE. This Group was also used for the Low Carbon Economy Roadmap and Transport White paper. Problem definition, objectives and design of policy options were presented to the Impact Assessment Steering Group in May 2011 and the final draft IA in July 2011.

The following DGs participated in the Impact Assessment Steering Group: AGRI, CLIMA, COMP, ECFIN, EMPL, ENTR, ENV, INFSO, JRC, LS, MARE, MARKT, MOVE, REGIO, RTD, SANCO, SG, TAXUD.

176.

1.2. Consultation and expertise


On 20 December 2010, the Directorate General for Energy launched a public consultation on the Energy Roadmap. The public consultation[1] was based on an online questionnaire with seven questions, some requiring comments and others in the form of multiple choice i. The public consultation was open until 7 March 2011. Some 400 contributions, half from organisations and half from individual citizens, were received. Several Member States sent a formal reply to the public consultation. Given the participation from a broad spectrum of organisations as well as citizens, this public consultation offered insights into a large range of stakeholder opinions. All of the Commission's minimum consultation standards were met. The full report presenting results of the public consultation can be downloaded from Europa website i.

177.

Public consultation questions and summary of replies


Question 1 How to ensure credibility: Many contributors emphasised the need for a stable, clear and predictable legislative framework to encourage the necessary investments in the energy sector which generally have a very long lead time. An appropriate analytical framework including transparency on modelling assumptions and results was mentioned by several respondents.

Question 2 The EU's position in a global policy context: More than half of all respondents chose 'global energy efficiency and demand developments' and 'global development of renewable energy' as the most important issues.

Question 3 Societal challenges and opportunities: Overall responses were fairly evenly distributed among the different choices. Public acceptance of new infrastructures was seen as important by many.


178.

Question 4 Policy developments at EU level: Roughly half of the respondents believe that energy efficiency is among the three most important issues needing more development at the EU level.


Question 5 Milestones in the transition: Across all industries and NGOs, intermediate targets, checkpoints and regular updates towards 2050 were recommended. However, the decarbonisation roadmap should be flexible enough to allow the route to be changed along the way.

Question 6 Key drivers for the future energy mix: About half of all respondents believe that global fossil fuel prices in relation to costs of domestic energy resources and long term security of supply will be the most likely key drivers of the future European energy mix.

Question 7 Additional thoughts and contributions: There was considerable divergence in opinions on the best way to decarbonise the energy sector in terms of market intervention as well as in the selection of a preferred technology option to be pursued.

In addition to the public consultation, representatives from the Directorate General for Energy and Commissioner Oettinger met numerous stakeholders individually and received many reports prepared by stakeholders on this topic. A comparison of stakeholder reports is presented in Annex 2.

An informal Energy Council took place on 2-3 May 2011 where ministers had a full-day discussion on the Energy Roadmap 2050. A meeting of Member State (MS) energy experts on the Roadmap also took place on 25 May 2011. The European Commission (EC) presented the problem definition, objectives and design of policy options of this Impact Assessment (IA) report. An Advisory Group of 15 highly-regarded experts mainly from academia and international institutions was established to support the work on the preparation of the Roadmap A presentation on the Roadmap was also given to the European sectoral social dialogue committee in the electricity sector on 14 December 2010.

The Commission contracted the National Technical University of Athens to model scenarios underpinning the IA analysis. Similarly to previous modelling exercises with the PRIMES model, the Commission discloses a lot of details about the PRIMES modelling system, modelling assumptions and modelling results which can be found in sections 4 and 5 as well as the annex 1 including an extensive section on macroeconomic, energy import prices, technology (capital costs of different technologies in power generation, appliances and transport) and policy assumptions. The PRIMES model was peer-reviewed by a group of recognised modelling experts in September 2011 with the conclusion that the model is suitable for the purpose of complex energy system modelling.

179.

1.3. Opinion of the Impact Assessment Board (IAB)


The IA report was discussed at the IAB hearing on 14 September 2011 and the IAB issued a positive opinion acknowledging the quality of the technical analysis and modelling underpinning the Roadmap and the Impact Assessment. The IAB recommended to improve the report in the following aspects: (1) to bring key findings of the evaluation of on-going policies into the IA report; i to consider an alternative policy scenario relying on a more relaxed assumption about the global climate deal; i to better describe scenarios and underlying assumptions; (4) to improve assessment of non-energy related impacts (employment, skills and knowledge gaps) and (5) to present stakeholder views in a more transparent way.

As a response to these suggestions, the evaluation part was reinforced; the issue of carbon leakage and external competitiveness was added to the problem definition as well as section 4.1. Methodology, while the part on competitiveness issues was expanded in Annex 1; policy options were described in more detail and the assessment of employment impacts was improved.

180.

2. Section 2: Problem definition 2.1. Context


(i) In the 2nd Strategic Energy Review (November 2008), the Commission undertook to prepare an energy policy roadmap towards a low carbon energy system in 2050. The Europe 2020 strategy includes a general commitment to establish a vision of structural and technological changes required to move to a low carbon, resource efficient and climate resilient economy by 2050.

(ii) The Commission's approach to decarbonisation is firmly grounded in the EU's growth agenda, set out in the Europe 2020 strategy, including the Resource Efficient Europe Flagship Initiative[4]. The Communication 'Energy 2020 - A strategy for competitive, sustainable and secure energy' paves the way to 2020 stressing the three pillars of the EU's energy policy: competitiveness, security of supply and sustainability, building on the Climate and Energy package adopted in June 2009.

(iii) The European Council (October 2009) supports an EU objective, in the context of necessary reductions according to the IPCC by developed countries as a group, to reduce GHG emissions by 80-95% by 2050 compared to 1990 levels[5]. The European Parliament similarly endorsed the need to set a long-term GHG emissions reduction target of at least 80% by 2050 for the EU and the other developed countries[6].

(iv) The European Council (February 2011) confirms this emissions reduction commitment and recognises that it will require a revolution in energy systems, which must start now. It requests that due consideration should be given to fixing intermediary stages towards reaching the 2050 objective.

(v) The Roadmap for moving to a competitive low carbon economy in 2050[7] makes the economic case for decarbonisation and shows that the targeted 80-95% GHG emissions reduction by 2050 will have to be met largely domestically. Intermediate milestones for a cost-efficient pathway, e.g. 40% domestic reduction by 2030, and sectoral milestones expressed as ranges of GHG emissions reductions in 2030 and 2050 were put forward.

(vi) The Commission is now preparing sectoral roadmaps exploring the dynamics within the sector and the interplay of decarbonisation[8] and other sectoral objectives. The Roadmap to a Single European Transport Area – Towards a competitive and resource efficient transport system[9] aims to introduce profound changes in passenger and freight transport patterns, resulting in a competitive transport sector which allows increased mobility, cuts CO2 emissions to 60% below 1990 levels by 2050 and breaks the transport system's dependence on oil. A Roadmap to a Resource Efficient Europe, also planned for 2011, builds on and complements other initiatives, focusing on increasing resource productivity and decoupling economic growth from resource use.

This IA is a key part of initiatives to deliver on a resource Efficient Europe, one of the 7 flagships of the Europe 2020 strategy[10]. It aims at further developing the decarbonisation analysis of the energy sector as presented in the Low Carbon Economy Roadmap in March 2011, with particular attention to all three EU energy policy objectives - energy security, sustainability and competitiveness.

181.

2.2. What is the problem?


The well-being of people, industry and economy depends on safe, secure, sustainable and affordable energy. Energy is a daily need in a modern world and is mostly taken for granted in Europe. The energy system and its organisation evolved over centuries if not millenaries using different fuels and distribution systems to cover basic needs such as food preparation, protection against winter temperatures and production of tools e.g. via metal melting. Over the last century this has concerned delivering heat and warm water as well as industrial and transport fuels and electricity to consumers. There has been a significant increase in energy production and consumption over the last 100 years providing more comfort and individual freedom but at the same time polluting the environment and (at least partially) depleting existing reserves. Our current energy system and ways of producing, transforming and consuming energy are unsustainable due to:

High GHG emissions of which the great majority is directly or indirectly linked to energy[11] which are not compatible with the EU and global objectives of limiting global climate change to a temperature increase of 2ºC to avoid dangerous impacts[12] (even though the EU contribution to global emissions is low and will decline in particular if other regions make no or little efforts on decarbonisation,[13] industrialised countries should keep their leading role in the fight against climate change);

Security of supply risks, notably those related to:

- high dependence on foreign sources of energy imported from a limited number of suppliers (EU27 currently imports 83.5% of its oil and 64.2% of its gas consumption; overall import dependency is around 54% and is projected to slightly increase by 2050), including supplies from politically unstable regions;

- gradual depletion of fossil fuel resources and rising global competition for energy resources;

- increasing electrification from more variable sources (e.g. solar PV and wind) which poses new challenges to the grid to ensure uninterrupted electricity deliveries;

- low resilience to natural or man-made disasters and adverse effects of climate change;

Competitiveness risks related to high energy costs and underinvestment. External competitiveness of the European industry vis-à-vis its international competitors is another crucial aspect determining the design and timing of EU energy and climate action. While it is important to sustain first mover advantage and industrial leadership it should also be assessed whether 'early' action comes at a cost of comparatively high carbon, fuel and electricity prices for industry compared to action undertaken in the rest of the world.

It will take decades to steer our energy systems onto a more secure and sustainable path. In addition, there is no silver bullet to achieve it. There is no single energy source that is abundant and that has no drawbacks in terms of its sustainability, security of supply and competitiveness (price). That is why the solution will require trade-offs and why the market alone under the current regulatory environment might fail to deliver. The decisions to set us on the right path are needed urgently as failing to achieve a well-functioning European energy market will only increase the costs for consumers and put Europe’s competitiveness at risk. Significant investments will however be needed in the near future to replace energy assets in order to guarantee a similar level of comfort to citizens at affordable prices; assure secure and competitive supplies of energy inputs to businesses and preserve the environment. The energy challenge is thus one of the greatest tests which Europe has to face.

Relying on more low-carbon, domestic (i.e. intra EU) or more diversified sources of energy, produced and consumed in an efficient way, can bring significant benefits not only for the environment, competitiveness and security of energy supply but also in terms of economic growth, employment, regional development and innovation. What are the barriers? Why is the shift to an energy system using low-carbon, more competitive and more diversified sources not, or too slowly happening?

182.

2.3. Underlying drivers of the problem


There are several factors that hamper the shift:

183.

2.3.1. General barriers


1) Energy market prices do not fully reflect all costs to society in terms of pollution, GHG emissions, resource depletion, land use, air quality, waste and geopolitical dependency. Therefore, user and producer choices are made on the basis of inadequate energy prices that do not reflect true costs for society.

184.

2) Inertia of the physical system


The majority of investments in the energy system are long-term assets, sometimes requiring long lead times, and having life times of 30-60 years, leading to significant lock-in effects. Any change to the system materialises only gradually. Current market structure and infrastructures can discourage new technology development, since infrastructure, market design, grid management and development require adaptation and modernisation which represent additional costs which face resistance from industry.

185.

3) Public perception and mindset of the users


General public perception of the risks related to the construction of new power plants (large-scale RES, nuclear, low-carbon fossil) and infrastructure needed to introduce large shares of renewables (which additionally implies new grid lines and large energy storage technologies) or of CO2 storage can be more negative than expert judgements. Public acceptance was also acknowledged as important by many respondents in public consultation. It can also take a long time and require adequate incentives or regulation to persuade people to change the way they heat their houses, transport themselves, etc.

186.

4) Uncertainty concerning technological, demand, prices and market design developments


The energy system is characterised by a large proportion of long-term fixed costs that need to be recovered over several decades. Uncertainty about future technologies, energy demand development, market integration and rules[14], carbon and fuel prices, availability of infrastructures can significantly increase investor risks and costs, and make consumers and businesses reluctant to invest. Private investors can cope well with some categories of risks but policy makers and regulators can contribute to decreasing the uncertainties as regards political and regulatory risks.

187.

5) Imperfect markets


There is weak competition in some Member States where markets are still dominated by incumbents. In particular, the absence or lack of effective non-discriminatory third party access to infrastructure can constitute an entry barrier for new entrants. Another factor is market myopia, i.e. the fact that long-term investments are not necessarily pursued by market actors who are generally drawn towards shorter-term gains.

Regarding new infrastructure investments, it can be difficult to clearly identify the beneficiaries, and therefore efficiently allocate the costs of new investments. In addition, in liberalised markets with various players, interdependencies might impose additional efforts to coordinate some investments (it is unrealistic to expect wind power plants to be constructed in the North Sea if no adequate grid is built).

In some Member States developing markets for energy efficiency services and decentralised RES are faced with a low number of actors on the supply side (lack of qualified labour force) as well as on the demand side (low levels of consumer awareness partly as a consequence of the ongoing rapid technological advances) and the lack of enabling regulatory framework. This has a particularly negative effect on the uptake of energy services companies (ESCOs) that can provide integrated energy saving solutions together with financing schemes. Renewable energy can also suffer from market designs that have been developed alongside the development and optimisation of centralised power generation and trading.

188.

2.3.2. Sector specific barriers


Besides these factors and based on an evaluation of ongoing policies[15], there are problems specific to energy efficiency, infrastructure, security of supply and low-carbon generation technologies which are discouraging investments.

189.

Energy efficiency


Though a number of initiatives were undertaken at EU level since the mid-1990s, the European Energy Efficiency Action Plan[16] created a framework of legislation, policies and measures with a view to realise the 20% energy efficiency and saving objective. After years of growth, the EU primary energy consumption has stabilized in 2005 and 2006 at around 1,825 Mtoe and decreased in 2007, 2008 and 2009 to reach around 1,700 Mtoe[17]. Energy intensity kept improving. For the first time, the latest business-as-usual scenario projections (PRIMES 2009) show a break in the trend of ever-increasing energy demand in the EU27[18].

However, the EU is far from reaching its 20% objective. The projections indicate that with the rates of implementation of the current energy efficiency policies in Member States only half of the objective might be achieved by 2020[19]. Furthermore, while the economic crisis contributed to this decrease in energy consumption, it has also negatively impacted energy efficiency investment decisions at all levels - public, commercial and private. As a response to this, the Commission has recently adopted two new initiatives - an Energy Efficiency Plan[20] and a Directive on Energy Efficiency - aiming at stepping up efforts towards the 20% target.

In addition to the above mentioned barriers, there are many examples of split incentives or principal-agent market failures in the energy sector where the decision maker may be partially detached from the price signals. For example, landlords are often the decision-makers about renovation of buildings, but it is usually tenants that pay the energy bills and benefit from their reduction, giving landlords little reason to invest.

190.

Internal market


The process of opening the EU energy markets to competition started ten years ago. It has allowed EU citizens and industries to benefit in terms of more choice, more competition for a better service and improved security of supply. Since July 2007, all consumers in all EU countries have been free to switch their suppliers of gas and electricity.

Independent national regulatory authorities have been established in each EU country to ensure that suppliers and network companies operate correctly and actually provide the services promised to their customers. An inquiry into the electricity and gas sectors published in January 2007[21] revealed that too many barriers to competition and too many differences across the Member States remain. In 2007 and 2008, a great deal of effort was put into enhancing competition on the wholesale market; significant progress was made through the regional initiatives. However, the Benchmarking Report adopted in 2009[22] still showed a mixed picture of the accomplishment of the internal market and revealed in particular that there are still high levels of concentration on the retail and wholesale markets and a lack of liquidity.

To remedy the situation, the Commission came forward with the third internal energy market liberalisation package. It foresees the effective separation of supply and production activities to make the market accessible for all suppliers, the harmonization of powers of national regulators, better cross-border regulation to promote new investments and cross-border trade, effective transparency, as well as assuring that EU and third country companies compete in the EU on an equal footing. For the electricity market, a target model has been agreed in the context of the Florence regulatory forum and for gas markets a target model is under development.

191.

Infrastructure


Tariff regulation - Transmission is a mostly regulated business at national level and cost allocation to final beneficiaries can be difficult for large trans-European infrastructure. Tariff regulation in most Member States has been based on the principle of cost-efficiency, allowing recovery of costs only for projects based on real market needs or cheapest available solutions, but some externalities, such as innovation, security of supply, solidarity aspects or other wider European benefits may not always be fully taken into account. For infrastructure networks that are entirely new, such as electricity highways or CO2 transport infrastructure, it is likely to be of public interest to ensure that the first investments are compatible with later, more efficient network solutions.

In the EU internal energy market, a key tool to promote interconnections is the trans-European energy networks (TEN-E) programme which has positively contributed to the development and operation of the internal energy market and increased security of supply[23]. Despite the progress achieved, the dramatic changes to the EU energy policy framework in recent years call for a review of the TEN-E framework. The programme has responded too slowly to the major energy and climate goals of today, and is poorly equipped to deal with the growing challenges that will arise from the 2020 and 2050 ambitions. In 2009, as the financial crisis unfolded, EU institutions agreed on the European Energy Programme for Recovery (EEPR)[24] which was endowed with a €3,980 million financial envelope in support of gas and electricity interconnection projects, offshore wind projects as well as carbon capture and storage projects.

192.

Security of supply


EU Energy import dependency for all fuels is 54%. More importantly, the EU is vulnerable to the increasing supply of some commodities by global oligopolies which can create internal and external imbalances. EU experiences of gas supply interruptions in early 2006, 2008, 2009 and 2010, as well as the EU's strong dependence on imports of petroleum products and the geopolitical uncertainty in many producer regions led to the adoption of the Regulation concerning measures to safeguard security of gas supply[25].

Since 1968, EU legislation imposes an obligation on Member States to maintain minimum stocks of crude oil and/or petroleum products that can be used in the event of a supply crisis and a new directive[26] adopted in September 2009 aligns stockholding obligations with those of the International Energy Agency.

Electricity blackouts in the EU in November 2006 highlighted the need to define clear operational standards for transmission networks and for correct maintenance and development of the network. Therefore, in order to ensure the functioning of the internal energy market, the EU established obligations for Member States to safeguard security of electricity supply and undertake significant investment in electricity networks[27].

193.

Low-carbon generation technologies


All low carbon technologies are reliant upon a strong carbon price or other regulatory measures. As well as continuous R&D funding, long-term market or regulatory signals to investors are needed.

194.

Renewables


Some renewables are currently at early development stage, insofar as they often have higher costs than alternatives, though they form part of a sector with rapid technological developments and significantly declining production costs resulting from early economies of scale and technology learning.

Renewable energy production has grown rapidly in the last ten years. The Green electricity Directive (2001/77) and the Biofuels Directive (2003/30) aimed to stimulate an increase in the consumption of renewable energy. The former established an overall EU target of 21% and national indicative targets for the RES shares in gross electricity consumption by 2010. The latter required that all Member States should ensure that at least 5.75% of their petrol and diesel for transport comes from renewable fuels. Despite significant growth, the latest EUROSTAT data indicate that 2010 targets will not be met.[28] The Renewable Energy Directive[29] sets out binding targets for all Member State to achieve the 20% renewable energy target for the EU by 2020 as well as a 10% target for the share of renewable energy in transport. It also addresses the problems of administrative barriers to the development of renewables and their integration in the grids and sustainability requirements for biofuels. According to the Communication on 'Renewable Energy: Progressing towards the 2020 target', Member States are on track to reach their overall renewable energy target as well as the sub-target for renewable energy in transport.

Table 1: Renewable energy developments and defined targets.

Share of renewable energy in…| Most recent data| Target 2010 (indicative)| Target 2020 (binding)

electricity generation| 13.4% (36 Mtoe)| 16. 6 % (48 Mtoe - 2008 )| 21%| no

transport| 0.3% (1 Mtoe)| 3. 5 % (11 Mtoe - 2008)| 5.75%[30]| 10%[31][3]

heating i| 9.1% (52 Mtoe)| 12 % (67 Mtoe - 2008 )| no target| no

Gross final energy consumption| 7.6% (89 Mtoe)| 10.6 % (132 Mtoe - 2009 )| no target i| 20%

Gross inland consumption| 5.8% (101 Mtoe)| 9.0% (153 Mtoe)| 12%| no

195.

Nuclear


The EU-27 has the largest number of commercial nuclear power stations in the world: some 150 nuclear reactors are in operation, providing around 30% of the EU's electricity and 60% of low carbon electricity. Although nuclear is a proven technology, in some MS it faces uncertainties regarding public acceptance due to risk perception and often also due to lacking implementation of available technical solutions for long term disposal of nuclear waste. The nuclear accident in Japan could further aggravate public acceptance problems in some MS while possible further increased safety requirements might affect the competitiveness of existing nuclear generation capacities in some MS.

Nuclear safety is and will remain one of the absolute priorities of the EU. A Directive establishing the basic framework for nuclear safety i adopted in 2009 provides a Community framework in order to maintain and promote the continuous improvement of nuclear safety. When this Directive will be implemented the EU will be the first major regional nuclear player with common binding nuclear safety rules. On 3 November 2010, the European Commission also proposed a Directive which sets safety standards for disposing spent fuel and radioactive waste.

196.

CCS


As a new and developing industry, CCS faces similar challenges to innovative renewable energy technologies. At present, it is in the early commercial-scale demonstration phase, and is ambitiously striving to be commercially viable soon after 2020. But facing a number of problems, its progress is currently challenged by issues that include financing and public perception concerns in some Member States.

The European Council of March 2007 urged to work towards strengthening R&D and developing the necessary technical, economic and regulatory framework to remove existing legal barriers and to bring environmentally safe CCS to deployment. In 2008, the European Council made a commitment to supporting the design, construction and operation of CCS in up to 12 large-scale demonstration plants by 2015. Demonstration of the technology in commercial plants is considered to be an essential step towards commercialisation of CCS to demonstrate the environmental safety and economic viability of the technology, which is also dependent on strong carbon prices. The CCS Directive i establishes a comprehensive legal framework to safely manage the environmental aspects of capture, transport and the geological storage of CO2. The revised ETS Directive ensures that safely stored CO2 is not regarded as emitted and provides therefore a financial incentive for CCS. In addition, 300 million allowances from the New Entrants Reserve (NER) shall be available to support commercial-scale CCS and innovative RES demonstration projects under the NER300 funding programme, thus complementing and going beyond funding already provided by the EEPR. CCS is also an important option for decarbonisation of several heavy industries i. Moreover, CCS has the potential to deliver carbon-negative power, if it is combined with biomass combustion or co-firing.

As the Energy Roadmap 2050 is a broad policy document without having the ambition of defining individual policy measures, this IA tries to present a broad picture of the challenges and barriers but will not propose solutions to all of them.

197.

2.4. Business as usual developments 2.4.1. Modelling approach


The Commission has carried out an analysis of possible future developments in a scenario of unchanged policies, the so-called “Reference scenario”. The Reference scenario was also used in the IA for the “Low-carbon economy 2050 roadmap” and IA for the 'White Paper on Transport'. The Reference scenario is a projection, not a forecast, of developments in the absence of new policies beyond those adopted by March 2010. It therefore reflects both achievements and deficiencies of the policies already in place. In order to take into account the most recent developments (higher energy prices and effects of the nuclear accident in Japan) and the latest policies on energy efficiency, energy taxation and infrastructure adopted or planned after March 2010, an additional scenario called Current Policy Initiatives scenario (CPI) was modelled.

Both scenarios build on a modelling framework including PRIMES, PROMETHEUS, GAINS and GEM-E3 models. The PRIMES model is a modelling system that simulates a market equilibrium solution for energy supply and demand. The model is organized in sub-models (modules), each one representing the behaviour of a specific (or representative) agent, a demander and/or a supplier of energy. GAINS complements PRIMES with consistent estimates of non-CO2 emissions and their contribution to reach the policy targets included in the reference scenario. PROMETHEUS is a stochastic world energy model used for determining fossil fuel import prices, while the results of the GEM-E3 general equilibrium model are used as inputs of macro-economic (e.g. GDP) and sectoral numbers (e.g. sectoral value added) for PRIMES. Several EU scenarios were established at different points in time using a framework contract with National Technical University of Athens (author and owner of the PRIMES model).

198.

2.4.2. Assumptions


The Reference scenario 2050 includes current trends and recent Eurostat and EPC/ECFIN long term projections on population and economic development. It takes into account the upward trend of import fuel prices in a highly volatile world energy price environment. Economic decisions are driven by market forces and technological progress in the framework of concrete national and EU policies and measures implemented by March 2010. The 2020 targets for RES and GHG will be achieved in this scenario, but there is no assumption on targets for later years besides annual reduction of the cap in the ETS directive.

The CPI scenario builds on the same macroeconomic framework and includes policy initiatives adopted after March 2010 or policy initiatives currently being planned as well as updated technology assumptions for nuclear and electric vehicles.

The main assumptions used for both scenarios are presented in table 2 and all assumptions and more detailed description of results can be found in Annex 1 (part A).


199.

Table 2: Main assumptions in the Reference scenario 2050 and Current Policy Initiatives Scenario


GDP growth rate: 1.7 % pa on average for 2010-2050

Oil price: 106 $/barrel in 2030 and 127 $/barrel in 2050 (in year 2008 dollars)[37]

Main policies included (Reference scenario): Eco-design and Labelling directives adopted by March 2010; Recast of the Energy Performance of Buildings Directive, EU ETS directive; RES directive (20% target); Effort Sharing Decision (non-ETS part of the 20% GHG target); Regulation on CO2 from cars and vans.

Main policies included (Current Policy Initiatives scenario) in addition to those already included in the Reference scenario 2050: Energy efficiency Plan; facilitation policies for infrastructure and updated investments plans based on ENTSO-e Ten Year Network Development Plan; Nuclear Safety Directive; Waste management Directive; revised Energy Taxation Directive

Consequences of the Japanese nuclear accident leading to abandon of nuclear programme in Italy, nuclear phase-out in Germany and in case of nuclear lifetime extension up to 20% higher generation costs reflecting higher safety requirements as well as introduction of a risk premium for new nuclear power plants; revisiting of progress on CCS in demonstration projects and policies and initiatives leading to slightly higher uptake of electric vehicles.

Costs for technologies: Technology parameters are exogenous in the PRIMES modelling and their values are based on current databases, various studies and expert judgement and are regularly compared to other leading institutions. Technologies are assumed to develop over time and to follow learning curves which are exogenously adjusted to reflect the technology assumptions of a scenario. Overall, mature fossil fuel, nuclear as well as large hydroelectric technologies exhibit rather stable technology costs, except for innovative concepts such as 3rd generation nuclear power plants or carbon capture and storage (CCS), where costs decline with further RTD and more technology experience. Similar developments are assumed for new renewable technologies, such as off-shore wind and solar PV as has been witnessed in the past for most energy technologies (e.g. on-shore wind or more recently solar energy).

Drivers: Within these framework conditions market forces drive energy and emission developments. Economic actors optimise their supply and demand behaviour while the simulation of energy markets in the model derives energy prices, which in turn influence the behaviour of energy actors (power generators, various industrial and service consumers, households, transport, etc). The Reference and CPI scenarios do not assume any additional policies. The model provides a simulation of what the interplay of market forces in the current economic, world energy, policy and technology framework would bring about if no new policies would be put in place.

All scenarios are built on assumptions of perfect foresight and 'representative' consumer leading to a very high certainty on regulatory framework for investors and rather optimistic deployment of technologies by households and services that will be challenging to ensure in practice.

200.

2.4.3. Energy developments


Energy consumption

Primary energy consumption peaked in 2006, from which point it decreases slightly up to 2050 (-4%). This is despite economic growth leading to a doubling of GDP between 2005 and 2050.

Final energy consumption continues rising until 2020, after which demand stabilises as more efficient technologies have by then reached market maturity and the additional energy efficiency of the appliances is sufficient to compensate for increased demand. The share of sectors remains broadly stable with transport remaining the biggest single consumer accounting for 32% in 2050; the industrial share increases slightly while that of households declines a bit.

In the CPI scenario, further energy savings are brought about mainly by energy efficiency measures for households and services sector and efficiency improvements in energy transformation in the short to medium term leading to further declines in final energy demand which remains 4-6% below the Reference scenario. There are marked changes also at the level of primary demand in 2020 (-5.0%); 2030 (-5.8%) and 2050 (-8.4%).

