WPS5098 Policy Research Working Paper 5098 Why Have CO2 Emissions Increased in the Transport Sector in Asia? Underlying Factors and Policy Options Govinda R. Timilsina Ashish Shrestha The World Bank Development Research Group Environment and Energy Team September 2009 Policy Research Working Paper 5098 Abstract Rapidly increasing emissions of carbon dioxide from the considered, three--economic development, population transport sector, particularly in urban areas, is a major growth, and transportation energy intensity--are challenge to sustainable development in developing responsible for driving up transport sector CO2 emissions countries. This study analyzes the factors responsible in Bangladesh, the Philippines, and Vietnam. In contrast, for transport sector CO2 emissions growth in selected only economic development and population growth developing Asian countries during 1980­2005. The are responsible in the case of China, India, Indonesia, analysis splits the annual emissions growth into Republic of Korea, Malaysia, Pakistan, Sri Lanka, and components representing economic development; Thailand. CO2 emissions exhibit a downward trend population growth; shifts in transportation modes; in Mongolia due to decreasing transportation energy and changes in fuel mix, emission coefficients, and intensity. The study also finds that some existing policy transportation energy intensity. The study also reviews instruments help reduce transport sector CO2 emissions, existing government policies to limit CO2 emissions although they were not necessarily targeted for this growth, particularly various fiscal and regulatory policy purpose when introduced. instruments. The study finds that of the six factors This paper--a product of the Environment and Energy Team, Development Research Group--is part of a larger effort in the department to study climate change and clean energy issues. Policy Research Working Papers are also posted on the Web at http://econ.worldbank.org. The author may be contacted at gtimilsina@worldbank.org. The Policy Research Working Paper Series disseminates the findings of work in progress to encourage the exchange of ideas about development issues. An objective of the series is to get the findings out quickly, even if the presentations are less than fully polished. The papers carry the names of the authors and should be cited accordingly. The findings, interpretations, and conclusions expressed in this paper are entirely those of the authors. They do not necessarily represent the views of the International Bank for Reconstruction and Development/World Bank and its affiliated organizations, or those of the Executive Directors of the World Bank or the governments they represent. Produced by the Research Support Team Why Have CO2 Emissions Increased in the Transport Sector in Asia? Underlying Factors and Policy Options Govinda R. Timilsina1* and Ashish Shrestha Development Research Group, The World Bank, 1818 H Street, NW, Washington, DC 20433, USA Key Words: Transport sector CO2 emissions, Driving factors for emission growth, Decomposition analysis. We sincerely thank Jean-Noel Guillossou, Baher El-Hifnawi, Mike Toman and two anonymous referees for their review of this paper. The Knowledge for Change Program (KCP) Trust Fund provided the financial support. The views expressed in this paper are those of the authors and do not necessarily represent the World Bank and its affiliated organizations. 1 Corresponding author; Fax: 1 202 522 1151; e-mail: gtimilsina@worldbank.org 1. Introduction Carbon dioxide (CO2) emissions released from fossil fuel consumption2 in Asia increased from 2,136 million tons in 1980 to 7,692 million tons in 2005, with an average annual growth rate of 5.3 percent (IEA, 2007a). The transport sector remains one of the main sources of CO2 emissions in most countries in Asia despite the fact that its share of total regional CO2 emissions remained constant at about 10% over the last 25 years. This is because China and India, which are responsible for approximately 80 percent of the region's total CO2 emissions, feature a relatively low share of emissions from the transport sector in their national CO2 emissions, thus skewing the regional aggregates. Nevertheless, since rising incomes are associated with higher levels of car ownership and usage (Webster et al, 1986a,b) and greater trip rates and distances (Schäfer, 2000), transport activity and resulting CO2 emissions could increase significantly in these countries along with economic growth and consumer clout. In most other Asian countries, the transport sector already accounts for a substantial share of total national CO2 emissions. Therefore, any attempt to address climate change in Asia must pay attention to transport sector emissions. The identification of key factors driving CO2 emissions is essential for the formulation of effective climate change mitigation policies and strategies. One approach to accomplish this objective is to decompose the growth of emissions into the possible affecting factors. 2 This study considers only fossil fuel consumption related CO2 emissions as our analysis is for the transport sector which emits CO2 through energy use. 2 Most existing studies are focused on the decomposition of national CO2 emissions and emission intensities. Examples include Wu et al (2005) and Wang et al (2005) for China, Kawase et al (2006) for Japan, Rhee and Chung (2006) for Japan and South Korea; Lise (2006) for Turkey, Diakoulaki et al (2006) for Greece, Saikku et al (2008) for 27 EU member States, Lee and Oh (2006) for APEC countries, Luukkanen and Kaivo-oja (2002a) for ASEAN countries; Luukkanen and Kaivo-oja (2002b) for Scandinavian countries, Ebohon and Ikeme (2006) for sub-Saharan African countries, and Han and Chatterjee (1997) for nine developing countries (Brazil, Chile, Colombia, India, Korea, Mexico, Philippines, Thailand and Zambia). Some existing studies are focused on the decomposition of manufacturing and power sector CO2 emissions or emission intensities (e.g., Liu et al (2007), Yabe (2004), Liaskas et al (2000), Schipper et al (2001), Chang and Lin (1998) and Bhattacharyya and Ussanarassamee (2004), Shrestha and Timilsina (1996), Nag and Kulshrestha (2000) and Shrestha and Marpuang (2006)). Note that power, industry and transport are the three major sectors responsible for fossil fuel related CO2 emissions in each country in the world. While the factors affecting CO2 emissions and emission intensities of the industry and power sectors have been analyzed in many countries, transport sector emissions and emission intensities have not been examined to the same extent, especially in developing countries. Nevertheless, a few studies also examine factors affecting transport sector emissions growth. For example, Lakshmanan and Han (1997) attribute the change in transport sector CO2 emissions in the US between 1970 and 1991 to growth in people's propensity to travel, population, and GDP. Lu et al (2007) decompose changes in CO2 emissions from highway vehicles in Germany, Japan, South Korea and Taiwan during 3 1990-2002 into changes in emission coefficient, vehicle fuel intensity, vehicle ownership, population intensity and economic growth. Scholl et al (1996) calculate how changes in transport activity, modal structure, CO2 intensity, energy intensity and fuel mix affect CO2 emissions from passenger transport in nine OECD countries between 1973 and 1992. Similarly, Schipper et al (1997) identify the relative contribution of activity, modal structure, and energy intensity to changes in energy use and carbon emissions from freight transport in ten industrialized countries from 1973 to 1992. Schipper et al (2000) attribute transport sector CO2 emission growth to transportation activity, modal structure, modal energy intensity and fuel mix. Kveiborg and Fosgerau (2007) decompose the historical growth in national Danish road freight traffic using a Divisia index decomposition method. Finally, Wu et al (2005) consider changes in transport energy intensity, average traveling distance, and number of vehicles (amongst numerous other factors) in their investigation of the underlying forces behind the stagnancy of China's energy-related CO2 emissions from 1996 to 1999. Understanding the factors affecting the growth of CO2 emissions from the transport sector is critical because of its increasing prominence as a source of emissions in most countries and its relevance to the preparation of climate change mitigation strategies. Hence, this study aims to address this gap by executing a Divisia decomposition analysis of CO2 emissions from the transport sector in 12 Asian countries during the 1980-2005 period. We attribute the growth of transport sector CO2 emissions over the last 25 years to six factors. These are: (i) fuel switching, (ii) modal shifting, (iii) change in emission coefficients, (iv) sectoral energy intensity change, (v) per capita economic growth and (vi) population growth. Among these, three factors -- change in 4 transport sector energy intensity, and per capita GDP and population growth -- are found primarily responsible for driving transport sector CO2 emissions in Asia. This paper is organized as follows: Section 2 examines CO2 emission trends and potential factors driving transport sector CO2 emission growth over the last 25 years. This is followed by a discussion on methodology and data in Section 3. The main results of the study are presented in Section 4, followed by a review of policies addressing transport sector CO2 emissions in Asia in Section 5. Finally, Section 6 offers key conclusions. 2. Potential Factors Driving the Transport Sector CO2 Emissions Growth Before discussing potential factors driving transport sector CO2 emission growth, we first highlight the trend of CO2 emissions in selected Asian countries. This is followed by a discussion of direct factors, such as fuel switching, modal shifting and changes in transportation energy intensity. Moreover, we analyze some trends, such as population growth and urbanization, and economic growth and motorization, which provide further insights on the causes of transport sector CO2 emission growth. 