This chapter begins with the assertion that energy use is the most important human-generated source of greenhouse gases and that worldwide energy consumption is expected to increase steadily in coming decades. Thus, attempts to limit greenhouse gas concentrations in the atmosphere must focus on energy supply and demand. The costs of reducing greenhouse gas emissions from fossil fuel combustion are central in the debate over appropriate climate change policy responses.

The chapter reviews historical energy use and the development of energy technology, describes alternative approaches to energy-industrial analysis, summarizes studies of energy use and carbon emissions in seven countries, and reviews the results from recent global studies of the costs of reducing carbon emissions.


Attempts to project future energy use and energy-related carbon emissions are fraught with uncertainties related to population growth, economic growth, and technological change. Two key uncertainties seem especially important. (1) The long-term trend in carbon intensity of energy use will be governed by the rate of development of competitive alternatives to fossil fuels (such as solar or wind energy). (2) Energy and industrial development in the less industrialized countries over the next century will greatly influence the growth rate of carbon emissions. Less industrialized countries contribute only about a quarter of world carbon emissions today. Due to high anticipated population and economic growth, they are expected to contribute more than 2/3 of the total by the year 2100. However, through the choice of infrastructure investments and introduction of new, more efficient technologies, the energy intensity of economic activity could be kept lower than the present level in industrialized countries. Study of the less industrialized countries is identified by the authors as a major research need.

Approaches to energy-industrial analysis: Three different approaches to analyzing energy demand are compared.

(1) The economic (or behavioral) approach relates energy demand to energy prices, other prices, incomes, available technologies, and often to a time trend which serves as a proxy for changes in technology, tastes, etc. It may include regional differences (e.g. in climate or infrastructure) or the existence of energy conservation programs.

(2) The engineering (or optimization) approach focuses on energy technology, looking at demand projections for end-use energy services (such as space heating, industrial process heat, and transportation). It generally projects the least-cost technology to be selected for a particular end-use. These models tend to overpredict the adoption of new technologies because they leave out behavioral factors and market barriers. Energy conservation programs have not generally led to as much energy conservation as these studies projected, apparently because of behavioral factors (such as inadequate information, hidden costs, time and effort required, different secondary features and preferences).

(3) The social-psychological approach looks at energy demand within a framework of the full range of human activities, including individual and social influences (e.g. background and beliefs, events, institutions, group influences). This is important for the long-term projection period (50-100 years) of the climate change debate because changes in attitudes or social organization may significantly affect future energy use, especially in developing countries where most of the growth in emissions is projected to occur.

Each approach can provide important insights, thus they can complement each other. Hybrid models that combine the engineering and economic approaches have become popular. Social-psychological analyses involve more detail, therefore are usually done on a more micro level than economic or engineering models.

Overview of energy use and cost projection methods: Methods of modeling energy use are described in considerable detail. Some key "challenges" which must be met for successful model building are highlighted, but no assessment of the accuracy or reliability of economic and engineering models is given. Some of the key challenges are: (1) choosing the level of aggregation (more disaggregation may be needed to capture the heterogeneity of decision-making within groups); (2) assumptions about future technologies; (3) negative feedbacks in carbon flows (carbon reduction in one country could lead to lower international energy prices or movement of production plants, leading to an increase in carbon emissions in other countries).

Strategic energy-sector planning: Approaches to long-term planning are suggested which could develop new ways of doing things that are more energy-efficient. Industrial ecology systematically analyzes interactions between human activities and the environment, seeking to optimize the total industrial materials cycle from virgin material to finished material to the ultimate disposal of wastes. By influencing infrastructure development, public policymakers can explicitly plan infrastructure that reduces the need for fossil fuel use (for example, emphasizing mass transit development over road development). This is especially possible in less industrialized countries where less infrastructure is already in place.

Country studies: Studies of energy use and carbon emissions are summarized for seven countries in differing stages of development: Brazil, China, India, Japan, South Korea, the Russian Federation, and the United States. More than half of present global carbon emissions originate in these 7 countries and that share is expected to increase.

In Brazil, most electricity is generated by hydropower; however, few new sites are available for hydroelectric generation. Sugar cane is used to make ethanol fuel for light vehicles. To constrain growth in carbon emissions, more emphasis on energy conservation and biofuels development will be needed.

China is self-sufficient in energy and is expected to be a net exporter of energy for several years to come. More than 75% of its energy comes from coal, creating serious pollution problems, particularly near industrialized urban areas. Old, inefficient technology is still in use in many areas and the efficiency of energy use is low. Energy conservation that emphasizes technological progress will be important both to reduce carbon emissions and for economic and social development.

In India, the energy system is heavily based on coal. In addition, more than one-fourth of the energy consumed comes from noncommercial sources, mainly biomass. Most biomass is collected or homegrown and is not recognized in official energy and economic accounts. The use of biomass fuels puts tremendous pressure on forests and creates serious air pollution. With high population growth, ample coal reserves, and ambitious development goals, it will be difficult for India to constrain its growth in carbon emissions.