The energy intensity of the economy and of different sectors decreases. Increased energy efficiency in the residential sector is due to the use of more efficient energy equipment (appliances, lighting, etc.) and buildings, being driven by the Eco-Design regulations and by better thermal integrity of buildings reflecting the Recast of the Energy Performance of Buildings Directive. Energy consumption in transport is decoupling significantly from underlying transport activity growth due to the use of more energy efficient vehicles; this development is largely driven by more fuel efficient cars, in particular hybrids, following the CO2 performance standards set by the CO2 from cars regulation[38].

There is considerable fuel switching in final and primary energy demand in the Reference scenario. In primary energy, the dominance of fossil fuels diminishes with its share falling from 83% and 79% in 1990 and 2005, respectively, to only 64% in 2050. While non fossil fuels (RES and nuclear) account for 36% of primary energy in 2050, they reach a significantly higher share in the 2050 electricity mix. Energy sources not emitting CO2 supply 66% of electricity output in 2050, with 40% RES and 26% nuclear.

201.

Graph 1: Reference scenario- Fuel shares in primary energy



In the CPI scenario, the share of nuclear is lower due to a change in nuclear assumptions. In this new policy environment gas and RES replace nuclear and thereby increase their share over Reference scenario levels.

202.

Power generation


The demand for electricity continues rising and there is a considerable shift towards RES with a strong increase in wind. Power generation and capacity from solids decrease throughout the projection period due to increasing carbon prices that reduce the competitiveness of this technology; gas power generation capacity increases, also as peak load activated during back-up periods due to the increased amount of RES in the system. As a result of the large increase in RES in power generation the load factor of the system decreases given the more widespread use of technologies that run only a limited number of hours per year. Investment in power generation increases over the projection period, driven by RES and gas.

The carbon intensity of power generation falls by over 75% in 2050 compared to 2010 levels, driven by the decreasing ETS cap and rising carbon prices. CO2 emissions from power generation decline by 2/3rd between 2010 and 2050, while electricity demand still increases. This strong decarbonisation is brought about by fuel switching to RES and nuclear, an increasing share of gas in fossil fuel generation and significant penetration of CCS after 2030. In 2050 18% of electricity is generated through power plants with CCS (solids and gas).


Electricity demand in the CPI scenario falls well below electricity use in the Reference scenario (by 6.5% in 2030 and 4.3% in 2050), reflecting measures in the Energy Efficiency Plan and the revised Energy Taxation Directive. The CPI scenario takes account of the post Fukushima policy change in Member States, notably the abandonment of the nuclear programme in Italy, and new initiatives, such as the nuclear stress tests that will tend to increase costs for new power plants and retrofitting. The CPI scenario has significantly lower CCS penetration primarily as a result of the ETS price being lower in the longer term and also as a consequence of the relatively moderate progress that has been made since 2009 (Reference scenario) towards the EU objective of having up to 12 large-scale CCS demonstration plants operational by 2015 in Europe.

203.

Table 3: Electricity related indicators in CPI scenario and differences from Reference


Heating

A strong increase in demand for distributed steam and heat can be observed between 2005 and 2020 following strong CHP promoting policies, as well as commercial opportunities that arise from gas and biomass based CHP technologies. In the longer term further demand for distributed heat in the tertiary and residential sectors slows down as a result of the trend towards electrification (i.e. heat pumps) and higher energy efficiency which limits the overall demand for heating. In industry the increase in demand for distributed steam is projected to continue in the future because the changes of industrial activity are favourable for sectors with high demand for steam such as chemicals, food, tobacco, and engineering.

In the CPI scenario, demand for distributed heat rises compared to current levels but is 1-2% lower than in the Reference scenario, reflecting the effects of more efficient heating systems used in houses.

204.

Transport


Transport accounts today for over 30% of final energy consumption. In a context of growing demand for transport, final energy demand by transport is projected to increase by 5% by 2030 rising further marginally by 2050. Transport growth is driven mainly by aviation and road freight transport. The EU transport system would remain extremely dependent on the use of fossil fuels. Oil products would still represent 88% of EU transport sector needs in 2030 and 2050 in the Reference scenario.

Energy consumption in transport is little affected by current energy policy initiatives (- 1.7% in 2030 and -5.7% in 2050). Changes from the Reference scenario are brought about in particular by the proposed new energy taxation system and through the somewhat more favourable policy environment for electric and plug-in hybrid vehicles.

205.

Policy relevant indicators (and targets)


Emissions - It is estimated that a continuation of current trends and policies (Reference scenario) would result in 40% reduction in energy-related CO2 emissions between 1990 and 2050 and 26% by 2030. All GHG emissions would fall 40% by 2050 (29% by 2030) which represents about half of the domestic efforts needed by a developed economy in the context of limiting climate change to 2°C[39]. Most emissions continue to be energy related emissions. Carbon intensity falls markedly. Producing one unit of GDP in 2050 would lead to only 21% of energy related CO2 emissions that were required in 1990.

In the CPI scenario emission reductions are broadly similar to those in the Reference scenario. CO2 emissions in 2050 are 41% below 1990 values and below those reached in the Reference case due to greater energy intensity improvements brought about by vigorous energy efficiency policies which overcompensates worsening carbon intensity due to lower availability of nuclear and CCS and lower ETS carbon prices. Total GHG emissions in 2050 decrease by 39% below the 1990 level (1 percentage point less than in the Reference scenario) mainly a result of changes of the carbon price over the next decades.

ETS prices under developments in the Reference scenario rise from 40 € (08)/tCO2 in 2030 to 52 € in 2040 and flattens out to 50 € in 2050. The ETS price in the CPI scenario is lower for most of the projection period reflecting efficiency and RES policies (by about 20% in 2025-2035) and ends at 51 € in 2050.[40]

RES target - The Reference scenario assumes that the RES target is reached in 2020; the RES share continues rising in the Reference scenario to reach 24% in 2030 and over 25% in 2050. Further penetration of RES progresses more slowly due to the assumed phasing out of operational aid to mature RES technologies. RES in transport contribute 10% in 2020 to comply with the RES directive; this share increases to 13 % by 2050. However, the pace of electrification in the transport sector is projected to remain slow in the Reference scenario: electric propulsion in road transport does not make significant inroads by 2050[41]. The CPI scenario has higher RES shares, e.g. 25% RES in final energy in 2030 and 29% in 2050.

The indicative 20% energy savings objective for 2020 would not be achieved under current policies - not even by 2050. The Reference scenario would deliver 10% less energy consumed in 2020 compared to the 2007 projections. The CPI scenario delivers significantly more. Energy consumption in 2020 is 14% below the 2007 projections further decreasing significantly up to 2050.[42]

Import dependency - Total energy imports increase by 6% from 2005 to 2050. The increase is rather limited despite decreasing indigenous production, as rising gas (+28% from 2005 to 2050) and biomass imports are compensated by a marked decline in coal imports while oil imports remain broadly stable. Import dependency rises above the present level (54%), reaching 58% in 2020 and flattening out to 2050 thanks to more RES and nuclear. It remains broadly unchanged in the CPI scenario.

Average electricity prices rise up to 2030 and stabilise thereafter. The price increase up to 2030 is due to three main elements: RES supporting policies, ETS carbon price and high fuel prices due to the world recovery after the economic crisis. Thereafter electricity prices remain stable because of the techno-economic improvements of various power generation technologies that limit the effects of higher input fuel prices and CO2 prices. In the CPI scenario, electricity prices are slightly higher (1% in 2030 and 4% in 2050) reflecting the lower share of nuclear as well as higher lifetime extension costs post Fukushima and high investments for new electricity generation capacity, especially RES.

Total costs of energy (including capital costs, energy purchases and direct efficiency investment costs) are rising fast over the projection period but are not equally distributed across sectors. Energy related expenditures in households rise strongly while the growth of energy related costs for services and industry is more moderate. Energy costs are rising faster than GDP and represent around 15.1% of GDP in 2030 (up from 10.5% in 2005) and 14.3% in 2050. The faster rate of growth relative to GDP reflects significant investments needs in energy production, transmission and distribution as well as demand based energy efficiency measures. Under the CPI scenario, system costs are slightly higher amounting to 15.3% and 14.6% in relation to GDP in 2030 and 2050, respectively, reflecting in particular greater investment requirements.

206.

2.4.4. Sensitivity analysis


Considering the high degree of uncertainty surrounding projections over such a long time horizon, a sensitivity analysis has been carried out with respect to two key parameters - energy imports prices and GDP. A high and a low case has been analysed for both variables.

207.

GDP


The two economic growth variants explore a High GDP case where GDP per capita is 0.4 percentage points (pp) higher than in the Reference scenario throughout the projection period (+15% increase in GDP level in 2050) and a Low GDP case with GDP per capita 0.4 pp lower (-14.7% in GDP level in 2050). GDP and economic activity have a significant influence on energy consumption in particular in industry and services.

The model based analysis shows that policy relevant indicators are rather insensitive against variations in GDP assumption, which is a significant result given the great uncertainty in making GDP projections for the next few years let alone the next four decades.

CO2 reduction becomes only slightly more difficult to achieve under significantly higher economic growth. Higher economic growth brings more opportunities for innovation and investment leading to improvements in both energy and carbon intensity. In a similar manner, low economic growth entails lower economic activity but fewer investments in low carbon and energy efficient technologies. There is thus only limited further emission reductions brought about by considerably lower GDP levels. RES shares in gross final energy consumption are pretty robust with respect to GDP levels with variation spanning just 1 percentage point in 2050. Import dependency is also unaffected by such significant changes in GDP levels. Policy relevant indicators regarding competitiveness are pretty much unaffected by economic growth; while ETS prices differ to some extent, the effects on electricity prices are marginal.

208.

Energy prices


Two energy price sensitivities were modelled – a High energy price case with the world oil price 28% higher in 2050 and a Low energy price case with the world oil price 34% below the Reference scenario in 2050. In the low price case, fossil fuel import prices remain broadly at the 2010 level; coal prices are stable, oil has a small peak around 2030, whereas gas prices remain weak over the next few years but recover to the 2010 level in the long run.[43]

High world energy prices reduce CO2 and GHG emissions, while low prices exert the opposite influence. However, there are several other effects via the fuel mix, electricity generation, ETS price adaptations with a given cap and CCS incentives that modify the overall effect. In total, differences in world energy prices exert only a minor influence on total GHG emissions in the EU given the existence of the EU ETS with a decreasing cap that is independent from GDP or world energy price developments.

High fossil fuel prices limit business opportunities for energy exporters given that EU imports would decrease, especially for natural gas. Conversely, with lower fossil fuel prices, significantly higher gas deliveries to the EU can be assumed. Import dependency increases with low world energy prices, whereas it stays below the Reference scenario in the High price case. Electricity prices are significantly lower in the Low price case, whereas they are significantly higher in the High energy price case. High energy import prices increase the EU’s external fuel bill substantially. On the contrary, lower fossil fuel prices give a boost to the EU economy improving its competitiveness, also through lower costs and inflation.

209.

2.4.5. Conclusion


The Reference scenario and CPI assume the overall GHG target, ETS cap and non-ETS national targets to be achieved by 2020 but thereafter GHG reductions fall short of what is required to mitigate climate change with a view to reaching the 2 °C objective. Import dependency, in particular for gas, increases over the projection period and electricity prices and energy costs are rising. So despite efforts over recent years, the long term effects of our current and planned policies are not sufficient to achieve the ambitious decarbonisation objective and to improve both security of supply and competitiveness. These conclusions are broadly consistent with other major stakeholder work such as the IEA World Energy Outlook 2010 (Current Policies scenario), the European Climate Foundation (baseline scenario); Power Choices (baseline scenario) and Greenpeace (baseline scenario). A more thorough comparison of stakeholder work is provided in Annex 2.

210.

2.5. The EU's right to act and EU added-value


The EU's competence in the area of energy is set out in the Treaty on the Functioning of the European Union, in Article 194[44]. EU competences related to combating climate change, including GHG emission reductions in energy and other sectors, are enshrined in Art. 191-193. The EU's role needs to respect the principles of subsidiarity and proportionality.

From an economic perspective, as is the case with the European carbon market, many energy system developments can best be achieved on an EU-wide basis, encompassing both EU and Member State action while respecting their respective competences. An EU wide European market can facilitate the balancing of the electricity system, reduce the need for back-up capacities and encourage RES production where it economically makes most sense. Large scale investments require big markets which also justify one EU wide approach. A bigger market can also better encourage the development of innovative products and systems mainly in the area of energy efficiency and renewables.

211.

2.6. Who is affected?


Everybody is affected. Energy consumers will be affected by higher energy costs (a combination of energy prices and amount of energy used) as well as by extra non-energy investment needed such as more efficient appliances, new types of vehicles, house renovations, etc. The energy industry will be directly concerned as it needs to heavily invest in the next two decades. Public authorities will also need to engage in discussions about the pros, cons and trade-offs of different options as each generation source has its drawbacks (solar and wind generation will require significant infrastructure investments; supply of sustainable biomass might be limited; nuclear faces public acceptance and waste problems and CCS still requires large-scale experience to be able to reduce costs and sufficiently decrease financial risks for private investors). Changes in the EU energy sector will also have a strong influence on third countries, notably fuel suppliers.

212.

3. Section 3: Objectives 3.1. General objective


The general objective is to shape a vision and strategy of how the EU energy system can be decarbonised by 2050 while taking into account the security of supply and competitiveness objectives.

213.

3.2. Specific objectives


To achieve the general objective, more specific objectives are being proposed:

– Assist political decision making for providing more certainty to investors as regards possible future policy orientations at the EU level by showing different decarbonisation pathways to 2050 as well as their main economic, social and environmental impacts;

– Show trade-offs among policy objectives as well as among different decarbonisation pathways and identify common elements in all decarbonisation pathways;

– Help policy makers set milestones after 2020.

The Roadmap 2050 should be based on the current key objectives of EU energy policy – sustainability, security of supply and competitiveness. Not all three objectives can be specified and quantified in the same manner. While the decarbonisation objective can be relatively easily defined and quantified, the other two are more complex. The goal of sustainability is linked in particular to the achievement of 80% domestic GHG reduction below 1990 in 2050, which implies a reduction of energy related CO2 emissions by 85%, consistent with the required contribution of developed countries as a group to limit global climate change to a temperature increase of 2ºC compared to pre-industrial levels. The goal of security of supply entails not only decreasing import dependency but also increasing supply diversity and continued stability of electricity grid. The competitiveness objective implies assuring a competitive energy sector, encouraging investments and achieving affordable energy costs for consumers as well as developing new technologies and ensuring a competitive clean technology manufacturing sector.

In general the objectives of energy policy are complementary and mutually reinforcing. For example, increased energy efficiency reduces GHG emissions, increases energy security and contributes towards achieving a competitive energy sector. A significant part of low carbon energy supply can be produced in the EU, thus also increasing energy security of supply. However, there are also some possible trade-offs. Some of them are presented below for illustration:

– Renewables do not require fuels to be imported and emit less or no GHG emissions, but may need public support (if necessary and proportionate) to be competitive; this increases costs to consumers. The merit order effect however reduces wholesale electricity prices.

– Although nuclear is a large provider of low carbon electricity in the EU, it faces in some MS acceptance and financing problems.

– CCS prevents CO2 emissions, but is comparatively resource inefficient in relation to unabated fossil fuel combustion. Up to 25% additional energy input may be needed for capture, transport and storage of CO2.

– Gas is the fossil fuel with the lowest carbon content but poses a challenge to security of supply especially for countries with undiversified supplies.

– The current tariff-setting for transmission and distribution networks is cost-based and should assure the lowest short term prices to consumers but is not yet supportive enough to new technologies enabling integration of RES and energy efficiency that have longer term benefits.

214.

3.3. Consistency with other European policies


The Energy Roadmap 2050 subscribes into the overall framework of decarbonisation as designed by the flagship initiative Resource efficient Europe and the Roadmap for moving to a competitive low carbon economy in 2050. All objectives are coherent with the objectives of the medium term strategy as described in the Communication Europe 2020 and Energy 2020 as well as with energy policy objectives as described in the Lisbon Treaty.

215.

4. Section 4: Policy options 4.1. Methodology


This is not a typical impact assessment in that it does not list policy options to meet certain policy objectives and then assesses impacts of these policy options to determine a preferable one. It rather examines a set of possible alternative future developments to get more robust information on how the energy system could achieve 85% reduction of energy related CO2 emissions compared to 1990 without selecting one of them as the preferred option. Nor does it seek to justify the decarbonisation target as this was the focus of the Low Carbon Economy Roadmap[45] . It is mainly concerned with analysing possible energy related pathways to reach decarbonisation in a 'global climate action' world. Lower import fossil fuel prices are introduced to reflect significant impacts on global fossil fuels prices in policy scenarios while fossil fuel prices are higher in the Reference scenario and CPI scenarios which project current trends and policies[46].

The Energy Roadmap assumes the implementation of the European Council's decarbonisation objective that includes similar efforts by industrialised countries as a group. The analysis presented focuses on energy consequences. A more comprehensive analysis of different global paths to decarbonisation was presented in the Low Carbon Economy Roadmap 2050[47], exploring the impacts of three global climate situations: a) business as usual; b) global climate action and c) fragmented action. Fragmented action assumes strong EU climate action that is however followed globally only by the low end of the Copenhagen pledges up to 2020 and afterwards the ambition level of the pledges is assumed to stay constant. It analyses impacts on energy intensive industries (EII) both in a global macroeconomic modelling framework to address carbon leakage issues and by means of energy system modelling to address the effects of fragmented action, including electricity costs for companies. Electricity costs are, in fact, higher in the fragmented action scenarios as compared to the global action scenarios due to higher energy import prices. On the other hand, carbon prices are lower under fragmented action.

A 'fragmented' action scenario including measures against carbon leakage was not analysed in this IA report as the challenges for the energy sector arising from decarbonisation are the biggest under the 'global climate action' assumption, given that fragmented action with measures against carbon leakage will deliver lower GHG reductions by 2050. Decarbonisation scenarios that accommodate action against carbon leakage under fragmented action could either go for lower ambitions in terms of GHG reduction for sectors with relevant leakage risks or could have measures included that compensate efforts for energy intensive industries. With action on carbon leakage the challenge for the transition in the energy system could be smaller given lower efforts in parts of the system. Such results are however modified through countervailing effects from lower world fossil fuel prices under global action that encourage somewhat higher energy consumption and emissions. In any case, the implementation of measures will be crucial. The real difference for industrial and thereby climate policy might come from the concrete design of policy instruments that is not discussed in this the Energy Roadmap Impact Assessment (e.g. special provisions on ETS for EII).

Section 5 provides an assessment of the environmental, economic and social impacts that is proportionate to the nature of the document proposed. The assessment is supported by modelling results and/or by academic research where possible. It is important to underline that modelling results are tentative and present impacts as illustrations rather than as conclusive evidence. A 40-year outlook is naturally steeped in uncertainty. Whereas some parameters such as population growth can be projected with a reasonable degree of confidence, the projection of other key factors such as economic growth, energy prices or technological developments over such a long time span incorporates a great deal of uncertainty.

The modelling framework used for decarbonisation scenarios is the same as for the Reference scenario (see section 2.4 and annex 1). A quantitative methodology is the core of this assessment. However, not all aspects could be modelled. For instance, significant environmental impacts that go beyond GHG emissions, such as impacts on biodiversity and air pollution, were not assessed quantitatively. For GDP and employment impacts, analysis done for the Communication on moving beyond 20% GHG reductions[48] and several recent studies were used. It was neither possible to assess impacts on different household income levels, nor distributional impacts at Member State level.

The methodology factors in uncertainties but ensures for a coherent approach based on proven technologies, applying the following limitations:

– Taking into account existing physical and capital infrastructure and limitations regarding physical and capital stock turn-over.

– Technological progress over time is assumed as typical in long term modelling. Potential break-through technologies depending on unforeseeable structural change have not been taken into account. Similarly, major lifestyle changes, beyond demand side effects of carbon pricing on behaviour, have not been taken into account in quantitative terms, as this goes beyond the capabilities of the quantitative modelling tools. [49]

– The modelling also could not take into account effects of the changing climate itself on the energy system. Effects can go in different directions and will depend on how climate changes in different parts of the EU (e.g. more demand for cooling, less demand for heating, impact on water availability for power plant cooling or hydroelectricity production).

Only by comparing results from different decarbonisation scenarios is it possible to extract more robust conclusions, how key parameters influence the results and how various parts interact with each other. By requiring similar levels of cumulative GHG emissions across scenarios, this analysis ensures comparability, as regards the objective of decarbonisation, given that emission mitigation aims at preventing dangerous levels of atmospheric GHG concentrations that is a matter of cumulative emissions. An identification of common features to all scenarios will be an important part of the analysis. The Commission's own scenario analysis will be complemented by MS and other stakeholders' work. An in-depth impact assessment report examining impacts of concrete policy measures will be submitted for any legislative proposal following this roadmap.

216.

4.2. Policy options


Several useful scenarios could be proposed for a decarbonisation analysis of the energy system. The design of scenarios was extensively discussed with various stakeholders. Stakeholders and the European Commission identified four main decarbonisation routes for the energy sector – energy efficiency impacting mostly on the demand side and RES, nuclear and CCS predominantly on the supply side (lowering the carbon intensity of supply). This finding is in line with the decarbonisation scenarios of a number of stakeholders, such as Eurelectric Power Choices, the Energy Roadmap of the European Climate Foundation and the work done at national level by some MS (such as the UK, DE and DK). The policy options (scenarios) proposed explore five different combinations of the four decarbonisation routes. Decarbonisation routes are never explored in isolation as the interaction of different elements will necessarily be included in any scenario that evaluates the entire energy system.

All decarbonisation scenarios achieve close to 85% energy related CO2 emissions by 2050 and it is carefully assessed what effect each policy option has in terms of security of supply, competitiveness of the energy sector and affordability of energy costs. All scenarios use the same assumptions about GDP developments as the Reference scenario. The scenarios achieving the European Council's GHG objective have lower fossil fuel prices as a result of lower global demand for fossil fuels reflecting worldwide carbon policies (oil price is 84 USD'08 per bbl in 2020; 79 in 2030 and 70 in 2050). In addition, most technology assumptions are the same as in the Reference scenario, although there are additional features and mechanisms to stimulate decarbonisation and technology penetration. For details please see Annex 1, pages 56-60.

217.

Table 4: Policy options/Scenarios


|| Option/scenario| Short description

Business as usual (Reference scenario[50])| The Reference scenario includes current trends and long-term projections on economic development (GDP growth 1.7% pa). It takes into account rising fossil fuel prices and includes policies implemented by March 2010. The 2020 targets for GHG reductions and RES shares will be achieved but no further policies and targets after 2020 (besides the ETS directive) are modelled. See also section 2.4 Sensitivities: a) a case with higher GDP growth rates, b) a case with lower GDP growth rates, c) a case with higher energy import prices, d) a case with lower energy import prices.

1bis| Current Policy Initiatives – CPI scenario (updated Reference scenario)| The Reference scenario includes only adopted policies by March 2010. Since then, several new initiatives were adopted or are being proposed by the EC. The EC outlined its future work programme on energy mainly until 2020 in the Communication 'Energy 2020 - A strategy for competitive, sustainable and secure energy'. This policy option analyses the extent to which measures adopted and proposed will achieve the energy policy objectives.[51] It includes additional measures in the area of energy efficiency, infrastructure, internal market, nuclear, energy taxation and transport. Technology assumptions for nuclear were revised reflecting the impact of Fukushima and the latest information on the state of play of CCS projects and policies were included. See also section 2.4.

|| Decarbonisation scenarios| All decarbonisation scenarios build on Current Policy Initiatives (reflecting measures up to 2020) and are driven by carbon pricing to reach some 85% energy related CO2 reductions by 2050 (40% by 2030) which is consistent with the 80% reduction of GHG emissions. Transport measures (energy efficiency standards, low carbon fuels, infrastructure, pricing and transport planning) as reflected in the Transport White Paper are included in all scenarios. All scenarios will reflect significant development of electrical storage and interconnections (with the highest requirements in the High RES scenario). Different fuels can compete on a market basis besides constraints for nuclear investment in scenario 6.

High Energy Efficiency| This scenario is driven by a political commitment of very high primary energy savings by 2050 and includes a very stringent implementation of the Energy Efficiency plan. It includes further and more stringent minimum requirements for appliances and new buildings; energy generation, transmission and distribution; high renovation rates for existing buildings; the establishment of energy savings obligations on energy utilities; the full roll-out of smart grids, smart metering and significant and highly decentralised RES generation to build on synergies with energy efficiency.

Diversified supply technologies[52]| This scenario shows a decarbonisation pathway where all energy sources can compete on a market basis with no specific support measures for energy efficiency and renewables and assumes acceptance of nuclear and CCS as well as solution of the nuclear waste issue. It displays significant penetration of CCS and nuclear as they necessitate large scale investments and does not include additional targeted measures besides carbon prices.

High RES| The High RES scenario aims at achieving a higher overall RES share and very high RES penetration in power generation, mainly relying on domestic supply[53].

Delayed CCS| This scenario follows a similar approach to the Diversified supply technologies scenario but assumes difficulties for CCS regarding storage sites and transport while having the same conditions for nuclear as scenario 3. It displays considerable penetration of nuclear.

Low nuclear| This scenario follows a similar approach to the Diversified supply technologies scenario but assumes that public perception of nuclear safety remains low and that implementation of technical solutions to waste management remains unsolved leading to a lack of public acceptance. Same conditions for CCS as scenario 3. It displays considerable penetration of CCS.

A more detailed presentation of assumptions for all scenarios can be found in Annex 1.

218.

5. Section 5: Analysis of impacts 5.1. Environmental impacts


Energy consumption and use of renewable energy

Primary energy consumption is significantly lower in all decarbonisation scenarios as compared to the Reference scenario. The biggest decline of primary energy consumption comes in the High Energy Efficiency scenario (-16% in 2030 and -38% in 2050) showing the effects of stringent energy efficiency policies and smart grid deployment. The decrease in energy consumption compared with the Reference scenario for all decarbonisation scenarios spans a range from 11-16% in 2030 and 30-38% in 2050. Compared with primary energy consumption in 2005 there is a very significant decrease of 32-41%. It is important to note that these levels of reduced primary energy demand do not come from reduced GDP or sectoral production levels (which remain the same in all scenarios). Instead they are mainly the result of technological changes on the demand and supply side, coming from more efficient buildings, appliances, heating systems and vehicles and from electrification in transport and heating. All decarbonisation scenarios over-achieve the 20% energy saving objective in the decade 2020-2030[54]. This result is consistent with other stakeholder work.

Not only the amount, but also the composition of energy mix would differ significantly in a decarbonised energy system. Low carbon energy sources are strongly encouraged but can follow various decarbonisation routes shown by rather wide ranges for shares of energy sources in primary energy while all satisfying the decarbonisation requirement by 2050. Moreover, all decarbonisation routes achieve the same cumulative GHG emissions in 2011- 2050.

219.

Table 5: Fuel shares in primary energy consumption


Renewables increase their share in primary energy substantially in all decarbonisation scenarios to reach at least 22% by 2030 and at least 41% by 2050. The RES share in primary energy is the highest in the High RES scenario (60% in 2050). The RES share is higher when calculated in terms of gross final energy consumption[55]- it represents at least 28% (2030) and 55% (2050) in all decarbonisation scenarios and rises up to 75% in 2050 in the High RES scenario. The share of renewables in power generation stands at 86% in 2050 in the High RES scenario and the share in power consumption is even higher at 97% in 2050.[56] RES share in power generation can be further increased by allowing for imports of renewable electricity from North Africa.

Nuclear developments have been affected by the policy reaction in some Member States after the nuclear accident in Fukushima. The share of nuclear varies depending on policy assumptions. In the Low nuclear scenario the nuclear share declines gradually to 3% by 2050. In the most ambitious nuclear scenario (Delayed CCS scenario), the share rises to 18%.

The share of gas is higher in the Current Policy Initiatives scenario compared to the Reference scenario, partly replacing nuclear. It increases slightly by 2050 in the Low nuclear scenario where the CCS share in power generation is around 32%. The oil share declines only slightly until 2030 due to the high dependency of transport on oil. However, the decline is significant in the last decade (2040-2050) when oil in transport is to a large extent replaced by biofuels and electricity. The share of solid fuels shrinks further to reach only 2-6% in all decarbonisation scenarios except in the Low nuclear scenario (10% in 2050).