2.1. CO2 Emissions Figure 1 presents the trend of transport sector CO2 emissions in the 12 Asian countries that are responsible for more than 95 percent of the total CO2 emissions from developing countries in the region. Aggregate transport sector CO2 emissions at the 5 regional level more than tripled from 210 million tons in 1980 to 745 million tons in 2005 with a robust average annual growth rate of 5.2 percent. Figure 1: Trend of Transport Sector CO2 Emissions in Asia China India Indonesia Bangladesh Mongolia Sri Lanka Korea T hailand Malaysia Pakistan Philippines Vietnam 350 30 300 25 250 20 Mt of CO2 Mt of CO2 200 15 150 10 100 50 5 0 0 1980 1983 1986 1989 1992 1995 1998 2001 2004 1980 1983 1986 1989 1992 1995 1998 2001 2004 Note: These graphs are based on total transport sector CO2 emissions including those from international aviation, pipeline transport and all non-specified categories, and excluding emissions from electricity, which is assigned to the power sector. Hence, emissions presented in these graphs may differ from the data presented in Tables 2 and 3 where we exclude international aviation and pipeline transport, and include CO2 emissions from electricity used by the transport sector. Source: IEA (2007a) Despite the increase in transport sector emissions in absolute terms, the share of the sector in the national total in China and India are significantly smaller than that in most Asian countries. Table 1 presents total national CO2 emissions and the sectoral emission mix for the abovementioned 12 Asian countries. Transport sector shares of total national CO2 emissions have increased in Republic of Korea, the Philippines and Vietnam; decreased in India, Indonesia, Pakistan and Sri Lanka; and remained more or less stable in Bangladesh, China, Malaysia, Mongolia and Thailand. Because the transport, power and industry sectors are the main contributors to national CO2 emissions, changes in the magnitude of the emissions from the other two sectors, particularly the 6 power sector, have a considerable impact on the transport sector's share of national CO2 emissions (see Table 1). For example, the transport sectors in China and India exhibit exceptionally low shares of total national emissions because power generation in these countries is heavily reliant on emission intensive fuels, mainly coal. On the other hand, while the transport sector share of total national emissions in Mongolia has hardly changed, the striking rise in its power sector's share is directly attributable to a drastic drop in industrial output (and industry sector emissions), reflecting the recession that followed the end of Soviet assistance in 1990-91. Low emissions from the power sector typically signify the exploitation of renewable resources such as hydro-power as opposed to fossil fuels. Of the twelve Asian countries considered, only Sri Lanka demonstrates a very low share of emissions from the power sector, and only until 1996, when it started to supplement its almost exclusive reliance on hydro-power with substantial usage of fossil fuels in its generation mix. Table 1: CO2 Emission Mix by Sector 1980 2005 Indus- Trans- Indus- Trans- Country Total Power try port Other Total Power try port Other (Mt of (Mt of (%) (%) (%) (%) (%) (%) (%) (%) CO2) CO2) Bangladesh 7 21 41 14 24 36 35 29 12 24 China 1,403 20 51 6 23 5,060 48 37 7 9 India 292 26 39 19 16 1,147 52 30 8 10 Indonesia 69 10 39 26 26 341 28 39 22 11 Korea 122 20 32 12 37 449 35 31 19 15 Malaysia 23 32 34 28 6 138 33 35 28 3 Mongolia# 12 48 25 11 16 10 70 8 12 10 Pakistan 26 16 37 25 22 118 30 37 22 11 Philippines 32 27 39 15 18 76 37 19 37 7 Sri Lanka 4 8 22 55 16 12 28 16 45 11 Thailand 34 33 23 28 16 214 30 37 26 7 Vietnam 14 24 36 14 26 80 24 37 25 14 Note: The note for Figure 1 also applies here. #: 1985 data are used instead of 1980. Other includes residential, commercial and agricultural sectors. Source: IEA (2007a) 7 2.2. Modal Mix in the Transport Sector One potential factor driving transport sector CO2 emissions growth could be modal shifting, from less emission intensive modes (in terms of CO2 emissions per passenger/freight kilometer), such as railway and water transportation, to more emission intensive modes, such as commercial airplanes and private road vehicles. Figures 2(a)-2(d) illustrate the evolution of the modal mix for passenger transport activity in China, Korea, Vietnam and Pakistan, respectively, in terms of passenger kilometers. Road and rail are the main modes of passenger transport in China from 1990 to 2006, but it is clear that there has been a shift from rail towards road (see Figure 2(a)). There has also been a steady increase in the reliance on air transport for passengers, while the use of waterways has declined considerably. It should be noted that international aviation is included in the modal data for China. Road is by far the dominant mode of passenger transport in Korea, although the share of subway in passenger transport has increased over the period 1994-2006 (see Figure 2(b)). On the other hand, the share of aviation in passenger transport has waned only slightly, but the share or rail has declined significantly. Air transport has gained a substantial share of passenger transport in Vietnam from 1990 to 2006, largely at the expense of road transport, which, nonetheless, remains dominant (see Figure 2(c)). The share of rail and navigation has also gone down, especially sharply in 2005-2006, when road regained some of the share it had been losing in previous years. In Pakistan, where there is virtually no water transport, and aviation only accounts for a miniscule and declining share, road is overwhelmingly the main mode 8 for passenger transport over the period 1990 to 2004 (see Figure 2(d)). Rail's share of passenger transport activity has been steadily declining over the same time frame. Figure 2: Modal Mix for Passenger Transport Railway Roads Aviation Shipping Subway 96 98 00 02 04 06 (a) China (1990-2006) (b) Korea (1994-2006) 100% 100% 80% 80% 60% 60% 40% 40% 20% 20% 0% 0% 1990 1992 1994 1996 1998 2000 2002 2004 2006 1994 1996 1998 2000 2002 2004 2006 (c) Vietnam (1990-2006) (d) Pakistan (1990-2004) 100% 100% 80% 80% 60% 60% 40% 40% 20% 20% 0% 0% 1990 1992 1994 1996 1998 2000 2002 2004 2006 1990 1992 1994 1996 1998 2000 2002 2004 Source: National Bureau of Statistics of China (2007); Korea National Statistical Office (2008); General Statistical Office of Vietnam (2005); Accountancy (2007); United Nations Statistics Division (2008a) Whereas road is the chief mode of passenger transport in all four of these countries, the modal mix for freight transport is rather different, except in Pakistan. Figures 3(a)-3(d) show changes in the modal mix for freight transport activity in China, 9 Korea, Vietnam and Pakistan, respectively, in terms of ton kilometers. Domestic navigation has supplanted rail as the dominant mode of freight transport in China over the period 1980 to 2006, and although the share of road is higher in 2006 compared to 1980, this share has been relatively stable in recent years (see Figure 3(a)). The use of aviation for freight transport remains negligible in China. In Korea, shipping is the main mode for freight transport from 1994 to 2000 (see Figure 3(b)). Shipping is also by the far the most important mode for freight transport in Vietnam, with a share above 80 percent for much of the 1990-2005 period, followed by road, and then rail (see Figure 3(c)). However, domestic navigation is not utilized for freight transport from 1990 to 2004 in Pakistan, where freight is overwhelmingly transported via road, which gained significant share from rail over this period (see Figure 3(d)). Note that modal mix data in terms of transportation services are available only for four countries (i.e., China, Korea, Vietnam and Pakistan) out of the 12 countries considered in this study. Moreover, since these data are taken from each country's national statistical agency, they are not comparable across countries due to the lack of international reporting standards for transportation data. For example, road transport activity indicated for China does not include public buses and taxis, whereas they are included in road transport activity for Korea. Furthermore, the temporal scope of the data also differs. Modal mix is available for China from 1990 to 2006, while it is only available for varying parts of that time frame for Korea, Vietnam and Pakistan. Hence, such data will not be useful in the decomposition analysis pursued later. Instead, we will 10 be using energy consumption data as a proxy for transportation services3 although passenger and freight kilometers (or any equivalent units) would be the desired measurement for transportation services. Figure 3: Modal Mix for Freight Transport Railway Roads Aviation Shipping Subway 96 98 00 02 04 06 (a) China (1990-2006) (b) Korea (1994-2000) 100% 100% 80% 80% 60% 60% 40% 40% 20% 20% 0% 0% 1990 1992 1994 1996 1998 2000 2002 2004 2006 1994 1995 1996 1997 1998 1999 2000 (c) Vietnam (1990-2005) (d) Pakistan (1990-2004) 100% 100% 80% 80% 60% 60% 40% 40% 20% 20% 0% 0% 1990 1992 1994 1996 1998 2000 2002 2004 1990 1992 1994 1996 1998 2000 2002 2004 Source: National Bureau of Statistics of China (2007); Korea National Statistical Office (2008); General Statistical Office of Vietnam (2005); Accountancy (2007); United Nations Statistics Division (2008a). 3 In energy literature, this is a common practice to measure modal mix in the transportation sector (see e.g., EIA, 2007; IEA, 2004) 11 Table 2 presents modal mix in terms of energy consumption and CO2 emissions. Road was the predominant mode of transportation in all countries in 1980, especially in Malaysia and Thailand. The role of road transportation was even more prominent in 2005 as most countries have increased their reliance on road transportation since 1980. Road transport accounted for more than 90% of transport sector CO2 emissions in 6 out of the 12 countries in 2005, and only China features a modal share of less than 70 percent of CO2 emissions for road transport. A few countries utilized domestic air transport to a large extent, e.g., more than 10 percent of transport sector fuel consumption and CO2 emissions in Sri Lanka and Vietnam in 1980 came from domestic air transport, but its share declined in all countries, except for China, by 2005. It should be noted, however, that reliable data for fuel consumption in domestic aviation is not available for half the countries considered (see Table 2). 12 Table 2: Modal Mix for Fuel Consumption in Transport Sector 1980 2005 Modal mix in terms of energy consumption Total Air Water Rail Road Total Air Water Rail Road Country Mtoe % % % % Mtoe % % % % Bangladesh 0.3 n.a. 29.5 13.2 57.3 1.5 n.a. 16.6 12.0 71.4 China 24.4 1.2 5.2 43.6 50.0 107.5 7.0 8.6 13.9 70.5 India 16.9 n.a. 2.0 38.6 59.3 33.7 n.a. 1.9 8.3 89.8 Indonesia 6.0 5.9 3.0 0.2 90.8 25.0 2.4 6.9 0.0 90.7 Korea 4.8 4.6 22.3 7.4 65.7 29.1 4.3 3.0 1.7 91.1 Malaysia 2.2 n.a. 0.0 n.a. 100.0 13.3 n.a. 0.0 0.0 100.0 Mongolia 0.44 n.a. 0.0 19.4 80.6 0.39 n.a. 0.0 26.9 73.1 Pakistan 2.2 n.a. 0.0 5.9 94.1 8.1 n.a. 0.0 3.5 96.5 Philippines 1.8 4.2 12.4 0.0 83.4 8.0 3.2 10.3 5.2 81.4 Sri Lanka 0.7 18.7 0.6 5.1 75.5 1.8 9.5 1.2 1.4 87.9 Thailand 3.2 n.a. 0.1 n.a 99.9 18.6 n.a. 0.4 0.6 99.0 Vietnam 0.6 15.1 0.0 9.7 75.2 6.8 2.5 4.5 0.5 92.4 Modal mix in terms of CO2 emissions Country Mt % % % % Mt % % % % Bangladesh 1.0 n.a. 29.9 13.5 56.6 4.5 n.a. 16.8 12.2 71.0 China 83.6 1.0 4.7 51.5 42.7 338.6 6.7 8.6 18.2 66.5 India 58.1 n.a. 1.9 45.0 53.1 108.6 n.a. 1.9 14.2 83.9 Indonesia 17.9 5.9 3.2 0.3 90.6 74.6 2.4 7.2 0.0 90.4 Korea 14.8 4.4 23.0 8.4 64.2 86.8 4.3 3.2 2.3 90.3 Malaysia 6.5 n.a. 0.0 n.a. 100.0 39.6 n.a. 0.0 0.1 99.9 Mongolia 1.3 n.a. 0.0 18.3 81.7 1.2 n.a. 0.0 29.0 71.0 Pakistan 6.8 n.a. 0.0 6.1 93.9 24.7 n.a. 0.0 3.5 96.5 Philippines 5.2 4.3 13.1 0.0 82.6 24.3 3.1 10.7 5.4 80.8 Sri Lanka 2.1 18.4 0.6 5.2 75.8 5.5 9.3 1.2 1.5 87.9 Thailand 9.6 n.a. 0.1 n.a. 99.9 56.5 n.a. 0.4 0.6 99.0 Vietnam 2.0 14.8 0.0 12.6 72.6 19.2 2.5 4.8 0.8 91.9 Note: Based on the available data and following normal practices in energy accounting systems, we considered four modes of transportation: road, rail, water and air. If data is available, road transportation, which is the primary mode for providing transportation services as well as energy consumption and associated emissions, can be disaggregated further into auto, bus, etc. Source: IEA (2007b, c 13 Rail transport has been an important mode of transportation in a number of the countries since 1980. China and India exhibited particularly high modal shares for rail in 1980, although rail was an important mode of transport in Bangladesh, Mongolia and Vietnam as well. However, by 2005, these two countries had registered massive declines in the modal share for rail, and in fact, only Mongolia and the Philippines show significant gains in the modal share for rail over the study period. The share of CO2 emissions from rail exceeds its share of fuel consumption appreciably in India, China and Vietnam, again on account of the electricity used being generated largely from coal. Domestic navigation represented a major mode of transportation in Bangladesh, Korea and the Philippines in 1980, but its share declined in all three countries by 2005, and especially drastically in Bangladesh and Korea. Substantial growth in the share of domestic navigation in energy consumption and CO2 emissions is only seen in China, Indonesia and Vietnam. In the case of Vietnam, fuel consumption in domestic navigation as reported by the IEA (see Table 2) is likely to be a significant underestimate as the data do not capture the overwhelming reliance of freight transport in Vietnam on coastal shipping (see Figure 3(c)). In most countries, there is found to be no or minimal reliance on waterways over the study period. 