CBS Newsletter
Summer 1997
pg. 6

Carbon Dioxide Emissions from Industrialized Countries

Return to CBS News Article

Figure 1: Carbon emissions per capita 1973 vs. 1991 by major end use. (Denmark comparison is 1972 and 1991)

With the Third Conference of the Parties (COP-3) in Kyoto approaching, there is a great deal of excitement over policies designed to reduce future carbon dioxide (CO2) emissions from fossil fuels. At COP-3, more than 130 nations will meet to create legally binding targets for CO2 reductions. Accordingly, we have analyzed the patterns of emissions arising from the end uses of energy (and electricity production) in ten industrialized countries, with surprising, and, in some cases, worrisome results. The surprise is that emissions in many countries in the early 1990s were lower than those in the 1970s in an absolute sense and on a per capita basis; the worry is that factors that reduced emissions are not having the same effects in the mid-1990s as they did in the past.

Using data from recognized authorities in these countries, we traced the evolution of economic output and human activities from the period 1970 to 1993. We analyzed the emissions from nearly three dozen energy end uses or economic subsectors. Figure 1 shows our first-stage results, CO2 emissions by end-use sector or subsector for ten countries in 1973 and 1991, normalized to each country's population. Note that in these calculations, electricity is counted at the annual average rate of CO2 emissions per unit of electricity delivered to the economy (excluding exports or imports).

Our findings can be viewed in three basic ways - by country, by sector, and by factor. Looking across countries, Figure 1 shows that emissions per capita fell in every study country, with the biggest declines in Sweden, Norway and France. This was largely because of the rise in non-fossil sources of electricity and the increased use of biomass in Sweden and Norway. Even in absolute terms, emissions from the major energy end uses and associated power and heat production fell in most countries.

We also analyze the changes in emissions by sector. Changes in the ratio of emissions to activity by sector (i.e., carbon intensity) are shown in Figure 2. In general, manufacturing showed the most consistent decline relative to activity, with carbon intensity falling between 25 percent (Denmark) and 67 percent (Sweden). Lower energy intensities were the main reason, but shifts away from solid fuels and oil also aided this decline. Emissions per capita from the residential sector fell in all but two countries, Japan and Finland, where the rise in household standards offset the impacts of lower energy intensities for end uses or switches in fuels. Emissions from services generally declined relative to output (and in many countries, in the absolute), depending both on how much space heating intensity fell and how much CO2 was released in producing electricity - the most rapidly growing energy carrier in this sector.

Two sectors behaved differently. Emissions from the personal transportation sector fell only in the U.S., a result of the great decline in fuel use/km for cars. In Japan and Europe, increased automobile use, modest declines in fuel use/km, and, in most countries, an increase in fuel use per passenger-km pushed per capita emissions up. In the freight transportation sector, per capita emissions rates grew strongly with GDP in all countries, despite the fact that energy intensities for trucking fell in six of our study countries.

To examine trends by factor, we then factorialized the changes in each sector using Laspeyres indices, which show us how changes in sectoral activity or output, sectoral structure (or mix of activities within the sector), the energy intensity of each activity, and the fuel mix associated with each activity, have changed emissions. This method is useful for isolating the different impacts of these underlying components of emissions, in particular as a starting point for understanding how policies and other driving forces affected emissions in the past. The procedure also identifies factors that may already constrain future emissions growth, and shows where policies might control further growth.

Table 1 gives our results, in each case showing the change in emissions between 1973 and 1991 from each factor. In all, falling energy intensities and changes in final fuel and utility fuel mix led to lower releases of carbon, while structural changes and increases in activity raised emissions. But the rate of decline in sectoral energy intensities has slowed in the 1990s; indeed, in the U.S., automobile fuel use/km has stopped falling and even rose slightly after 1991. This means that a key component that restrained or even reduced emissions weakened through the early 1990s.