Final energy demand declines similarly to primary energy demand. In the High Energy Efficiency scenario the reduction compared to the Reference scenario is -14% in 2030 and -40% in 2050. The decrease in the decarbonisation scenarios is at least -8% in 2030 and -34% in 2050. Sectors showing higher reductions than the average are residential, tertiary and generally also transport. There is a lot of structural change in the fuel composition of final energy demand. Given that it is highly efficient and emission free at use, electricity makes major inroads already under Current Policy Initiatives (increase by 9 pp in 2005-2050). The electricity share soars further in the decarbonisation scenarios reaching 36% - 39% in 2050 (almost doubling from current levels and becoming the most import final energy source), reflecting also its important role in decarbonising heating and transport. The crucial issue for any decarbonisation strategy is therefore the full decarbonisation of power generation.

Energy intensity reduces by at least 67% in the Delayed CCS scenario (2005-2050). It reduces by 70% in the High RES and Low nuclear scenarios and by 71% in the Energy Efficiency scenario in 2005-2050 (against a 53% improvement in the Reference scenario).

220.

Emissions


All decarbonisation scenarios achieve 80% GHG reduction and close to 85% energy related CO2 reductions in 2050 compared to 1990 as well as equal cumulative emissions over the projection period. In 2030, energy-related CO2 emissions are between 38-41% lower, and total GHG emissions reductions are lower by 40-42%.

221.

Impacts on biodiversity, air pollution and other environmental impacts


The ranking of the different policy options as regards impacts on biodiversity, air pollution, water use and other environmental impacts depends on the implementation of different energy mixes. Some overall trends are presented below while some impacts are analysed in the Resource Efficiency Roadmap 2050 but with much less focus on energy.

In most scenarios, air pollution can be expected to decrease significantly, as this often goes hand in hand with GHG emissions. However, in some cases (especially if the energy mix leads to the development of small unregulated biomass plants), particulate matter (PM) and gaseous emissions could rise, causing local air pollution and regional acidification issues, although the overall effects can be expected to remain positive[57].


All options will impact land use and consequently biodiversity and other land-related ecosystem services. Indeed, any new infrastructure, be it in terms of grid development, power plant installations (nuclear, CCS, fossil), renewable infrastructure (sitting of wind mills, hydropower dams) will lead to land use changes and fragmentation, with potential negative impacts on biodiversity and on the services we receive from ecosystems. However, if the infrastructure development follows well established environmental rules, these potentially negative consequences can be limited[58]. Therefore, the pathways as such do not necessarily lead to land use and biodiversity problems, as this will depend on implementation. Consequences of mostly domestic RES are presented in terms of needs for domestic biomass[59] giving details for each scenario on the total use of biomass and biofuels in transport). The maximum amount of biofuels in 2050 would reach 300 Mtoe for use within the EU and 20 Mtoe for bunkers. The other decarbonisation scenarios have around 270 Mtoe including bunkers.[60] Still, there are also impacts of CO2 emissions related to land use, land use change and forestry due to increased bioenergy use.[61] As the biomass needed for energy will not only come from forests/forest-based industries, biowaste and residues, this will require considerable additional amounts of agricultural land.

In terms of water use, the consequences will depend on the energy mix. New hydropower projects (including pumped storage), the cultivation of some energy crops, and increased demand for water for cooling in the nuclear energy sector might exacerbate existing water shortages, increasing potential impacts on river morphology and groundwater availability, all this in a context of increasing EU temperatures and reduced water availability.

222.

5.2. Economic impacts


Economic growth

The current report is part of a joint Commission analysis related to the transition to a low-carbon economy by 2050. Previous assessment by the Commission shows that the costs by 2020 of putting the EU economy on a path that meets the long-term requirements for limiting climate change to 2°C would be limited compared to business-as usual, at around 0.2%-0.5% of GDP[62], with access to international carbon credits. Using the additional revenues from auctioning CO2 emissions allowances in EU ETS sectors and tax revenues from the non-ETS sectors to decrease labour costs would improve overall macroeconomic results leading to 0.4%-0.6% increase in GDP by 2020.

As regards the differentiated impact of policy options on economic growth, the long-term perspective implies that it is very difficult to go beyond a qualitative assessment. The Reference and CPI scenarios have higher fuel costs which do not generate much economic growth but require fewer investments in new technologies. On the contrary, the decarbonisation scenarios entail much higher investment in equipment and energy efficiency while lowering expenditure on fuels. These investments can generate further GDP growth and technologies may be exported worldwide if the EU keeps its front-runner position. Thus, policy scenarios which drive forward energy efficiency measures and investments in renewable energy technology have the potential to generate new industries, jobs and substantial economic growth. Although it is difficult to assess in details, such investments could also protect the EU economy against external energy price shocks[63].

An assessment of the macro-economic impact of the European decarbonisation objectives towards 2050 was performed in the European Climate Foundation's 2050 Roadmap[64]. It shows an annual GDP growth of 0.1% below the baseline scenario until 2015 but a reversal of the trend afterwards resulting in GDP being 2% above the baseline in 2050. Marginally positive effects remain under different sensitivity cases.

223.

Energy system costs


The total energy system costs are costs for the entire energy system including capital cost, (for energy using equipment, appliances and vehicles), fuel and electricity costs, and direct efficiency investment costs (house insulation, control systems, energy management, etc) [65]. They exclude disutility costs[66] and auction payments[67].

224.

Table 6: Average annual total energy system cost (without auctioning and disutility)


Average annual total energy system costs 2011-|||

Bln. EUR'| Ref| CPI| High Energy effic.| Div. supply techn.| High RES| Delayed CCS| Low nuclear

Capital cost| 1104

Energy purchases| 1311

Direct efficiency inv. costs| 161

Total cost for final consumers excl. all auction payments and disutility| 2552

Absolute Difference to Reference||||||

Bln. EUR'||| High Energy effic.| Div. supply techn.| High RES| Delayed CCS| Low nuclear

Δ Capital cost||| 149

Δ Energy purchases||| -| -| -| -| -312

Δ Direct efficiency inv. costs||| 133

Δ Total cost for final consumers excl. all auction payments and disutility| -| -| -29

Depending on the decarbonisation scenario, there are no or little additional average annual energy system costs due to the pursuit of major decarbonisation as part of a global effort compared with the Reference and CPI scenarios. Diversified supply technologies and Delayed CCS scenarios have the lowest level of average annual energy system costs, representing even a cost saving of around 90 bn €(08) compared with CPI (around 50bn € compared to the Reference scenario) mainly due to large fossil fuel import savings. Those two scenarios have the highest nuclear share[68].

The modelling results suggest that the highest total energy system costs will occur in the High Energy Efficiency scenario. Unlike the majority of other scenarios, the modelling of the High Energy Efficiency scenario does not rely entirely on economic optimisation in determining the level of energy consumption, but rather projects the impact of a set of energy efficiency measures (building retrofit etc.). In addition, the scenario pushes the limits of what the chosen measures can achieve (by assuming that the whole European building stock is fully refurbished; by making use of distributed renewable energy solutions as one of the more expensive renewable energy solutions; by amortising long-lived measures over a short time). Furthermore, it has to be taken into account that all policy scenarios already include considerable energy efficiency improvements and the cost difference merely indicates an increasing marginal cost for moving from a high to a very high level of energy efficiency (see Annex 1, part B for details). Finally, the modelling reflects significant transaction costs for energy efficiency investments in the form of relatively high weighted average costs of capital.

Cumulative auction payments are lowest in the High Energy Efficiency scenario due to the reduced energy consumption, decreasing emissions and therefore the necessity to buy ETS permits. The scenario with the highest auction revenues is Delayed CCS where the delay in the use of CCS leads to high carbon prices to ensure the achievement of the decarbonisation target in later years, which is made more challenging by the fact that CCS has not been able to move down the cost curve earlier. The auction revenues represent an equivalent of around 1% of total cumulative energy system costs.

All scenarios show higher annual costs in the last two decades 2031-2050 reflecting mainly increased investments in transport equipment as the major transition to electric and plug in hybrids vehicles is projected after 2030. In the High RES scenario costs are also linked to significant expansion of RES based power generation capacity.

The ratio of energy system costs to GDP is similar across the scenarios: ranging from around 14.1% to 14.6%, the costs of the Diversified supply technologies and delayed CCS scenarios being at the lower end of the range.

225.

Table 7: Cumulative system costs related to GDP 2011-2050


| Cumulative system costs related to GDP

Reference| 14.37%

CPI| 14.58%

High Energy Efficiency| 14.56%

Diversified supply technologies| 14.11%

High RES| 14.42%

Delayed CCS| 14.06%

Low nuclear| 14.21%

The external fuel bill arising from the net imports of fossil fuels decreases below 2005 levels in all decarbonisation scenarios by 2050. This result stems from the pursuit of major decarbonisation as part of a global climate effort with fossil fuel import prices expected to be much lower. The actual imports of fossil fuel due to energy efficiency and penetration of RES will be much lower too. These combined effects reduce the expenditure for each fossil fuel and thereby the total external fuel bill of the EU. The decrease of the fuel bill from 2005 in the decarbonisation scenarios is smallest in the Low nuclear scenario at 31% and highest in the High RES scenario at 43% with RES replacing most fossil fuels. Compared with the current level, all decarbonisation scenarios increase their fuel bill in 2030, but to much lower levels than the Reference and CPI scenarios. Savings in the external fuel bill are most striking in 2050. Compared with the CPI scenario, the EU economy could save in 2050 between 518 and 550 bn € (08) by taking this strong decarbonisation route under global climate action.

226.

Impacts on competitiveness


Average prices of electricity are rising compared to 2005 in all scenarios including Reference and CPI scenarios (by a range of 41% in the High Energy Efficiency scenario to 54% in the Low nuclear scenario in 2030 and by 34% in the Diversified supply technologies to 82% in the High RES scenarios in 2050). Electricity prices are calculated in such a way that total costs of power generation, balancing, transmission and distribution are recovered, ensuring that investments can be financed. The residential sector has the highest user price and industry the lowest as is currently the case. Decarbonisation scenarios have lower fuel costs but tend to have higher capital investment costs that offer more business opportunities for domestic investments instead of fuel imports.

Due to depressed demand for electricity, the High Energy Efficiency scenario shows the lowest prices in 2030 for all sectors – even slightly lower than in the Reference scenario (which however exhibits a significant price increase from today's level). In 2050, electricity prices are lowest in the Diversified supply technologies scenario for all sectors, except industry, which faces slightly higher prices compared with the Reference and Current Policy Initiatives. In 2050, average electricity costs are highest in the High RES scenario while the Low nuclear scenario has the highest prices in 2030.

In this exercise, potential macroeconomic benefits from the development of 'green technology' manufacturing and services sectors have not been quantified for the various policy scenarios.

227.

Energy related costs for companies


Electricity prices for industry are the lowest among all sectors. The lowest increase occurs in the Diversified supply and Delayed CCS scenarios and the highest increase, similarly to average prices developments, in the High RES scenario. As the whole analysis is performed under the hypothesis of 'global climate action', the whole world would decarbonise and would have to bear carbon prices, so the question of industrial competitiveness would not arise. More information on electricity costs is provided in Annex 1 (part B, point 2.7). If no global climate deal is reached and the EU is reducing emissions significantly more than other countries, certain industries supplying low carbon technologies will benefit from improved competitiveness due to higher internal demand and first mover advantage. However, for energy intensive industries it would be difficult to realise the prescribed GHG reductions without affecting their international competitiveness through higher carbon, fuel and electricity prices. This might be even more pronounced if reductions need to be achieved with CCS, which is a technology that has no other benefits than reducing GHG emissions.

Energy related costs in relation to sectoral value added rise from 5.8% in 2005 to 7.8% in 2030 in the Reference/CPI cases and to around 7.5% in the decarbonisation scenarios. In 2050, under current policies, this indicator declines to 7.5% and even more so in the decarbonisation scenarios falling to less than 7%.

Energy intensive industries face particularly high energy costs for their highly energy consuming production processes. Energy related costs in relation to sectoral value added for five industrial sectors (iron and steel, non-ferrous metals, non metallic mineral products, chemicals, paper and pulp industries) would rise under current trends, but would be markedly lower under global decarbonisation. Following lower world energy prices and due to energy efficiency improvements, the ratio of energy costs to value added would return to the 2005 level by 2050 in most decarbonisation scenarios, except for the Energy Efficiency scenario, which exhibits an even lower ratio.

228.

ETS carbon prices


The ETS allowance price rises moderately from the current level until 2030 and significantly in the last two decades providing support to all low carbon technologies and energy efficiency. After 2020, the same carbon value applies also to non- ETS sectors assuring cost-efficient emissions abatement in the whole economy post 2020. Concrete policy measures such as those pushing energy efficiency and/or those enabling penetration of renewables depress demand for ETS allowances which subsequently lead to lower carbon prices. Carbon prices are the lowest in the High Energy Efficiency scenario with lowest energy demand followed by the High RES scenario (in 2030 and 2040) and Diversified supply technologies[69] (in 2050). Delay in penetration of technologies (CCS) or unavailability of one decarbonisation option (nuclear) put an upwards pressure on demand for allowances and ETS prices.

Table 8: ETS prices in €'08/t CO2

|| 2050

Reference| 50

CPI| 51

High Energy Efficiency| 234

Diversified supply technologies| 265

High RES| 285

Delayed CCS| 270

Low nuclear| 310

229.

Impacts on infrastructure


Infrastructure[70] requirements differ between scenarios. Decarbonisation scenarios require increasingly more sophisticated infrastructures (mainly electricity lines, smart grids and storage) than Reference and CPI scenarios. The High RES scenario necessitates additional DC lines mainly to transport wind electricity from the North Sea to the centre of Europe and more storage.

Table 9: Grid investment costs (investments in transmission grid including interconnectors and investments in distribution grid including smart components).

(Bln Euro '05)| 2011-| 2021-| 2031-| 2011-2050

Reference| 1269

CPI| 1357

High Energy Efficiency| 1518

Diversified supply technologies| 1712

High RES| 2195

Delayed CCS| 1717

Low nuclear| 1793

The model assumes that grid investments, that are prerequisites to the decarbonisation scenarios in this analysis, are undertaken and that costs are fully recovered in electricity prices. Reality might differ in the sense that the current regulatory regime might be more short to medium term cost minimisation oriented and might not provide sufficient incentives for long-term and innovative investments. There might also be less perfect foresight and lower coordination of investments in generation, transmission and distribution as the model assumes.

230.

Impacts on internal market and competition


Electricity markets might change substantially with an increasing share of generation with close to zero marginal costs. A competitive market would in this situation lead to almost zero prices which would be insufficient to pay for upfront capital investments[71]. A different market design might be needed. While a specific regime for RES (e.g. feed-in tariffs) may be justified in certain situations (e.g. for new RES which are not yet competitive), every effort is needed to ensure that RES is integrated into the energy market, through support, regulatory and infrastructure policies. This is even more the case when RES becomes a significant share of overall energy production (especially in the high RES scenario).

Innovation and R&D

A goal of the Europe 2020 strategy[72] (underpinned by the Communication on the Innovation Union[73]) is to increase innovation in Europe and focus R&D and innovation policies on tackling major societal challenges such as climate change. The EU27 is already a world leader in some segments of low-carbon and energy efficient technologies (nuclear power plants, wind turbines, some energy efficient appliances, etc). All policy scenarios involve significant improvement in efficiency and cost parameters of new technologies as compared to the Reference scenario due to more economies of scale and faster learning rates. The deployment of CCS and some RES in the decarbonisation scenarios, for instance, implies a rate of capacity growth and innovation that is at least as great as that seen for energy technologies in the 20th century[74]. As a consequence all policy options are expected to further boost research and innovation, thereby also improving competitiveness. However, the magnitude of innovation between different policy options might differ. Moreover, impacts expected on innovations can hardly be grasped by current models.

231.

Impacts on third countries


Impacts on third countries, mainly oil and gas importing countries would be significant. Imports in decarbonisation scenarios decrease sharply (besides gas imports in the Low nuclear scenario). In addition, global decarbonisation efforts lead to lower fossil fuel prices. So, under these particular circumstances the export revenues from European customers are 31 to 43% lower in 2050 than in 2005. In the mid-term, in 2030 all decarbonisation scenarios have a higher fuel bill compared to 2005 by at least 35%, but to much lower levels than the Reference and Current Policy Initiative scenarios[75]. (See also section on Energy system costs).

There is no major impact on electricity trade, which remains marginal with third countries. The increased global use of biomass for energy purposes might have impacts on food prices and input costs of other biomass-using industries.

232.

Impacts on prices for biomass and land prices


Bioenergy is expected to be an important part of any low-carbon energy strategy. This might have impacts on prices for biomass from agriculture and forest-based industries either directly through increased demand for energy use, or through increased demand for land and thus higher land prices. As most of the biomass used for energy has competing uses (food and feed, renewable raw materials), food prices and input costs of other biomass-using industries are likely to increase.

233.

5.3. Social impacts


Impacts on employment

The social dimension of decarbonisation is crucial as transition to a low carbon economy will require an in depth change in several sectors, affecting companies, employment and working conditions. Education and training need to be addressed at an early stage in order to avoid unemployment in some sectors and labour shortages in others. More knowledge should be gathered about the social implications of deep and long-term decarbonisation as no studies are available yet. Consultations, also in the context of the social dialogue, could improve the follow-up work on the decarbonisation roadmaps[76], including decarbonisation of the energy sector.

Employment effects of decarbonisation policies up to 2020 are generally ambiguous and difficult to assess. A direct positive effect of relative growth in the 'green' technology sector is that some subsectors like energy efficiency in buildings are usually assumed to have a relatively high labour intensity. Indirect positive effects for employment may include increased innovation resulting from stricter environmental policy, increased export potential for green technologies, as well as less fossil fuel imports. Negative effects may include transition costs, such as inflexibilities in the labour market to respond to changes in skill demand. There is uncertainty as to whether positive or negative effects would dominate.

However, most studies that evaluate the net employment effects of the EU's 20-20-20 targets record impacts of typically ±1%[77]. A recent extensive macroeconomic study suggests that net employment effects for meeting the EU's targets for 2020 will be small and positive, leading to an average increase in employment demand of up to 0.3%[78]. The two scenarios with the most ambitious targets (30% GHG emission reductions by 2020, achieving the 20% energy efficiency target) have the highest net effects on employment. Similarly, a 2009 study[79] finds modestly positive net employment effects of up to 0.1% for supporting policies to meet the 2020 RES targets. An assessment of net employment effects of the European decarbonisation objectives towards 2050 was performed in the European Climate Foundation's 2050 Roadmap[80]. It expects net employment to initially be marginally negative and turn positive at a later stage: employment in the decarbonisation scenario is 0.06% below the baseline by 2020 and 1.5% higher than the baseline in 2050. An estimate of net employment effects until 2030 and some quantitative examples of job creation in certain sectors are provided in the IA report on Low Carbon Economy Roadmap[81].The net impact on jobs can be an increase by 0.7% compared to the Reference scenario, corresponding to 1.5 million jobs by 2020.

The overall effects of the increased investment in green technologies on the labour market are thus expected to be fairly modest relative to the effects of other developments such as globalisation, technical progress and demographic change. On a sectoral level, a small increase in jobs in the engineering and construction sectors and a decrease in the energy supplying sectors might arise. The effects on the energy-intensive sectors are ambiguous. Higher energy prices may lead to losses in competitiveness on the one hand while there would also be increased demand for goods from the sector (such as steel and concrete) on the other. However, by focussing on sectoral gains and losses, potentially significant impacts at a more micro level may not be captured in these studies. Also, regional differences may be significant.

As the whole analysis was done in a global climate effort context, there are no job losses due to carbon leakage. However the decision by companies to relocate production away from the EU may be related to other factors such as access to markets or raw materials or secure access to energy sources with long-term price guarantees.

234.

Quality of jobs


The more investments are made in new technologies – many of which are likely to be energy saving or related to new forms of energy generation – the more demand there will be for people in higher skilled jobs (especially professional and associate professional ones). In this way, the greening of the economy can stimulate the demand for highly skilled (and high waged) workers, although the extent to which this will occur even under the most optimistic of scenarios is relatively modest when compared to the business as usual scenario.

235.

Affordability


Affordability of energy services as regards costs for fuel and electricity as well as for equipment, appliances, insulation and transport services is one of the essential elements of the analysis. The sector most concerned is households. All decarbonisation scenarios show significant fuel savings compared to the Reference and CPI scenarios but also higher costs for energy appliances and insulation.

Energy related expenditures of households for heating, cooling, lighting, cooking, appliances i.e. excluding transport services, almost double from around 2000 EUR'08 today to 3800-3900 EUR'08 in 2050 in the Reference and CPI scenarios reflecting rising fuel and electricity prices and increasing direct household investments in energy efficiency. Expenditures per household amount to around 4500 EUR'08 in most decarbonisation scenarios in 2050, with expenditure per household reaching some 4800 €(08) and almost 4900 €(08) in the Energy Efficiency and High RES scenarios respectively. It is important to note that per capita income in 2050 will also almost double from today's level, but also that households will be composed of fewer members reflecting aging and changing lifestyles. Energy costs for stationary uses per household exceed the Reference/CPI case level by 16-17% in 2050 in most decarbonisation scenarios. They are 25-27% higher in the Energy Efficiency and High RES scenarios, as these scenarios are particularly investment intensive.

However, energy expenditures including expenses for transport services as a percentage of household expenditure show a different picture. They rise over time in all scenarios from 10% in 2005 to around 16% in 2030, stabilising thereafter to around 15-16% by 2050. Among the decarbonisation scenarios, the costs of the Delayed CCS and the Diversified Supply Technology scenarios, similar to the Reference and CPI scenarios, are at the lower end of this range, whereas the High RES and Energy efficiency scenarios show 2050 costs at the upper end. To the extent that vulnerable consumers would incur similar expenditure increases, in particular the necessary upfront investment to realise later savings may pose an affordability challenge for them.

236.

Security of supply


Import dependency, one of the indicators of security of supply, does not change substantially in 2030 in decarbonisation scenarios compared to the Reference scenario and Current Policy Initiatives scenario due to declining gross inland consumption and imports. There is however a substantial decrease in 2050, driven by increased use of domestic resources, mainly renewables. Import dependency is only 35% in the High RES scenario[82] (compared to 58% in the Reference and CPI scenarios) and 39-40% in the other decarbonisation scenarios besides the Low nuclear scenario (45% due to significant use of fossil fuels with CCS). Decarbonisation will significantly reduce fossil fuel security risks.

Large scale electrification combined with more decentralised power generation from variable sources brings other challenges to high quality energy service at any time. However, there are no standardised indicators for the time being. Moreover, adequate stability of the grid is a precondition for modelling, which is why differences in indicators on the stability of the grid are rather small across scenarios[83].

237.

Safety and public acceptance


Safety concerns might be raised against some power generation technologies as well as against infrastructure and exploration of energy fuels. The public in general perceives technological risks as more important than expert judgement would suggest. Across Europe, public acceptance of different generation technologies and infrastructures differs, but none of them is 100% accepted by local communities where they are (going to be) located. A better and more targeted communication with the concerned public and stakeholders might be needed in the future to assure the EU's energy needs.

238.

Table 10: Selected results of scenario analysis


|| Current trends| Decarbonisation scenarios

Reference scenario| Current Policy Initiatives| High Energy Efficiency| Diversified Supply Techno-logies| High Renewables| Delayed CCS| Low nuclear

Primary energy demand reduction (in % from 2005)[84]|| -5.| -10.| -20.| -| -17.| -16.| -18.5

|| -3.| -11.| -40.| -33.| -37.| -32.| -37.7

Electrification| 2030 20.2 -| 25.1 29.| 24.5 29.| 25.2 37.| 26.0 38.| 25.4 36.| 26.0 38.| 25.7 38.5

Fuels (in %)|||||||||

Renewables in gross final energy| 8,| 23.| 24.| 27.| 27.| 31.| 28.8

-| 25.| 57.| 54.| 75.| 55.| 57.5

CCS in power generation| 2.| 0.| 0.| 0.| 0.| 0.| 2.1

-| 17.| 7.| 20.| 24.| 6.| 31.9

Nuclear energy in primary energy| 14,| 14.| 12.| 11.| 13.| 9.| 13.| 8.4

-| 16.| 13.| 13.| 15.| 3.| 17.| 2.6

Fuels in electricity generation (in%) RES CCS NUC| 2030 2050 2030 2050 2030 14.3 - 0.0 - 30.5 -| 40.5 40.3 2.9 17.8 24.5 26.| 43.7 48.8 0.8 7.6 20.7 20.| 52.9 64.2 0.7 20.5 18.6 14.| 51.2 59.1 0.8 24.2 21.2 16.| 59.8 86.4 0.6 6.9 15.8 3.| 51.7 60.7 0.7 19.0 21.5 19.| 54.6 64.8 2.1 31.9 13.4 2.5

Average electricity prices (in EUR'08 per MWh, after tax)[85]| 109,| 154,| 156,| 154,| 159,| 164,| 160,| 168,2

-| 151,| 156,| 146,| 146,| 198,| 151,| 157,2

Annual energy system costs related to GDP (in % 2011 – 2050)|| -| 14.| 14.| 14.| 14.| 14.| 14.| 14.21

Import dependency (in %)| 52,| 56.| 57.| 56.| 55.| 55.| 54.| 57.5

-| 57.| 58.| 39.| 39.| 35.| 38.| 45.1

239.

Source: PRIMES modelling


Table 11: Summary of impacts

|| 1 Reference scenario| 1bis Current Policy Initiatives| 2 High Energy Efficiency| 3 Diversified supply technologies| 4 High RES| 5 Delayed CCS| 6 Low nuclear

240.

Environmental impacts


Energy consumption/Energy intensity||| + + +| +| + +| +| + +

RES share|| +| + +| + +| + + +| + +| + +

Energy related CO2 emissions| =| + + +| + + +| + + +| + + +| + + +

241.

Economic impacts


Economic growth|| =| =| =| =| =| =

Competitiveness|| =| +| +| +| +| +

Energy security (import dependency and imports from third countries)|| =| + +| + +| + + +| + +| +

242.

Social impacts


Employment|| =| + +| +| + +| +| +

Quality of jobs|| =| + +| + +| + +| + +| + +

Affordability|| =| -| =| -| =| =

Legend:

= equivalent to Reference scenario

+ to +++ improvement compared to Reference scenario

- to - - - worsening compared to Reference scenario

243.

5.5 Sensitivity analysis


It is clear that the robustness of modelling results is affected by the assumptions underlying the modelling scenarios. As outlined in section 2.4, sensitivity analysis has been carried out for the Reference scenario by varying two key parameters – GDP and energy import prices. The conclusions on GDP analysis are quite robust showing that key policy indicators do not vary significantly with GDP given feedback mechanisms and the architecture of EU energy and climate policies (ETS). Following this pattern, a similar outcome might be expected for policy scenarios even though it has not been demonstrated by current analysis. This holds also for variations in energy import prices, although the results are somewhat less stable regarding certain indicators, such as import dependency. Impacts of additional variations in import price assumptions in decarbonisation scenarios (very high oil price and oil shock scenarios) were analysed in the Low Carbon Economy Roadmap.

Constant climate conditions were assumed over time. This simplification may be justified given that all decarbonisation scenarios assume that the climate targets are met. However, even when temperature changes are limited to 2 degree Celsius, some climate impacts will occur.[86] In addition, changes in temperature will lead to changes in energy demand patterns for heating and cooling. It can hence be expected that decarbonisation leads to further positive economic impacts with regard to energy security and competitiveness by avoiding parts of the expected damage and adaptation costs in the energy system due to climate change impacts.

Other assumptions are embedded in the design of policy scenarios. Policy scenarios assume different costs and timing of technology (delay of CCS, faster penetration of RES) and can therefore be interpreted as sensitivity analysis on R&D and learning curves for main technologies. Changes in other sectors such as a higher uptake of electricity in transport, were implicitly studied in this report by assuming that the main thrust of the policies included in the 2011 White Paper on Transport is also pursued in these decarbonisation scenarios. No additional transport related policies were examined.