2.3. Transport Sector Fuel Mix How fuel substitution occurs within a mode of transportation is another factor that explains transport sector CO2 emissions growth. However, there exist very limited fuel choices for the transport sector. Motor gasoline and diesel are the main fuels used in the 14 transport sector (see Table 3). A comparison of the fuel mix between 2005 and 1980 suggests significant substitution of gasoline with diesel in some of the countries ­ Pakistan, the Philippines, Thailand and Vietnam. According to GTZ4 (2007), these are the countries where gasoline is taxed while diesel is subsidized. Since the price of diesel has been maintained at an artificially low level by the governments via price controls for socioeconomic reasons, the substitution of gasoline with diesel in these countries is not surprising. Minimal substitution between gasoline and diesel is observed to have occurred in the other countries. Obviously, CO2 emissions largely originate from the combustion of diesel and motor gasoline, but even though diesel has slightly higher carbon content as compared to gasoline, the substitution of gasoline with diesel does not change CO2 emissions significantly as diesel provides better fuel economy as compared to gasoline. A comparison of diesel and gasoline shares in Table 6 also demonstrates this fact. 4 Deutsche Gesellschaft fur Technische Zusammenarbeit 15 Table 3: Transport Sector Fuel Mix 1980 2005 Heavy Fuel Heavy Fuel Electricity Electricity Kerosene Kerosene Bio-fuels Bio-fuels Gasoline Gasoline Aviation Aviation Natural Natural Diesel Diesel Fuels Fuels LPG Coal LPG Coal Gas Gas Oil Oil Country Total Total Fuel mix in terms of energy consumption Mtoe % % % % % % % % % % Mtoe % % % % % % % % % % Bangladesh 0.3 0.0 0.0 73.3 0.0 0.0 19.2 0.0 7.4 0.0 0.0 1.5 0.0 0.0 79.0 0.0 0.0 21.0 0.0 0.0 0.0 0.0 China 24.4 1.2 0.9 13.4 0.0 0.0 43.9 0.0 1.9 38.7 0.0 107.5 7.0 1.6 40.2 0.5 0.0 44.2 0.1 2.6 3.8 0.0 India 16.9 0.0 1.2 55.0 0.0 0.0 9.6 0.0 1.9 32.3 0.0 33.7 0.0 2.5 66.4 0.0 0.3 27.5 2.0 1.4 0.0 0.0 Indonesia 6.0 5.9 0.0 42.5 0.0 0.0 50.3 0.0 1.1 0.2 0.0 25.0 2.4 0.0 41.3 0.0 0.0 55.5 0.1 0.7 0.0 0.0 Korea 4.8 4.6 0.7 60.5 3.5 0.0 17.8 0.0 12.8 0.0 0.0 29.1 4.3 0.8 51.6 14.6 0.0 25.5 1.1 2.0 0.0 0.0 Malaysia 2.2 0.0 0.0 39.5 0.0 0.0 60.5 0.0 0.0 0.0 0.0 13.3 0.0 0.0 38.2 0.0 0.0 61.7 0.0 0.0 0.0 0.0 Mongolia 0.4 0.0 2.8 20.3 0.0 0.0 76.9 0.0 0.0 0.0 0.0 0.39 0.0 2.3 19.5 0.0 0.0 69.4 0.0 0.0 8.8 0.0 Pakistan 2.2 0.0 0.1 71.4 0.0 0.0 26.2 0.0 2.2 0.0 0.0 8.1 0.0 0.0 84.1 0.0 0.0 15.6 0.0 0.2 0.0 0.0 Philippines 1.8 4.2 0.0 17.6 0.0 0.0 75.9 0.0 2.1 0.0 0.2 8.0 3.2 0.1 54.4 0.0 0.0 37.5 0.0 4.8 0.0 0.0 Sri Lanka 0.7 18.7 0.0 64.4 0.0 0.0 16.9 0.0 0.0 0.0 0.0 1.8 9.5 0.0 63.3 0.0 0.0 27.2 0.0 0.1 0.0 0.0 Thailand 3.2 0.0 0.0 52.2 1.2 0.0 46.5 0.0 0.0 0.0 0.1 18.6 0.0 0.0 68.6 1.9 0.0 29.2 0.2 0.0 0.0 0.0 Vietnam 0.6 15.1 0.0 15.5 0.0 0.0 59.7 0.0 0.0 9.7 0.0 6.8 2.7 0.5 55.6 0.0 0.0 41.2 0.0 0.0 0.0 0.0 Fuel mix in terms of CO2 emission Mt % % % % % % % % % % Mt % % % % % % % % % % Bangladesh 1.0 0.0 0.0 74.0 0.0 0.0 18.1 0.0 7.8 0.0 0.0 4.5 0.0 0.0 80.1 0.0 0.0 19.9 0.0 0.0 0.0 0.0 China 83.7 1.2 3.2 12.1 0.0 0.0 37.0 0.0 1.8 44.6 0.0 338.6 6.7 5.1 39.6 0.4 0.0 40.7 0.0 2.6 4.8 0.0 India 58.1 0.0 3.0 49.7 0.0 0.0 8.1 0.0 1.8 37.3 0.0 108.6 0.0 8.6 63.8 0.0 0.0 24.7 1.4 1.4 0.0 0.0 Indonesia 17.9 5.9 0.0 43.9 0.0 0.0 48.7 0.0 1.2 0.3 0.0 74.6 2.4 0.0 42.9 0.0 0.0 53.8 0.0 0.8 0.0 0.0 Korea 14.8 4.4 1.7 60.7 3.0 0.0 16.7 0.0 13.4 0.1 0.0 86.8 4.3 1.3 53.6 12.9 0.0 24.8 0.9 2.2 0.0 0.0 Malaysia 6.5 0.0 0.0 41.1 0.0 0.0 58.9 0.0 0.0 0.0 0.0 39.6 0.0 0.1 39.8 0.0 0.0 60.1 0.0 0.0 0.0 0.0 Mongolia 1.3 0.0 0.4 21.9 0.0 0.0 77.7 0.0 0.0 0.0 0.0 1.2 0.0 0.3 20.1 0.0 0.0 67.1 0.0 0.0 12.5 0.0 Pakistan 6.8 0.0 0.1 72.6 0.0 0.0 24.9 0.0 2.4 0.0 0.0 24.7 0.0 0.0 85.0 0.0 0.0 14.8 0.0 0.2 0.0 0.0 Philippines 5.2 4.3 0.0 18.5 0.0 0.0 74.7 0.0 2.3 0.0 0.2 24.3 3.1 0.2 55.7 0.0 0.0 35.9 0.0 5.1 0.0 0.0 Sri Lanka 2.1 18.4 0.0 65.5 0.0 0.0 16.1 0.0 0.0 0.0 0.0 5.5 9.3 0.0 64.6 0.0 0.0 26.0 0.0 0.1 0.0 0.0 Thailand 9.6 0.0 0.0 53.9 1.1 0.0 44.9 0.0 0.0 0.0 0.1 56.5 0.0 0.1 70.2 1.7 0.0 27.9 0.2 0.0 0.0 0.0 Vietnam 2.0 14.8 0.0 15.8 0.0 0.0 56.8 0.0 0.0 12.6 0.0 19.2 2.6 0.8 57.0 0.0 0.0 39.5 0.0 0.0 0.0 0.0 Note: CO2 emission from electricity consumption in transport is computed by multiplying the electricity consumption by country and year specific emission coefficient for electricity (derived using IEA data on electricity output and CO2 emissions from electricity production). Source: IEA (2007b, c) 16 China and India were both highly dependent on coal as fuel for rail transport in 1980 (coal comprised well over 30 percent of fuel consumption for transportation in these two populous nations (see Table 3)). By 2005, coal had already been completely phased out as railway fuel in India, and its share in the fuel mix had declined to under 4 percent in China. Over the same period, considerable growth in the share of diesel in the fuel mix is observed in both countries, especially China. Some of this is the result of the direct substitution of coal with diesel in rail transport. However, since rail transportation itself has been significantly replaced with road transport, the substitution of coal with diesel only accounted for a small part of the gains in diesel consumption in China and India. Most of the increased diesel demand was due to the growth of road transportation. The combustion of aviation fuels represented another notable source of emissions from transportation in 1980, but only in a few countries such as Sri Lanka and Vietnam. Reliable aviation fuel consumption data is not available for all countries; however, the share of aviation fuel in total transport sector fuel consumption as well as in emissions declined in the countries for which data is available, except for China and the Philippines. Utilization of electricity for transportation in 1980 was negligible in all countries except China, India, Korea and Mongolia, However, the share of electricity has increased slightly in China, India and Korea (along with the Philippines and Vietnam), whereas it has declined in Mongolia. Despite the small share of electricity in the energy mix in China and India, it is nonetheless an important source of CO2 emissions due to the high CO2 emission coefficient for electricity generation in these countries. Liquefied petroleum gas (LPG) was not a major fuel for transportation in any of the countries, except Korea, where its share rose from 3.5 percent of the energy mix in 17 1980 to over 14 percent in 2005 and is used mainly in taxis, buses and trucks (Liu et al, 1997). China and Thailand, however exhibited far more modest increases in the use of LPG over the study period. Natural gas was not exploited as fuel for transportation in any of the countries in 1980, and only India and Korea had managed to incorporate it noticeably into their fuel mix by 2005. While none of the countries used biofuels for transportation in 1980, biofuels comprised a very small share of the fuel mix in India by 2005. Coal was one of the most important fuels for transportation, primarily for rail, in China and India, and to a lesser extent, Vietnam, in 1980; but these countries managed to either dramatically reduce or eliminate entirely their reliance on coal by 2005. Mongolia, on the other hand, which did not use coal as a fuel for transportation at all in 1980, had incorporated it into its transport fuel mix by 2005. Finally, the use of kerosene as a share of total fuel consumption for transportation was and remains negligible, if not nil, in all countries considered. 2.4. Transportation Energy Intensity Figure 4 displays transportation energy intensity for all the Asian countries considered (except Mongolia5) over the period 1980-2005. It can be seen that transportation energy intensity, which is the ratio of total fuel consumption for transportation in an economy to its gross domestic product, varies significantly across 5 Mongolia is by far the most transportation energy intensive country considered in the study, outdistancing the country ranked second, Malaysia, by a very large margin, but transportation energy intensity in Mongolia has been declining, from 172 ktoe per billion US$ in 1985 to 82 ktoe per billion US$ in 2005. 