EU Energy Outlook 2050: How will Europe evolve over the next 30 years?

First published on Energy Brainpool in 2021

There are significant market changes: rising CO2 certificate prices lead to higher profitability of renewable energies, keyword: Power Purchase Agreements (PPAs). What do these developments indicate for power prices, revenue potential, and risks for photovoltaics and wind?

The electricity markets in Europe are subject to constant change, which makes current price scenarios indispensable. This is the only way to correctly assess market developments, assets and contracts, investment decisions, PPAs or business models.

The “EU Energy Outlook 2050” shows the development of the “Energy Brainpool” scenario for EU-27, UK, Norway, and Switzerland. The actual processes in the individual countries can vary considerably. In order to make well-founded decisions, detailed modelling of the individual national markets and the existing influencing factors, including sensitivity analyses, is essential.

What does the power plant fleet of the future look like?

The power plant fleet in Europe has developed over many decades and was particularly dominated by fossil generation capacities (see Figure 1). Many of the power plants on the market have already reached an advanced age. They will have to be replaced by 2050, including all nuclear power plants. The only exceptions are power plants already under construction.

The current climate debate is showing results. By now, a total of 10 EU states have decided to phase out coal in order to limit the negative effects of high emissions. Well-known and proven technologies are ready to be used in the future: Gas-fired power plants, renewable energies and nuclear power plants.

Wind power and photovoltaics in particular continue to have great growth potential. These technologies are now competitive, thanks to the sharp drop in costs over the past ten years. This is also evident from the increasing number of PPA-based projects, especially for solar installations. Experts expect this development to continue. In the “EU Energy Outlook 2050”, the share of these volatile renewable energies (vRES) will rise to around 65 percent of the total supply by 2050. Renewables account for 76 percent of the power plant fleet.

At the European level, gas-fired power plants will be the main source of controllable fossil generation capacity in the future. This is due to the lower emissions compared to coal-fired power plants. Even with carbon capture storage (CCS), the latter continue to lose importance.

The capacities of nuclear and coal-fired power plants will be reduced by more than 55 percent by 2050. Germany, France, Great Britain, Spain, the Netherlands, Finland, Italy, Ireland, Portugal and Denmark have announced coal exits for the future. As a result, the currently installed output of hard coal in particular will fall sharply to around 18 percent by 2050.

In overall terms, the share of generation capacity from controllable thermal power plants will be reduced from 47 percent to around 24 percent by 2050. This will have a considerable impact on the structure of power prices, which will increasingly be influenced by vRES prices.

Why does electricity demand rise until 2050?

The demand for electricity will rise by around 31 percent by 2050 (see Figure 2). Power demand increases mainly due to national hydrogen strategies, enhanced electrification in households, and the expansion of electric cars. According to the plans of the European Commission, most of the economic growth is taking place in the tertiary service sector, which also needs more electricity. In the industrial sector, increased efficiency can prevent a significant increase in electricity consumption.

The amount of electricity produced from coal-fired power plants is declining sharply, decreasing about 58 percent by 2030 and about 91 percent by 2050. However, production from gas-fired power plants will increase by around 25 percent by 2050. Wind and solar power plants will generate around 45 percent of the electricity in 2050. Around 36 percent of the electricity comes from controllable fossil-fuel power plants. The remaining electricity is produced by controllable, renewable energies such as biomass power plants or storage lakes. 79 percent of the electricity is generated emission-free. As a result, the climate targets set would be missed.

The long-term development of commodity prices

The development of the most important commodity prices until 2050 is based on the “Sustainable Development Scenario” of the IEA’s World Energy Outlook 2021 (IEA, 2021). In this scenario, three goals are defined: The stabilization of climate change, clean air, and universal access to modern energy. In particular, it is assumed that the majority of industrialized countries will achieve the net zero emissions target in 2050, thus limiting the increase in global average temperature to 1.65 °C.

Compared with today’s level, the prices for gas, oil and hard coal in this scenario fall continuously until 2030 (see Figure 3). Gas prices in particular are currently at an extraordinarily high level; the more pronounced will the decline therefore be in the coming years.