244.

6. Section 6: Comparing the options


This section provides an assessment of how the policy options will contribute to the realisation of the policy objectives, as set in Section 3, in light of the following evaluation criteria:

– effectiveness – the extent to which options achieve the objectives of EU energy policy[87];

– efficiency – the extent to which objectives can be achieved at least cost;

– coherence – the extent to which policy options are likely to limit trade-offs across the economic, social, and environmental domains.

245.

Effectiveness


As regards effectiveness, the three objectives of energy policy – sustainability, security of supply and competitiveness - were taken into account. All policy scenarios were designed to reach 85% reduction of energy related CO2 emissions in 2050, so all are effective in that sense. It should be noted that some scenarios are highly dependent on success of new technologies that are still under demonstration or only partly proven commercially (CCS, off-shore wind, 3rd generation nuclear etc). For the other two objectives the question of most suitable indicators arises. As regards security of supply, all policy scenarios improve import dependency, the best being the High RES scenario with 35% import dependency in 2050 and the least effective the Low nuclear scenario with 45% in 2050 (as compared to 58% in the Reference scenario). However, in a more electrified world, stability of the grid might be of much higher concern with major challenges ahead that can be met as demonstrated by the modelling of the scenarios. As regards competitiveness, some scenarios show a small decrease in electricity prices as compared to the Reference and CPI scenarios (High Energy Efficiency, Diversified supply technologies) while some others show increases (High RES and to a lesser extent Low nuclear). ETS prices are significantly higher than in the Reference and CPI scenarios with the highest values in Delayed CCS scenario and lowest in High Energy Efficiency scenarios where decarbonisation is triggered also by specialised measures. The model triggers adequate investments which are driven by specific policies or carbon prices and investment decisions are based on perfect foresight assumption. All decarbonisation scenarios foster innovation and R&D.

246.

Efficiency


In terms of efficiency, the analysis demonstrates that the costs of decarbonisation of the energy system are not substantially higher compared to the Reference scenario and most decarbonisation scenarios even show a lower annual average cost than the CPI scenario. The least costly scenarios are Delayed CCS and Diversified Supply Technologies scenarios with significant penetration of nuclear.

247.

Coherence


All policy scenarios are coherent with other EU long term objectives (on climate, transport, etc). There is no clear winner among policy options scoring the best in all criteria and several trade-offs will need to be taken into account. The role of this analysis is not to select one preferred pathway but rather to identify the pros and cons of different options and identify common elements from all of them.

248.

Table 12: Comparison of policy scenarios to the Reference scenario


|| 1bis. Current Policy Initiatives| 2. High Energy Efficiency| 3. Diversified supply technologies| 4. High RES| 5. Delayed CCS| 6. Low nuclear

249.

Effectiveness


Sustainability| =| + + +| + + +| + + +| + + +| + + +

Security of supply| =| + +| + +| + + +| + +| +

Competitiveness| =| +| +| +| +| +

250.

Efficiency


Additional annual average total costs relative to Reference scenario in bn EUR'| -| -| -29

Additional annual average total costs as % of GDP| 0.21%| 0. 19%| -0.26%| 0.05%| -0.31%| -0.16%

251.

Coherence


Trade-offs between economic, social and environmental impacts|| Scenario reducing the most energy consumption and significantly improving import dependency but rather costly for households and difficult to implement when it comes to behavioural changes| Scenario with lowest cost from the economic actors' point of view, significant energy efficiency gains and renewables shares but depending on success (technological progress of CCS and some RES as well as public acceptance of nuclear and CCS)| Scenario showing the highest penetration of RES; highest decrease in import dependency and second strongest reduction of energy consumption pushing innovation in new technologies, but rather costly and leading to highest electricity prices| Scenario with lowest costs scoring well on security of supply, RES penetration and competitiveness but the least effective in terms of energy efficiency; rather strong reliance on nuclear being contingent on absence of further public acceptance problems| Scenario scoring well on costs, RES shares and energy efficiency but still with high consumption of fossil fuels and dependency on their imports. Heavily dependent on technological progress and acceptance of CCS

Legend:

= equivalent to Reference scenario

+ to +++ improvement compared to Reference scenario

- to - - - worsening compared to Reference scenario

Conclusions

The Commission services conducted a model-based analysis of decarbonisation scenarios exploring energy consequences of the European Council's objective to reach 80% GHG reductions by 2050 (as compared to 1990), provided that industrialised countries as a group undertake similar efforts. These scenarios explore also the energy security and competitiveness dimension of such energy developments. Businesses as usual projections show only half the GHG emission reductions needed; increased import dependency, in particular for gas; and rising electricity prices and energy costs. Several decarbonisation scenarios highlighting the implications of pursuing each of the four main decarbonisation routes for the energy sector – energy efficiency, renewables, nuclear and CCS - were examined by modelling a high and low end for each of them. The model relies on a series of input assumptions and internal mechanisms to provide the outputs.

252.

The most relevant assumptions and mechanisms of the model


Ø All scenarios were conducted under the hypothesis that the whole world is acting on climate change which leads to lower demand for fossil fuel prices and subsequently lower prices.

Ø The model assumes perfect foresight regarding policy thrust, energy prices and technology developments which assures a very low level of uncertainty for investors, enabling them to make particular cost-effective investment choices without stranded investments. There is also no problem with uncertainty on whether all the infrastructure and other interrelated investment needed to make a particular investment work will be in place in time.

Ø Regulatory framework in model allows for investments to be built and costs fully recovered.

Ø The model assumes a 'representative' or average household or consumer while in reality there is a more diversified picture of investors and consumers.

Ø The model assumes continuous improvements of technologies.

The model-based analysis has shown that decarbonisation of the energy sector is feasible; that it can be achieved through various combinations of energy efficiency, renewables, nuclear and CCS contributions; and that the costs are affordable. The aim of the analysis was not to pick preferred options, a choice that would be surrounded with great uncertainty, but to show some prototype of pathways to decarbonise the energy system while improving energy security and competitiveness and identify common features from scenario analysis.

253.

Common elements to scenario analysis


Ø There is a need for an integrated approach, e.g. decarbonisation of heating and transport relies heavily on the availability of decarbonised electricity supply, which in turn depends on very low carbon investments in generation capacity as well as significant grid expansions and smartening.

Ø Electricity (given its high efficiency and emission free nature at use) makes major inroads in decarbonisation scenarios reaching a 36-39% share in 2050 (almost doubling from the current level and becoming the most important final energy source). Decarbonisation in 2050 will require an almost carbon free electricity sector in the EU, and around 60% CO2 reductions by 2030.

Ø Significant energy efficiency improvements happen in all decarbonisation scenarios. One unit of GDP in 2050 requires around 70% less energy input compared with 2005. The average annual improvement in energy intensity amounts to around 2.5% pa.

Ø The share of renewables rises substantially in all scenarios, achieving at least 55% in gross final energy consumption in 2050, up 45 percentage points from the current level (a high RES case explores the consequences of raising this share to 75%).

254.

Ø The increased use of renewable energy as well as energy efficiency improvements require modern, reliable and smart infrastructure including electrical storage.


Ø Nuclear has a significant role in decarbonisation in Member States where it is accepted in all scenarios (besides Low nuclear and High RES), with the highest penetration in case of CCS delay.

Ø CCS contributes significantly towards decarbonisation in most scenarios, with the highest penetration in case of problems with nuclear investment and deployment. Developing CCS can be also seen as an insurance against energy efficiency, RES and nuclear (in some Member States) delivering less or not that quickly.

Ø All scenarios show a transition from high fuel/operational expenditures to high capital expenditure.

Ø Substantial changes in the period up to 2030 will be crucial for a cost-efficient long term transition to a decarbonised world[88]. Economic costs are manageable if action starts early so that the restructuring of the energy system goes in parallel with investment cycles thereby avoiding stranded investment as well as costly lock-ins of medium carbon intensive technology.

Ø The costs of such deep decarbonisation are low in all scenarios given lower fuel procurement costs with cost savings shown mainly in scenarios relying on all four main decarbonisation options.

Ø Costs are unequally distributed across sectors, with households shouldering the greatest cost increase due to higher costs of direct energy efficiency expenditures in appliances, vehicles and insulation.

Ø The external EU energy bill for importing oil, gas and coal will be substantially lower under decarbonisation due to a substantial reduction in import quantities and prices dependent on global climate action lowering world fossil fuel demand substantially.

Some policy relevant conclusions can be drawn based both on the results of the scenario analysis as well as on a comparison of the hypothetical situation of ideal market and technological conditions needed for modelling purposes and what is found in the much more complex reality.

255.

Implications for future policy making


Ø Successful decarbonisation while preserving competitiveness of the EU economy is possible. Without global climate action, carbon leakage might be an issue and appropriate instruments could be needed to preserve the competitiveness of energy intensive industries.

Ø Predictability and stability of policy and regulatory framework creates a favourable environment for low carbon investments. While the regulatory framework to 2020 is mainly given, discussions about policies for 2020-2030 should start now leading to firm decisions that provide certainty for long-term low-carbon investments. Uncertainty can lead to a sub-optimal situation where only investment with low initial capital costs is realised.

Ø A well functioning internal market is necessary to encourage investment where it is most cost-effective. However, the process of decarbonisation brings new challenges in the context, for example, of electricity price determination in power exchanges: deep decarbonisation increases substantially the bids based on zero marginal costs leading in many instances to prices rather close to zero, not allowing cost recovery in power generation. Similarly, the necessary expansion and innovation of grids for decarbonisation may be hampered if regulated transmission and distribution focuses on cost minimisation alone. Building of adequate infrastructure needs to be assured and supported either by adequate regulation and/or public funding (e.g. financed by auctioning revenues).

Ø Energy efficiency tends to show better results in a model than in reality. Energy efficiency improvements are often hampered by split incentives, cash problems of some group of customers; imperfect knowledge and foresight leading to lock-in of some outdated technologies, etc. There is thus a strong need for targeted support policies and public funding supporting more energy efficient consumer choices.

Ø Strong support should be given to R&D in order to bring costs of low-carbon technologies down and to minimize potential negative environmental and social side-effects.

Ø Due attention should be given to public acceptance of all low carbon technologies and infrastructure as well willingness of consumers to undertake implied changes and bear higher costs. This will require the engagement of both the public and private sectors early in the process.

Ø Social policies might need to be considered early in the process given that households shoulder large parts of the costs. While these costs might be affordable by an average household, vulnerable consumers might need specific support to cope with increased expenditures. In addition, transition to a decarbonised economy may involve shifts to more highly skilled jobs, with a possibly difficult adaptation period.

Ø Flexibility. The future is uncertain and nobody can predict it. That is why preserving flexibility is important for a cost efficient approach, but certain decisions are needed already at this stage in order to start the process that needs innovation and investment, for which investors require a reasonable degree of certainty from reduced policy and regulatory risk.

Ø External dimension, in particular relations with energy suppliers, should be dealt with pro-actively and at an early stage given the implications of global decarbonisation on fossil fuel export revenues and the necessary production and energy transport investments during the transition phase to decarbonisation; new areas for co-operation could include renewable energy supplies and technology development.

256.

7. Monitoring and evaluation


The Roadmap is not a one-off exercise and will be regularly updated taking into account the most recent developments. In addition, the Commission will constantly monitor a set of core indicators which are already available and are being currently used. Other indicators might be added at a later stage.

257.

Table 12: Key indicators and their relevance


Key indicators| Relevance

Share of RES in gross final energy consumption| 10.3% (2008)| Increase in RES use in the economy

Share of renewable energy in transport| 3.5% (2008)| Increase in RES use in the transport

Energy intensity| 165.48 (toe/M€ '00)| Increase in energy efficiency

Gross inland consumption (by fuel)| 1703 Mtoe ec.europa.eu/energy/publications/statistics/doc| Changes in the overall demand and composition of energy mix over time; existing indicative energy saving objective for 2020

Energy per capita| 3403 kgoe/cap| Evolution of energy consumption relative to population growth

Final energy consumption (by fuel and by sector)| 1114 Mtoe ec.europa.eu/energy/publications/statistics/doc| Decrease in absolute energy consumption and effectiveness of energy efficiency policies as well as sectoral developments

Electricity generation| 3210 TWh| Electrification of the economy

Energy related CO2 emissions| 4055 MT CO| Trends in the emissions from the energy sector; lion's share in total GHG emissions

Import dependency for all fuels| 54%| Vulnerability to imports from third countries

Electricity prices| ec.europa.eu/energy/observatory/electricity ec.europa.eu/energy/observatory/reports| Competitiveness of European industry and affordability for households

Diesel and petrol prices in different MS| ec.europa.eu/energy/observatory/oil/bulletin_en| Evolution in prices of transport fuels and their convergence across the EU 27

Total GHG emissions compared to -17.4% ec.europa.eu/clima/policies/g-gas/docs| Meeting climate targets

258.

8. Annexes


Annex Scenarios - assumptions and results

Annex Report on Stakeholders scenarios

Annex 1 Scenarios – assumptions and results

259.

Part A: Reference scenario and its sensitivities and Current Policy Initiatives scenario. 49


260.

1. Assumptions. 49


261.

1.1 Macroeconomic and demographic assumptions. 49


262.

1.2 Energy import prices. 51


263.

1.3 Policy assumptions. 56


264.

1.4 Assumptions about energy infrastructure development 64


265.

1.5 Technology assumptions. 64


266.

1.6 Other assumptions. 73


267.

2. Results. 75


268.

2.1 Reference scenario. 75


269.

2.2 Economic growth sensitivities. 84


270.

2.3 Energy import price sensitivities. 92


271.

2.4 Current Policy Initiatives scenario. 98


This document describes in detail the assumptions and results of the Reference scenario 2050 and its sensitivities, Current Policy Initiatives scenario and decarbonisation scenarios developed for the purposes of the Energy Roadmap 2050.

The Commission contracted the National Technical University of Athens to model scenarios underpinning the Impact Assessment analysis. Similar to previous modelling exercises with the PRIMES model, the Commission discloses a lot of details about the PRIMES modelling system, modelling assumptions and modelling results. In this tradition, the Commission services, based on the modelling results and analysis on specific topics from NTUA, have drafted the following comprehensive overview of the macroeconomic, world energy price, policy, technology and other assumptions as well as the detailed results of the current trend scenarios including sensitivities (Part A) and the various decarbonisation scenarios (Part B). This is complemented with the attachments to this document giving further details.

The PRIMES model was peer-reviewed by a group of recognised modelling experts in September 2011 with the conclusion that the model is suitable for the purpose of complex energy system analysis.

Reference scenario is based on the scenarios up to 2030 published in the report 'Energy Trends to 2030: update 2009', but extends the projection period to 2050. It includes current trends on population and economic development and takes into account the highly volatile energy import price environment. Economic decisions are driven by market forces and technology progress in the framework of concrete national and EU policies and measures implemented until March 2010. These assumptions together with the current statistical situation derived from the Eurostat energy balances represent the starting point for projections which are presented from 2010 onwards in 5 year steps until 2050. The 2020 targets on RES and GHG will be achieved, but there is no assumption on targets for later years. Sensitivities on higher/lower economic growth and higher/lower energy import prices were undertaken in order to assess the robustness of policy relevant indicators with respect to these framework conditions for EU energy policy.

The overall policy context has developed since the Reference scenario was established in 2010. Therefore an additional trend scenario has been modelled including policies that are being prepared with a view to the 2020 Energy Strategy. The Current Policy Initiatives scenario includes the same macroeconomic and demographic assumptions as the Reference scenario, slightly updated energy import prices (only for 2010 with repercussions on 2015), revised cost-assumptions for nuclear following post Fukushima reactions and policies either adopted after March 2010 or being currently proposed by the Commission.

In addition to their role as a trend projection, the Reference and the Current Policy Initiatives scenarios are benchmarks for energy scenarios achieving the European Council's objective to reduce GHG by 80-95% below the 1990 level as part of industrialised countries as a group undertaking such a reduction effort. Comparisons of other scenarios with the Reference scenario concern questions related to the additional policies with respect to those already implemented in the Member States. Distinct from this, comparisons of the Current Policy Initiatives scenario with decarbonisation scenarios address further policies that might be envisaged in addition to those being proposed in the context of the 2020 Energy Strategy. Such comparisons on the basis of the Current Policy Initiatives scenario deal with new policies that might be debated under a 2030 horizon, which is an important milestone year on the decarbonisation pathways to 2050.

Decarbonisation scenarios in the Energy Roadmap 2050 have been designed to provide more detail on the analysis of the energy sector that was presented in the Low Carbon Economy Roadmap. Scenarios showing different energy related decarbonisation pathways reach the 85% domestic energy related CO2 emission reductions by 2050 as compared to 1990 which is consistent with the required contribution of developed countries as a group to limit global climate change to a temperature increase of 2ºC compared to pre-industrial levels. All decarbonisation scenarios developed for the Low carbon Economy Roadmap show around 85% reductions of energy related CO2 emissions.

The scenarios modelled for the 2050 Energy Roadmap investigate in great depth the main strategic directions (energy efficiency, RES, CCS and nuclear) towards a decarbonised European energy system. In doing so, they reflect for each of these directions or main ways of decarbonisation a low and a high end option. This underlines the fact that there are many different pathways for reaching the same level of decarbonisation in the EU.

All numbers included in this report, except otherwise stated, refer to European Union of 27 Member States.

272.

Part A: Reference scenario and its sensitivities and Current Policy Initiatives scenario 1. Assumptions


273.

1.1 Macroeconomic and demographic assumptions


The population projections draw on the EUROPOP2008 convergence scenario (EUROpean POPulation Projections, base year 2008) from Eurostat, which is also the basis for the 2009 Ageing Report (European Economy, April 2009)[89]. The key drivers for demographic change are: higher life expectancy, low fertility and inward migration.

The macro-economic projections reflect the recent economic downturn, followed by sustained economic growth resuming after 2010. The medium and long term growth projections follow the “baseline” scenario of the 2009 Ageing Report (European Economy, April 2009), which derives GDP growth per country on the basis of variables such as population, participation rates in the labour market and labour productivity.[90] Based on the Ageing Report the Commission services developed a common Reference scenario, the macroeconomic part of which is referred to below. Further details relating notably to the sectoral value added can be found in the report 'EU Energy Trends to 2030'.[91] The same macroeconomic assumptions were already used for the 'Roadmap for moving to a competitive low-carbon economy in 2050' of March 2011.[92]

The Reference scenario assumes that the recent economic crisis has long lasting effects, leading to a permanent loss in GDP. The recovery from the crisis is not expected to be so vigorous that the GDP losses during the crisis are fully compensated. In this scenario, growth prospects for 2011 and 2012 are subdued. However, economic recovery enables higher productivity gains, leading to somewhat faster growth from 2013 to 2015. After 2015, GDP growth rates mirror those of the 2009 Ageing Report. Hence the pattern of the Reference scenario is consistent with the intermediate scenario 2 “sluggish recovery” presented in the Europe 2020 strategy[93].

The average growth rate for EU-27 is only 1.2% per year for 2000-2010, while the projected rate for 2010-2020 is recovering to 2.2%, similar to the historical average growth rate between 1990 and 2000. GDP increases in line with the Ageing Report developments, depicting declining growth rates over time as well as great variation among Member States. Recovering from the crisis (reflected by only 0.6% pa GDP growth in 2005-2010), EU-27 GDP is expected to rise 1.7% per annum (pa) from 2010 to 2050, and more specifically by 2.0% up to 2030 and only 1.5% pa after 2030. EU-12 growth is considerably higher in 2010-2030 (2.7% pa) but significantly smaller post 2030 due to shrinking and ageing population (0.9% pa).

The recent economic crisis has added sustainability problems to the public finances. Overall, as an effect of both economic crisis and the ageing of the population, without fiscal consolidation the gross debt-to-GDP ratio for the EU as a whole could reach 100 percent as early as 2014 and 140 percent by 2020[94],[95]. The recent economic crisis might therefore limit the public funding available for low carbon investments.

Sensitivities – Higher and Lower GDP cases

Considering the high degree of uncertainty surrounding projections over such a long time horizon, a sensitivity analysis has been carried out with respect to GDP developments. A high and a low case have been analysed. The GEM-E3 model was deployed to simulate higher and lower expansion paths for GDP growth, while all other assumptions, including world fossil fuel prices, have remained the same.

274.

Table 1: EU-27 GDP in real terms in the high and low economic growth variants, compared to the Reference scenario GDP


Table 2: Average annual growth rate for the EU-27

The two economic growth variants are designed to provide insights into the energy system developments stemming from alternative outcomes on economic drivers of energy consumption. In the high growth variant, GDP per capita is 0.4 percentage points higher than in the Reference case throughout the projection period, whereas it would be 0.4 pp lower in the low growth case. These variants examine the energy consequences of alternative economic developments broken down by economic sector in particular with regard to activities of energy intensive sectors versus less intensive ones.

Higher GDP growth would be driven mainly by enhanced activities of the services sector, with particular high value added growth in market services and trade, as these sectors are not very energy intensive. By comparison, industrial value added would exhibit less additional growth with expansion rates lower than that of GDP Both energy intensive and less energy intensive industrial sectors would however still show healthy additional growth.

In the low economic growth variant, all economic sectors would suffer to a similar extent with value added in most cases being 14-15% lower in 2050 than in the Reference case. One exception would be agriculture where the decrease in output with respect to the Reference case would be smaller.

275.

1.2 Energy import prices


The energy projections are based on a relatively high oil price environment compared with previous projections and are similar to reference projections from other sources i. The baseline price assumptions for the EU27 are the result of world energy modelling (using the PROMETHEUS stochastic world energy model) that derives price trajectories for oil, gas and coal under a conventional wisdom view of the development of the world energy system.

International fuel prices are projected to grow over the projection period with oil prices reaching 88$’08/bbl in 2020, 106$’08/bbl in 2030 and 127 $08/barrel in 2050 with 2% inflation (ECB target) this corresponds to some 300 $ in 2050 in nominal terms.

Gas prices follow a trajectory similar to oil prices reaching 62$’08/boe in 2020, 77$’08/boe in 2030 and 98 $(08)/boe in 2050 while coal prices increase during the economic recovery period to reach almost 26$’08/boe in 2020 and stabilize at around 30$’08/boe.[97]

The price development to 2050 is expected to take place in a context of economic recovery and resuming GDP growth without decisive climate action in any world region. Prices were derived with world energy modelling that shows largely parallel developments of oil and gas prices whereas coal prices remain at much lower levels.

276.

Figure 1: Reference scenario fossil fuel price assumptions


The evolution of the ratio of gas and coal prices can to a great extent influence the investment choices taken by investors in the power sector. A relatively low gas to coal price ratio up to the year 2000, together with the emergence of the gas turbine combined cycle technology, led to massive investments in gas fired power plants. The investments decreased afterwards due to significant gas price increases. The ratio between gas and oil prices remains stable over time as gas prices continue to follow oil prices. The gas to coal price ratio is projected to rise steadily over time as the coal prices in the world modelling results do not follow oil prices but remain around 30$’08/boe from 2030 onwards. While this ratio will increase over time, investment decisions will also be highly dependent on the expectations about future carbon prices.

277.

Figure 2: Ratios of fossil fuel prices


Sensitivities: Higher and lower energy import prices

Considering the high degree of uncertainty surrounding projections over such a long time horizon, a sensitivity analysis has been carried out with respect to developments in energy imports prices. A high and a low case have been analysed. When undertaking the price sensitivities in 2011, the energy price figures for 2010 were updated from the estimates made in early 2009 for the Baseline/Reference scenario (in the same way as in the Reference case).[98] Global developments as regards shale gas are taken into account in this analysis.

The world energy model PROMETHEUS was deployed to derive the alternative prices trajectories. This stochastic model is particularly well suited given the great uncertainty regarding future world economic developments and the extent of recoverable resources of fossil fuels. Two different world energy price developments have been examined. The high world fossil fuel price development is driven by somewhat higher global GDP growth than under reference developments, especially in China, giving rise to higher energy consumption. Moreover, there are somewhat less optimistic assumptions on reserves regarding unconventional oil, which has the highest marginal costs. This favours stronger market power of key exporting countries and thereby higher prices. On the contrary, the low world energy prices derive from markedly more subdued world economic growth combined with higher fossil fuel reserves and consequently less market power of key export players.

The sensitivities below are more symmetrical around the Reference case, including a High Price case with oil prices exceeding the Reference case level by 28% in 2050 and a Low Price case, in which the oil price in 2050 is 34 % lower than in the Reference case.

The price trajectories for oil, gas and coal shown in table 3 for the high energy price scenario stem from the following developments mirrored in the world modelling analysis:

· There is sustained economic growth in many Asian economies (notably China) following their reaction to the recent crisis, which has been to support domestic market expansion as a counterweight. The result has been that economic growth in the large Asian economies like China and India has barely been affected by the world economic slowdown. Since these are large consumers of coal the effect of this economic activity revision is particularly pronounced on short to medium term coal prices.

· There appears to be pronounced delays in oil productive capacity expansion with many plans being constantly revised. In addition, the recent accident in the Gulf of Mexico has resulted in a moratorium on deep water development in that area and is likely to result in delays in other parts of the world as well, in response to increased environmental concerns.

· There is increased concern that oil reserves and prospects for undiscovered resources are overstated. This may be particularly the case in OPEC countries where resource endowment is used as a criterion for production quota allocations.

· In view of the oligopolistic nature of world oil markets the tighter supply conditions usually translate into disproportionate increases in resource rents. Likewise such conditions imply greater vulnerability to short term supply disruptions leading to price spikes and resulting in higher average prices.

· The higher oil prices result in substitution of oil for gas in markets where the two fuels compete. The reduction in oil discoveries also implies a reduction in future reserves of associated gas. On the other hand gas price increases are moderated by an increasing share of unconventional gas from shales, as technology improves and the interest in its potential spreads beyond North America.

The low energy price scenario has been based on the following hypothetical background:

· There is currently great uncertainty on economic development including regarding excessive debts. It cannot be excluded that the recovery observed in 2009 and 2010 could prove to be relatively short lived, potentially leading to a "W shaped recession”).Whereas the reference scenario assumes a strong recovery of the world economy in the 2011-2014 period predicated on a rapid absorption of excess productive capacity (both capital and labour) and a strong resumption of investment in anticipation of fast growth in demand, developments could be less favourable. In particular, credit expansion could be hampered by the persistence of creditor exposure to uncertainty and increasing concern over the scope and timing of adjustments aimed at addressing imbalances (including sovereign debt). Consequently the investment boom may fail to materialize leading to some permanent loss of potential GDP (in the longer term world GDP is 7% lower in the modelled environment, which explains particularly low world fossil fuel prices).

· There is also uncertainty about energy resources and a more optimistic view could be adopted on this world energy price driver. In the low price variant, undiscovered conventional oil resources are set at their upper ten percentile value following USGS and PROMETHEUS assessments (in the reference scenario median values were used).

· In addition, the low price variant also assumes an increase in exploration activity outside the Gulf region as a response to security of supply concerns. This results in a more rapid translation of the resource basis into larger quantities of exploitable reserves. The main impact of this assumption is to bring forward the market easing emanating from greater resource abundance.

· The variant assumes rapid improvements in the knowledge and technologies associated with unconventional (shale) gas extraction. These in turn lead to enhanced interest in shale gas resources beyond North America leading to their more rapid incorporation into the exploitable resource base of some regions of the world. The assumptions concerning shale gas are the key driver for the high oil to gas price ratio that characterizes the low price variant.

278.

Table 3: Energy import prices in the Reference scenario and low and high price variants


Figure 3: Sensitivity for international fuel prices

Similarly, to these sensitivities, the Current Policy Initiatives Scenario is based on slightly higher short term energy import prices reflecting 2010 developments.

279.

1.3 Policy assumptions


Policy measures included in the Reference scenario are resumed in the following table:

|| Measure|| How the measure is reflected in PRIMES

280.