18 countries and over time. Bangladesh registers as the country with the lowest, whereas Malaysia features the highest, transportation energy intensity of these eleven countries throughout the study period. Most of the countries demonstrate an overall upward trend in transportation energy intensity, that is, their fuel consumption for transportation per unit of economic output has increased, yet the two largest economies in the region, China and India (especially China), show steadily improving transportation energy intensity over the study period, which should mitigate some of the increased CO2 emissions expected from their swift economic growth. Transportation energy intensity in Pakistan and Sri Lanka at the end of the study period are found to be close to the level at the beginning, although they fluctuate in between. Despite the overall deterioration in transportation energy intensity in most of the other countries, a noticeable improvement can be observed in the last year of the study period for every country other than Bangladesh and China. Figure 4: Transportation Energy Intensity in Selected Asian Countries (1980-2005) 60 Bangladesh China 50 India 40 Indonesia (ktoe/billion US$) Korea 30 Malaysia Pakistan 20 Philippines Sri Lanka 10 Thailand Vietnam 0 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 Source: IEA (2007b, c, f, g) 19 2.5. Population Growth and Urbanization The Asian countries included in this study alone currently account for over half of the world's population and are projected to keep growing to comprise 3.6 billion out of the 7.2 billion total world population in 2015 (World Bank, 2008). Aside from population growth, the other notable demographic change in Asia is that it is also going through a period of rapid urbanization. Table 4 lists the population of Asian countries in 1990 and 2006, the percentage of urban dwellers, and the growth rates of total and urban populations. Urban population growth can be seen to be significantly higher than total population growth in all of the countries except Mongolia and Sri Lanka. Table 4: Total and Urban Population Size and Growth Rate Population Avg. Annual Growth 1990 2006 (1990-2006) Total Urban share Total Urban share Total Urban (millions) (%) (millions) (%) (%) (%) Bangladesh 113 20 156 26 2 3.6 China 1,135.2 27 1,311.8 41 0.9 3.5 India 849.5 26 1,109.8 29 1.7 2.5 Indonesia 178.2 31 223 49 1.4 4.4 Korea, Rep. 42.9 74 48.4 81 0.8 1.3 Malaysia 18.1 50 26.1 68 2.3 4.3 Mongolia 2.1 57 2.6 57 1.3 1.3 Pakistan 108 31 159 35 2.4 3.3 Philippines 61.2 49 86.3 63 2.1 3.8 Sri Lanka 17 17 19.9 15 1 0.2 Thailand 54.3 29 63.4 33 1 1.6 Vietnam 66.2 20 84.1 27 1.5 3.3 Source: World Bank (2008). While developed and developing countries alike continue to urbanize, the rate of urbanization is especially fast in developing countries where the majority of people have not been city-dwellers traditionally. The year 2007 was historic in that it commemorated the first time that half of the world's population lived in cities, a share that is anticipated 20 to grow to 58 percent by 2025 because world urban population is projected to expand at almost double the rate of total world population over this period (ESCAP, 2007). Tokyo has been the largest urban area in the world for some time now, but more interestingly, of the twenty mega-cities (defined as a city with a minimum of 10 million inhabitants) around the world, twelve of them are found in Asia. South and Central Asian cities such as Dhaka, which has been growing at an average annual rate of approximately 5 percent since 1975 and is expected to be home to almost 10 percent of Bangladesh by 2015, Karachi and Delhi have been the fastest growing over the past quarter century (ibid.). As Asian countries continue to grow and urbanize, increasing motorization can be expected to generate higher levels of CO2 emissions and place additional stresses on the transport infrastructure. 2.6. Motorization6 and Economic Growth One of the key factors that explain why there occurred a modal shift towards road transportation from other modes of transportation is the rate of motorization. Most Asian countries have experienced significant growth in their road transport fleets, particularly in urban areas, resulting in soaring transportation energy use and CO2 emissions. Estimates of the stock of passenger cars in Asian countries are presented in Table 5. Rapid growth in the passenger car fleet can be seen in almost every country considered, including approximately 19 percent growth in China, the most populous country in the world, and 6 Motorization, as per the commonly utilized UN definition, is measured here as the number of private cars per 1000 people in a country. 21 around 10 percent or more growth in two of the other most populous countries: India and Indonesia. Table 5: Passenger Car Fleet in Selected Asian Countries (`000 cars) Average Annual 1990 1995 2000 2005 Growth Rate Bangladesh 42 51 70 90a 5.6% India 2,694 3,841 6,143 8,619b 9.4% Pakistan 559 773 1,066 1,167b 5.8% Sri Lanka 174 229 335 444b 7.5% Indonesia 1,313 2,107 3,039 5,494 10.0% Malaysia 1,715 2,609 4,213 5,500b 9.4% Philippines 1,471 2,333 2,156 3,982b 8.0% Thailand 1,222 1,913 2,665 2,993a 6.6% Vietnam 33 26 44 n.a. 2.9% China 1,622 4,179 8,537 17,359a 18.5% Korea 2,075 6,006 8,084 10,621a 12.4% Mongolia 8 24 44 80a 17.8% Source: ESCAP (2007b) Note: a data are for 2004; b data are for 2003 While all these countries presently have low levels of personal motorization (mostly 2-wheelers in many cases), it is highly probable that these levels will increase dramatically as they achieve economic growth and given the expected increase in urban population (ADB, 2006). The World Energy Outlook 2007 (IEA, 2007h) warns that India is getting closer to the tipping point of US$3000 GDP per capita (based on purchasing power parity, or PPP), after which vehicle ownership rates accelerate swiftly. It also estimates that per capita GDP (PPP) in India will rise to US$13,000 by 2030, and greater buying power of that magnitude can be anticipated to assert itself partly in higher demand for 4 wheel vehicles, such as, cars and sports utility vehicles (SUVs) (ESCAP, 2007a). The Asian Development Bank (ADB, 2006) projects the active population of cars and SUVs in China to grow to 15 times its 2005 size to 193 million in 30 years under a business-as-usual (BAU) scenario; the analogous figure for India is 13 times to 80 million in 2035. 22 Table 6 displays the level of motorization, along with per capita GDP, in the Asian countries considered in this study. The level of motorization in developing Asian countries lags well behind those in developed countries, but since economic growth and energy-demand (and CO2 emissions) growth are linked, the expected expansion of developing Asian economies at an average rate of more than 5 percent per year through 2030, with China, India, Indonesia and Malaysia leading the way, means that the current levels of motorization are likely only a fraction of what they will be in two or three decades as these populous and rapidly growing economies go through a phase of rapid motorization (ADB, 2006). The demand for passenger cars in Asia is growing much faster than per capita income in every country considered other than Bangladesh; sometimes astonishingly so, as in the case of China, Korea and Mongolia. Recent motorization data for Vietnam is not available. Table 6: Motorization (Number of passenger cars per 1,000 persons) and Gross Domestic Product per Capita (US$) Average Annual 1990 2005 Growth Rate Cars per GDP per Cars per GDP per Cars per GDP per 1,000 capita 1,000 capita 1,000 capita people people people Bangladesh 0.4 282 0.6a 422 2.9% 2.9% India 3.2 381 8.0b 713 7.4% 4.2% Pakistan 5.0 506 7.7b 820 3.4% 1.7% Sri Lanka 9.8 479 21.8b 1,253 6.4% 3.7% Indonesia 7.2 685 24.7 1,244 8.5% 2.9% Malaysia 96.1 2,432 225.1b 5,098 6.8% 3.8% Philippines 24.1 724 49.8b 1,163 5.7% 1.4% Thailand 21.8 1,572 46.7a 2,797 5.6% 3.5% Vietnam 0.5 98 n.a. 621 n.a. 5.9% China 1.4 358 13.4a 1,766 17.4% 9.1% Korea 48.4 6,153 222.9a 16,454 11.5% 4.7% Mongolia 3.6 567 30.4a 800 16.5% 0.2% Source: ESCAP (2007b); UN Statistics Division (2008b) Note: a data are for 2004; b data are for 2003 23 3. Methodology to Determine Factors Affecting CO2 Emission Growth 3.1 Methodology In this section, we derive the methodology to decompose transport sector CO2 emission growth to the contributing factors. Data needed to implement the methodology are also discussed here. Total CO2 emission from the transport sector in a country in year t (CO2t) is the summation of CO2 emissions from all fuels used in all transport modes in that year, i.e., CO2 t CO2 ijt (1) ij where, subscripts i, j and t refer to fuel type (e.g., gasoline, diesel, electricity), transportation mode (e.g., road, rail, air and water) and year, respectively. In order to decompose the emission to the potential factors affecting it, Equation (1) can be expressed as: CO 2 ijt FC ijt FC jt TS jt TS t GDPt CO 2 t POPt (2) ij FC ijt FC jt TS jt TS t GDPt POPt where, FC refers to fuel consumption, TS represents transport services (e.g., passenger kilometers, tons kilometers or any equivalent measurement representing transport services7) and GDP is used to measure economic output. Unfortunately, data for transportation services are not available for the countries and for the time horizon considered in the study. We use an alternative approach as shown in Equation (3) to decompose the emission to the potential factors affecting it. 7 Includes transport services provided to all sectors (e.g., households, industry, government). 24 CO 2 ijt FC ijt FC jt FC t GDPt CO 2 t POPt t (3) ij FC ijt FC jt FC t GDPt POPt As implied in Equation (3), we represent modal mix by energy consumption by mode instead of transportation services provided by the mode. Equation (3) can also be rewritten as: CO 2 t EC ijt FM ijt MM jt EI t PC t POPt (4) ij where, EC is emission coefficient or CO2 intensity of a fuel (i.e., CO2/FC), FM refers to fuel mix (i.e., share of a fuel in a transportation mode), MM represents modal mix (i.e., share of fuel consumption by a mode in total transport sector energy consumption); EI is the transportation energy intensity (i.e., FC/GDP), PC is economic activity as captured by per capita GDP, and POP is population. The growth of emissions is often decomposed into the potential driving factors using different methods, such as the Laspeyres or Divisia methods. While studies such as Lin et al (2008), Diakoulaki and Mandaraka (2007), Diakoulaki et al (2006), and Ebohon and Ikeme (2006) use the refined Laspeyres techniques, studies such as Liu et al (2007), Hatzigeorgiou et al (2007) and Wang et al (2005) use the Arithmetic Mean Divisia Index (AMDI) and the Logarithmic Mean Divisia Index (LMDI) techniques. In this study we follow the LMDI approach, which, unlike the AMDI approach, provides a residual-free decomposition and can accommodate the occurrence of zero values in the data set8 (Ang, 2004). Although the refined Laspeyres methods also have these virtues, their formulae become increasingly complex when the number of factors exceeds three, and the linkages between the additive and multiplicative forms can not be established easily. 8 In this approach zero values are replaced with a small positive constant. 25 Using LMDI (Ang, 2005), the additive decomposition of the change in transport sector CO2 emissions from year t-1 to t is expressed as: ECijt FM ijt MM ijt CO 2 t CO 2t 1 wij ln ~ wij ln ~ wij ln ~ ij ECijt 1 ij FM ijt 1 ij MM ijt 1 (5) EI t PCt POPt wij ln ~ wij ln ~ wij ln ~ ij EI t 1 ij PCt 1 ij POPt 1 where ~ CO 2 ijt CO 2 ijt 1 wijt for CO 2 ijt CO 2 ijt 1 ln CO 2 ijt ln CO 2 ijt 1 (6) CO 2 ijt for CO 2 ijt CO 2 ijt 1 Similarly, the multiplicative decomposition of the change in the transport sector CO2 emissions from year t-1 to t (again, following Ang (2005)) is given as: CO 2 t ~ ECijt ~ FM ijt exp vij ln exp vij ln CO 2 t 1 ij ECijt 1 ij FM ijt 1 ~ MM ijt ~ EI t exp vij ln exp vij ln (7) ij MM ijt 1 ij EI t 1 ~ PCt ~ POPt exp vij ln exp vij ln ij PCt 1 ij POPt 1 where ~ vijt L (CO 2 ijt , CO 2 ijt 1 ) L (CO 2 t , CO 2 t 1 ) (8) with ab L(a , b) for a b ln a ln b (9) a for a b The first term on the right hand side of Equations (5) and (7) represents the emission coefficient (EC) effect. Note that only the coefficient of electricity is changing 26 due to variations in electricity generation mix over time. Emission coefficients (i.e., carbon contents) for other fuels are assumed to be constant over time. The second and third terms represent the fuel mix (FM) or fuel switching and the modal mix (MM) or modal shift effects, respectively. The fourth term represents the transportation energy intensity (EI) effect. And finally, the fifth and sixth terms represent the economic activity or per capita GDP (PC) effect and population (POP) effect, respectively. We have carried out the decomposition analysis on an annual basis over the twenty five year period between 1980 and 2005. 