Since the last update of the World Energy Outlook, the future assumed gas prices have been slightly lowered, while the price paths for coal and oil have remained virtually unchanged. However, the CO2 price assumed for the EU has risen sharply compared to the WEO 2020, from the equivalent of just under EUR 114/t CO2 in 2040 to over EUR 140/t CO2. This 23 percent increase – as discussed in the next section – directly affects expected electricity prices.

In the WEO 2021, the IEA has also introduced a new scenario, the “Announced Pledges Scenario” (APS). In contrast to the “Sustainable Development Scenario” (SDS), only those emission reductions to which governments have already committed themselves in the form of pledges will be realized. Global CO2 emissions are therefore not reduced until 2030, and in 2050 emissions are still more than twice as high as in SDS. With regard to commodities, the same development of CO2 prices is assumed as in SDS. However, since gas, coal and oil are used even more after 2030, prices in this scenario are higher than in SDS, which additionally drives up average electricity prices.

A sensitivity scenario can be calculated and delivered soon for the new WEO scenario APS upon request.

The development of average power prices

Primary energy and CO2 prices are of particular relevance for the development of average, unweighted power prices between 2022 and 2050. Due to rising CO2 prices, power prices will increase continuously from 2030. However, this development is dampened by the high feed-ins from wind and photovoltaic power plants. These can only be partially offset by an increasingly flexible demand for electricity. This leads to hours with low and, more frequently, negative power prices.

Compared to the last edition of the EU Energy Outlook from June 2021, the calculated power prices have increased by an average of 10 percent between 2030 and 2050. The reason for this is the increase in assumed CO2 prices shown above, based on the current WEO. The actual developments in the individual countries differ considerably in some cases. This is shown by the fluctuation ranges in Figure 4. Due to the development of commodity prices, countries with a low expansion of renewable energies in particular record a stronger increase in power prices.

If we look at power prices on a monthly basis, we can see the seasonality and volatility of the power market (see Figure 5). For the winter, the analyses show rising prices due to the temperature sensitivity of electricity demand.

On the other hand, power prices are usually significantly lower in the summer. This effect is reinforced by the increasing share of solar power generation, which has a price-reducing effect.

What revenues can wind turbines achieve?

The sales value is the average volume-weighted power price that wind power plants can achieve on the spot market. Only generation hours with positive power prices are taken into account (including 0 EUR/MWh).

As Figure 6 shows, the sales value of wind energy will rise continuously by 2030. However, the annual increase is low, also due to the fact that capacities

continue to grow steadily. Simultaneous generation reduces power prices in these hours (merit order effect). The sales volumes (share of generated volumes at power prices >=0 EUR/MWh) will decline only slightly on average in the EU, and in some countries will also decline very significantly. Sales revenues are derived from the product of sales values and sales volumes.

The many hours in which controllable fossil-fuel power plants set the price – despite the high share of renewable energies –make rising positive revenue streams possible. The fluctuation range of the markets shows how different the country-specific average revenue opportunities of wind turbines are.

Energy Brainpool defined, among other things, the indices sales value and volume in the white paper “Valuation revenues of fluctuating renewable energy sources“. These indices enable a realistic determination of the revenue potential of fluctuating, renewable energies in the electricity market.

What revenues can photovoltaik systems (solar) achieve?

The development of sales values for solar energy is in line with the trend for sales values for wind energy, but at a lower level (see Figure 7). The reason for this is the significant simultaneity effect of solar energy: the majority of electricity is generated during the day in summer. In hours when a lot of solar power is generated, the price of electricity drops and thus, the revenues fall.

The sales volumes for solar energy remain almost constant on EU average, but also decline in individual countries. The large fluctuation margin of the solar sales values in the individual countries shows how strongly the revenue opportunities vary. However, it should be noted that in a sunny country, high revenues are possible even with low sales values. The reason for this is that the plants are operating at better utilization rates.

Solar thermal plants for electricity generation are a marginal technology in the scenario and are not being expanded on a large scale.

Increase in price volatility in detail

In the scenario, many factors lead to a significant increase in price volatility. In the figure above, price volatility is shown with boxplots describing annual demand-weighted baseload prices and quantiles of hourly prices in the considered year. The generation costs of controllable, fossil power plants increase due to the development of rising commodity prices and prices for emission certificates. On the other hand, the expansion of fluctuating, renewable energies has a price-reducing effect.