Regulatory measures


|| Energy efficiency

Ecodesign Framework Directive| Directive 2005/32/EC | Adaptation of modelling parameters for different product groups for Ecodesign and decrease of perceived costs by consumers for labelling (which reflects transparency and the effectiveness of price signals for consumer decisions). As requirements and labelling concern only new products, the effect will be gradual (marginal in 2010; rather small in 2015 up to full effect by 2030). The potential envisaged in the Ecodesign supporting studies and the relationship between cost and efficiency improvements in the model's database were cross-checked.

Stand-by regulation| Regulation No 1275/2008

Simple Set-to boxes regulation| Regulation No 107/2009

Office/street lighting regulation| Regulation No 245/2009

Household lighting regulation| Regulation No 244/2009

External power supplies regulation| Regulation No 278/2009

TVs regulation (+labelling)| Regulation No 642/2009

Electric motors regulation| Regulation No 640/2009

Circulators i regulation| Regulation No 641/2009

Freezers/refrigerators regulation (+labelling)| Regulation No 643/2009

Labelling Directive| Directive 2003/66/EC| Enhancing the price mechanism mirrored in the model

Labelling for tyres| Regulation No 1222/| Decrease of perceived costs by consumers for labelling (which reflects transparency and the effectiveness of price signals for consumer decisions)

Energy Star Program (voluntary labelling program)|| Enhancing the price mechanism mirrored in the model

Directive on end-use energy efficiency and energy services| Directive 2006/32/EC| National implementation measures are reflected

Buildings Directive| Directive 2002/91/EC| National measures e.g. on strengthening of building codes and integration of RES are reflected

Recast of the EPBD| Directive 2010/31/EU| New building requirements are reflected in technical parameters of the model, in particular through better thermal integrity of buildings and requirements for new buildings after 2020

Cogeneration Directive| Directive 2004/8/EC| National measures supporting cogeneration are reflected

|| Energy markets

Completion of the internal energy market (including provisions of the 3rd package)| ec.europa.eu/energy/gas_electricity| The model reflects the full implementation of the Second Internal market Package by 2010 and Third Internal Market Package by 2015. It simulates liberalised market regime for electricity and gas (decrease of mark-ups of power generation operators; third party access; regulated tariffs for infrastructure use; producers and suppliers are considered as separate companies) with optimal use of interconnectors.

EU ETS directive| Directive 2003/87/EC as amended by Directive 2008/101/EC and Directive 2009/29/EC| The ETS carbon price is modelled so that cumulative cap for GHGs is respected i. The permissible total CDM amount over 2008-2020 is conservatively estimated at 1600 Mt. Banking of allowances is reflected The ETS cap is assumed to continue declining beyond 2020 as stipulated in legislation, however with an effective domestic emission decrease lower than the linear decrease rate of 1.74%) to result in a 50% cumulative decrease of actual emissions instead of 70% which could stem from the Directive as a maximum reduction of EU emissions if no use of international credits would be allowed beyond 2030 i; currently no provision for the use of international credits post 2020 have been fixed and in the reference scenario world without global action, the higher ETS price might trigger greater use of such credits, which would also be in greater supply with higher ETS prices. ETS prices are derived endogenously on the basis of allowances, international credits, emissions reflecting developments of energy consumption while taking account of banking.

RES directive| Directive 2009/28/EC| Legally binding national targets for RES share in gross final energy consumption are achieved in 2020; 10% target for RES in transport is achieved for EU27 as biofuels can be easily traded among Member States; sustainability criteria for biomass and biofuels are respected using the full detail of the biomass model linked to the PRIMES energy system model; cooperation mechanisms according to the RES directive are allowed and respect Member states indications on their 'seller' or 'buyer' positions. RES subsidies decline after 2020 starting with the phasing out of operational aid to new onshore wind by 2025; other RES aids decline to zero by 2050 at different rates according to technology. Increasing use of RES co-operation mechanisms is assumed and should help to reduce RES costs. Policies on facilitating RES penetration will continue.

GHG Effort Sharing Decision| Decision 406/2009/EC| National targets for non-ETS sectors are achieved in 2020, taking full account of the flexibility provisions such as transfers between Member States. After 2020, stability of the provided policy impulse but no strengthening of targets is assumed.

Energy Taxation Directive| Directive 2003/96/EC| Tax rates (EU minimal rates or higher national ones) are kept constant in real term. The modelling reflects the practice of MS to increase tax rates above the minimum rate due to i.a. inflation.

Large Combustion Plant directive| Directive 2001/80/EC| Emission limit values laid down in part A of Annexes III to VII in respect of sulphur dioxide; nitrogen oxides and dust are respected. Some existing power plants had a derogation which provided them with 2 options to comply with the Directive: either to operate only a limited number of hours or to be upgraded. The model selected between the two options on a case by case basis. The upgrading is reflected through higher capital costs.

IPPC Directive| Directive 2008/1/EC| Costs of filters and other devices necessary for compliance are reflected in the parameters of the model

Directive on the geological storage of CO| Directive 2009/31/EC| Legal framework regulating the geological storage of CO2 allowing together with EEPR and NER300 CCS demonstration support (see below) economic modelling to determine CCS penetration

Directive on national emissions' ceilings for certain pollutants| Directive 2001/81/EC| PRIMES model takes into account results of RAINS/GAINS modelling regarding classical pollutants (SO2, NOx). Emission limitations are taken into account bearing in mind that full compliance can also be achieved via additional technical measures in individual MS.

Water Framework Directive| Directive 2000/60/EC| Hydro power plants in PRIMES respect the European framework for the protection of all water bodies as defined by the Directive, which limits the potential deployment of hydropower and might impact on generation costs.

Landfill Directive| Directive 99/31/EC| Provisions on waste treatment and energy recovery are reflected

|| Transport

Regulation on CO2 from cars| Regulation No 443/| Limits on emissions from new cars: 135 gCO2/km in 2015, 115 in 2020, 95 in 2025 – in test cycle. The 2015 target should be achieved gradually with a compliance of 65% of the fleet in 2012, 75% in 2013, 80% in 2014 and finally 100% in 2015. Penalties for non-compliance are dependent on the number of grams until 2018; starting in 2019 the maximum penalty is charged from the first gram.

Regulation EURO 5 and Regulation No 715/| Emissions limits introduced for new cars and light commercial vehicles

Fuel Quality Directive| Directive 2009/30/EC| Modelling parameters reflect the Directive, taking into account the uncertainty related to the scope of the Directive addressing also parts of the energy chain outside the area of PRIMES modelling (e.g. oil production outside EU).

Biofuels directive| Directive 2003/30/EC| Support to biofuels such as tax exemptions and obligation to blend fuels is reflected in the model The requirement of 5.75% of all transportation fuels to be replaced with biofuels by 2010 has not been imposed as the target is indicative. Support to biofuels is assumed to continue. The biofuel blend is assumed to be available on the supply side.

Implementation of MARPOL Convention ANNEX VI| 2008 amendments - revised Annex VI| Amendment of Annex VI of the MARPOL Convention reduce sulphur content in marine fuels which is reflected in the model by a change in refineries output

Regulation Euro VI for heavy duty vehicles | Regulation (EC) No 595/| Emissions limits introduced for new heavy duty vehicles.

Regulation on CO2 from vans[102]| Part of the Integrated Approach to reduce CO2 emissions from cars and light commercial vehicles.| Limits on emissions from new LDV: 181 gCO2/km in 2012, 175 in 2016, 135 in 2025 – in test cycle

281.

Financial support


TEN-E guidelines| Decision No 1364/2006/EC| The model takes into account all TEN-E realised infrastructure projects

EEPR (European Energy Programme for Recovery) and NER 300 (New entrance reserve) funding programme| For EEPR: Regulation No 663/2009; For NER300: EU Emissions Trading Directive 2009/29/EC Article 10a(8), further developed through Commission Decision 2010/670/EU[103]| Financial support to CCS demonstration plants; off-shore wind and gas, innovative renewables and electricity interconnections is reflected in the model. For CCS, - the following envisaged demonstration plants are taken into account for commissioning in 2020: Germany 950 MW (450MW coal post-combustion, 200MW lignite post-combustion and 300MW lignite oxy-fuel), Italy 660 MW (coal post-combustion), Netherlands 1460 MW (800MW coal post-combustion, 660MW coal integrated gasification pre-combustion), Spain 500 MW (coal oxy-fuel), UK 3400 MW (1600MW coal post-combustion, 1800MW coal integrated gasification pre-combustion), Poland 896 MW (306MW coal post-combustion, 590MW lignite post-combustion); investment in further plants depends on carbon prices

RTD support (7th framework programme- theme 6)| energy research under FP| Financial support to R&D for innovative technologies such as CCS, RES, nuclear and energy efficiency is reflected by technology learning and economies of scale leading to cost reductions of these technologies

State aid Guidelines for Environmental Protection and 2008 Block Exemption Regulation| Community guidelines on state aid for environmental protection| Financial support to R&D for innovative technologies such as CCS, RES, nuclear and energy efficiency is reflected by technology learning and economies of scale leading to cost reductions of these technologies

Cohesion Policy – ERDF, ESF and Cohesion Fund|| Financial support to national policies on energy efficiency and renewables is reflected by facilitating and speeding up the uptake of energy efficiency and renewables technologies.

Rural development policy - EAFRD| Council Regulation (EC) No. 1698/| Financial support for supply and use of renewable energy to farmers and other actors in rural areas, financial support to investments increasing energy efficiency of farms

282.

National measures


Strong national RES policies || National policies on e.g. feed-in tariffs, quota systems, green certificates, subsidies and other cost incentives are reflected

Nuclear|| Nuclear, including the replacement of plants due for retirement, is modelled on its economic merit and in competition with other energy sources for power generation except for MS with legislative provisions on nuclear phase out. Several constraints are put on the model such as decisions of Member States not to use nuclear at all (Austria, Cyprus, Denmark, Estonia, Greece, Ireland, Latvia, Luxembourg, Malta and Portugal) and closure of existing plants in some new Member States according to agreed schedules (Bulgaria 1760 MW, Lithuania 2600 MW and Slovakia 940 MW). The nuclear phase-out in Belgium and Germany is respected while lifetime of nuclear power plants was extended to 60 years in Sweden. Nuclear investments are possible in Bulgaria, the Czech Republic, France, Finland, Hungary, Lithuania, Romania, Slovakia, Slovenia, Spain and UK For the modelling the following plans on new nuclear plants were taken into account: Bulgaria (1000 MW by 2020 and 1000 MW by 2025), Finland (1600 MW by 2015), France (1600 MW by 2015 and 1600 MW by 2020), Lithuania (800 MW by 2020 and 800 MW by 2025), Romania (706 MW by 2010, 776 MW by 2020 and 776 MW by 2025), Slovakia (880 MW by 2015). Member States experts were invited to provide information on new nuclear investments/programmes in spring 2009 and commented on the PRIMES baselines results in summer 2009, which had a significant impact on the modelling results for nuclear capacity.

In addition to these measures, the Current Policy Initiatives Scenario includes the following policies and measures:

Area| Measure| How it is reflected in the model

Internal market||

Effective transposition and implementation of third package, including the development of pan-European rules for the operation of systems and management of networks in the long run| The modelling approach mirrors completion of the internal market, but has to account for existing interconnector limitations. Better market integration is reflected by having higher net transfer capacities in the near future and additional interconnectors in the longer term which lead to higher price convergence in multi-country market coupling in both electricity and gas markets (for details see below). In the gas market, more diversification (see also point 1) and higher degree of competition lead to lower oligopoly mark-ups and lower prices.

Regulation on security of gas supply (N-1 rule, necessity for diversification)| Compliance with N-1 rule and the necessity for diversification induce higher costs in the model for gas companies.

Regulation on Energy market integrity and transparency (REMIT)| The model simulates well functioning energy markets

Infrastructure||

Facilitation policies (faster permitting; one stop shop)| All these policies induce shorter lead times and slightly lower costs allowing faster infrastructure deployment.

Infrastructure instrument| More funding available from the EU budget

Updated investments plans based on ENTSO-e Ten Year Network Development Plan| Interconnection capacity reflects projects in the TYNDP by 2020.

Smartening of grids and metering| Smart grids and meters will lead to higher costs mainly for distribution but will allow for more energy efficiency in the system and decentralised RES

Energy efficiency| measures proposed in the Energy efficiency Plan – implementation compared to scenario 3[104] less vigorously and at a more moderate rate|

Obligation for public authorities to procure energy efficient goods and services| Cost perception parameters for non market service sector adapted accordingly

Planned Ecodesign measures (boilers, water heaters, air-conditioning, etc)| Adaptation of modelling parameters for different product groups. As requirements concern only new products, the effect will be gradual (rather small in 2015 and up to full effect by 2030/2035 as e.g. boilers can have a very long lifetime)

High renovation rates for existing buildings due to better/more financing and planned obligations for public buildings| Change of drivers (ESCOs, energy utilities obligation in point 13, energy audits point 14) influence stock – flow parameters in the model reflecting higher renovation rates, with account being taken of tougher requirements for public sector through specific treatment of the non-market services sector

Passive houses standards after 2020 (already in the Reference scenario)| Higher penetration of passive houses standards compared to the Reference scenario (around 30-50 KWh/m2 depending on a country which might to a large extent be of renewable origin)

Greater role of Energy Service Companies| Enabling role of ESCOs is reflected via altered economic parameters leading to more energy efficient choices (see also point 10)

Obligation of utilities to achieve energy savings in their customers' energy use of 1.5% per year (until 2020)| Induce more energy efficiency mainly in residential and tertiary sectors by imposing an efficiency value for grid bound energy sources (electricity, gas, heat)

Mandatory energy audits for companies| Induce more energy efficiency in industry (see also point 10)

Obligation that, where there is a sufficient demand authorisation for new thermal power generation is granted on condition that the new capacity is provided with CHP; Obligation for electricity DSOs to provide priority access for electricity from CHP; Reinforcing obligations on TSOs concerning access and dispatching of electricity from CHP| To a large extent already reflected in the Reference scenario 2050 Further facilitation of CHP penetration in the model

Obligation that all new energy generation capacity reflects the efficiency ratio of the best available technology (BAT), as defined in the Industrial Emissions Directive| High energy efficiency to a large extent already reflected in the Reference scenario 2050 as a response to ETS carbon prices; energy efficiency improves furthermore in power generation along with new investment from more efficient vintages

Other measures (better information for consumers, public awareness, training, SMEs targeted actions)| Induce faster energy efficiency improvements

Nuclear||

Nuclear Safety Directive| Harmonisation with international standards

Waste Management Directive| Cost for waste management reflected in generation costs

Consequences of Japan nuclear accident| Stress tests and other safety measures reflected through higher costs for retrofitting (up to 20% higher generation costs after lifetime extension compared with Reference scenario) and introduction of risk premium for new nuclear power plants. Nuclear determined on economic grounds, subject to non nuclear countries (except for Poland) remaining non-nuclear

CCS||

Slower progress on demonstration plants| Downward revision of planning for some CCS demonstration plants compared to the Reference case; some plants might be commissioned later depending on carbon prices. Change regarding potential storage sites in BE and NL.

Oil and gas||

Offshore oil and gas platform safety standards| Standards slightly increase production costs for oil and gas in the EU

Taxation||

Energy taxation Directive (revision 2011)| Changes to minimum tax rates for heating and transport sectors reflect the switch from volume-based to energy content-based taxation and the inclusion of a CO2 tax component. Where Member States tax above the minimum level, the current rates are assumed to be kept unchanged. For motor fuels, the relationships between minimum rates are assumed to be mirrored at national level even if the existing rates are higher than the minimum rates. Tax rates are kept constant in real terms.

Transport||

A revised test cycle to measure CO2 emissions under real-world driving conditions (to be proposed at the latest by 2013) [105]| Implementation of CO2 standards for passenger cars (95 g CO2/km) by 2020. Starting with 2020 assume autonomous efficiency improvements as in the Reference scenario.

Update of the CO2 standards for vans according to the adopted regulation i| Implementation of CO2 standards for vans (175 g of CO2 per kilometre by 2017, phasing in the reduction from 2014, and to reach 147g CO2/km by 2020).

Other parameters||

Energy import prices|| Short-term increase to reflect the evolution of prices up to 2010

Technology assumptions| Higher penetration of EVs reflecting developments in 2009-2010 national support measures and the intensification of previous action programmes and incentives, such as funding research and technology demonstration (RTD) projects to promote alternative fuels.| Slightly higher penetration of EVs Assumed specific battery costs per unit kWh in the long run: 390-420 €/kWh for plug-in hybrids and 315-370 €/kWh for electric vehicles, depending on range and size, and other assumptions on critical technological components i.

283.

1.4 Assumptions about energy infrastructure development


Regarding infrastructure representation, the scope of the modelling was increased by undertaking the determination of electricity interconnectors in a two stages approach. The aim is to represent market integration cost-effectively given many different scenarios modelled. The purpose of stage 1 is to determine electricity trade in the internal market based on a simpler version of PRIMES determining the equilibrium with all countries linked through endogenous trade, which due to its great technology detail on power generation requires very long computing times for each run. Stage 2 concerns the fully detailed modelling on the basis of the outcome of stage 1. The very long computing times for each model run under endogenous trade require a cost-effective approach, given that many iterations need to be performed between demand and supply and for meeting carbon targets. Running all countries in parallel in stage 2, involving many iterations, ensures delivery of modelling results in time.

Data about NTCs and interconnection capacities were taken from ENTSOe databases. Information on new constructions was taken from the latest “Ten-year network development plan 2010-2020”, complemented, where necessary, with information from the Nordic Pool TSOs and the Energy Community (for South East Europe). Some of the planned new constructions would justify increase of NTCs values until 2020, as mentioned in the ENTSOe’s TYNDP document. Other mentioned new constructions regard directly the building of new interconnection lines which are introduced as such in the model database.

Market integration leads to more electricity trade, which in turn needs infrastructure that is also dealt with in the modelling. Several test modelling runs were undertaken. It turned out that for the Reference and Current Policy Initiatives scenarios, the 2020 interconnection capacity would allow for most intra-EU electricity trade up to 2050 provided that a few identified bottlenecks would be dealt with. Such areas are the southern and eastern connections of Germany, the area linking Italy, Austria and Slovenia, the linkages of Balkans with northern neighbours and the linkages within Balkans. Some NTC additions should be also made for the linkages Denmark-Sweden and Latvia-Estonia. With lower electricity demand due to the assumed strong energy efficiency policies, these results also hold for the Current Policy Initiatives scenario.

Other infrastructure is dealt with in a less sophisticated way given that this is not so much in the focus of the energy system model at the European level. For CCS infrastructure (CO2 storage and transport) as well as for the sites of power plants, e.g. nuclear or RES installations (the sites - not the generation as such, see below) non-linear cost supply curves have been applied that take account of increasing costs, leading to higher costs once the most suitable and cheapest sites have been used.

Details on the modelling approach taken can be found in the Attachment 2 on interconnections.

284.

1.5 Technology assumptions


Technology parameters are exogenous in the PRIMES modelling and their values are based on current databases, various studies[108] and expert judgement and are regularly compared to other leading institutions. Technologies are assumed to develop over time and to follow learning curves which are exogenously adjusted to reflect the technology assumptions of a scenario. For some technologies, in particular, for off-shore wind and nuclear, the database of realised projects is very limited which can lead to significant differences depending on how many projects and what projects were included and where projects are being built.

The energy efficiency and other characteristics of the existing stock for a technology in a given period depend on previous investments. This ensures that as in real life changes in the characteristics of the technology stock happen only gradually depending on the type and magnitude of new investment as well as the rate of retirement of obsolete equipment. The market acceptance of a technology is also modelled and depends on the maturity of a technology; the more mature a technology the higher its market acceptance. Nuclear is however a special case driven mainly by political considerations at government levels and acceptance by citizens.

In order to ensuring comparability across scenarios, technology assumptions regarding capital and operational costs as well as technology performance over time have to remain the same across scenarios, except for cases, in which there were specific policies on technology progress (e.g. targeted support to one specific technology). In addition to these genuine technology parameters, the uptake of technologies is also influenced by other modelling parameters reflecting policy intensity, such as carbon and renewables values; these are discussed in later chapters. Current trend and decarbonisation scenarios differ regarding enabling policies, impacting also on technology uptake, as well as economies of scale in technology deployment, bringing lower energy costs. Technology specific parameters as such remain the same across scenarios.

The modelling cycle ending with the Energy Roadmap started in 2009 with the update of the Baseline, meaning that capital costs assumptions for 2010 and their evolution up to 2050 are based on information available in 2009/2010.. The Low Carbon Economy Roadmap and the Transport White Paper of spring 2011 were based on the same technology assumptions. It is clear that markets and technology costs as well as performance parameters evolve over time. Therefore, such assumptions need periodical update, which will be done again for the next modelling cycle starting in 2012.

285.

Power generation


Power generation technologies are characterised by capital costs, variable and fixed operation costs and by efficiencies. These characteristics are assumed to change over time due to technological improvements (impacting predominantly on capital costs). The assumptions for the Reference scenario for 2010 have been compared to other studies (e.g. IEA[109] and US DOE[110]), where possible[111]; all costs have been transformed into EUR[112].

As can be seen in Figure 4 the capital costs in PRIMES are within the range of other studies.

Figure 4: Capital costs in EUR/kWh in 2010[113]

The costs of technologies evolve over time in the Reference scenario reflecting learning curves and economies of scale. There are ample possibilities for solar technologies, both thermal and PV, to see costs decreasing over time, which is also the case for CCS technologies. These are not yet mature technologies and can therefore still follow steep learning curves. By comparison, the possibilities of wind onshore to further decrease its costs are rather limited with some potential still existing for small wind turbines., Figure 5 shows cost developments for mainstream onshore wind at medium size. As can also be seen in that figure, capital costs for off-shore wind can be expected to decrease significantly over time.

286.

Figure 5: Development of capital costs over time in the Reference scenario



The effective cost of a technology depends also on subsidies that may be paid by governments for environmental reasons to encourage specific innovative technologies that may require state aids for some time. In the case of renewables, Member States have support schemes that encourage the uptake of renewables technologies depending often on cost differences with conventional power generation technologies. This implies dependence of such aids on the progress in the cost reduction for renewables technologies, which are becoming increasingly cost competitive over time.

The Roadmap modelling assumes that such existing operational aid to RES for power generation is being phased out according to the maturity of the individual technology subgroups. In the longer term, only innovative and still costly RES technologies, such as solar PV, wave, tidal and off-shore wind at difficult sites, would receive aids. While for the more mature technologies (onshore wind) such aid is assumed to have been phased out rather early in the modelling (by 2025), the phasing-out of operational aid is completed by 2050 for other technologies. As RES technology costs come down, sometimes ahead of expectations, governments curtail the aid they grant.

In any case, the operational aids modelled only foster the uptake of RES technologies that are not yet fully commercial. Renewables support is modelled via support to capital costs. This support is relevant only for the investment decision but does not reduce electricity costs, given that the full costs of RES deployment are paid for by electricity consumers. In a large number of Member States this is currently done via feed-in tariffs, the salient features of which (all electricity consumers pay for the support to specific technologies) are captured by the electricity modelling undertaken in these scenarios. It is important to note that the current trend and decarbonisation scenarios have the same levels of operational aids that decrease over time.[114]

287.

Distributed Heat and Steam


Distributed heat in PRIMES can come either from CHP or district heating boilers. There are several technologies to produce steam, but distribution technologies are rather standard. For CHP there are ten different technologies that are applicable to different power generation technologies; the CHP technologies relate to the different technical options to extract the steam e.g. extraction, back-pressure or condensing technologies. The CHP technologies are considered mature, therefore no new learning effects are assumed. The higher penetration of CHP technologies in the different scenarios is based on policy drivers.

288.

Demand side technologies


Demand side technologies are mainly related to buildings, appliances, industrial equipment and transport vehicles. The penetration of new technologies can have important effects on energy efficiency improvement as well as on fuel switching. Technology parameters are exogenous with assumptions being based on results of various studies. The PRIMES data is compared regularly to other sources. For electric appliances PRIMES technologies were compared to the EuP Preparatory studies set out in directive 2005/32/EC and to the IEA Energy Technology Perspectives 2008, as well as the “Study on the Energy Savings Potentials in EU Member States, Candidate Countries and EEA Countries Final Report”[115]. The comparison proved that the assumptions taken in the PRIMES model are comparable to the developments of BAT and BNAT available from the EuP preparatory studies.

There is a very large number of different energy uses and technologies to provide the energy services (heating and cooling, light, motion, communication, etc) that consumers want when purchasing equipment and energy carriers.

In the PRIMES modelling, consumers always have the possibility of choosing between several vintages of the same technology, which are characterised by different prices and efficiencies. Throughout the projection period technologies become more mature and their market acceptance may grow, due to increased market maturity and policies.

Figure 6: Examples of developments of electric appliances in PRIMES compared to other literature sources[116]


The technologies in the above table only show a small variety of the technologies available in the model; further technologies and fuels for the technologies are available both for the services and residential demand as well as for industry and agriculture. The data has been compiled and updated over the years based on numerous sources including data from NEMS, the MURE database, industrial surveys, EU project reports and IEA studies.

For households PRIMES includes five different dwelling types, differentiated according to the main energy pattern[117] which each have energy services provided to them such as: space heating, water heating, cooking, cooling, lighting and other needs. Because of the very large variety of housing types both within and between countries, PRIMES uses curves for the possibilities of changes in thermal integrity of buildings relating marginal costs with energy efficiency improvements. Specific numbers for a typical household/dwelling type can therefore not be provided explicitly.

289.

Transport


For transport vehicles the same mechanisms apply as for appliances; a consumer can choose different vintages of the same kind of vehicle at different costs and efficiency. Also for transport, a comparison with a variety of literature sources was carried out, which proves that the estimates of PRIMES are in line with other estimates.

Table 4: Comparison of costs and efficiencies from different literature sources with PRIMES[118]

The amounts of biofuels in the fuel mix of the Reference scenario are determined by the relative costs of the fuels taking account of tax differentials and biofuel quotas. The PRIMES model currently does not distinguish between dedicated biofuel vehicles and vehicles that allow only for blending; the fuel and vehicle stock mix simulate the inclusion of dedicated vehicles implicitly.

The Current Policy Initiatives Scenario relies on the same technology assumptions besides nuclear in power generation which has been revised upwards reflecting the follow-up to the Japanese nuclear accident.

290.

1.6 Other assumptions


Discount Rates

The PRIMES model is based on individual decision making of agents demanding or supplying energy and on price-driven interactions in markets. The modelling approach is not taking the perspective of a social planner and does not follow an overall least cost optimization of the energy system. Therefore, social discount rates play no role in determining model solutions. However, social discount rates can be used for ex post cost evaluations.

Discount rates pertaining to individual agents play an important role in their decision behaviour. Agents’ decisions about capital budgeting involve the concept of cost of capital, which is depending on the sector - weighted average cost of capital (for firms) or subjective discount rate (for individuals). In both cases, the rate used to discount future costs and revenues involves a risk premium which reflects business practices, various risk factors or even the perceived cost of lending. The discount rate for individuals also reflects an element of risk averseness.

Table 5: Discount rates for the different actors[119]

291.

Discount rates


Industry| 12%

Private individuals| 17.5%

Tertiary| 12%

Public transport| 8%

Power generation sector| 9%

292.

Degree days against the background of climate change


The heating degree days, reflecting climate conditions, are kept constant at the 2000 level, which is higher than the long term average without assuming any trend towards further warming. The degree days in 2000 were fairly similar to the ones in 2005. This simplification allows comparison of recent statistics with the projection figures, without the need for climate correction.

There are also other energy related impacts from climate. However, future climate change depends on future emissions worldwide, atmospheric concentration and the sensitivity of the climate system to such concentration increases. Future developments in these areas are surrounded by substantial uncertainty. Given this uncertainty and the focus of this impact assessment on the various energy system impacts this quantitative analysis has assumed constant climate conditions over time. This simplification should be borne in mind when considering the following detailed results under constant climate, which is likely to change more, the more pronounced the global emission increase. All the decarbonisation scenarios in Part B assume meeting the climate targets, which are expected to prevent dangerous climate change. However, even when temperature changes are limited to 2 degrees Celsius, some climate impacts will occur. A literature review on climate change impacts in the European energy supply sector[120] has identified the following main impacts:

· Cooling water constraints for thermal power generation (especially during heat waves), with nuclear appearing to be particularly strongly affected[121]

· Damage to offshore or coastal production facilities due to sea level rise and storm surges

· Damage to transmission and distribution lines due to storm events, flooding

· Lower predictability of hydropower availability

· Affected yield in renewable energy sector (hydropower in Southern Europe, possibly biofuels due to diseases and forest fires, possibly faster biomass plant growth in certain areas)

· Melting permafrost affecting energy production and distribution in cold climates

· Damages and output constraints in wind energy due to storms and increased average wind speed

In addition, changes in temperature might lead to changes in energy demand patterns for heating and cooling.

