The Future Energy Supply of Society: Options, Risks and Choices

by Executive Director Robert Priddle

Lecture held at a seminar in the Academy on March 16, 2000.

Robert Priddle, born 1938, is Executive Director at The International Energy Agency (IEA). He has also been Director at UK Petroleum Science and Technology Institute in Edinburgh and has had some prominent offices at the British Ministry of Energy.

Let me start by telling an anecdote on one of the keywords of the title: the word "option". It was the only word that caused problems in the Conference of IEA Ministers in Paris in May 1999. Initially draft Ministerial Communiqué says: "Ministers asked the Secretariat to continue assessing the full range of energy issues and options, including renewable, energy and nuclear power." After a long discussion, the word options was changed into "choices" and everybody was satisfied.

Why? Because there are evidently differences between these two words when translated into different languages: it seems that the German translation of the word "option" means that you are indicating which is the right course to take. German speakers did not want to suggest the IEA should make decisions for them.

But in English: options mean a range of possibilities; it is a "choice" which means a conscious decision.

So, in proposing to add one word to the title of the conference as the topic of my speech: "The Future Energy Supply of Society: Options, Risks and Choices", I am suggesting that, while we need to look at all the options and evaluate the risks, we also need to move on and make some positive choices.

Main Trends in World Energy (1995-2020)

Let me first outline the key trends of the world energy demand supply, using the results of the IEA's long-term outlook exercise.

Our latest long-term outlook is the 1998 edition of the World Energy Outlook. It was carried out on the basis of a "business-as-usual" scenario, projecting forward the trends, which could be econometrically derived from the latest data available - at that date, up to the end of 1995. We did not, in that base case, take into account the impacts of new policies and measures that might be implemented subsequently. Currently we are revising the model to prepare the 2000 edition of the outlook.

The 1998 World Energy Outlook projects strong growth in energy demand, in line with economic growth. Our projections show that world energy demand could grow by 65 per cent between 1995, the base year of our analysis, and 2020. This is equivalent to an average annual rate of 2 per cent over the outlook period.

A key message of this outlook is that fossil fuels will continue to dominate the energy mix. Indeed, 95 per cent of the additional energy demand between 1995 and 2020 will be met by fossil fuels. In absolute terms, some 92 per cent of total primary energy demand in 2020 will be fossil fuel-based.

The outlook also stresses the increasingly important role of developing countries in the energy scene. Two-thirds of the increase in energy demand over the period 1995-2020 will occur in China and the other developing countries. By 2020, China could be the second largest energy-consuming region in the world, after North America.

Global CO₂ emissions will rise by 70 per cent over the same period, higher than the increase in energy demand because of the rapid increase in coal consumption in developing countries and constraints on nuclear power generation in the developed world.

World Energy Demand Projections by Fuel

The share of coal in the energy mix remains almost unchanged. Some three-quarters of additional demand for coal will be in the power generation sector. Growth is fast in China and South Asia (mostly India), which are expected to base their growth on further development of their indigenous supplies of coal.

Oil continues to be the dominant energy source. Most of the increase in oil demand will stem from additional demand for transportation.

Demand for gas increases fast, particularly in the OECD regions. Natural gas is the preferred fuel for many applications, especially for new power stations. Gas consumption nearly doubles over the outlook period.

World nuclear power remains almost static. During the outlook period, the retirement of several units will offset the increase in nuclear output from new or refurbished plants.

Hydropower is expected to show moderate growth, with most of the increase coming from outside the OECD.

Finally, the use of other renewable energy sources increases steadily, but remains at low levels. Under business-as-usual assumptions, the share of non-hydro renewables will be around 1 per cent of the primary energy mix by 2020.

Oil Supply

On the supply side, we do not see the world running out of oil. There will be enough supplies to meet demand. However, there are two important issues that arise from our analysis of global oil supply.

The first is that the oil-importing countries' dependence on a small number of suppliers will rise. We expect an increased reliance on oil imports from the Middle East, as oil production outside this region is expected to reach a peak within the first half of the outlook. The share of Middle East OPEC in world crude oil production increases, to reach almost 50 per cent sometime in the second half of the outlook period. This suggests an increased potential for oil supply disruptions.

The second issue is that we expect unconventional oil to become increasingly important as we get closer to 2020. The expansion of unconventional oil production will require significant investment. Because of the long lead times involved, there is a risk of a mismatch between supply and demand.