3.2 Data The study required a large set of data on energy consumption and emissions by fuel and by mode for the study period of 25 years. We explored every potential source of data, such as national energy statistics, UN energy statistics, and US Department of Energy databases. While national statistical agencies collect data at the level of detail needed in a few countries (e.g., Korea), they do not provide such data in most of the countries. Moreover, mixing data from different sources with different conventions and assumptions used for collection and aggregation would cause an artificial change in the trends. Therefore, we uses transport sector energy consumption data by fuel type and mode from the International Energy Agency (IEA, 2007b, 2007c). No source other than the IEA provides data at the required details and time series needed for the study. Fuels included are biofuel (i.e., ethanol), natural gas, liquefied petroleum gases (LPG), motor gasoline, aviation fuels (i.e., aviation gasoline, kerosene and jet fuel), 27 diesel oil, heavy fuel oil, coal, kerosene and electricity. The modes of transportation considered are domestic aviation, road, rail and domestic navigation9. We have excluded energy consumption in oil and gas pipeline transport. Emission coefficients are based on the carbon contents of fuels and are obtained from Intergovernmental Panel on Climate Change (IPCC) Guidelines for National Greenhouse Gas Inventories (IPCC, 2006) for all fuels except electricity. Emission coefficients for electricity are derived using IEA data on electricity output and CO2 emissions from electricity production (IEA, 2007a, 2007d, 2007e). While CO2 emission factors for other transportation fuels (e.g., gasoline, diesel, etc.) remain the same throughout the study period, CO2 emission factors for electricity vary with time. This is because, the carbon content of a fossil fuel is not expected to change over time, but CO2 emission coefficients of an electricity grid change over time as the electricity generation mix and thereby the input fuel mix for electricity generation changes every year. Moreover, the CO2 emission coefficient of a particular fossil fuel is the same for all countries, whereas the coefficient for electricity varies across the countries depending on their electricity generation mix. CO2 emissions are estimated by type of fuel and mode, using the corresponding fuel consumption and associated emission factors. Data on gross domestic product (GDP), expressed in 2000 constant dollar measured at purchasing power parity, and population are also taken from the IEA (IEA, 2007f, 2007g). 9 In energy statistics, energy consumption by international aviation and maritime transportation are not considered part of national energy consumption. These are treated separately under their international conventions (i.e., International Civil Aviation Organization and International Maritime Organization). 28 Data for China does not include Hong Kong or Macau. Korea refers to the Republic of Korea. Data for Mongolia was only available for the period 1985-2005. 4. Results and Discussion All countries, with the exception of Mongolia, experienced significant growth in transportation sector CO2 emissions during the 1980-2005 period. However, there remain significant differences in the magnitude of emissions growth and the factors driving it. Table 7 summarizes the results of the additive decomposition of transport sector CO2 emissions growth into fuel switching, modal shifting and changes in emission coefficients10, transportation energy intensity, per capita GDP and population. Detailed year-by-year additive decomposition results for each country can be found in the Appendix. Figure 5 displays indexed time-series charts of the multiplicative decomposition of CO2 emissions and its driving factors in each of the 12 countries considered. 10 Only the emission coefficients of electricity changes due to different electricity generation mix over time; emission coefficients of other fuels remain constant throughout the study period. 29 Table 7: Annual Average CO2 Emission Change and Responsible Factors (1980-2005) Factors Influencing the CO2 Emissions Change Average CO2 GDP (PC) Modal Mix Coefficient Population Per Capita Emission Emission Fuel Mix Intensity Energy Change (POP) (MM) (FM) (EC) (EI) (`000 Country tCO2) Influencing Factors¥ China 10,199 173 -664 -27 -4,553 13,591 1,679 FM, PC, POP Bangladesh 140 0 0 0 31 61 48 EI, PC, POP India 2,022 80 -291 30 -2,896 3,508 1,592 FM, EC, PC, POP Indonesia 2,271 -14 1 0 392 1,256 637 MM, EI, PC, POP Korea 2,882 32 -88 -10 -254 2,758 445 FM, PC, POP Malaysia 1,322 2 0 0 215 647 458 FM, EI, PC, POP Mongolia* -4 14 -7 0 -45 17 16 MM, EI Pakistan 719 4 0 0 -75 360 430 FM, PC, POP Philippines 761 37 6 0 236 192 290 FM, MM, EI, PC, POP Sri Lanka 137 -1 0 0 -19 120 37 PC, POP Thailand 1,874 27 -18 0 251 1,272 343 FM, EI, PC, POP Vietnam 746 5 -4 0 236 398 112 FM, EI, PC, POP Note: The modal mix effect, as defined in this study, considers only four modes: road, rail, water and air. If necessary data is available to further disaggregate road transportation into auto, bus, etc., modal mix might be found to influence CO2 emission growth. In 2000 US$ PPP. ¥ Factors in the same direction as average emission change. * Analysis for Mongolia is from 1985 to 2005. As can be seen from Table 7, economic growth (i.e., per capita GDP growth) and population growth are the critical factors in the growth of transportation sector CO2 emissions in all countries except Mongolia. Deterioration in transportation energy intensity is also found to contribute to emissions growth in Bangladesh, Indonesia, Malaysia, the Philippines, Thailand and Vietnam. Change in transportation energy intensity is not selected as an influencing factor in China, India, Korea, Pakistan and Sri Lanka because it is not directly contributing to emission growth in these countries, but improvements in transportation intensity, nonetheless, have had a tremendous impact by reducing emissions from where they would have been otherwise. This is particularly clear in the case of China and India, where average emission growth would have been 45% higher (14,753 kilotons of CO2 rather than 10,199 kilotons) and 143% higher (4,918 30 kilotons of CO2 rather than 2,022 kilotons), respectively, if there were no reduction in transportation energy intensity. The same applies for modal mix effects in China and India, where average emission growth would have been 6.5% and 14.3 % higher, respectively, otherwise. This may appear to be a counter-intuitive result given the shift in China and India away from rail to road (see Table 2), but the reliance of rail in these countries on coal, and coal based electricity renders road a less carbon intensive mode of transport than rail11. Fuel switching contributes directly to emission growth in most of the countries considered (i.e., China, India, Korea, Malaysia, Pakistan, the Philippines, Thailand and Vietnam), but its impact is relatively small (see Table 7). For example, in India and the Philippines, where the fuel switching effect is most influential, it still only accounts for 4% and 4.9%, respectively, of the average growth in emissions. In Indonesia and Sri Lanka, where fuel switching serves as a brake on emission growth, its effect is negligible in magnitude. Although fuel switching is a common phenomenon in many Asian countries during the 1980-2005 period (see Table 3), interestingly, the fuel switching effect is not found to play a major role in driving transport sector CO2 emissions in these countries. This is because fuel substitution mostly occurred between diesel and gasoline, and their CO2 emission coefficients are not significantly different. Moreover, the effects of fuel mix and modal mix are much smaller relative to the impacts of economic and population growth. 11 Note however that the carbon intensity of rail would, in fact, be lower than road on a passenger-km or ton-km basis. 31 The emission coefficient effect has no impact on emission growth in all of the countries considered, with the exception of China, India and Korea. Considering that only the emission coefficient of electricity changes over time and that electricity use for transportation remained minimal in most of the countries, this is not surprising. The emission coefficient effect exacerbated emission growth in India, while it moderated it in China and Korea, but the effect is seen to be nominal in all cases. In all countries considered aside from Mongolia, the effects of fuel switching, modal shifting and changes in emission coefficients were eclipsed by the effects of the overwhelming growth in economic activity and population. In Mongolia, the only country considered where transport sector CO2 emissions display a declining trend (an average reduction of 4 kilotons of CO2 per year over the study period), the amelioration of emissions is largely due to improvements in transportation energy intensity and, to a lesser extent, modal mix. A sharp decline in per capita GDP, reflecting the financial crisis in the region in 1997, can be observed for Indonesia, Korea, Malaysia and Thailand in 1997-1998, followed by renewed growth (see Figure 5d, 5e, 5f and 5k). CO2 emission growth also follows this pattern in these countries. Deteriorating transportation energy intensity is found to have exacerbated CO2 emission growth from economic activity in Indonesia, Malaysia, and Thailand. On the other hand, improvement of transportation energy intensity has restrained emission growth in Korea as well as China (see Figure 5b). 