It can hence be expected that decarbonisation has also positive economic impacts with regard to energy security and competitiveness by avoiding parts of the damage and adaptation costs in the energy system due to climate change.

In any case, given our lack of knowledge – perhaps for a considerable time to come - about how the EU 2050 GHG emission objective will be met and how global GHG emission will develop over time and therefore lacking information on future atmospheric concentrations and their impacts on temperatures in the Member States, the simplifying assumption has been made in this analysis that heating degree days remain constant.

293.

Exchange rates


All monetary values are expressed in constant, 2005, terms (without inflation). The economic modelling in PRIMES is based on euros. The dollar exchange rate for current money changes over time; it starts at the value of 1.45$/€ in 2009 and is assumed to decrease to 1.25 $/€ by 2020 and to remain at that level for the remaining period.

294.

2. Results 2.1 Reference scenario


Energy consumption and supply

Primary energy consumption peaked in 2006 at a level only marginally different from the year before. Given that 2005 numbers in the PRIMES output have been fully calibrated to 2005 Eurostat energy statistics, the following comparisons start from 2005, being virtually the peak year of energy consumption so far[122]. With ongoing energy efficiency policies – even in the absence of any further policy intensification as depicted in the Reference case- total energy demand decreases slightly up to 2050 (-4% from 2005). This is despite post-crisis economic growth leading to a doubling of GDP between 2005 and 2050 (on an EU-27 average of 1.6 % per year). Therefore, energy intensity drops considerably with one unit of GDP in 2050 requiring only less than half the energy needed in 2005.

Final energy consumption continues rising until 2030, after which demand stabilises as more efficient technologies have by then reached market maturity and the additional energy efficiency of the appliances is sufficient to compensate for increased demand for energy services (heat, light, motion, etc). The share of sectors remains broadly stable with transport staying the biggest single consumer accounting for 32% in 2050; the industrial share increases slightly while that of households declines a bit.

295.

Figure 7: Final energy demand indicators


The energy intensity of different sectors decreases, as does the overall energy intensity of the economy. Increased energy efficiency in the residential sector is due to the use of more efficient energy equipment (appliances, lighting, etc.) and buildings as well as behavioural changes. The strong improvement in the energy efficiency of energy equipment is driven by the Eco-Design regulations and by better thermal integrity of buildings reflecting the Recast of the Energy Performance of Buildings Directive. While these improvements are sufficient to ensure a decrease in final energy demand over the projection period in the residential sector, the increased efficiency is not sufficient to compensate for higher needs in the tertiary sector.

In the transport sector, the correlation between GDP growth and transport activity is found to decouple somewhat when using satellite transport modelling tools. Energy consumption is decoupling much more significantly due to the use of more energy efficient vehicles, in particular hybrids. The CO2 from cars regulation is instrumental for this development. This scenario takes a conservative view regarding the development of alternative energy carriers such as electric and fuel cell cars; it does not assume strong policies leading to a shift towards electric mobility or plug-in hybrid vehicles in addition to the existing CO2 from cars regulation. The CO2 emissions per kilometre driven decrease rapidly up to 2020 but as the regulation is not strengthened after 2020 in this scenario, improvements thereafter are due to stock renewal and some autonomous efficiency improvements brought about by markets as has been the case in the past. The penetration of biofuels in the Reference scenario is limited to road transportation; overall biofuels in liquid fuels achieve a share of 10% by 2050. The amount of RES in transport meets the 10% target in 2020 to comply with the RES directive and increases to 13.3% by 2050.

Figure 8: Energy and Activity in transport; composition of private vehicle stock[123]

There is considerable fuel switching in final energy demand, especially in the residential and tertiary sectors where the use of fossil fuels (solids, petroleum products and gas) decreases while there is a strong tendency towards electrification. The share of RES in final energy consumption increases markedly, reflecting the RES Directive. RES penetration continues with ongoing enabling policies (priority access, streamlined authorisation) whereas operation aid to mature RES technology is progressively reduced in this Reference case.

Also on the primary energy level, there is significant restructuring. This can be seen from the pronounced shifts in the shares of individual fuels up to 2050 (in terms of primary energy):

296.

Figure 9: Fuel mix development


· RES gain 13 percentage points (pp) from 2005 (15 pp from 1990); making it the third most important primary energy source (after oil and gas) in 2050 (when it reaches 20% of primary energy consumption);

· Nuclear increases 2 pp from 2005 (4 pp from 1990), becoming more important than solid fuels (16% share in 2050);

· Oil loses 5 pp (6 pp on 1990); oil share in 2050 amounts to only 32%;

· Solids lose 7 pp from 2005 (16 pp from 1990) reaching just 11% by 2050;

· Gas declines least of all fossil fuels (-3 pp from 2005 to 2050); the gas share in 2050 is still higher than in 1990 (3 pp) because of the significant inroads made up to now; gas will represent more than a fifth of the primary EU energy mix in 2050 (21%);

· The dominance of fossil fuels diminishes with their share falling from 83% and 79% in 1990 and 2005, respectively to only 64% in 2050.

While non fossil fuels (RES and nuclear) account for 36% of primary energy in 2050, they reach a significantly higher share in the 2050 electricity mix. Energy sources not emitting CO2 supply 66% of electricity output, with 40% RES and 26% nuclear. In addition, 18% of electricity would come from CCS plants, which do however still emit some CO2.

Power generation changes substantially in the projection period; the demand for electricity continues rising and there is a considerable shift towards RES. As can be seen in Figure 10 installation of capacity and generation from wind increase steadily throughout the period. The incentives due to the RES target until 2020 are sufficient to make wind power generation competitive with other technologies. Power generation and capacity from solids decrease throughout the scenario due to the carbon prices that reduce the competitiveness of this technology; gas power generation capacity increases, also as peak load activated during back-up periods due to the increased amount of RES in the system. As a result of the large increase in RES in power generation the load factor of the system decreases due to the more widespread use of technologies that run only a limited number of hours per year, such as wind.

Investment in power generation increases over the projection period, driven by new investments in RES and gas.

297.

Figure 10: Electricity generation and net generation capacity


The carbon intensity of power generation reduces by over 75% in 2050 compared to 2010 levels, driven by the decreasing ETS cap and the rising carbon prices (see Figure 11). In 2050 17.8% of electricity is generated through power plants equipped with CCS. This corresponds to a CCS share in fossil fuel power generation of over 50%. More than 50% of the potential emissions from the power generation sector are captured. The efficiency of thermal electricity production rises throughout the projection period due to the renewal of the power plants, in particular investment in gas and in spite of CCS being widely used in power generation. CHP plants are assumed to be integrated into the competitive electricity markets, facilitated by the CHP Directive and their share in electricity generation will rise. Incentives for CHP focus on electricity, which implies that an increase in electricity production from CHP power plants does not necessarily imply an increase in CHP capacity, given that there is some flexibility in the power to heat ratio.

Figure 11: Power generation indicators [124]


The price of electricity peaks in 2030 and decreases slightly thereafter. The price increase up to 2030 is due to three main elements: the policies inducing investment in RES, the ETS carbon price and the high fuel prices due to the world recovery after the economic crisis. Thereafter electricity prices do not increase further, indeed decline slightly, because of the technical improvements of technologies that limit the effects of higher input fuel prices. Moreover, taxes on fuels and ETS auction payments sink beyond 2030. This is due to the declining cap and the introduction of CCS in particular, which limits emission quantities and therefore auction revenues from the ETS despite rising ETS prices. Whereas the CO2 price increases, the average levy on electricity production, including the carbon free and decarbonised parts, declines in the long term. Moreover, there is a continued decrease in the use of diesel oil in power generation, which Member States may tax for environmental reasons.

298.

Figure 12: Cost components of average electricity price


Distributed Heat

Demand for distributed heat demand rises in the Reference scenario throughout the projection period; a strong increase can be observed between 2005 and 2020 reflecting the strong CHP promoting policies in all Member States, as well as commercial opportunities that arise from gas based and biomass based CHP technologies. It is assumed that the same policies continue at least until 2020 as part of the implementation of the 20-20-20 policy package. Among the CHP promoting drivers worth mentioning are: the CHP directive (facilitating absorption of CHP-electricity by wholesale markets), national policies including feed-in tariffs and the ETS-carbon prices. CHP growth is limited by the geographic possibilities of the distribution system. District heating powered by boilers is a less attractive option, except in cases exploiting local resources e.g. biomass, and existing distribution networks.

In the longer term further demand for distributed heat in the tertiary and residential sectors seem to slow down as a result of the trend towards electrification (i.e. heat pumps) and higher energy efficiency which limits the overall demand for heating. In industry the increase in demand for distributed steam is projected to continue in the future because the changes of industrial activity are favourable for sectors with high demand for steam such as chemicals, food, drink, tobacco, engineering and other industries. Furthermore the development of the market for distributed steam and the possibilities of selling electricity to the wholesale market favours the construction of CHP units of different sizes and technologies in these industrial sectors

Figure 13: Heat demand by sector

Transport

Transport accounts today for over 30% of final energy consumption. In a context of growing demand for transport, final energy demand by transport is projected to increase by 5% by 2030 to represent 32% of total final energy consumption. This development is driven mainly by aviation and road freight transport. At the same time, however, the energy use of passenger cars would drop by 11% between 2005 and 2030 due to the implementation of the Regulation setting CO2 emission performance standards for new passenger cars[125]. After 2030 transport energy demand would increase only marginally up to 2050.

The EU transport system would remain extremely dependent on the use of fossil fuels. Oil products would still represent 88% of the EU transport sector needs in 2030 and in 2050 in the Reference scenario.

299.

Energy Imports/ Security of Supply


Total energy imports increase 6% from 2005 to 2050. The increase is rather limited despite decreasing indigenous production, as rising gas and biomass imports are compensated by a marked decline in coal imports while oil imports remain broadly stable.

Gas imports continue to rise (28% from 2005 to 2050) due to declining production and despite decreasing consumption.

Import dependency rises only slightly above the present level (54%), reaching 58% in 2020 and flattening out to 2050 thanks to more RES and nuclear.

300.

Emissions


Energy related CO2 emissions decline much faster than energy consumption, giving rise to some decarbonisation of the energy system because of fuel switching to RES and nuclear at the expense of solid fuels and oil:

· Carbon intensity falls markedly. Producing one unit of GDP in 2050 would lead to only 30% of energy related CO2 emissions that were required per unit of GDP in 2005 and to just one fifth of what the CO2/GDP indicator was in 1990.

· Energy related CO2 emissions sink 40% below the 1990 level in 2050; thus the reference scenario represents about half of the efforts needed by a developed economy if a global deal to limit climate warming to 2°C will be achieved.

· CO2 emissions from electricity and heat generation fall almost 70% between 1990 and 2050 when they will make up 14 % of all GHG emissions (down from 27 % in both 1990 and 2005);

· Total GHG emissions decrease slightly less (39%) by 2050 from 1990. Whereas non-CO2 emissions fall somewhat more, the total emission decline is hampered by the very moderate decrease of CO2 from industrial processes (CO2 not related to fuel combustion).

Figure 14: CO2 emissions [126]

The contribution to the emission reductions is driven by the ETS sectors which decrease emissions by 48% between 2005 and 2050; on the contrary the non-ETS sectors reduce by 21% compared to 2005. The share therefore shifts from 56% of emissions in ETS sectors in 2005 to 46% in 2050. Most emissions continue to be energy related emissions; energy related CO2 emissions decrease by 39% in the time period from 2005 to 2050 whereas non-energy related CO2 emissions increase by 3%.

301.

Policy relevant indicators (and targets)


The indicative 20% energy savings objective for 2020 would not be achieved under current policies - not even by 2050. The reference case would deliver 10% less energy consumed in 2020 compared to the 2007 projections.

The reference case assumes that the RES target is reached in 2020; the RES share (as defined in the RES directive: as a percentage of gross final energy consumption) continues rising to reach 24% in 2030 and 25% in 2050; further penetration of RES is limited due to the assumed phasing out of operational aid to mature RES technologies (see below). On the basis of final energy, the RES share gains nevertheless 17 pp between 2005 and 2050 (13 pp on the basis of primary energy).

The ETS carbon price rises from 40 € (08)/tCO2 in 2030 to 52 € in 2040 and flattens out to 50 € in 2050 (after having triggered some emission reducing restructuring in ETS sectors to comply with the dynamic requirements of the Directive).

These CO2 prices seem high enough to trigger significant CCS investment from 2040 onwards; whereas the CCS share in gross power generation reaches only 2% in 2030, it rises to 12% in 2040 and 18% in 2050 (this percentage is 15% in net power terms). CCS is mainly applied on solid fuel power generation, but also to gas power plants towards the end of the projection period; by 2050 half of solid fuel power capacities are equipped with CCS and 17% of gas power plants. Generation by solid fuel CCS plants represents 10% of net total power generation in 2050; the share of gas based CCS is 5% in 2050.

The reference case assumes the overall GHG target, ETS cap and non-ETS national targets to be achieved by 2020 but thereafter GHG reductions fall short of what is required to mitigate climate change with a view to reaching the 2 °C aim. While the reference case development lead to only 40% less GHG emissions from 1990, more than twice as much might be needed, i.e. minus 80-95% by developed economies.

302.

2.2 Economic growth sensitivities


Economic activity is a key driver of energy consumption and therefore emissions. It can be expected that higher GDP growth rates will lead to higher energy consumption and CO2 emissions and vice versa in the case of lower economic growth.

303.

Final energy demand


In fact, final energy consumption in the high economic growth case is 7.3% higher in 2050 than in the Reference case. This increase is however much lower than the increase in GDP (+15.0%) due to important energy intensity improvements. These improvements are linked in particular to the structure of the additional economic activity, which takes place mainly in less energy intensive sectors, such as market services and trade. Moreover, higher economic growth allows faster capital turnover so that more energy efficient equipment enters the capital stock sooner. Better capacity utilisation in case of high economic growth can also add to this improvement in energy intensity. Higher household income also allows for faster replacement with new, more energy efficient, appliances and cars, although the overall demand of energy services would increase via more purchase of higher performing items.

CO2 emissions from final energy demand rise slightly less than energy consumption thanks to some fuel switching to zero carbon (electricity and heat) or low carbon fuels (gas). In 2050, CO2 emissions in final demand are 6.9% higher than in the Reference case (while energy demand and GDP are 7.3% and 15% higher, respectively).

Figure 15: Final energy consumption broken down by sector in different economic growth cases (in Mtoe) Additional energy consumption is most pronounced in the services/agriculture sector where demand in 2050 is 14.9% higher than in the Reference case. Again, CO2 emissions rise less than energy consumption thanks to fuel switching connected especially with more use of electricity[127]. In 2050, CO2 emissions from this sector exceed the Reference level by 12.6%, falling nevertheless well below current levels (see table 6).

With less pronounced expansion of economic activities in industry there is lower, but still considerable, growth in final energy demand. Increased industrial activities require more energy inputs so that industrial energy demand exceeds Reference case levels in 2050 by 9.9%. Energy consumption growth in industry is fossil fuel intensive with higher demand for carbon rich coal in certain branches, which – under constant CO2 policies via the EU ETS - leads to higher CO2 emissions, which exceed the Reference case level in 2050 by 12.0%. It is however worth noting that even with such high economic growth, industrial CO2 emissions in 2050 remain below today's level.

304.

Table 6: CO2 emissions from final energy demand sectors in different economic growth cases (in Million tonne CO2)


|| 2050 low growth| 2050 Reference| 2050 high growth

Industry| 476

Services/agriculture| 178

Households| 303

Transport| 1061

Total final demand| 2018


Energy consumption of households rises much less in comparison to the Reference case (by 1.9% in 2050) because many energy services, such as heating and cooking are very income inelastic once certain comfort levels have been reached. Moreover, increased purchases of appliances in the context of higher incomes concern items with lower specific energy consumption compared with the existing stock, a process that is being made more pronounced with eco-design Regulations. Household CO2 emissions in 2050 are just 2% higher than in the Reference case, but still a third lower than today.

Transport energy demand exceeds Reference case levels by only 5.5% in 2050. The reason is similar to that for households. Except for holiday trips, passenger transport activity tends to grow slower than private incomes. On the contrary, freight transport activity is much more influenced by the level of economic activity. In the absence of major possibilities for fuel switching under current trends and policies, higher transport energy demand translates directly into higher CO2 emission (5.3% higher than Reference in 2050), keeping emissions at current levels in 2050.

The improvement of carbon intensity in final energy demand under high economic growth (lower CO2 growth than growth of final energy demand) is mainly due to fuel switching towards electricity, which has been an ongoing trend with higher incomes and structural change in the economy (e.g. more ICT based services). Higher economic growth would lead to 8.8% higher electricity consumption (compared with Reference) in 2050 with CO2 consequences for power generation.

Higher GDP growth leads to higher demand for heat (+ 7% in 2050) in line with overall increase of final energy demand but significantly lower than increase in GDP (+15%). The growth comes mainly from industry and tertiary sectors reflecting higher economic activity in these two sectors. Residential demand is rather stable (+1%) as heat is an essential need and not very elastic to changes in household income. Supply increases from both CHP and district heating units.

Lower economic growth entails lower energy consumption and emissions in all sectors. With GDP in 2050 remaining 14.7% below the Reference case level, there would be a reduction of final energy demand by only 8.4%. Consequently, energy intensity (of final demand) would deteriorate compared with the Reference case (and even more so in the high growth case). Slower capital turnover in case of sluggish economic growth is one reason for this as well as a lot of energy uses being rather income inelastic, such as home heating and cooking. CO2 emission would decline to a somewhat smaller extent than energy consumption (only by 7.8% in 2050 compared with Reference). Low carbon content fuels reduce somewhat more than the more carbon intensive ones, leading also to a slight worsening of carbon intensity of final energy demand.

Energy demand in services/agriculture would fall almost as much as GDP in 2050 compared with the Reference case (-14.3%). The decline in CO2 emissions would be similar (-13.8%). Industrial energy consumption and emission decrease also markedly with lower economic growth; they are down on the Reference case in 2050 by 13.6% and 15.1%, respectively. CO2 emissions reduce somewhat more than energy consumption, as fossil fuel demand drops slightly more than demand for electricity and steam that are carbon free at use.

By comparison, households and transport reaction to lower GDP is much less pronounced. Household energy consumption and CO2 emissions are both down 2% on the Reference case 2050 level (i.e. substantially less than the decline in GDP: almost -15%). Given that freight transport reacts rather strongly to lower economic activity while passenger transport decreases comparatively little with lower income, transport energy consumption falls 5.7% below the 2050 reference case. CO2 emissions sink by almost the same percentage (-5.5%), as possibilities for fuel switching are limited in a Reference case environment without intensified climate or renewables policies.

Lower economic growth leads to a rather strong reduction of electricity demand, which remains 9.7% below the Reference case level in 2050, still exhibiting healthy growth from current levels.

Demand for distributed heat decreases by 10% in 2050 compared to the Reference scenario mainly due to sharp decreases in tertiary (-14%) and industry (-12%) sectors reflecting lower economic activity. Residential demand reacts much less (-3%) as heat is an essential need and not very elastic to changes in household income. There is a shift from CHP production (-11% in 2050 following lower electricity demand) to higher district heating units production (+10%).

305.

Electricity generation


Electricity demand is particularly sensitive to variations in economic activity. With limited possibilities for electricity imports this translates into a similar requirement on the generation of electricity in the EU. In the high economic growth case with 15% higher GDP in 2050, gross electricity generation exceeds the 2050 reference case level by 9.2%. Similarly, 14.7% lower economic activity in 2050 entails 10.2% less electricity generation in 2050.

Whereas the level of electricity generation strongly depends on the magnitude of economic growth, its structure changes much less with lower or higher GDP in 2030 and 2050. In 2030, the RES share in electricity varies within a margin of 1 percentage point around 40.5% in the Reference case (see table 7 on fuel shares in generation). This range becomes somewhat larger in 2050 (around 2 percentage points). With unchanged support for RES, higher economic growth encourages in particular nuclear and fossil fuel generation, leading to a somewhat lower RES share in power generation; it should be noted that the absolute level of RES based electricity generation is significantly higher with high economic growth (+5.3% in 2050 compared with Reference).

306.

Table 7: Electricity related indicators under different economic growth assumptions


High economic growth brings about higher ETS prices (see table 7), which in turn encourage CCS deployment. Combined with a higher share of fossil fuel based power generation, this leads to CCS shares in power generation that are higher than Reference in 2030. The increase is particularly pronounced in 2050, when 20% of total power generation would be equipped with CCS. On the contrary, with low economic growth leading to low ETS prices as well as lower fossil shares in power generation, CCS amounts to only 12% in 2050.

Electricity prices are rather insensitive with respect to variations in economic growth. Higher economic growth increases the 2030 average electricity price slightly by 1.2%, while lower economic growth would lead to an electricity price that is 0.4% below the Reference case price. These electricity price modifications relate to the significant changes in ETS prices brought about by variations in allowances demand due to growth of energy demand and changing fossil fuel inputs to power. In 2050, when the variation in ETS prices from the Reference case is pretty small, the variations in electricity prices become marginal or even undetectable (electricity prices: minus 0.2% with low GDP growth and 0.0% with high growth). Consequently, different economic growth patterns do not alter the Reference case result that shows strongly rising electricity prices up to 2030 in the context of higher fixed costs following the restructuring of the power generation system for reaching the RES and GHG targets, with a stabilisation of prices in the following two decades.

307.

Primary energy consumption and energy intensity


As was discussed in the part on final energy demand, certain parts of energy consumption react only to a limited extent to variations in economic growth; this concerns in particular the household sector and also passenger transport. Combined with more favourable conditions for improving energy efficiency under high economic growth (bringing about, together with structural change in economic activity, 5.8% better energy intensity), this leads to primary energy demand rising much less than GDP. Compared with the Reference case, primary energy demand increases 3.4% in 2030 while GDP is 6.3% higher, in 2050 primary energy exceeds the Reference case by 8.4% with the economy being 15.0% larger in terms of GDP.

Also in the case of lower economic growth, the effects on primary energy consumption are moderated by the less income elastic consumption sectors (households, where heating needs remain largely the same, as well as passenger transport having rather unchanged needs for commuting, shopping and similar travelling). Moreover, lower capital turnover with lower economic growth limits the opportunities for investing in energy efficient items. As a result, energy intensity worsens by 6.4% in 2050. Consequently, energy consumption sinks significantly less than GDP. With 7.7% lower GDP in 2030 compared with the Reference case, primary energy is down 5.0%; in 2050 with 14.7% lower GDP compared with Reference there is a decline of primary energy by just 9.3%.

These energy intensity effects (the improvement of 5.8% compared with Reference in 2050 under high economic growth and the 6.4% deterioration under sluggish GDP growth) limit the impacts of alternative developments of GDP on CO2 emissions. Another countervailing (or reinforcing) factor could come from changes in the fuel mix. Different economic growth patterns exert somewhat different influences on individual fuels.

308.

Fuel mix and carbon intensity


Under high economic growth, oil and gas consumption grow less than overall energy consumption. Nuclear reacts in a more pronounced way (above average) given its exclusive use in power generation, which in turn is more sensitive to variations of GDP. Also the reaction, to higher economic growth, of solids being mostly used in power generation is fairly marked in 2050, given the absence of strong CO2 limitation policies. On the assumption of unchanged RES support schemes, RES are not particularly encouraged by higher economic growth.

On the other hand, RES are not particularly discouraged by lower economic growth. The negative effects of such GDP losses on nuclear and solids are much stronger, exceeding the percentage changes of total energy consumption. Oil and gas sink largely in line with the reduction in total energy demand.

This leads to the following fuel shares in 2050:

· Oil reaches shares between 31% and 32.5% under high and low economic growth, respectively;

· The gas share amounts to 20% in both growth cases;

· Solids account for 12% under high and 10.5% under low economic growth;

· The nuclear share reaches 17.5% under high and 16% under low economic growth

· RES increase their share to 19.5% with high GDP and even 21% with lower economic expansion;

When evaluated in terms of gross final energy consumption (definition in the RES Directive), the RES shares amount to 25% under high and to 26% under low economic growth, which represents increases from the 2005 level of between 16 and 18 pp in the high and low GDP case, respectively. The RES share in transport is also pretty robust across economic growth cases amounting to 13% in 2050 under the different GDP assumptions, up half a percentage point from its level in 2030.

While there are only limited changes of fuel shares across economic growth cases, the evolution of fuel shares over time, especially regarding RES, is pretty dynamic. Fossil fuels in total lose around 16 percentage points between 2005 and 2050, with somewhat higher losses for solids and oil. RES gain between 12.5 and 14 percentage points under high and low growth, respectively, while nuclear accounts for the remaining gain.

309.

Figure 16: Development of the fuel mix under high and low economic growth


The overall result of these changes in the fuel mix is that the carbon intensity improves with higher economic growth, i.e. one unit of energy consumed results in slightly less CO2 emissions under high growth (1.32 t CO2/toe in 2050) than in the Reference case (1.36 t CO2/toe for the same year). The opposite effect on carbon intensity comes about under lower economic growth, in which case one tonne of oil equivalent energy consumption is associated with CO2 emissions of 1.45 tonnes, which equates to a 6.4% worsening.

310.

Total CO2 emissions


These effects on energy and carbon intensity and the existence of the ETS with a given emission cap mean that GDP-induced changes in CO2 emissions are much less significant than underlying changes in GDP. With 15% higher GDP in 2050, CO2 emissions are only 5.3% higher (both on Reference in 2050). Similarly, a GDP drop of 14.7% leads to CO2 emissions that are only 3.3% lower in 2050. For 2030, a GDP rise on Reference by 6.3% is associated with a 1.2% increase of CO2 emissions, while a GDP loss of 7.7% entails 2.3% lower CO2 emissions (compared with Reference case).

It can be concluded that emission results are pretty robust with respect to variations of GDP. This reduces greatly one possible uncertainty regarding policy objectives on emissions, as there are mechanisms (ETS, effects on energy intensity) that limit the effects of variations in GDP levels on energy consumption and on CO2 emissions. This is important given the great uncertainty in projecting economic activity for the coming years, let alone over the next four decades.

While there are such energy and carbon intensity effects, limiting the impact of economic activity on CO2 emissions, alternative economic developments would still alter the expected decline in CO2 emissions up to 2050. Such a decline of emissions materialises under Reference case policies and is also brought about by Current Policy Initiatives and even more so in decarbonisation scenarios.

Emissions reduce somewhat more over time with lower economic growth and somewhat less with higher economic growth. Variations in CO2 reductions from 1990 levels are however marginal in 2030 (around 1 percentage point more or less CO2 reduction from Reference case level in 2030 with higher or lower growth), while GDP varies 6-8 percentage points. In 2050 variations in the policy relevant indicator: CO2 reductions from 1990 around what would materialise in the reference case are still rather small (plus/minus 2-3 percentage points) - with GDP varying 15 percentage points around the reference case level.

Table 8: CO2 reduction below 1990 (index 1990 =100) and major drivers

1990 = 2050

|| High growth| Reference| Low growth| High growth| Reference| Low growth

GDP| 236

Energy consumption| 96

CO2 emissions| 58

Again, the possibilities for technically achieving GHG gas targets are not overly dependent on the level of economic growth. In any case, it needs to be borne in mind that GHG reduction requires innovation and investment, which is harder to finance in a low economic growth environment. Overall, emission reduction may be rather facilitated with sustained economic growth.

Finally, regarding the GHG emission reductions in ETS and non-ETS sectors, there is as to be expected with a given emission cap particularly little variation across economic growth cases for ETS sectors, whereas the GDP growth cases are somewhat more contrasted regarding non-ETS emissions.