This is the IEA's view on long-term oil supply. It depicts the trend and does not attempt to predict the precise course of events from year to year. The current tightening of oil supply is one such a deviation from the trend, but that is another topic. We are concerned that the current restraints on world oil supply and the sharp rise in oil prices could feed inflation and slow economic growth, thereby giving rise to problems, particularly for developing countries.

Gas Supply

In our outlook, we have also examined the availability of natural gas to meet the growing demand. Gas reserves are not expected to set limits on gas production, but there is a need for continuing exploration. The international transit of natural gas will need to be facilitated to allow the further use of gas world-wide.

We also expect the gas import dependency of OECD Europe and the Asia-Pacific region to rise significantly. Europe would have to import 60 per cent of its gas by 2020. This is the combined effect of high gas demand (to fuel, mostly, new power generation schemes) and a decline in European production, starting in the second half of the outlook. The security of gas supply in Europe needs attention.

Different Supply Options and their Risks

In the light of this background, I would now like to explore how these trends relate to our common concern with sustainable development. At the same time, I would like to link the discussion somewhat closer to the specific challenges faced by the Swedish energy sector. We all agree that sustainable development is very desirable; and the definition contained in the Brundtland Report of 1987 has stood the test of time: "Development that meets the needs of the present without compromising the ability of future generations to meet their own needs." However, we all struggle when it comes to the choices necessary to operationalise such general objectives in the energy area.

One promising approach is to consider the issues in terms of successful management of risk. A commitment to sustainable development involves a commitment to the management of risk in all its dimensions, in particular to the management of large-scale risk and the preservation of economic and environmental options for present and future generations. In this perspective, the task is to improve the resilience to error of the economic, environmental and social systems.

The two main concepts involved are aversion to risk and the high costs of sudden and discontinuous change. The latter concerns the costs for those who will have to deal with them in all probability, future generations. The availability of their options, of their possibility to choose freely between different courses of action unconstrained by unsustainable past policy decisions, is the essential criterion for decision-making in this perspective.

Three Dimensions of Risk Management in the Energy Sector

At the IEA, we are interested in the implications of such an approach of multi-dimensional risk management for the energy sector. The three dimensions of risk about which I would like to talk to you today in the context of the global energy background and Swedish energy policy in particular are:

• The risk of supply interruptions,
• The risk of environmental degradation; especially global climate change, and
• The risk through large-scale accidents due to technical failures or human error.

The first issue, management of the risk of supply interruptions has several elements; including achieving stable and transparent relations between consumers and suppliers and the diversification of energy supply. The resolution to fight the temptation "to put all your eggs into one basket", that is to go all out for what may appear to be the economic or political least-cost solution in the short-run, is an essential part of risk minimisation.

In recent years, we have all become familiar with risk management in the environmental area, notably in the context of the global effort to contain the risks of climate destabilisation, with the associated economic and social welfare costs. We attempt to lower greenhouse gas emissions in order to stabilise their concentrations so as to reduce the likelihood of potentially severe environmental damage. The political process has, very significantly, put an implicit value on the benefits and costs involved by adopting the commitments in the Kyoto Protocol (even if it would be fair to say that the political judgement was made without much idea of the costs involved).

Last but not least, I come to the classic and most common form of risk management, the effort to ensure the safe and accident-free technical operation of the plants and machinery involved in the production and consumption of energy. Customarily, this sort of management relies on clear definition of management responsibility, backed by technical standards, regulation and inspections. Applications include the extraction of fuels, their transport over large distances, storage and processing, the running of power plants, as well as the setting of standards for technical end-use equipment. Due to the sheer complexity of its technical operation and the continued absence of a comprehensive solution to the problem of waste disposal, nuclear energy merits special attention in this context.

There are two common elements in all three areas of risk management. First, in all three areas the reduction or the avoidance of risk involves cost. Second, in all three cases the public and the private sector have to act together, to manage the risk and, possibly, to share the costs. I might add a third element. All three dimensions of risk, security of energy supply, climate change and accident prevention, are elements in the decision of the Swedish government to phase out nuclear power. In the remainder of my talk, I will discuss that decision using this analytical framework.