32 Figure 5: Transport Sector CO2 Emissions Growth and Driving Factors in Selected Asian Countries 4.5 a) Bangladesh 8 b) China CO2 CO2 Fuel Mix 7 Fuel Mix Modal Mix Modal Mix Emission Coefficient 6 Emission Coefficient 3.5 T ransport Energy Intensity T ransport Energy Intensity Per capita GDP Per capita GDP Index (1980=1) Index (1980=1) Population 5 Population 2.5 4 3 1.5 2 1 0.5 0 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 c) India 4.3 d) Indonesia CO2 CO2 2.4 Fuel Mix Fuel Mix Modal Mix 3.8 Modal Mix Emission Coefficient Emission Coefficient T ransport Energy Intensity 3.3 T ransport Energy Intensity Per capita GDP Per capita GDP Index (1980=1) Index (1980=1) 1.9 Population Population 2.8 1.4 2.3 1.8 0.9 1.3 0.4 0.8 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 7 e) Korea f) Malaysia CO2 5.8 CO2 6 Fuel Mix Fuel Mix Modal Mix Modal Mix Emission Coefficient Emission Coefficient 5 T ransport Energy Intensity 4.8 T ransport Energy Intensity Per capita GDP Per capita GDP Index (1980=1) Index (1980=1) Population Population 4 3.8 3 2.8 2 1.8 1 0 0.8 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 33 Figure 5: Transport Sector CO2 Emissions Growth and Driving Factors in Selected Asian Countries (continued) g) Mongolia 4.1 h) Pakistan CO2 1.5 3.6 Fuel Mix Modal Mix Emission Coefficient 3.1 T ransport Energy Intensity Per capita GDP Index (1980=1) Index (1980=1) 1 Population 2.6 2.1 CO2 0.5 Fuel Mix Modal Mix 1.6 Emission Coefficient T ransport Energy Intensity 1.1 Per capita GDP Population 0 0.6 1985 1987 1989 1991 1993 1995 1997 1999 2001 2003 2005 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 i) Philippines j) Sri Lanka CO2 CO2 4.5 Fuel Mix 2.7 Fuel Mix Modal Mix Modal Mix Emission Coefficient Emission Coefficient T ransport Energy Intensity T ransport Energy Intensity 3.5 Per capita GDP Per capita GDP Index (1980=1) Index (1980=1) 2.2 Population Population 2.5 1.7 1.5 1.2 0.5 0.7 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 k) Thailand l) Vietnam 2004 CO2 10 CO2 5.5 Fuel Mix 9 Fuel Mix Modal Mix Modal Mix Emission Coefficient 8 Emission Coefficient 4.5 T ransport Energy Intensity T ransport Energy Intensity Per capita GDP 7 Per capita GDP Index (1980=1) Index (1980=1) Population Population 3.5 6 5 2.5 4 3 1.5 2 1 0.5 0 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 34 As mentioned above, improvement of transportation energy intensity has also had a tremendous impact on emission growth in India. However, the drastic improvement in transportation energy intensity between 1995 and 1996 (see Figure 5c) is the product of a statistical adjustment in the reporting of diesel consumption rather than the result of any real change (Zhou, 2007). It is thus likely that both CO2 emission growth and transportation energy intensity for India ought to be revised downwards from 1980 to 1995. Similarly, an abrupt change in diesel consumption is reported for the Philippines between 1993 and 1994 (see Figure 5i). While no documentation could be found, it is highly likely that this is another instance of a statistical adjustment to the energy consumption data, and CO2 emission growth and transportation energy intensity for the Philippines ought to be revised upwards from 1980 to 1993. Figure 5g presents an indexed time-series chart of the decomposition results for Mongolia, the only country where transport sector CO2 emission changes were primarily driven by changes in transportation energy intensity. Transportation energy intensity in Mongolia declined drastically after 1988, but especially from 1990-91 onwards, without Soviet assistance after the demise of the Soviet Union. Although Mongolia suffered a devastating recession after the collapse of the Soviet Union, its primary trading partner, its economy recovered and started growing again after 1994. However, its transportation energy intensity remained low as the country continued to suffer from chronic energy shortages. This explains the net reduction in CO2 emissions from the transport sector as 35 compared to 1985. Further studies are needed to precisely determine the causes of decreases in the transportation energy intensities in Mongolia.12 5. Policy Initiatives to Address CO2 Emissions Growth The results of our decomposition analysis show that economic activity, population growth and transport sector energy intensity are the main factors driving transport sector emissions growth in Asia. Developing Asian countries are not expected to slow down their economic growth to control their CO2 emissions because these countries have neither mandatory nor voluntary commitments to reduce CO2 emissions under the Kyoto Protocol. Thus, the main strategy to limit the growth of CO2 emissions in the transport sector in these countries in the future would be the decoupling (or weakening) of the growth of CO2 emissions from economic growth, which has not been the case historically. Rapid switching to clean fuels and shifting over to public transportation, including rail and water transportation, could help achieve this objective. Fiscal instruments, such as subsidies for public transportation, clean fuels and clean vehicles, would be helpful in triggering fuel switching and modal shifting activities. Moreover, regulatory instruments, such as vehicle efficiency standards, vehicle occupancy standards, congestion charges, investments in road maintenance and congestion reduction, would also be required to reduce transportation energy intensity and thereby reduce transport sector CO2 emissions. 12 In order to determine the reasons for transportation energy intensity change, this indicator itself can be decomposed into its driving factors, such as fuel efficiency of transportation by mode and transport service intensity of the economy. 36 In the following sub-sections, we summarize policy initiatives undertaken by Asian countries that help address energy consumption and associated emissions from the transport sector. 5.1. Fiscal Instruments to Reduce Fuel Consumption Many countries in Asia have traditionally subsidized fuels used for transportation13. Historically, some Asian countries, such as Indonesia and Malaysia, have featured the some of lowest fuel prices in the region due to high subsidization, but both raised retail prices sharply between November 2004 and November 2006 (GTZ, 2007). The Philippines has also introduced drastic price increases, whereas Vietnam has pursued a policy of continuous reductions in subsidies. In both cases, gasoline has recently begun to be taxed although diesel remains subsidized. Thailand raised diesel prices by 76 percent to the point where it is almost subsidy-free, like gasoline in the country. Korea, which has long featured some of the highest fuel prices in the world, continues to raise taxes on diesel and gasoline. China has also raised its fuel prices by more than 40 percent between November 2004 and November 2006 so that they are no longer subsidized and hover near U.S. levels. India and Pakistan both maintain high non- 13 GTZ (2007) has been conducting worldwide surveys of diesel and gasoline prices since 1991 and uses U.S. retail prices (minus the U.S. highway tax) as benchmark values for non-subsidized fuels. U.S. prices are taken as benchmark values because the high degree of competition in the U.S. fuel market implies pricing that reflect commercially calculated full-cost prices. Most Asian countries tend to price diesel below this benchmark, indicating significant national subsidies for diesel, and some do so for gasoline as well. 37 subsidized prices for gasoline, but India has relaxed price controls on diesel since the early part of this decade. Even though Bangladesh has increased the price of state- subsidized diesel by over 30 percent, smuggling from Bangladesh to India (and, similarly, from Iran to Pakistan) continues to subvert state fuel policy in these countries. Increasing diesel prices, or removing subsidies from diesel, could have environmental and ethical ramifications. First, diesel is mainly used in vehicles for public transportation which emit less CO2 emissions than automobiles in terms of passenger kilometer traveled. Rising diesel price could encourage switching over to private transportation from public transportation and hence increase CO2 emissions. Second, since public transportation is the main mode of road transportation for low income households, removing subsidies from diesel or increasing taxes on diesel would put extra burden on those households. Fuel pricing is not the only means by which governments in Asia have attempted to curtail fuel consumption and CO2 emissions growth. Other policies, such as providing incentives to low emission vehicles, are also practiced in some Asian countries. For example, the Chinese government updated excise tax rates, cutting rates by 5 to 3 percent on small-engine (1.0-1.5 liter) vehicles while raising rates by 8 to 20 percent on vehicles with larger-engines (more than 4 liters), to further encourage the manufacture of smaller- engine vehicles (ICCT, 2007). This was supplemented by the elimination of the preferential 5 percent tax rate on SUVs. Similarly, in 2007, Thai authorities approved a range of incentives for the production of eco-cars14: Eco-car manufacturers were granted 14 Cars with engine displacement of under 1,300cc if running on petrol or less than 1,400cc if running on diesel, and fuel consumption that is better than 20 kilometers per liter. 38 a maximum eight-year exemption from corporate income tax payments and machinery import duties, and excise rates for eco-cars were set at 17%, compared with a rate of 30 ­ 50% for other types of cars in Thailand (Economist Intelligence Unit, 2008). Due to the relative price inelasticity of demand for automotive fuel in Asia, pricing instruments alone may not be effective in curtailing transport sector fuel consumption and associated emissions substantially (ADB, 2006). Other policy instruments, particularly regulatory policy instruments, would also be needed. 5.2. Regulatory Instruments to Improve Fuel Economy Regulatory instruments, such as vehicle fuel economy standards, have been applied in many developed countries, where they are regarded as successful tools to reduce fuel demand and CO2 emissions from the transportation sector (Pew, 2004). In developing Asia, fuel economy standards have already been implemented in China and South Korea, and India is expected to join them in the near future. China first began regulating fuel economy standards for its passenger vehicle fleet in 2005 and followed that with even more stringent standards in 2008 in order to decrease its dependence on foreign oil and to encourage foreign automakers to introduce more efficient vehicle technologies to the local market (ICCT, 2007). The standards, which are structured to be more stringent for the heavier vehicle classes to provide incentives for manufacturers to produce lighter vehicles, specify maximum allowable fuel consumption by 16 weight categories and cover passenger cars, SUVs and multi-purpose vans (MPVs) but not commercial vehicles and pickup trucks (Pew, 2004). As the Chinese standards are 39 not based on fleet average, every single vehicle model sold in China is required to conform to the standard for its weight class in terms of maximum liters consumed per 100-km (Pew, 2004). A recent study by the China Automotive Technology and Research Center finds that the standards implemented in 2005 improved overall passenger vehicle fuel efficiency from 26 mpg in 2002 to 28.4 mpg in 2006, or by about 9 percent, in spite of increases in gross weight and engine displacement (ICCT, 2007). South Korea announced, in 2004, that it will implement mandatory fuel economy regulations ­ the Average Fuel Economy (AFE) standard ­ to replace the existing system of voluntary standards in 2006 for domestic cars and in 2009 for imports. The standards are set at 34.4 CAFE-normalized15 mpg for vehicles with engine displacement less than 1,500 cubic centimeters (cc) and 26.6 mpg for those over 1,500 cc (ICCT, 2007). However, despite being mandatory, there are no fiscal or criminal penalties for violating the AFE standard, with only a public shaming to serve as deterrent (Pew, 2004). The new AFE standard is partly a reaction to declining average fuel economy resulting largely from the growing popularity of SUVs, but since fuel economy standards remain static from 2006 onwards while the market share of larger, more powerful vehicles continues to rise gradually, South Korea is the only country with mandatory fuel economy standards where fleet average fuel economy is expected to deteriorate through 2012 (ICCT, 2007). In India, the "Mashelkar Committee" proposed voluntary fuel economy standards for the automotive industry in 2002 (Mashelkar, 2002). Manufacturers, however, failed to commit to the voluntary scheme within the four year grace period granted by the 15 U.S. Corporate Average Fuel Economy (CAFE) test method. Other methods include New European Drive Cycle (NEDC), and Japan 10-15 Cycle. 