Back to search

ENERGIX-Stort program energi

Uncertainties in the European Energy Market: Modeling approaches and policy issues

Alternative title: null

Awarded: NOK 7.1 mill.

The European energy market is exposed to a number of uncertainties; economic, climatic, technological and policy. Uncertain future energy prices and income growth affect demand for energy. Uncertain weather has impact on supply of renewable energy from water, wind and solar. Uncertainties with respect to future energy technologies affect current investment decisions and therefore also future supply of energy. Finally, policy uncertainty, for example, with respect to future policy goals and instruments, affects the business opportunities of the energy industry and hence has impact on both current and future decisions. Because all sources of uncertainty affect ? directly or indirectly ? all energy markets, a consistent treatment of uncertainty requires a modeling framework that encompasses the entire energy industry, both demand and supply in several market segments. Our starting point is a large scale deterministic model of the European energy market (LIBEMOD) that provides a description of the markets for eight energy goods; three types of coal, natural gas, oil, two types of bio energy and electricity. Some of these energy goods are traded in European markets, whereas others are traded in world markets. The model takes into account interaction effects between different markets, for example, natural gas can be used to produce electricity and can also be consumed by end-users, for example, for the purpose of heating and cooking. The numerical model LIBEMOD determines all prices and quantities in the energy markets, and also emissions of CO2 by country and sector. In the first part of the project we transformed the deterministic LIBEMOD model to a stochastic model, which can be used to analyze effects of economic, climatic, technological and policy uncertainty. We find that relative to the case of no uncertainty, policy uncertainty with respect to future climate policy increases investment in international transmission capacity. This is in line with economic theory. In the economics literature, learning curves, that is, the relationship between the total unit cost of a technology and the corresponding R&D expenditures ? frequently referred to as the rate of progress ? is studied and estimated. In 2015 we reviewed this literature to identify the potential of cost savings of renewable energy technologies. We find that the rate of progress is about 8-10 percent for solar power, and between 10 and 25 percent for bio power, depending on type of bio technology. For wind power, earlier estimates were around 10 percent, whereas currently the saving potential for wind power is typically seen as minor. For Carbon Capture and Storage (CCS), no estimate for the rate of progress was found ? this simply reflects the very limited number of full scale CCS plants in operation world wide. The EU aims at cutting its GHG emissions by 40 percent (relative to 1990) by 2030, and by (at least) 80 percent by 2050. In the EU Roadmap study, the prediction for the power generation sector is an emissions reduction of 95 percent by 2050. Such an ambitious target will require huge investments in carbon-free electricity. Because the potential for hydro is rather limited in Europe, there is necessary to phase in substantial capacity in wind- and solar power. These are, however, intermittent technologies; at night, when there is no sun, there will be no solar production. With heavy investments in solar and wind power, it is crucial to identify the correlation between wind and sun in Europe. If there is much wind when there little sun, then these two technologies can to some extent substitute each other. We have used metrological data from NASA, combined with European load data, to study the questions presented above. In general, we find a small correlation between wind and sun in all European countries. However, the correlation is negative, that is, when there is little/no solar power, the lack of supply from this technology can to some degree be replaced with wind power. The two countries with the highest negative correlation are Norway and Iceland. Slightly more than half of the European countries have positive correlation between sun and consumption of electricity. Most of these countries are found in Southern Europe. Italy, Spain, Greece, and Portugal are the countries with the highest positive correlation. In these countries, there are good conditions for development of solar power as solar radiation is high. In most European countries, there is positive correlation between wind and consumption of electricity. This correlation is particularly high among the Nordic countries, with Norway at the top of the list.

Policy should take account of factors like economic performance, energy security and emission of harmful materials, as well as uncertainty, in a systematic, coherent way when deciding which policy to implement in order to reach a specified target. This re quires development of a framework that assesses different concerns when energy decisions are taken under uncertainty. The aim of this project is to develop a detailed stochastic, numerical model of the energy industry that covers all countries in Europe t hat can be used for policy analyses when agents make decisions under uncertainty. Both domestic policies, for example, initiatives to spur Norwegian production of clean energy, as well as EU policies, for example, the triple 20 percent targets and the roa d to a low carbon European society, can be examined with our model. The model provides a detailed description of the European energy industry where energy carriers are extracted/produced, there is trade in energy between European countries and end-users , as well as power plants, use energy. There is inter-fuel competition, and electricity can be produced by a number of technologies; fossil-fuel based, hydro, nuclear and renewable. We use a method frequently referred to as scenario aggregation to trans form a deterministic numerical model into a stochastic model where several agents make decisions under uncertainty simultaneously. Our method ensures a consistent modeling of uncertainty, and is in line with both statistical theory and economic theory of decisions under uncertainty. This is not the case for some well known methods that have been applied in the literature, in particular the scenario method where a few pictures of the future are analyzed, and also Monte Carlo simulations. These methods assu me implicitly that agents always know which future that will materialize.

Publications from Cristin

No publications found

No publications found

No publications found

No publications found

Funding scheme:

ENERGIX-Stort program energi