Nuclear Power in the World Energy Outlook

Let me first offer a few words on the global nuclear outlook as we portrayed it in the World Energy Outlook. Global nuclear electricity generation in 2020 is expected to be at about the same level as in 1995, the base year of our projections. However, the nuclear power profile of each region varies.

A significant number of new nuclear plants are expected to be constructed in Asian countries, with China and South Korea leading the growth.

The Transition Economies once had very ambitious nuclear programmes, but lack of necessary funding to construct new plants (or in some cases to complete unfinished reactors) is likely to lead to a decline in nuclear power in these countries.

Within the OECD, nuclear power declines significantly, as new plant build is limited and several nuclear reactors reach the end of their design life.

The most dramatic decline is expected in North America, where some of the oldest plants are located. By 2020, nuclear electricity generation could shrink to half of today's level and its share in total electricity generation could fall to one third of what it is today.

In OECD Europe, the decline is less pronounced. A number of nuclear reactors have been completed recently, and plant upgrades in some countries will also add extra capacity. The share of nuclear in the electricity generation mix is nevertheless projected to be halved by 2020.

In the OECD Pacific, nuclear electricity generation increases substantially, with several new plants constructed in Japan. The generating share of nuclear increases. Retirements in this region are less important, since the growth in nuclear in this region is fairly recent compared to OECD Europe and North America - nuclear capacity in OECD Pacific increased by more than 40 per cent between 1990 and 1997.

The reasons for this decline of nuclear in most OECD countries are partly economic - competition from fossil fuels, notably natural gas in CCGTs - and partly political. As you will see on the next slide, several OECD countries have policies (often buttressed by specific laws) to restrict nuclear power.

Restrictions on Nuclear Power

This map illustrates the political aspect of the issue. It provides a picture of the restrictions imposed by several OECD member governments to restrain nuclear plant build or to shut down existing plants. For most of these countries the duration of the restrictions is indefinite.

Australia, Austria, Denmark, Greece, Ireland and Norway do not have nuclear power and will not permit new plant construction. Italy shut down its only nuclear plant in 1990. Poland discontinued the construction of a nuclear power plant. Belgium, Germany, Netherlands, Spain, Sweden and Switzerland all use nuclear power but have decided not to build new plants or intend to phase out nuclear power.

On the other hand, there are countries that make use of nuclear power and have no plans to eliminate this option. This group includes Japan, Korea, France, Canada, the United States and the United Kingdom.

Economics of New Plants

On the economic side of the issue, let me show you some results of the electricity generating cost study that was published jointly by the IEA and the NEA in 1998. In this study, 18 countries - most of them OECD Member countries - were invited to provide cost estimates for plants that will be commissioned in the medium term.

The graph on the left shows the number of countries in which the three main technologies considered in the study were cheapest by a margin of 10 per cent or more. It includes 11 countries out of a total of 18 countries that provided data for two options, Results are given at 5 per cent discount rate and at 10 per cent. The results confirm the current cost advantage of fossil-fuelled power generation. In particular, they show the strong competitiveness of gas-fired power generation (CCGTs).

Power Plant Economics - Effect of a Carbon Value

However, if new policies to reduce GHG emissions were adopted, in particular carbon taxes, then the competitive position of nuclear would obviously improve.

The chart shows the effect of a carbon value on the relative economics of fossil fuel and carbon-free electricity generation, including capital costs.

The figures are indicative only, but certain features warrant attention.

The introduction of a carbon value would most strongly affect the position of coal, since this is the most carbon-intensive of fossil fuels. This estimate shows that a $25-$35 carbon value would bring coal-fired generating costs up to the level of nuclear generating costs.

Gas-fired combined cycle gas turbine plants are - and are likely to remain - the favoured option for new power generation schemes in many OECD countries where gas is available. As you can tell from the chart, this type of generation is less sensitive to a carbon value. This is the combined effect of high conversion efficiency and the lower carbon content of natural gas.

Given the nuclear moratoria, the long licensing and construction periods and the absence, as yet, of a clear basis of establishing a carbon value, construction of new nuclear plants in the near-term will be limited. Over the longer-term, however, (especially if commitments to reduce GHG emissions become truly binding and are extended beyond the Kyoto time horizon) nuclear power could clearly improve its competitive position vis-à-vis fossil fuels, especially if fossil fuel prices begin to rise and if other ways to reduce emissions prove insufficient.

This is the wider context of the decision by the Swedish government to begin implementing the phase-out nuclear power, whose implications I would now like to explore.