40 government. In 2007, it was reported that the Indian government had resolved to create mandatory fuel efficiency standards for vehicles of every class and type, including cars, scooters, bikes, trucks, buses and three-wheelers; and that manufacturers may be penalized for neglecting to meet these standards (Times of India, 5 Oct 2007). Government projections estimated that the country could save up to $36 billion if fuel efficiency is improved by 50% by 2030 in all sectors (CSE, 2008). 5.3. Policy Instruments to Promote Fuel Switching Aside from fuel economy regulations and pricing instruments to control fuel consumption, policy instruments to promote switching to alternative fuels have also been implemented in some countries and municipalities. The International Association of Natural Gas Vehicles (2008) reports extraordinary average growth of approximately 50 percent a year in the CNG vehicle population in Asia since 2000. Consequently, almost 2.5 million of the total 7 million compressed natural gas (CNG) vehicles estimated to be in service around the world in 2007 are found in developing countries in Asia (IANGV, 2008). This is, to some extent, by design. For example, in 2002, the Supreme Court in India directed the Union government to give priority to the transport sector over other sectors in India with regard to the allocation of CNG (ADB, 2006). In Thailand, the government, aiming to displace at least 10 percent of gasoline and diesel consumed with natural gas to help reduce the country's dependence on oil, scrapped import duties and lowered excise taxes on CNG-compatible cars in 2006 (Economist Intelligence Unit, 2008). CNG, which is sold by the state monopoly, PTT, is also priced to be more 41 attractive than other fuels and costs only about 30 percent as much as unleaded regular gasoline. In addition, ceiling prices have been set to ensure that the price of CNG does not exceed 55 percent of the price of unleaded regular gasoline in 2007, 60% in 2008, and 65% in 2009 (IANGV, 2007). At the municipal level, more than 10,000 CNG buses are already running in Delhi (IANGV, 2008), where India's Supreme Court mandated CNG as the fuel for public transport in 1998 to improve urban air quality (ADB, 2006). Another four cities in India ­ Ahmedabad, Lucknow, Kanpur, and Hyderabad ­ have implemented CNG programs for urban transport and 22 more cities will be enabled to do so if the national gas grid is implemented (ADB, 2006). Elsewhere in the region, Dhaka requires its 3-wheelers to run on CNG (as does Delhi), while Shanghai requires a fraction of its new buses to be constructed to run on CNG. CNG has also started to be used in buses and taxis in cities such as Xi'an and Beijing (Schipper and Ng, 2004). Most of the approximately 10 million LPG vehicles in use worldwide in 2004 were concentrated in a few countries, and four countries of them ­ Korea, Japan, Poland, Turkey, and Australia ­ account for more than half the world on-road transport consumption of LPG (ADB, 2006). In Asia, Korea and Japan actively promote the use of LPG. More than 10 percent of all registered vehicles in Korea, the world's largest automotive LPG consumer, are LPG-fueled due to the rapid market growth brought about by a large excise tax advantage over gasoline and diesel (which has since been reduced) (DOE, 2002). The world's second largest automotive LPG market is Japan, where the government offers grants for conversion or purchase of LPG vehicles and installation of filling stations. LPG is primarily utilized for taxis, buses and trucks in Korea, and mainly 42 for taxis in Japan, where more than 90 percent of taxis run on LPG, because the capital costs of conversion are more easily recovered by high mileage commercial vehicles (Liu et al, 1997). Introduction of biofuels would be another option to replace gasoline and diesel in the transport sector. Current biofuels production capacity in Asia is very small. Of the 2.2 billion liters of biodiesel produced globally in 2004 (accounting for around 0.33% of worldwide diesel production), only a small fraction of that was made in Asia, specifically in the Philippines and in Malaysia (ADB, 2006). 33 billion liters of ethanol were produced worldwide, of which 0.2 billion liters came from Thailand, a major producer of sugar cane and cassava, two important feedstocks for the manufacture of ethanol. However, investment in new biodiesel production capacity, with coconut oil, palm oil, and jatropha as feedstock, is growing. Production of coconut biodiesel has already been started at the commercial scale in the Philippines, and Malaysia is expanding its palm oil biodiesel production capacity. Thailand, where a National Biofuels Committee, supported by the ministries of finance, agriculture, industry and energy, is responsible for policy direction, strategy planning and implementation (Gonsalves, 2006), has taken aggressive steps towards displacing substantial quantities of fossil fuels with biofuels in its transport sector. The Thai government, which mandated that all government vehicles use gasohol by 2006, more recently passed a tax incentive package to encourage motorists and truckers to switch to ethanol or compressed natural gas from gasoline or diesel since it is negatively affected by rising oil prices as a net crude oil importer (Reuters UK, 2008). Following three years of strong sales growth of E10, propelled by lower government duties and 43 surcharges compared to gasoline, Thailand announced the launch of E85 (an 85:15 mix of ethanol and gasoline) in October 2008. Retail prices for E85 will be 30-40 percent cheaper than premium gasoline, and the excise tax on E85 vehicles will be lower than that on E20 vehicles (currently 25 percent) and regular vehicles (30-50 percent) (USDA, 2008). Biodiesel production also enjoys tax incentives in Thailand: the government does not assess taxes on it and subsidizes it via an oil fund financed by surcharges from domestic retail prices of diesel and gasoline (Reuters UK, 2007). From February 1, 2008 onward, the Thai government began to enforce compulsory production of B2 (2 percent biodiesel in diesel) throughout the country and guarantee the sale quantity and prices to manufacturers (USDA, 2008). The mandate is expected to escalate to B5 in the near future. In support of this program, it also grants B2 manufacturers a subsidy of 0.30 Baht/liter from the State Oil Fund. 5.4. Investments in Public Transport Infrastructure Even though public transportation is more sustainable than private cars in economic, social and environmental terms, many Asian cities have not been able to deliver the urban transport services needed to meet growing demand. Most Asian cities offer some inexpensive public transport options, but these are often overcrowded and dirty, and consequently, unable to draw commuters who can afford private transport. Despite the availability of formal bus services, more passenger trips are carried out on informal buses and paratransit vehicles (ADB, 2006). Both these public buses and 44 paratransit vehicles are typically responsible for significant amounts of local and pollutants and GHG emissions. A number of cities in the region have constructed rail-based public transport systems, which provide high capacity along with high quality service, in recent years. Some, such as Manila and Shanghai, have put in place light rail systems, while others, such as Beijing, Delhi, Nanjing and Shanghai, have opted for subway systems, and Bangkok has invested in both light rail and subway systems (ESCAP, 2007). Rail-based mass transit systems have long been operational in the Korean cities of Seoul, Busan, Daegu and Incheon, the Chinese cites of Beijing and Tianjin, and the Indian city of Kolkata. Furthermore, they are under construction in Bangalore and Mumbai; and have been proposed for Karachi and Dhaka. Despite their high public profile, rail-based public transport systems are responsible for a relatively small share of overall trips (usually less than 10%), yet they have always required large subsidies for construction and operation (ADB, 2006). Even with the subsidies, they are only viable with high passenger flows, and, thus, are best suited for large, densely populated cities. Another option that is gaining popularity in Asian cities is the bus rapid transit (BRT) system, characterized by "high-capacity (usually articulated) buses operating in exclusive segregated bus lanes, with rapid loading and unloading of passengers at stations that provide electronic fare pre-payment and obstacle-free waiting areas and level access to the buses" (ADB, 2006). Whereas rail-based public transport systems have larger passenger capacity, higher speeds and emit less air pollutants than bus systems (Litman, 2008), their construction and operation costs are often prohibitive. These modern bus corridors have proven that carrying capacities close to that of rail-based systems can be 45 delivered without subsidizing operations as they are often the product of public-private partnerships where the public sector funds only the infrastructure costs. As the infrastructure costs of BRT systems are far lower than those for rail-based mass transit, e.g., US$1-15 million per km compared to US$50 million to over US$200 million per km (Wright and Fulton, 2005 (citing Wright, 2004)), this makes them especially attractive as an alternative to rail-based systems in developing countries. Twenty BRT systems were already operational in Asia as of July 2008, with another fifty planned or under construction (CAI-Asia, 2009). The ADB (2006) reports that more cities in Asia are planning to implement BRT systems than rail-based mass transit systems. While existing BRT systems are most common in Japan, they are already operational in cities in China, India, Indonesia and South Korea as well. Most of the BRT systems that are planned, under construction or under consideration are in Chinese cities, but they are also anticipated in Thailand, Philippines, Sri Lanka, Bangladesh and Pakistan (CAI-Asia, 2009). The potential of BRT to reduce GHG emissions is recognized, as evidenced by the approval of the baseline and monitoring methodologies of the BRT in Bogotá, Columbia for the Clean Development Mechanism in July 2006. 