In Japan, strong economic growth has boosted energy consumption; however, Japanese industry has developed energy-saving technology and has shifted to an energy-economizing industrial structure. The population growth rate is low. Japan has almost no indigenous energy resources. Its dependence on imported fossil fuels and already highly efficient energy system will make it difficult to further constrain carbon emissions.

Korea is a newly industrialized country, characterized by remarkable economic growth since the late 1960s. Energy efficiency and conservation are emphasized in government economic policy. Korea is highly dependent on imported fossil fuels. Increased prosperity has led to rapid increases in vehicle ownership in the 1990s. Improvements in energy efficiency and substitution of nuclear power for fossil fuels appear to be likely ways for Korea to slow its increase in carbon emissions.

The Russian Federation has abundant reserves of fossil fuels and an unusually high dependence on natural gas. Russia is one of the largest energy exporters in the world. The population growth rate is low. Economic activity has declined since 1990, resulting in a large decline in carbon emissions. Greenhouse gas emissions are expected to remain below 1990 levels for the next 5-10 years, perhaps longer, because economic growth during recovery is expected to be concentrated in non-energy-intensive industries.

The United States has an abundance of domestic energy resources and a modest population growth rate. In 1990, it had the highest energy consumption and carbon emissions in the world. Carbon emissions are expected to grow steadily unless policies are adopted to prevent it. To reduce emissions growth, the U.S. would need to focus on energy conservation and substituting natural gas for coal in electricity generation.

Global studies which project the costs of reducing greenhouse gas emissions: Results are reported from an extensive study organized by the Energy Modeling Forum at Stanford University (the "EMF 12 study") entitled Controlling Carbon Emissions: Costs and Policy Options. The study used 13 standardized scenarios, and was implemented by 14 modeling teams employing a wide variety of technoeconomic models. The population and economic growth assumptions used in the baseline scenario are similar to those used by the IPCC. Other key input assumptions concern (1) the price and availability of energy resources, and (2) the costs and efficiencies of current and future technologies for energy supply and use. (The chapter's authors note that, in the long run, the assumptions made about the costs of substitutes for conventional oil and gas determine the cost of controlling carbon emissions.)

The costs of different levels and rates of carbon dioxide emissions control were projected. All models project that market intervention will be required to achieve the emissions targets in all regions. For the more stringent carbon limits (e.g. the "20% reduction scenario") many models project that carbon taxes of hundreds of dollars per ton would be required. Because of China's projected economic growth, its participation in the control program would be essential.

Incremental reductions in emissions become more costly as the absolute level of emissions in a given year is reduced. Sensitivity analyses show that even a 10-year delay in the target date for global stabilization of emissions can reduce the peak cost projections dramatically. Cost-effective mitigation paths are generally characterized by emissions that grow in the early years, peak, and then decline in later years. (This is because premature retirement of capital stocks is avoided, technological change benefits increase, and time-cost discounting is greater.)

Impacts of climate change on energy and industrial systems: With an abrupt change of focus, this section looks briefly at how climate change may affect energy demand and supply. The projected impacts on demand depend strongly on how much energy in a region is used to heat and cool buildings. In cool areas with limited air conditioning, demand for heating would probably decline. Where cooling is more important, or becomes more important, the increase in cooling demand overwhelms the decrease in heating demand. The use of air conditioners is still low in most countries, but experience in a number of tropical countries has shown that when they become affordable their proliferation can be very rapid.

Projections of climate change impacts on energy supply are more problematic because current climate models do not make reliable regional projections of moisture levels. Future moisture levels will affect wood fuel resources and hydroelectric power resources. Climate model results showing reduced snowpacks and greater flood intensities suggest possible problems in the management of hydroelectric power facilities.


The scope and depth of this chapter are impressive. It contains an enormous amount of information about the modeling of energy use and the projected costs of reducing carbon emissions. It should provide a useful overview, for researchers and policy analysts, of recent results of economic and engineering models. In addition, it summarizes some of the weaknesses and uncertainties inherent in the models. It draws some broad conclusions about strategies and costs of reducing carbon emissions.


Although the chapter points out some uncertainties in the models, it makes no attempt to evaluate their reasonableness and does not address the enormous errors that are inherent in long-term projections. It mentions that the social-psychological approach may help explain certain failures in the engineering and economic models. But it then must rely on such technoeconomic models to draw conclusions about effective strategies and future costs of reducing carbon emissions. [These weaknesses are further elaborated in the critique of conventional methods of modeling future energy use, which is provided in the next chapter (Vol. 2, Ch. 5).]

The brief final section on how climate change may affect energy use deserves a more detailed analysis in a separate chapter (if such information is available). Indeed, this is another major uncertainty in the projection of energy demand.

Back to Table of Contents

Roger Pielke, Jr.'s Book Reviews and Publications | Roger Pielke, Jr.'s Home Page | ESIG Home Page