Non-ETS GHG emissions reduce comparatively little up to 2030 under high economic growth and stay almost flat thereafter, whereas there is still a slight decrease in the reference case. Nevertheless, these changes are much lower than the underlying changes in GDP. In case of sluggish economic development, non-ETS emissions would continue declining through 2050. However, the reduction from Reference in 2050 is much lower than the decline of GDP.

311.

Energy imports and external dependency


Total net energy imports increase 16% from 2005 to 2050 under high economic growth, whereas they decline 4% with low GDP (Reference case: 7% increase). Increasing imports of both gas and biomass contribute to the import rise in both economic growth cases, whereas imports of solid fuels decline both under high and low GDP assumptions; oil imports increase with higher economic growth and decline under low economic growth.

Despite different developments of energy imports in quantitative terms, import dependency as a percentage of total supply stays constant at 58% in 2050 in the different economic growth cases, marginally up from 57% in all cases in 2030 and an estimated 54% in 2010.

312.

Conclusions on economic growth variants


The model based analysis shows that policy relevant indicators are pretty robust against variations in economic growth assumption, which is a significant result, given the great uncertainty in making GDP projections for the next few years let alone the next four decades.

· CO2 reduction becomes only slightly more difficult in technical terms under significantly higher economic growth. Moreover, it is important to note in this context that higher economic growth brings also more opportunities for innovation and investment in low carbon technologies, thus facilitating climate change mitigation and dealing with the competitiveness and energy security aspects. This result stems from improvements in both energy and carbon intensity facilitated by the ETS emission cap in place.

· In a similar manner, the countervailing effects through energy and carbon intensity are also present in the case of low economic growth so that there is only limited further emission reduction brought about by considerably lower GDP levels.

· RES shares are pretty robust with respect to GDP levels with variation spanning just 1 percentage point in 2050 for the RES share in gross final energy demand (overall indicator of the RES Directive). Similar results hold for the RES share transport and to a slightly lesser extent for the RES-E share.

· High economic growth gives rise to more energy intensity improvements, but would render absolute energy saving objectives with respect to e.g. a statistical year more difficult to achieve (with opposite conclusions under low economic growth). Energy saving objectives, such as the current one for 2020, are measured in absolute terms (without reference to GDP). However, energy consumption reacts to economic growth; it rises with higher GDP and declines in the opposite case.

· Policy relevant indicators regarding competitiveness are pretty much unaffected by economic growth; while ETS prices differ to some extent the effects on electricity prices are marginal.

· Exposure to external dependency measured as share of energy imports in energy supplies is also unaffected by such significant changes in GDP levels.

313.

2.3 Energy import price sensitivities


Two such sensitivities were modelled spanning a fairly wide range around the Reference case price trajectories (see assumptions part above for details). The world oil price in 2050 is assumed to be 28% higher than the Reference scenario in the high price case, whereas it stays 34% below Reference in the low price case. In the low price case, fossil fuel import prices remain broadly at the 2010 level; coal prices are stable, oil has a small peak around 2030, whereas gas prices remain weak over the next few years but recover to the 2010 level in the long run.

Higher import prices bring about higher end-user prices discouraging energy use in the various sectors and vice versa. Moreover, such developments change the competitive position of individual fuels and technologies given all the other cost elements in addition to fuel input costs in the formation of end-user prices (e.g. capital costs and taxes). Effects are therefore differentiated according to fuel and sector. For example, electricity prices are less affected than end user prices of e.g. gas. Similarly, the percentage increase of end user prices following higher import prices is much more pronounced in e.g. industry compared with transport where existing high excise taxes moderate the increase in percentage terms.

314.

Energy consumption


Under higher energy import prices (oil price up 28% on Reference in 2050), final energy consumption decreases by just 2.3% in 2050 from the Reference case. The decline spans from 1.1% in transport to 4.7% in services/agriculture where more electricity use, encouraged by higher prices of competing fuels, improves the energy intensity of the sector (electricity at use having a very high efficiency). Total electricity use in final energy consumption rises 1.0% compared with Reference in 2050.

Primary energy demand decreases 2.0% in 2050 compared with Reference, mirroring also the price induced effects in the energy transformation sectors notably in power generation as well as price inelastic parts, such as energy use as feedstock in the petrochemical industry.

Whereas higher energy prices exert only a limited effect on the level of energy consumption, the influence on the fuel mix is important. Gas demand reacts most strongly to rising prices given its use as a major input fuel for power generation where it competes with coal, RES and nuclear, which are either not affected (RES, nuclear) or less affected by rising fossil fuel import prices (see assumption part above). Gas demand in 2050 would fall by 14.7% in 2050 compared with Reference. Oil demand would also decline by 3.3% in 2050. The limited reaction is due to the concentration of oil use on petrochemicals and transport, where price reactions are small due to lack of substitutes and high existing tax levels in transport.

The use of solid fuels is encouraged despite higher import prices as gas competitiveness suffers in particular as a result of the more pronounced price increases (this is also linked to the cost structure in power generation where fuel costs are relatively more important for gas given its lower capital costs than for coal plants). Solid fuel use would increase 2.2% over Reference in 2050. Nuclear benefits also from higher fossil fuel import prices gaining 4.4% on Reference in 2050. Renewables win most, reaching 5.4% higher use compared with Reference in 2050.

In the case of low energy import prices (-34% in 2050 from Reference), final energy demand would increase by only 4.2% above Reference in 2050. The increase would be particularly high in services/agriculture where higher gas and oil use would be encouraged to the detriment of electricity. As electricity loses competitiveness, it contributes less to overall energy intensity improvements. Similar effects occur in households, while demand rises in the other sectors stay well below average. Electricity demand under low prices sinks 2.2% in 2050 compared with Reference.

Lower fossil fuel energy import prices entail 5% decrease in heat and steam demand, mainly due to decrease in industry (-9%). There is also a shift from CHP generation that looses 5% in 2050 to district heating units (+14%).

Primary energy consumption increases 2.8% in 2050 compared with Reference. The lower increase than for final energy is linked to lower electricity demand, which entail somewhat lower electricity generation and therefore transformation losses.

Again, with limited effects on overall energy consumption there are considerable changes in the fuels mix. Gas consumption increases 23.0% over Reference in 2050, while oil demand rises 4.8%. Solids, RES and nuclear, having all power generation as major areas of use, are discouraged, also because their prices do not fall (RES, nuclear) or to a lesser extent (solids). Solids use drops 7.3% below Reference, nuclear declines by 7.6%, while RES reduce least below Reference in 2050 (-6.6%).

315.

Fuel mix


Consequently, the fuel mix would be somewhat altered both in the high and in the low energy import price cases.

Table 9: Shares of energy sources in primary energy consumption (in %)

|| 2050

||| High price| Ref.| Low price| High price| Ref.| Low price

Oil| 37.| 32.| 32.| 32.| 31.| 31.| 32.4

Gas| 24.| 20.| 22.| 25.| 17.| 20.| 24.4

Solids| 17.| 13.| 12.| 11.| 11.| 11.| 10.2

Nuclear| 14.| 15.| 14.| 13.| 17.| 16.| 15.0

RES| 6.| 19.| 18.| 17.| 21.| 19.| 18.1

The marked variations in the import prices give rise to rather limited changes in the fuel share trends. Fossil fuels, especially solids, lose importance under both high and low prices, while RES make substantial inroads und nuclear progresses in the long term under high prices.

Variations across scenarios regarding the fuel shares are most important for gas, for which high import prices could lead to a considerable decline in its contribution (falling to only 17.7% in 2050 whereas low import prices could help maintain its current share in 2050 and even increase it somewhat in 2030 to over a quarter).

316.

Power generation


Fossil fuel import prices render direct use of fuels more expensive. They result in lower percentage price increases for electricity given the rather small part of fuel input costs in total electricity costs. Under high fossil fuel prices, electricity production is encouraged, whereas it falls below Reference case under low import prices.

RES and nuclear benefit from high fossil fuel import prices. Low fossil fuel prices affect in particular nuclear penetration. The RES share remains stable due to rather unchanged production still benefiting from RES support and sinking overall electricity generation (compared with Reference). The fossil fuel share in power generation would go down to only 30% in 2050 under high energy import prices, down from 55% in 2005.

CCS penetration would be somewhat encouraged by lower fossil fuel prices and corresponding higher ETS carbon prices to ensure meeting the emission cap, leading to almost 4 percentage points more deployment in 2050. On the contrary high fossil fuel prices would delay its introduction so that the CCS share would be about 3.5 percentage points lower in 2050 compared with Reference.

317.

Table 10: Electricity related indicators in different energy import price cases


Electricity prices are lower than Reference with low import and therefore power generation input prices, while the opposite is the case with high import prices. The time profile of prices remains the same as under Reference case developments (see above).

318.

CO2 emissions


The changes in the fuel mix and CCS penetration have important effects on CO2 emissions and ETS prices. With significantly more zero carbon power generation under high fossil fuel prices, ETS prices in the high import price case are somewhat below Reference, despite lower CCS penetration and somewhat higher electricity generation, given that with a constant ETS cap there is less demand for allowances. Under low import prices the opposite trends materialise and ETS prices are higher. In total, significant changes in the level of world energy prices exert only a small influence on ETS prices, as long as the coal to gas price ratio does not change significantly.

In the high fossil fuel price case, larger use of zero carbon fuels, moderated by effects on coal and lignite consumption as well as lower CCS deployment and slightly lower ETS prices bring about a marginal improvement in carbon intensity of primary energy consumption (1.35 t CO2/toe instead of 1.36 t CO2/toe in 2050 in Reference). Combined with the 2.0% improvement of energy intensity under high import prices, this leads to a 2.7% reduction of CO2 emissions below Reference in 2050.

In the low fossil fuel price case, in which oil and gas prices remain virtually flat at 2010 level through 2050 rather than increasing as in the Reference case, there is a more marked increase of CO2 emissions from Reference in 2050 (6.6%), in particular due to increases of non-ETS emissions with lower fuel prices. This is due to energy intensity deteriorating 2.8% compared with Reference in 2050 combined with a worsening of carbon intensity by 3.7%. Carbon intensity rises to 1.45 t CO2/toe in 2050 as a result of delayed CCS and higher shares of gas and of oil in the long run.

It can therefore be concluded that important further rises in oil and gas import prices, under a given emission cap for power and energy intensive industries, lead to only minor changes in CO2 emissions via limited effects on energy intensity and marginal effects through changes in fuel mix and technology deployment. The CO2 effects of lower fossil fuel prices (virtual stabilisation of fossil fuel import prices) appear to be proportionately more pronounced in the long term than those from further price increases above Reference case levels.

Table 11: CO2 reduction below 1990 (index 1990 =100) and major drivers

1990 = 2050

|| High prices| Reference| Low prices| High prices| Reference| Low prices

Oil ($(08) / barrel)| 84

Energy consumption| 109

CO2 emissions| 71.| 74.| 76.| 57.| 59.| 63.5

|||||||

Higher world energy prices bring lower CO2 emissions including in the sectors subject to ETS, which in turn reduces both demand for allowances and their price, given the fixed cap. Conversely, lower fossil fuel prices increase emissions and therefore demand for allowances, leading to higher ETS prices.

In total, differences in world energy prices exert only a minor influence on total CO2 emissions in the EU. There are feedback mechanisms via ETS carbon prices. High fossil fuel prices reduce demand and CO2 emissions and thereby carbon prices. With low fossil fuel prices there is upward pressure on CO2 emissions and carbon prices increase under ETS.

319.

Energy imports


Net energy imports fall 6.9% below Reference in 2050 under high import prices. Gas imports are particularly sensitive to variations in price levels (-15.7% on Reference in 2050) given the competitive environment in power generation and most final demand sectors, where ample substitution possibilities exist. Oil use is less flexible (transport, petrochemicals) so that oil imports decline by only 3.4% in 2050. Solid fuel imports are even less affected (-2.0%), while imports of biomass increase (4.9%) given higher demand.

Low energy prices encourage significantly higher net energy imports, which in 2050 exceed the Reference level by 11.0%. Gas is the main driver for this increase, with imports being 25.2% higher than Reference in 2050. Again, oil and coal imports react more moderately, rising 5.0% and 6.5%, respectively. With lower RES consumption, biomass imports would fall 11.5% below Reference in 2050.

Import dependency in the high price case would stay at the current level throughout the projection period reaching 54% in 2030 and 55%. Under low energy prices import dependency would increase slightly reaching 58% in 2030 and 62% in 2050 (up over 4 percentage points from Reference).

320.

Energy costs


Higher and lower fossil fuel import prices impact strongly on the EU's external energy bill. With fossil fuel prices exceeding significantly the Reference level (e.g. oil by 41% and 28% in 2030 and 2050, respectively), the EU has additional costs over Reference for fossil fuel imports of 158 bn € (08) in 2030 and of 148 bn € (08) in 2050. The average annual extra fuel bill over the next 40 years amounts to 131 bn (08); it is worth noting that this is per year and in real terms.

In the low fossil fuel import price sensitivity, i.e. in case energy import prices remain essentially at the 2010 level, there are considerable external fuel bill savings. The costs for importing oil, gas and coal would decrease by 88 bn € (08) in 2030 and by 230 bn € (08) in 2050 with respect to Reference developments, in which fossil fuel prices rise considerably. The average annual import cost saving in 2011-2050 would amount to 108 bn € (08).

Total energy system costs, i.e. the amount that the rest of the economy has to pay to the energy system for the provision of energy, including capital, fuel and other costs, amounts to 2582 bn € (08) on average in each year from 2011 to 2050. This amount does not include auctioning payments, as these expenditures for individual sectors are not costs for the economy as a whole, since the auctioning revenues are recycled back to the economy. Moreover, this cost concept excludes so called disutility costs.[128]

With higher energy import prices, total energy system costs are 187 bn € (08) per year larger throughout the period 2011 to 2050. Under the hypothesis of low world fossil fuel prices, average annual energy system costs would decrease by 155 bn € (08) per year over the same period.

321.

Conclusions on import price sensitivities


High world energy prices reduce CO2 and GHG emissions, while low prices exert the opposite influence. However, there are several other effects via fuel mix, electricity generation, ETS prices (given the same ETS cap across scenarios) and CCS incentives that modify the overall effect while working in different directions.

High fossil fuel prices lead to slightly higher electricity demand given the small reaction of electricity prices to increasing fuel input prices in the presence of large unrelated cost blocks such as capital costs, levies and taxes. Combined with a significant increase in the share of zero carbon (non-fossil) fuels there is lower demand for ETS allowances and therefore the ETS price decreases somewhat.

Lower fossil fuel prices give rise to the opposite effects. Energy consumption and CO2 emissions rise, however moderated by lower competitiveness of non-fossil, carbon free fuels. As an overall result, the effect of this fuel shift outweighs the effects through lower electricity production and lower CCS share, bringing about higher demand for allowances and slightly higher ETS prices.

The sensitivity cases show that significant changes in world energy prices exert only a small influence on ETS prices as long as the gas to coal price ratio does not change significantly.

This conclusion on rather limited effects of significant changes in world energy prices on EU GHG emission can also be derived by considering the above results on energy and carbon intensities. Important further rises in oil and gas import prices lead to only minor changes in CO2 emissions via limited effects on energy intensity and marginal effects through changes in fuel mix and technology deployment (carbon intensity). The CO2 effects of lower fossil fuel prices (virtual stabilisation of fossil fuel import prices) appear to be proportionately more pronounced in the long term than those from further price increases above Reference case levels. Regarding total GHG emission, the CO2 effects from changes in fossil fuel prices would be limited through countervailing effects of high fossil fuels prices through reduced carbon prices.

High fossil fuel prices limit business opportunities for energy exporters given that EU imports would decrease, most so for natural gas. Conversely, with lower fossil fuel prices, significantly higher gas deliveries to the EU can be assured. Import dependency increases with low world energy prices, whereas it stays below Reference at the current level throughout the projection period.

Electricity prices are significantly lower than Reference under low fossil fuel import prices, whereas they are significantly higher in the case that high energy import prices prevail.

Moreover, high energy import prices increase the EU’s external fuel bill substantially, thereby weakening the competitiveness of the EU economy. Income that would have been used to buy domestically produced goods and services would be diverted to energy exporters with only a small part being recycled into higher EU exports into these countries. On the contrary, lower fossil fuel prices give a boost to the EU economy improving its competitiveness, also through lower costs and inflation.

The external energy bill of the EU becomes significantly larger with high world energy prices (+132 bn € (08) per year over the next 40 years), whereas this bill was reduced by 109 bn € (08) annually in the case that fossil fuel prices remained broadly at the level seen in 2010. Similarly, total energy system costs would be significantly larger with high fossil fuel prices, whereas the rest of the economy would need to pay to the energy system a significantly lower amount in case of low world energy prices.

322.

2.4 Current Policy Initiatives scenario


This scenario reflects the Current Policy Initiatives (CPI) that are being discussed or undertaken in the EU context with a view to the 2020 Energy Strategy. This scenario does not attempt to give a full appreciation of all the results that might be expected from the Energy Strategy, nor does it mirror in detail the – future – policy adoption and implementation; it reflects the measures being proposed and discussed (for details see above under assumptions). While the measures focus on the medium term, the CPI scenario modelling evaluates also the long term consequences up to 2050 and provides thereby another benchmark for comparison with decarbonisation scenarios.

323.

Energy demand


Primary energy consumption under CPI declines pretty strongly between 2005 and 2020 (-6.9%) and continues to do so through 2030 when it will have fallen well below the 1990 level. There is a further decline up to 2050 (-11.6% from 2005), in which year energy consumption would be 8.4% lower than in the Reference case. There are also marked changes from Reference in 2020 (-5.0%) and 2030 (-5.8%).

These energy savings from 2005 levels are brought about by a decline in final energy demand, especially in the households and services/agriculture sectors, and by efficiency improvements in energy transformation resulting from the implementation of measures in the Energy Efficiency Plan. Bottom up energy efficiency measures reverse the trend of ever increasing final consumer demand witnessed so far in statistics and many trend scenarios, including the Reference scenario in the period up to 2020.

Total final energy demand reduces 1.3% from 2005 by 2020. Reductions by 2030 amount to 3.2%; thereafter final demand starts growing again slightly through 2050. Nevertheless, in 2050, CPI final demand stays 5.3% below Reference (even 5.6% for 2020 as CPI includes many energy efficiency policies to be implemented over the next few years).

Households show the greatest decrease below 2005 levels: by 6.1% up to 2020 as well as by 8.5% and 10.0% until 2030 and 2050, respectively. In 2020 household energy consumption is 8.9% below the Reference case, while this decline in 2050 amounts to 3.8%. This decline compared with Reference in 2050 is smaller given that large parts of the energy efficiency potential captured in CPI in the earlier years is taken up the Reference case in later years. Energy efficiency measures linked especially to Eco-design regulations and savings obligations on energy providers with respect to their customers are instrumental for this pronounced decline in CPI. Moreover, the effects on final consumer prices stemming from the proposed Energy taxation directive contribute towards reducing energy consumption.

Energy demand in services and agriculture also decreases significantly by 5.5% and 6.7% in 2005-2020 and 2005-2030, respectively. After 2030, final energy demand in this sector would resume its rising trend reflecting growing economic activity. In any case, demand in services/agriculture falls well below Reference case levels through 2050, with demand being 7.0% lower in 2050 and even 7.8 % lower in 2020. Eco-design measures, faster renovation rates for existing - especially public - buildings, promotion of energy service companies as well as energy savings obligations are key policy measures to bring about such savings. The new energy taxation directive also contributes to this decline.

Energy consumption in industry also declines from 2005 levels: by 2.3% up to 2020 and by 3.7% up to 2030. Thereafter, industrial energy demand starts growing slightly without reaching again the current level. Industrial energy demand stays below Reference scenario levels: by 5.5% in 2030 and 5.1% in 2050. Energy service companies, eco-design and energy savings obligations are among the drivers for bringing about such savings, which are somewhat moderated by healthy production growth and by the feedbacks through lower ETS prices regarding certain industrial branches. Such feedbacks stem from energy/electricity savings that reduce the demand for ETS allowances and therefore ETS prices (see below).

324.

Figure 17: Final Energy Consumption by sector in Current Policy Initiatives and Reference Scenarios (in Mtoe)


Transport energy consumption is comparatively little affected by current energy policy initiatives. Energy consumption continues to increase, exceeding the 2005 level by 5.6% in 2020. After 2025, transport energy consumption starts declining slowly, returning the 2005 level by 2050. Compared with Reference, consumption remains below the levels reached throughout the projection period (by 1.7% in 2030 and 5.7% in 2050). Changes from Reference are brought about in particular by the proposed new energy taxation system and through the somewhat more favourable policy environment for electric and plug-in hybrid vehicles, while CO2 standards exert only a limited influence given that the CO2 from cars regulation is already included in the Reference case.

While final energy demand for oil, gas and coal would continuously decline up to 2050, demand for electricity, heat and RES would increase. Most important in absolute terms is the increase in electricity demand, which rises 43% between 2005 and 2050. Nevertheless, electricity demand in CPI falls well below electricity use in Reference, reflecting measures in the Energy Efficiency Plan and revised Energy taxation Directive. CPI electricity consumption is down on Reference by 6.5% in 2030 and 4.3% in 2050.

Demand for distributed heat is rising compared to current level but is 1-2% lower than in the Reference scenario reflecting effects of measures in the Energy Efficiency Plan, in particular more efficient heating systems in houses. Heat demand in residential sector is 7% lower in 2020 compared to the Reference scenario. The difference is much lower towards the end of the projection period (1-2%) as the measures included in the Energy Efficiency Plan target short to medium term.

325.

Power generation


Rising electricity demand over time will require a similar increase in power generation and a lot of new investment in power generation and grids. Even though energy efficiency measures bring about lower electricity demand and production compared with Reference (see table 12) gross electricity production is expected to increase 41% by 2050 under CPI. Electricity based on RES is expected to make major inroads reaching a share in power generation of close to 50% in 2050.

326.

Table 12: Electricity related indicators in CPI scenario and differences from Reference


The CPI scenario takes account of the post Fukushima policy change in Member States, notably the abandoning of the nuclear programme in Italy and the new nuclear approach in Germany modifying somewhat the previously decided nuclear phase-out Moreover, it includes other changes and new initiatives, such as the nuclear stress tests that tend to increase costs for new power plants and retrofitting.[129]

The slightly higher nuclear share in 2020 reflects lower total electricity production and the modification in the nuclear phase-out provisions between the German nuclear law before the extension of nuclear plant lifetimes in autumn 2010 (mirrored in the Reference case) and the new schedule. The new phase-out schedule includes faster closure of nuclear plants in the next few years, compensated by slightly higher capacity around 2020, keeping cumulative allowed nuclear generation (in TWh) at the same level.

Fossil fuel based power generation falls significantly throughout the projection period; its share diminishes from 55% to just over 30% in 2050. Solid fuels lose most, with losses for gas based power generation remaining rather limited.

The CPI scenario has significantly lower CCS penetration in 2020 compared to the quite optimistic national plans as envisaged in 2009 (Reference scenario) and rather moderate recent progress in demonstration plants. This concerns also potential storage sites. In medium term, lower ETS price in the CPI scenario, reflecting lower energy demand due to additional energy efficiency measures, affects commercial viability of CCS. In the long term, lower numbers compared with Reference are also a result of the strong decline in solid fuels and gas based power generation.

ETS prices are lower in CPI compared with Reference in the medium to long term. The CCS incentive through carbon prices is reduced by 20% from 40 €/tCO2 to 32€/t CO2 in 2030. Consequently, the CCS share in CPI in 2030 amounts to 1% and rises thereafter significantly with high ETS prices to reach 8% in 2050. The energy efficiency measures in CPI cut electricity and fuel demand and the need for allowances, which in a context of an unchanged ETS cap leads to lower ETS prices. This limits - as a side effect - also the incentives for CCS.

Average electricity prices are slightly higher than Reference over the projection period (0.8% in 2030 and 4.0% in 2050) reflecting the lower share of nuclear post Fukushima and high investments for new electricity generation capacity, especially RES.

327.

Fuel mix


These changes in the demand side and in power generation have significant impacts on primary energy consumption and the fuel mix. Primary energy demand declines 200 Mtoe up to 2050, when it remains 150 Mtoe below the Reference case level.

In the long term to 2050, both fossil fuels and, to a limited extent, nuclear reduce their importance in the fuel mix, with solids undergoing the greatest decline (minus 8 percentage points in 2005-2050). The share of nuclear is lower also in comparison to the Reference scenario due to changes in nuclear assumptions. RES are the clear winner of this structural change, making them in 2050 the second most important fuel after oil. RES gain 16 percentage points from today's level in terms of primary energy and about 20 percentage points when accounted for in terms of gross final energy demand.

Oil remains the most important fuel throughout the projection period as the fuel mix in transport remains largely unchanged. Nevertheless oil loses 5 percentage points by 2050.With primary energy demand declining, the fuels used most in sectors that are least affected by current energy policies, such as oil in transport, are able to score a slightly higher share in the fuel mix compared with Reference.

Post Fukushima changes in nuclear (discussed above) reduce the role on nuclear compared with Reference. In this new policy environment gas and RES replace nuclear and thereby increase their share over Reference scenario levels.

These changes towards a significantly greater RES contribution bring about an important decline in carbon intensity over time (by a third between 2005 and 2050). However, with respect to Reference, there is a certain increase in carbon intensity, given that CPI relies less on nuclear and that CCS penetrates more slowly. Carbon intensity in 2050 exceeds the Reference case level by 7.7%.

328.

Table 13: Fuel mix of primary energy consumption in CPI and Reference


CO2 and GHG emissions

In spite of this deterioration of carbon intensity there is a somewhat greater CO2 reduction in CPI than in Reference; CO2 emissions in 2050 are slightly lower than in the Reference scenario. This development is due to greater energy intensity improvements brought about by vigorous energy efficiency policies, which overcompensates the worsening of carbon intensity due especially to lower use of nuclear and CCS.

This energy intensity effect on CO2 emissions is somewhat moderated by the effect of energy efficiency on carbon intensity via ETS prices. Declining ETS prices, triggered to some extent by lower energy demand, give rise to lower incentives for investing in e.g. CCS and nuclear, thereby giving rise to somewhat higher carbon intensity.

329.

Table 14: CO2 emissions and drivers in CPI and Reference scenarios


Energy intensity improvements are particularly pronounced in the earlier years of the projection period thanks to vigorous new energy saving measures targeting in particular the short and medium term. In total CO2 emissions reduce 40% between 2005 and 2050, up one percentage point from what would be achieved under reference case developments. With respect to 1990 CO2 emissions in CPI decline by 41.3% up to 2050. The Reference scenario has a decrease of 40.4%.

Total GHG emissions in 2050 decrease 38.6% below the 1990 level, which is slightly less than in the Reference case (-39.7%), given the significantly lower carbon price until just before 2050, reflecting especially successful energy efficiency policies. This means, on the other hand, that total GHG emissions reduce faster in CPI than in Reference in the time horizon to 2020 and also to 2030.

330.

Energy imports / security of supply


Lower energy demand and the changes in the political environment after the Japanese nuclear accident of March 2011 give rise to significant changes in EU energy production, which is down on Reference by 9.0% in 2050. Nuclear production sinks 25.8% compared with Reference in 2050, while RES production is 7.8% higher. Also gas production is seen in a more favourable light (+4.0%).

Despite lower indigenous production, energy imports are 7.5% lower in 2050 than in the Reference scenario due to the policy measures, notably on energy efficiency, included in CPI. Nevertheless, net energy imports are expected to broadly stabilise throughout the projection period (peaking in 2015, when they exceed the 2005 level by 6.4%, before declining 7.5% up to 2050).

Biomass and natural gas imports increase significantly, whereas oil imports decline moderately and solids see their imports sink considerably. Gas imports in 2050 are expected to be 26% higher than they were in 2005. Oil imports decrease 6% over this period, while solid fuel imports plummet 56%.

Import dependency remains broadly unchanged from Reference case and also current levels. Up to 2020, this indicator rises from 54% at present to reach 56%. This is one percentage point less than in Reference, reflecting the impact of efficiency measures mainly on imported fuels. In 2030, import dependency reaches 57.5%, up one percentage point on Reference, which is largely a result of lower nuclear availability. In 2050, this indicator amounts to 58% in both CPI and Reference.