Nuclear Power and Security of Energy Supply

Phasing out of nuclear power eliminates one option and thus reduces energy diversification. Given the continuing uncertainties as to whether renewable energy can constitute a cost-effective substitute, it is likely that the phase-out of nuclear power will increase the risk to security of supply. The security of uranium supply, vital for nuclear power generation is, by contrast, more political than physical or economic.

On the other hand, the Nordic electricity market offers important security benefits for all the participants because of the complementarity of the fuel mix in generation throughout the region: currently hydro, nuclear, coal and gas. There is also likely to be a small but growing proportion of renewable sources in the mix. Diversification will be reduced, however, if both coal-fired power (in Denmark) and nuclear energy (in Sweden) are phased out in the same time period.

Nuclear Power and Climate Change

Concerning the second dimension, global environmental impacts, the abolition of nuclear power will also increase the risk connected with the emission of greenhouse gases. The IEA's World Energy Outlook very clearly indicates how difficult it is for all Member countries to reduce their greenhouse gas emissions substantially. If Sweden were to substitute coal-fired generation for base-load nuclear power, these difficulties would increase. Even substitution with gas would increase greenhouse gas emissions noticeably. The nuclear phase-out will make the achievement of Sweden's Kyoto target more challenging, possibly more costly and probably both.

Sweden has taken account of these risks by imposing conditions on the closure of any further reactors. The Government has acknowledged that nuclear phase-out gives rise to problems of employment, welfare, competitiveness and the environment. Very sensibly, Parliament has not sought to impose a deadline by which all reactors must be closed.

Conditions on further closure have been included in legislation. These require more efficient energy use, electricity conservation, conversion from electricity and increased electricity supply from other sources at acceptable cost. It is our understanding that only if these conditions are met will Barsebäck 2 be allowed to close, even though a planning date of 1 July 2001 has been suggested. The Swedish government is actively supporting the achievements of the required conditions through a 9.2 billion SEK programme to develop renewables and improve energy efficiency.

It is obviously important, because of the risks that I have discussed, that the procedure for implementing Parliament's decision, and the analysis of the conditions set out in the legislation, be as transparent as possible, with independent analysis of the relevant information.

Accident Risk

Third, the phase-out of nuclear power, of course, reduces the risk of future nuclear accidents. This dimension of risk, or at least the perception of it, is customarily at the centre of the discussion about nuclear energy.

The probabilities of accidents are low but the potential damage can be very large. This is a structure of risk that sits particularly uncomfortably with a risk-averse public. It has at all times required special safeguards and the setting of strict technical safety guidelines. The evidence is that this risk in Sweden is one of the lowest in the world. In fact, Sweden is an example of how to run a safe, nuclear industry.

The phase-out of nuclear power will, unfortunately, do little to reduce the other major associated risk - the unsolved issue of disposal and long-term storage of nuclear waste. Sweden is making progress in resolving this issue; but a rapid phase-out of nuclear power could intensify the problem by producing a great deal of waste arising from decommissioning.

The decision of the Swedish government to forego the nuclear option thus reduces risk in one dimension and increases it in two others. The relative valuations of these risks is at the heart of the discussion about sustainable development in Sweden.

Ultimately, the necessary trade-offs should be the outcome of broad-based, diligent and transparent societal discussion. The information needed for such processes includes both "objective", technical information and some measurement of "subjective" preferences concerning the exposure to different kinds of risk. Willingness to pay for their reduction and management is an important test, which puts heavy emphasis on an objective evaluation of these costs. Sweden has not rushed into the nuclear phase-out. A very difficult decision process is being handled in a sensitive and responsible way.

Concluding Question and Answer

I would like to end my presentation by remarking how the nuclear debate in Sweden now appears to the outsider. In a sense, the question has now changed. It is now not so much will nuclear be phased out as "when will renewables become economic?" That is, indeed, another very important - and difficult - issue. And therein lies another sort of risk. No-one can guarantee that renewables will become economic. The risk is that there will be continued pressure to force the pace of change, notwithstanding Parliament's conditions.

I am not going to pronounce on the right choice. Ultimately, only Sweden itself can and must decide this question. What the IEA can offer, however, is a detached approach to the issues decision-makers need to take appropriately into account. I hope that this talk may have made a small contribution in this respect.