  Denmark Norway Sweden Finland W. Germany Japan USA France UK Italy
Manufacturing
Actual 100% 49% 35% 79% 75% 84% 77% 56% 64% 91%
Activity 134% 103% 120% 155% 134% 186% 146% 127% 102% 160%
Structure 100% 111% 99% 161% 91% 83% 84% 97% 96% 110%
Carbon Intensity 81% 49% 31% 49% 63% 57% 65% 48% 66% 57%
Energy Intensity 70% 79% 63% 76% 70% 61% 67% 69% 64% 54%
Fuel Mix 111% 60% 74% 68% 113% 111% 106% 103% 117% 103%
Utility Mix 99% 100% 89% 85% 84% 88% 94% 77% 93% 102%
Residences
Actual 78% 42% 37% 121% 95% 176% 103% 67% 96% 119%
Activity 103% 108% 106% 108% 103% 114% 119% 109% 103% 104%
Structure 136% 73% 131% 165% 162% 165% 128% 150% 115% 133%
Carbon Intensity 57% 62% 28% 123% 57% 101% 69% 43% 71% 106%
Energy Intensity 59% 82% 65% 70% 66% 127% 73% 73% 76% 83%
Fuel Mix 100% 77% 77% 123% 100% 103% 105% 62% 103% 130%
Utility Mix 97% 99% 87% 66% 89% 81% 92% 86% 90% 102%
Travel
Actual 122% 172% 134% 173% 160% 193% 106% 154% 152% 187%
Activity 125% 163% 122% 158% 143% 165% 131% 156% 154% 192%
Structure 96% 100% 101% 106% 104% 121% 103% 100% 106% 102%
Carbon Intensity 102% 108% 110% 104% 108% 102% 81% 97% 95% 91%
Energy Intensity 100% 108% 110% 103% 107% 101% 82% 99% 95% 90%
Fuel Mix 103% 100% 99% 100% 101% 101% 99% 98% 101% 102%
Freight
Actual 167% 138% 145% 131% 135% 163% 149% 160% 131% 222%
Activity 133% 126% 114% 120% 144% 137% 141% 105% 147% 180%
Structure 93% 96% 106% 111% 120% 130% 114% 126% 103% 115%
Carbon Intensity 113% 95% 119% 99% 77% 87% 95% 117% 89% 108%
Energy Intensity 112% 90% 121% 100% 77% 86% 95% 121% 88% 104%
Fuel Mix 101% 106% 98% 99% 99% 101% 101% 95% 101% 114%
All Sectors
Actual 97% 82% 56% 104% 93% 117% 97% 79% 91% 119%
Activity 121% 122% 119% 147% 135% 171% 141% 128% 114% 150%
Structure 117% 100% 109% 140% 114% 101% 103% 114% 104% 114%
Carbon Intensity 73% 68% 45% 74% 64% 71% 69% 58% 75% 78%
Energy Intensity 68% 88% 75% 79% 70% 77% 70% 78% 76% 70%
Fuel Mix 107% 77% 81% 88% 109% 109% 109% 93% 110% 110%
Utility Mix 98% 100% 89% 85% 88% 89% 93% 84% 93% 102%
Residual -5% -1% -21% -14% -11% -10% -6% -11% 0% -11%

The analysis also shows where per capita emissions differ from country to country. The most important differentiating factor is gross domestic product (GDP) per capita, followed by structrural differences, energy intensity, and fuel mix (including differences in utility fuel mix). The U.S. has a slightly higher than average ratio of CO2 emissions to primary energy consumed; somewhat higher-than-average energy intensities in travel, services, and manufacturing; and above all, larger homes and more driving per capita, which account for higher emissions there. Sweden and Norway, on the other hand, are much colder than average, and manufacture very energy-intensive raw materials, but these factors are offset in large part by the use of low-CO2 electricity generated almost exclusively from hydro and nuclear sources.

These findings raise important issues for the Kyoto meeting. Should countries with higher than average GDP growth rates be expected to cut their emissions/GDP more than those with lower growth? Do past reductions in emissions that cannot be repeated easily (such as big reductions in fossil fuel use for electricity generation or great cuts in the energy intensity of space heating) be taken into account in the discussion of future restraint or reductions? Should differences in emissions that arise because of climate, house size, or geography be subject to negotiation?

It is apparent that reducing energy intensities by improving energy efficiencies is crucial to further emissions cuts. Most of the energy efficiency programs of the 1980s have run their course, and real energy prices are for the most part stable or falling. Thus, that the rate of intensity reduction has fallen is no surprise. The Nordic countries have announced and indeed implemented modest carbon taxes. Others, including the Nordics but also the United Kingdom, Germany, and the Netherlands have begun to raise road fuel taxes or restructure vehicle taxation. All countries have placed great weight on new technology to restrain carbon emissions. Just what combination of taxation, efficiency-related programs, and new technology will maintain or even reduce emisions in the face of continued economic growth continues to vex authorities and experts alike.

At present we are updating our figures to 1994 and even 1995 as data permit. We will soon be able to show what path emissions have taken since 1990, the year adopted for reference in many international climate negotiations. Using the three-pronged approach (country, sectoral, and factorial analysis), we will also see what components of emissions still show restraint since 1990, and which not.

Return to CBS News Article

—Lee Schipper and Mike Ting

Info icon

Mike Ting
Energy Analysis Program
(510) 486-5916; fax (510) 486-6996

This work is supported by the DOE's, Office of Building Technologies, State and Community Programs; and the Environmental Protection Agency, Climate Change Branch.


EETD Newsletter Home Page
CBS Newsletter Home Page
Table of Contents for this Issue