331.

Conclusions on Current Policy Initiatives scenario


As a result of current policy initiatives, energy consumption is expected to be reduced significantly. The decline in both final and primary energy consumption is most pronounced in the medium term, for which most of the measures have been designed. The implementation of the Energy Efficiency Plan brings important reductions in final energy demand, especially in the household and services/agriculture sectors.

In terms of primary energy, consumption sinks throughout the projection period, falling below the 1990 level by 2030 with a continuing decline thereafter. In 2050, energy demand decreases 12% below the 2005 level. As a result, energy intensity improves 1.8% pa, which is 0.2 percentage points up from the number in the Reference case.

This decline in energy consumption is connected with significant changes in the fuel mix, which are also linked, among other things, to post Fukushima changes in the policy environment for nuclear energy in several Member States. Compared with Reference, the contribution of nuclear and solid fuels declines, while oil, gas and in particular RES account for higher shares in primary energy consumption in 2050.

In a comparison over time, fossil fuels lose as much as 16 percentage points from 2005 to 2050, of which solid fuels account for 8 percentage points, oil for 5 and gas for 3 percentage points. Renewables are the clear winner, benefiting from several policies not even directly targeting RES and of course those measures included in the 2008 Energy and climate package. The RES share in primary energy rises 16 percentage points, while the nuclear share remains almost constant (only a slight decrease post Fukushima).

The RES share in gross final energy consumption increases 20 percentage points from 2005 by 2050 when it reaches 29%. Also the RES shares in transport and power generation rise considerably reaching 49% and 20% in 2050, respectively. Taking a 2030 perspective, the overall RES share in final demand grows 16 percentage points to reach 25% in 2030 under current policy initiatives. RES in transport account for 13%. RES contribute 44% to power generation.

Electricity generation also falls compared with Reference, given successfully implemented energy efficiency policies, but would exceed the 2005 level by 41% in 2050. Again, there are significant changes in the generation mix, which also explain to a large extent the fuel mix changes at the primary energy level. Almost half of power generation in 2050 would be based on RES, up from just 14% in 2005. Nuclear loses around 10 percentage points share in power generation in 2005-2050 given strongly rising electricity production and the recent changes in the policy environment for nuclear. The share of fossil fuel based electricity generation diminishes from 55% in 2050 to just over 30% in 2050 mainly due to reductions in solid fired power generation.

These changes in power generation towards lower solid fuel contribution compared with Reference entail lower demand for ETS allowances giving rise to lower ETS prices thus also providing fewer incentives for CCS. As an overall result of these simultaneous changes, the ETS price falls 20% below the Reference level in 2030. In 2030 almost 1% of gross power generation undergoes CCS, while this share rises to 8% in 2050.

Developments of the fuel mix and the CCS penetration bring about a 0.9% pa decline in carbon intensity from 2005 to 2050. This decline in carbon intensity is marginally smaller than the one under Reference developments, reflecting in particular post Fukushima changes for nuclear and lower medium term ETS prices following strong energy efficiency measures, which, as an indirect effect, limit CCS penetration.

Nevertheless, energy related CO2 emissions reduce slightly more than under Reference developments. CO2 emissions in CPI sink 41.3% while the decline amounts to 40.4%. Total GHG emissions in CPI reduce 38.6% below 1990 by 2050.

Total energy imports broadly stabilise throughout the projection period, despite significant increases in biomass and natural gas imports. Oil and notably solid fuels import decline. Import dependency remains broadly unchanged from Reference case and also current levels.

The CPI scenario involves higher system costs stemming notably from the additional investment triggered through additional energy efficiency requirements and the restructuring of the energy and transport systems including the lower nuclear contribution due to upward revised costs and more Member States renouncing the nuclear option. Moreover the inclusion of the Energy taxation directive adds to these additional costs. Taking into account the fuel savings from energy efficiency measures as well as the taxation induced savings, energy system costs in the period 2011 to 2050 increase by an annual amount of bn 37 €(08). These cost estimates do not consider possible changes in the utility levels of consumers regarding the behavioural changes induced that are, in any case, not directly measurable and can only be captured in the modelling indirectly via the concept of compensating variations.

Average electricity prices rise at only a slightly faster pace compared with Reference developments. In 2030, the average electricity price exceeds Reference by only 1%; this price increase becomes 4% in 2050.

[1] ec.europa.eu/energy/strategies/consultations

[2] Questions 1, 5 and 7 were open questions and 2, 3, 4 and 6 were multiple choice.

[3] ec.europa.eu/energy/strategies/consultations

[4] COM(2011) 21, 26 January

[5] European Council, Brussels, 29/30 October 2009, Presidency conclusions. 15265/1/09

[6] European Parliament resolution of 4 February 2009 on "2050: The future begins today – Recommendations for the EU's future integrated policy on climate change; resolution of 11 March 2009 on an EU strategy for a comprehensive climate change agreement in Copenhagen and the adequate provision of financing for climate change policy; resolution of 25 November 2009 on the EU strategy for the Copenhagen Conference on Climate Change (COP 15)

[7] COM(2011)112, 8 March

[8] Both roadmaps provide analysis under global climate action assumption.

[9] COM(2011)144, 28 March

[10] COM(2010) 2020, EUROPE 2020 - A strategy for smart, sustainable and inclusive growth

[11] Energy related emissions account for almost 80% of the EU’s total greenhouse gas emissions with the energy sector representing 31%; transport 19%; industry 13%; households 9% and others 7 %.

[12] Other important issues related to the environmental impacts of our energy system include air pollution, water pollution, wastes and impacts to ecosystems and their services. Indeed, negative trends in land, water (fresh and marine) and air quality depend on how energy is generated and used: combustion processes, especially in the case of small unregulated biomass plants, give rise to gaseous emissions and cause local air quality and regional acidification; fossil and nuclear fuel cycles (as well as geothermal production) emit some radiation and generate waste of different levels of toxicity; intensification of biomass use (and of biomass imports) may lead to forest degradation; bioliquids may lead to GHG emissions and direct and indirect land use driving prices for food up globally; last but not least, large hydropower dams flood land and may cause silting of rivers.

[13] International Energy Agency, World Energy Outlook 2010. The EU contribution would decline from 13% of global CO2 at present to 8% in 2035 if all world regions are only pursuing current policies.

[14] As regards market developments, questions about adequacy and intensification of incentives for investments; future of support schemes for RES and other technologies; support mechanisms/regulations for energy efficiency; etc might arise.

[15] SEC(2010) 1346 final, COMMISSION STAFF WORKING DOCUMENT State of play in the EU energy policy

[16] COM(2006) 545.

[17] 2009 Eurostat data are the latest official data.

[18] The scenarios of the 'Energy trends 2030' (update 2009) are accessible at the following address: ec.europa.eu/energy/observatory/trends_2030/doc

[19] COM (2011) 109

[20] Communication Energy Efficiency Plan 2011, SEC(2011) 280 final, SEC(2011) 277 final, SEC(2011) 275 final, SEC(2011) 276 final, SEC(2011) 278 final, SEC(2011) 279 final

[21] COM(2006) 851.

[22] COM(2010) 84.

[23] SEC(2010) 505.

[24] Regulation (EC) No 663/2009 of the European Parliament and of the Council of 13 July 2009 establishing a programme to aid economic recovery by granting Community financial assistance to projects in the field of energy.

[25] Regulation 994/2010

[26] Council Directive 2009/119/EC of 14 September 2009 imposing an obligation on Member States to maintain minimum stocks of crude oil and / or petroleum products.

[27] Directive 2005/89/EC of the EP and of the Council of 18 January 2006 concerning measures to safeguard security of electricity supply and infrastructure investment.

[28] COM(2009) 192, The renewable energy progress report.

[29] 2009/28.

[30] Relates to share of biofuels and other renewable fuels in petrol and diesel for transport

[31] The 2020 target can be fulfilled through the use of renewable energy in all types of transport. Energy use in maritime and air transport counts only for the numerator, not the denominator.

[32] 'Heating' is a catch-all term for energy consumption that is neither for transport nor in the form of electricity.

[33] A 1997 White Paper established an indicative target of 12% of primary energy consumption in 2010, which was used to derive the 21% target for RES in power generation in 2010

[34] Council Directive 2009/71/Euratom of 25 June 2009 establishing a Community framework for the nuclear safety of nuclear installations.

[35] Directive 2009/31/EC on the geological storage of carbon dioxide adopted as part of the Climate and Energy Package in 2009

[36] According to recent Technology Roadmap from IEA/ UNIDO, CCS could reduce CO2 emissions by up to 4.0 gigatonnes annually by 2050 in industrial applications, accounting for 9% of the reductions needed to halve energy-related CO2 emissions by 2050.

[37] Short-term projections for oil, gas and coal prices were slightly revised according to the latest developments in the Reference scenario as compared to the version used in the low carbon economy roadmap.

[38] Regulation on CO2 from cars 2009/443/EC

[39] This includes also some energy-related non-CO2 emissions, e.g. methane emissions from coal mining and losses in gas distribution networks and F-Gas emissions related to air conditioning and refrigeration. While the former are estimated to decrease under current trends, the latter are projected to increase considerably. For a more detailed analysis of the overall GHG reduction efforts needed and of trends in non-CO2 emissions see the Impact Assessment of the Roadmap for moving to a competitive low carbon economy in 2050 (SEC(2011)288).

[40] Correspondingly, a higher amount of banking of ETS allowances beyond 2020 takes place in the CPI scenario compared to the Reference scenario, rising from around 2000 Mt to 2700 Mt in 2020 and reducing more slowly in the post-2020 period. For a detailed interplay of ETS, other policies, carbon prices and ETS allowance banking see SEC(2010)650 part 2.

[41] The Reference scenario does not cover the European Commission CARS 21 (Competitive Automotive Regulatory System for the 21st century) initiative and the recent initiatives of car manufacturers as regards electric vehicles.

[42] The results diverge slightly from the assessment done for the Energy Efficiency Directive. In fact, measures of the Energy Efficiency Directive were taken but they are expected to produce effects over a longer period of time. Also the stringency of energy efficiency measures is assumed to be slightly lower. However, a more vigorous implementation of the Energy Efficiency Directive is assumed in decarbonisation scenarios which all surpass the indicative 20% target in the decade 2020-2030.

[43] Global developments as regards shale gas are taken into account when projecting global gas prices.

[44] Article 194:

1. In the context of the establishment and functioning of the internal market and with regard for the need to preserve and improve the environment, Union policy on energy shall aim, in a spirit of solidarity between Member States, to:

(a) ensure the functioning of the energy market;

(b) ensure security of energy supply in the Union;

(c) promote energy efficiency and energy saving and the development of new and renewable forms of energy;

(d) promote the interconnection of energy networks.

[45] COM (2011)112

[46] Please see IA on Low carbon economy Roadmap for the analysis of impacts of decarbonisation on energy import prices SEC(2011)288.

[47] Impact assessment report SEC(2011)288 final, section 5)

[48] European Commission: Communication Analysis of options to move beyond 20% greenhouse gas emission reductions and assessing the risk of carbon leakage (COM(2010) 265 final). Background information and analysis, Part II (SEC(2010) 650).; ec.europa.eu/clima/documentation/international

[49] For details and the implications on the cost and benefit quantifications please refer to Annex 1, part A, point 1.4 and part B, points 1.4 and 2.7.

[50] Used also in the Low Carbon Economy Roadmap and Transport White Paper.

[51] This analysis does not prejudge the final outcome of the legislation process on these policies and will not be able to deliver a quantitative assessment of the consequences of the Energy 2020 strategy.

[52] Scenario 3 reproduces 'Effective and Widely Accepted Technologies' scenario used in Low Carbon Economy roadmap and Transport White Paper on the basis of scenario 1bis.

[53] Global climate action requires that each region uses its RES potential. Moreover, geopolitical and security of supply risks can justify the reliance on domestic energy sources.

[54] The scenarios are based on model assumptions, which are consistent with the input for the 2050 Low Carbon Economy Roadmap. Recognising the magnitude of the decarbonisation challenge, which implies a reversal of a secular trend towards ever increasing energy consumption, this Energy Roadmap has adopted a rather conservative approach as regards the effectiveness of policy instruments in terms of behavioural change. However, the Roadmap results should not be read as implying that the 20% energy efficiency target for 2020 cannot be reached effectively. Greater effects of the Energy Efficiency Plan are possible if the Energy Efficiency Directive is adopted swiftly and completely, followed up by vigorous implementation and marked change in the energy consumption decision making of individuals and companies. In modelling terms this means a significant lowering of the discount rate used in energy consumption decision making of hundreds of millions of consumers.

[55] As specified in the RES directive for the calculation of the 20% target by 2020.

[56] With much more variable supply and demand some electricity produced needs to be stored. Losses, linked to storage, lead to lower consumption than production of electricity. When calculating the RES-E share in line with the RES directive (focussing on gross final energy consumption i.e. excluding energy losses to pumped storage and hydrogen storage), the RES share in electricity consumption amounts to 97%.

[57] For a detailed analysis see SEC(2011)288, section 5.2.14.

[58] For example by making sure that rich habitats are not fragmented, ensuring the integrity of Natura 2000 sites and the coherence and connectivity of its network. Green Infrastructure developments can lead to win-win situations, where negative environmental impacts of energy-related infrastructure can be mitigated while adaptation to climate change is enhanced, as well as public acceptance of alternative energy projects.

[59] Annex 1, table 37, pages 83

[60] The European Environment Agency assessed the amount of biomass that could be used in an environmental sustainable way in EU-25 by 2030 at 295 Mtoe.

[61] .For a detailed analysis of these interactions see SEC(2011)288, sections 5.1.4, 5.2.7 and 5.2.10.

[62] SEC(2010) 650, Commission Staff Working Document accompanying the Communication from the Commission to the European Parliament, the Council, the European Economic and Social Committee and the Committee of the Regions - Analysis of options to move beyond 20% greenhouse gas emission reductions and assessing the risk of carbon leakage: Background information and analysis.

[63] For further analysis of the role of energy price shocks see SEC(2011)288.

[64] "Roadmap 2050: a practical guide to a prosperous, low-carbon Europe; Volume 1 – Technical and Economic Analysis" (European Climate Foundation, 2009)

[65] As discussed in Annex 1, this represents a cautious approach. Whereas investment costs are displayed at their actual maximum levels, future benefits are priced in at a lower level.

[66] Disutility costs are a concept that captures losses in utility from adaptations of individuals to policy impulses or other influences through changing behaviour and energy consumption patterns that might bring them on a lower level in their utility function. The PRIMES model has a micro-economic foundation which allows it to deal with utility maximisation and to calculate such perceived utility losses via the concept of compensating variations. While these costs capture relevant short term transition costs, their relevance and appropriate calculation over a long time horizon is challenging. This concept has to assume that preferences and values remain the same, even over 40 years, and it compares utility with a hypothetical state of no policy or no change in framework conditions. Examples of such decreases in utility are lowering thermostat in space heating, reducing cooling services in offices, switching lights off, staying at home instead of travelling, using a bicycle instead of a car, etc.

[67] Auction payments are expenditures for individual sectors, and are not considered as costs for the economy as a whole, since the auctioning revenues are assumed to be recycled back into the economy in a neutral way. However, one could also have taken account of the shadow costs in making public transfers and it is not guaranteed that this transfer would be purely neutral for the economy, as shown by the discussions on the optimal reallocation of auction revenues (see above).

[68] When taking a macroeconomic view, i.e. by excluding auctioning revenue that are recycled to the economy, and excluding disutility costs, the Delayed CCS scenario has lower costs than the Diversified supply technologies scenario. However, when the economic actors' perspective is taken, i.e. auctioning and disutility costs are included, the lowest system costs materialise in the Diversified supply technology scenario (for details see Annex 1, part B, point 2.7).

[69] The difference in ETS prices compared to Effective and Widely accepted technologies presented in the Low Carbon Economy Roadmap is due to additional energy efficiency measures, the revised Energy Taxation Directive and changed assumptions for nuclear after Fukushima. The share of nuclear is considerably lower than in decarbonisation scenarios presented in the Low Carbon Economy Roadmap. Current Policy Initiatives and all policy scenarios in this exercise are based on revised assumptions on nuclear (abandonment of the nuclear programme in Italy, change of nuclear policy in Germany, no new nuclear plants in Belgium and upwards revision of costs for nuclear power plants). Moreover, electricity demand is lower due to stringent energy efficiency measures. In addition, assumptions on the potential of electricity in transport were revised, following more closely the scenarios developed in the White Paper on Transport leading to lower utilisation rate of nuclear power plants than in the Low Carbon Economy Roadmap Scenarios. Electric vehicles flatten electricity demand and thus incentivise baseload power generation.

[70] A dedicated infrastructure modelling was performed with the PRIMES model and the main results are presented in Annex 1.

[71] The modelling does not show this situation arising because the model assumes full cost recovery of capital investments in all scenarios

[72] Europe 2020 COM(2010) 2020

[73] EU 2020 Flagship Initiative Innovation Union SEC(2010) 1161

[74] The fastest previous scale-up was for electricity generation from nuclear power, which expanded at a rate of approximately 25-30% per year between 1960 and 1980 globally. The decarbonisation scenarios almost all envisage a major roll-out of CCS starting after 2030 and reaching average rates of up to 36% per year in 2030-2040 (20% pa in 2030-2050); similarly but closer to now, certain RES technologies could be soaring, especially from 2010 to 2030 at average annual rates of up to 20% and 15% per year for off-shore wind and solar electricity, respectively.

[75] No further analysis has been done as regards the impact of increased revenues of oil and gas exporting countries on imports from the EU.

[76] The social dimension might be better tackled in a decarbonisation roadmap treating all the interdependencies among sectors such as energy, transport, industry and agriculture than in a sectoral roadmap dealing with energy only.

[77] See literature review section in the report "Studies on Sustainability Issues- Green Jobs; Trade and Labour" (2011) commissioned by the European Commission, DG Employment.

[78] "Studies on Sustainability Issues- Green Jobs; Trade and Labour" (2011) commissioned by the European Commission. The leading objective has been to analyse the employment consequences of the implementation of policies to achieve the key EU environmental targets of a 20% cut in emissions of GHG by 2020 compared to 1990 levels (increasing to 30% if other countries make similar commitments), a 20% increase in the share of renewable energy, and the objective of a 20% cut in energy consumption (the 20-20-20 targets).

[79] 'EmployRES: The impact of renewable energy policy on economic growth and employment in the European Union' (2009), commissioned by the European Commission, DG Transport and Energy

[80] "Roadmap 2050: a practical guide to a prosperous, low-carbon Europe; Volume 1 – Technical and Economic Analysis" (European Climate Foundation, 2009)

[81] SEC(2011) 288 final page 44 and 90-91

[82] High RES scenario relies mainly on domestic sources of renewable energy.

[83] Please see more specialised indicators in Annex 1, part B, section 2.5.

[84] Results for primary energy consumption should not be confused with the energy saving targets for 2020 which is calculated against the projected consumption for 2020. Relating this savings objective to energy consumption in 2005, similar to the calculations in the scenarios, would be equivalent to a saving target of 14% in 2020.

[85] The price projections ensure full recovery of costs associated with electricity supply in order to depict scenarios in which the investment in production, storage, grids, taxes, etc are fully covered by revenues from selling electricity. In that sense they are not forecasts of future electricity prices, as systems may evolve, in which, contrary to the overall practice today, such investments are partly remunerated by other schemes.

[86] A literature review on climate change impacts in the European energy supply sector as part of the European Commission contract 'Climate proofing EU policies' has identified the following main impacts:

• Cooling water constraints for thermal power generation (especially during heat waves), with nuclear appearing to be the most vulnerable technology

• Damage to offshore or coastal production facilities due to sea level rise and storm surges

• Damage to transmission and distribution lines due to storm events, flooding

• Unpredictable hydropower potential

• Affected yield in renewable energy sector (hydropower in Southern Europe, possibly biofuels due to vector diseases and forest fires)

• Melting permafrost affecting energy production and distribution in cold climates

• Damages and output constraints in wind energy due to storms and increased average wind speed

[87] It has been considered more useful to check scenarios against objectives of the EU energy policy than against those of the Roadmap that focus on instruments and processes to deliver more certainty to investors.

[88] Scenarios for the Low Carbon Economy Roadmap of March 2011 show the additional costs of delayed action.

[89] European Commission, DG Economic and Financial Affairs: 2009 Ageing Report: Economic and budgetary projections for the EU-27 Member States (2008-2060). EUROPEAN ECONOMY 2|2009, ec.europa.eu/economy_finance/publications The “baseline” scenario of this report has been established by the DG Economic and Financial Affairs, the Economic Policy Committee, with the support of Member States experts, and has been endorsed by the ECOFIN Council.

[90] European Commission, DG Economic and Financial Affairs: 2009 Ageing Report: Economic and budgetary projections for the EU-27 Member States (2008-2060). EUROPEAN ECONOMY 2|2009, ec.europa.eu/economy_finance/publications

[91] EU energy trends to 2030, Directorate General for Energy in collaboration with Climate Action DG and Transport DG, 2010

[92] COM(2011)112, 8 March 2011

[93] Communication from the Commission: Europe 2020. A strategy for smart, sustainable and inclusive growth. COM(2010)2020, Brussels, 3.3.2010.

[94] European Commission, DG Economic and Financial Affairs: Sustainability Report 2009. EUROPEAN ECONOMY 9|2009, ec.europa.eu/economy_finance/publications

[95] European Commission, DG Economic and Financial Affairs: Public Finances in EMU 2010. EUROPEAN ECONOMY 4|2010, ec.europa.eu/economy_finance/publications
$/barrel in 2007 prices for 2030, equivalent to 134 $/barrel in 2008 prices. The IEA World Energy Outlook 2009 assumed 115 $/barrel in 2008 prices for 2030.
$/€, there will be a somewhat faster increase in energy prices in euros than in dollar.
$(08)/barrel, representing a doubling from Reference case in that year. In the following years, the genuine oil shock case depicts some oil demand reaction and a subsequent gradual decline of oil prices towards Reference case levels without reaching those, not even in 2050 (still being 18% higher). On the contrary, an alternative development was also examined, in which the oil prices would stay at the high 212 $/barrel level throughout the rest of the projection period. In the latter case, the 2050 oil price exceeds the Reference case level still by two thirds. (Results can be found in the above mentioned Impact Assessment and are not repeated here).

For the other sectors (aviation and industry), the baseline follows a conservative approach which reflects the specifications in the directive on the evolution of auctioning shares and the provisions for free allocation for energy intensive sectors based on benchmarking.

[102] On 28 October 2009 the European Commission adopted a new legislative proposal to reduce CO2 emissions from light commercial vehicles (vans). The draft legislation is closely modelled on the legislation on the CO2 emissions from passenger cars (Regulation 443/2009) and it is part of the Integrated Approach taken by the Commission in its revised strategy to reduce CO2 emissions from cars and light commercial vehicles (COM(2007) 19 final). Not including this proposal in the 2050 Reference scenario could lead to an increased bias towards vans, which is not justified given the likelihood of its adoption towards the end of 2010/beginning of 2011.

[103] NER covers 300 million allowances set aside in the new entrants reserve of the EU ETS for the co-financing of commercial demonstration projects of environmentally safe CCS as well as innovative RES technologies

[104] All measures included in the scenario underpinning the IA for the Energy efficiency Directive are included. Energy (saving) results can differ given different framework conditions flowing from all the additional assumptions above. Moreover, it should be considered that scenario 3 Energy Efficiency should show contrasted results in terms of energy consumption so that a significant individual contribution of energy efficiency towards decarbonisation can be identified. Scenario 1bis includes some adjustments to reflect somewhat less optimistic expectations for penetration of energy efficiency products/renovation of buildings.

[105] In Europe, the New European Driving Cycle is the official driving cycle used for vehicle type approval. According to a study carried out for the Commission in 2009, there is some discrepancy (typically 10-20%) between the fuel consumption as measured on the NEDC and that in real world driving. Source: Sharpe, R.B.A. (2009) Technical options for fossil fuel based road transport, Paper produced as part of contract ENV.C.3/SER/2008/0053 between European Commission Directorate-General Environment and AEA Technology plc; eutransportghg2050.eu/cms/assets, p.9

[108] NEMS database and reports, IEA studies, industry surveys, EU project reports, etc.

[109] IEA (2010), Projected Costs of Generating Electricity, 2010 Edition. IEA, NEA, OECD, Paris

[110] Energy Information Administration, Annual Energy Outlook 2010, December 2009, DOE/EIA-0383 (2009)

[111] Definitions in the studies may not totally overlap, in particular for fixed and variable costs.

[112] The exchange rates used are: 1.34USD/EUR (USD2010 to EUR2010).

[113] Abbreviations in the figure: ST Coal: Steam Turbine Coal; CCS: Carbone Capture and Storage; PC with CCS: pulverised coal with CCS; IGCC: Integrated Gasification Combine Cycle; GTCC: Gas Turbine Combined Cycle; PV: photovoltaic.

[114] Greater deployment of RES or other low carbon technologies in decarbonisation scenarios is due to carbon prices/values as well as other specific changes (including higher RES values) depending on the scenario, but does not involve greater operational aid.

[115] Eichhammer et al. (2009), Study on the Energy Savings Potentials in EU Member States, Candidate Countries and EEA Countries Final Report, Fraunhofer ISI and ENERDATA and ISIS and Technical University Vienna and WI, March 2009.

[116] Due to the variety of appliances available (in particular for boilers) the values here are chosen as examples and due to lack of data it is possible that the typical appliances of the different sources do not correspond entirely to the PRIMES technology.

[117] Please refer to the PRIMES model description available at :

www.e3mlab.ntua.gr/e3mlab/PRIMES%20Manual/The_PRIMES_MODEL_2010.pdf

[118] Note: for EV 1l/100km is approximately 8.5kWh/100km; an exchange rate USD to EUR of 1.2USD/EUR has been used.

[119] The discount rate for private individuals includes risk aversion; risk premiums are added for other actors and are technology specific.

[120] As part of the European Commission contract 'Climate proofing EU policies'.

[121] Interim results of the FP7 project 'European RESPONSES to climate change'

[122] … and perhaps ever – except for much higher economic growth materialising (see below under sensitivities)

[123] Freight transport does not include international maritime.

[124] The percentage of emissions captured is calculated as the ratio between the total emissions captured and the potential emissions of thermal power plants, which are the remaining emissions plus the emissions captured.

[125] Regulation (EC) No 443/2009 of the European Parliament and of the Council of 23 April 2009 setting emission performance standards for new passenger cars as part of the Community’s integrated approach to reduce CO2 emissions from light-duty vehicles, OJ L 140, 5.6.2009, p. 1–15.

[126] The split between ETS and non-ETS emissions reflects over the whole period the ETS scope as valid from 2013 onwards.

[127] However, it should be noted that such higher electricity demand could lead to higher CO2 emissions, depending on the fuel input structure, which are accounted for under power generation (see below)

[128] Disutility costs are a concept that tries to capture losses in utility from adaptations of individuals to policy impulses or other influences through changing behaviour and energy consumption patterns that might bring them on a lower level in their utility function. The PRIMES model, having a micro-economic foundation, deals with utility maximisation and can calculate such perceived utility losses via the concept of compensating variations (amount of additional income that would bring the individual on the same level of utility as experienced before the change). However, this concept has to assume that preferences and values remain the same, even over 40 years, and has to compare utility with a hypothetical state of no policy or no change in the framework conditions. Numbers in particular in the longer term are uncertain. The numbers shown above relate to costs that reflect actual payments.

[129] There are slightly higher risk premiums for new nuclear investment in this scenario, considering that investors might factor into their decisions the possibility that the policy reaction to any hypothetical further nuclear accident may affect the nuclear plants under investment consideration, even though such an accident could happen rather far away geographically. Requiring thereby a slightly higher return on investment to cover this political risk has also certain effects on new nuclear investment. As a result of these changes in the policy environment, the nuclear share is somewhat lower than Reference in the long term, for which the Italian withdrawal from nuclear is particularly important. Moreover, lower ETS prices in CPI reduce the economic advantages connected to nuclear investments.