As a result, from today’s perspective, extreme prices occur much more frequently and become a normal part of the electricity price structure of the day-ahead market. The high extreme prices rise continuously over time, while the low extreme prices remain at an almost constant level after 2030. The reason for this is the flexibility options such as electrolysers, heat pumps, and electric mobility, which are becoming increasingly important in the future electricity supply.

Fluctuations due to weather risks in the determination of the sales values of fluctuating generators

In Germany and other European markets, due to the subsidies of wind and solar power, the focus so far was solely on the influence on the quantities produced, in case of the weather risks associated with fluctuating renewable energies. The guaranteed feed-in tariff and market premium meant that all price risks were irrelevant. For wind turbines, for example, this meant that high wind volumes generated high revenues and little wind led to low revenues. In order to estimate revenues, an expected quantity (e. g. P50 quantity) was consequently multiplied by the fixed subsidies.

However, this situation changes for the increasing number of generators that sell their electricity on the market without any subsidies. Their revenues are now based on volatile power prices. Since these prices are also affected by weather, we have to consider weather impacts in both, volume and price respects. Herein, we will demonstrate the existence of an anti-correlation between the two weather effects, which tends to stabilize revenues and reveals that weather risks are prone to systematic overestimation.

This anti-correlation is illustrated by the modelling results of a scenario calculation for the year 2021 using the weather years 2005 to 2016. Figure 9 shows the percentage fluctuations in producible volumes and sales revenues around the respective mean value. Multiplying the producible volume (in MWh) by the sales revenue (in EUR/MWh) gives the yearly revenues of the plant (in EUR/MW/a). Their deviations from the mean value are given as percentages as well as in EUR/MWh. Basically, they refer to revenue fluctuations of the produced volumes that are expected on a long-term average (i.e. P50 volume). A glance at the figures reveals a pattern: windy years show high volumes with low sales revenues, and windless years show low volumes with higher sales revenues. This can be traced back to the cannibalization effect of renewable energies, which can help stabilize annual revenues.

For example, producible volumes in the weather year 2007 are more than 16 percent above the P50 value, but sales revenues in EUR/MWh are 8 percent lower (see Figure 9). The annual revenue of the plant therefore fluctuates only by + 7.5 percent. This translates into + 3.12 EUR/MWh deviation from the revenues planned with the P50 volume as the long-term average.

In contrast, producible volumes in the weather year 2010 will be 10 percent lower, which corresponds roughly to the P90 volume. However, the lower volumes are more than offset by the rise in sales revenues of over 11 percent, and annual revenues remain stable (+ 0.7 percent). However, if expected revenues of a plant are calculated through multiplication of the P90 volume (of the weather year 2010) by the average sales revenue only, the weather risk is systematically overestimated and the stabilizing anti-correlation is ignored.

Looking at Figure 10, however, when comparing the weather years 2010 and 2016, it also becomes clear that this anti-correlation is not equally present in every weather year. It can be cancelled out by simultaneous solar feed-in. For example, in 2016, despite low annual volumes, wind power feed-in was distributed more strongly over hours with simultaneously high solar feed-in than in 2010, so that sales revenues hardly increased at all.

Overall, there are weather year specific fluctuations in revenues that reflect both weather-related volume and price risks. If the volumes of P90 (e. g. 2010) or P50 weather years (e. g. 2009) are used to estimate weather risks, it is advisable to consider these in combination with expected price effects. Otherwise, weather risks may be overestimated.

The values shown alter substantially in the future due to changing power plant parks and thus shifting cannibalization of renewable energy. Read more in our white papers “Power-Purchase-Agreements I & II“.

Fluctuations due to different scenario assumptions

Energy Brainpool offers a variety of different scenarios. Figure 11 shows the different trends of the scenarios. The fluctuations relate both to the assumptions on the development of commodity prices, the power plant fleet, e-mobility and other flexibility options.

Figure 12 shows the corresponding results of the power prices of the respective scenarios.

*EU-27 plus United Kingdom, Norway and Switzerland, depending on the evaluation, the significant states were selected to determine the mean value.

Sources:

• [1] EU Reference Szenario, 2016: Energy, transport and GHG emissions – Trends to 2026

• [2] IEA, 2021: World Energy Outlook

• [3] entso-e, 2021