6. Conclusions This study examines the growth of the transport sector CO2 emissions and determines the underlying factors in twelve Asian countries over 25 years between 1980 and 2005. To identify the driving factors, we decompose the emission growth into fuel switching, modal shifting, per capita economic growth, population growth and changes in 46 emission coefficients and transportation energy intensity using the logarithmic mean Divisia index (LMDI) approach. We find that population growth, per capita economic growth and change in transportation energy intensity are generally found to be principal drivers of transport sector CO2 emission growth in Asian countries, whereas fuel switching, modal shifting and change in emission coefficients are not found to have a sizeable influence on the growth of transport sector CO2 emissions. The per capita economic growth effect and the population growth effect are found to be primarily responsible for driving transport sector CO2 emissions growth over the study horizon in all countries except Mongolia. The transportation energy intensity effect is found to be the main driver of the reduction of CO2 emissions in Mongolia. However, improvement in transportation energy intensity is also found to restrain the growth of transport sector CO2 emissions in some countries, significantly in China and India, while deterioration of transportation energy intensity exacerbates emission growth in other countries. Some Asian countries, such Korea, China and India, have already implemented or are planning to implement, regulatory instruments, such as stringent fuel economy standards. This policy would reduce transportation energy intensity and thereby help reduce CO2 emissions. Transportation energy intensity can also be reduced by improving vehicle occupancy rates through carpooling, which has not been promoted widely in Asia, or through the use of high occupancy public transit systems. China and Thailand are using fiscal instruments (e.g., favorable tax treatment to the production of vehicles with smaller engine). Korea maintains one of the highest tax rates on gasoline and diesel in the world. Several Asian countries have eliminated or reduced subsidies on fuels used by 47 private vehicles. However, automotive fuel demand is relatively price inelastic and hence pricing policies need to be supplemented with other policy instruments, such as the incentives offered in China and Thailand to promote the proliferation of smaller vehicles. Congestion pricing, a fiscal instrument to lessen traffic congestion, has not been attempted in any of the countries considered although it has been applied with great success in Singapore. The fuel switching and modal shifting effects are not found to have a sizeable influence on the growth of transport sector CO2 emissions in any of the Asian countries studied. However, policy instruments to encourage these underutilized approaches could significantly reduce emissions. Some countries are giving preferential treatment to less carbon intensive fuels, such as compressed natural gas, liquefied petroleum gas and biofuels, through favorable taxation and vehicular and blending mandates (e.g., mandating CNG buses in Delhi and Shanghai and 3-wheelers in Dhaka and Delhi). If these policies are successful in inducing large scale switching from gasoline and diesel to less carbon intensive fuels, such as natural gas and electricity, then the fuel mix effect can be expected to have a significant impact on transport sector CO2 emissions. Some countries, such as Thailand, have used subsidies, exemption of import duties, and lower excise taxes to promote switching to cleaner fuels. Moreover, Korea and China have already introduced fuel economy standards, which others Asian countries could follow. The study does not detect modal shifting to be a main factor in the growth of transport sector CO2 emissions, but the analysis is limited by the unavailability of data specific to the various types of transport within road transport, such as cars, minibuses, buses and BRT systems. As road is overwhelmingly the most popular mode of transport 48 in Asia, disaggregated data on intra-modal shifting in road transport would enable a more accurate depiction of the impact of modal switching on transport sector CO2 emissions. Moreover, several Asian cities have recently constructed light rail (e.g., in China, India, Korea, Philippines, Malaysia and Thailand) and Bus Rapid Transit systems (e.g., in Jakarta, Seoul and Beijing) to provide convenient and price competitive alternatives to private road vehicles. 49 References Accountancy, 2007. Economic survey of Pakistan 2006-2007. Available at: . An, F., Sauer, A., 2004. Comparison of passenger vehicle fuel economy and GHG emission standards around the world. Pew Center on Global Climate Change. Ang, B.W., 2004. Decomposition analysis for policymaking in energy: which is the preferred method? Energy Policy, 32 (), 1131-1139. Ang, B.W., 2005. The LMDI approach to decomposition analysis: a practical guide. Energy Policy, 33 (7), 867-871. ADB, 2006. Energy efficiency and climate change considerations for on-road transport in Asia. Asian Development Bank (ADB), Philippines. Bhattacharyya S.C., Ussanarassamee, A., 2004. Decomposition of energy and CO2 intensities of Thai industry between 1981 and 2000. Energy Economics, 26 (5), 765-781. CAI-Asia, 2009. Bus rapid transit (BRT) overview. Clean Air Initiative Asia (CAI-Asia). Available at: . CSE, 2008. Avert the great guzzle: fuel economy regulations: setting the principles right (draft version). Center for Science and Environment (CSE), Delhi. Chang, Y.F., Lin, S.J., 1998. Structural decomposition of industrial CO2 emission in Taiwan: an input-output approach. Energy Policy 26 (1), 5-12. Diakoulaki, D., Mandaraka, M., 2007. Decomposition analysis for assessing the progress in decoupling industrial growth from CO2 emissions in the EU manufacturing 50 sector. Energy Economics, 29 (4), Modeling of Industrial Energy Consumption, Pages 636-664. Diakoulaki, D., Mavrotas, G., Orkopoulos, D., Papayannakis, L., 2006. A bottom-up decomposition analysis of energy-related CO2 emissions in Greece. Energy, 31 (14), 2638-2651. DOE, 2002. LPG around the world. Alternative Fuel News. Volume 6, Number 2. U.S. Department of Energy (DOE), October 2002. Ebohon, O.J., Ikeme, A.J., 2006. Decomposition analysis of CO2 emission intensity between oil-producing and non-oil-producing sub-Saharan African countries. Energy Policy, 34 (18), 3599-3611. Economist Intelligence Unit, June 10th 2008. "All is not well with Thailand's ambitious eco-car scheme." Available at: . Energy Information Administration (IEA), 2007. International Energy Outlook 2007. US Department of Energy, Washington, DC. ESCAP, 2007a. Review of Developments in Transport in Asia and the Pacific 2007: Data and Trends. United Nations Economic and Social Commission for Asia and the Pacific (ESCAP), Thailand. ESCAP, 2007b. Statistical abstract of transport in Asia and the Pacific. United Nations Economic and Social Commission for Asia and the Pacific (ESCAP), Thailand. Available at: . 51 General Statistical Office of Vietnam, 2005. Statistical Yearbook of Vietnam 2005. Statistics Documentation Centre, Hanoi. Available at . Gonsalves, J.B., 2006. An assessment of the biofuels industry in Thailand. United Nations Conference on Trade and Development GTZ, 2007. International Fuel Prices, 2007, Fifth Edition, Deutsche Gesellschaft fur Technische Zusammenarbeit (GTZ), Germany. Han, X., Chatterjee, L., 1997. Impacts of growth and structural change on CO2 emissions of developing countries. World Development, 25 (3), 395-407. Hatzigeorgiou, E., Polatidis, H., Haralambopoulos, D., 2008. CO2 emissions in Greece for 1990-2002: A decomposition analysis and comparison of results using the Arithmetic Mean Divisia Index and Logarithmic Mean Divisia Index techniques. Energy, 33 (3), 492-499. ICCT, 2007. Passenger vehicle greenhouse gas and fuel economy standards: a global review. International Council on Clean Transportation (ICCT), July 2007. IEA, 2007. International Energy Agency (IEA) Database Vol. 2007, Release 01: (a) CO2 Emissions from Fuel Combustion; (b) Energy Balances of Non-OECD Member Countries ­ Extended Balances (c) Energy Balances of OECD Member Countries - Extended Balances; (d) Energy Statistics of Non-OECD Countries - Basic Energy Statistics; (e) Energy Statistics of OECD Countries - Basic Energy Statistics; (f) Energy Balances of Non-OECD Member Countries ­ Indicators; (g) Energy Balances of OECD Member Countries ­ Indicators. Available at: . 52 IEA, 2007h. World energy outlook 2007. International Energy Agency (IEA), Paris. IEA, 2004. 30 years of energy use in IEA countries. International Energy Agency (IEA), Paris. p. 123. IANGV, 2007. Current status of NGV in Thailand. International Association for Natural Gas Vehicles (IANGV). Available at: . IANGV, 2008. Natural gas vehicle statistics. International Association for Natural Gas Vehicles (IANGV). Available at: . IPCC, 2006. Guidelines for National Greenhouse Gas Inventories. Intergovernmental Panel on Climate Change (IPCC). pp. 3-16. Kawase, R., Matsuoka, Y., Fujino, J., 2006. Decomposition analysis of CO2 emission in long-term climate stabilization scenarios. Energy Policy, 34 (15), 2113-2122. Kenworthy, J.R., 2003. Transport energy use and greenhouse gases in urban passenger transport systems: a study of 84 global cities. Presented to the international Third Conference of the Regional Government Network for Sustainable Development, Notre Dame University, Fremantle, Western Australia, September 17-19, 2003. Korea National Statistical Office, 2008. Available at: . Kveiborg, O., Fosgerau, M., 2007. Decomposing the decoupling of Danish road freight traffic growth and economic growth. Transport Policy, 14, 39­48. 53 Lakshmanan, T., Han, X., 1997. Factors underlying transportation CO2 emissions in the USA: a decomposition analysis. Transportation Research Part D, 2(1), 1-15. Lee, K., Oh, W., 2006. Analysis of CO2 emissions in APEC countries: A time-series and a cross-sectional decomposition using the log mean Divisia method. Energy Policy, 34 (17), 2779-2787. Liaskas, K., Mavrotas, G., Mandaraka, M., Diakoulaki, D., 2000. Decomposition of industrial CO2 emissions: The case of European Union. Energy Economics, 22 (4), Pages 383-394. Lin J., Zhou, N., Levine, M., Fridley D, 2008. Taking out 1 billion tons of CO2: The magic of China's 11th Five-Year Plan. Energy Policy, 36 (3), 954-970. Lise, W., 2006. Decomposition of CO2 emissions over 1980-2003 in Turkey. Energy Policy, 34 (14), 1841-1852. Litman, Todd. 2008. Evaluating public transit benefits and costs: best practices guidebook. Victoria Transport Policy Institute. Available at: . Liu, E., Yue, S.Y, Lee, J., 1997. A study on LPG as a fuel for vehicles. Research and Library Services Division, Legislative Council Secretariat, Government of Hong Kong. Liu, L.C., Fan, Y., Wu. G., Wei, Y.M., 2007. Using LMDI method to analyze the change of China's industrial CO2 emissions from final fuel use: An empirical analysis. Energy Policy, 35 (11), 5892-5900. 54 Lu, I.J., Lin, S.J., Lewis, C., 2007. Decomposition and decoupling effects of carbon dioxide emission from highway transportation in Taiwan, Germany, Japan and South Korea. Energy Policy, 35 (6), 3226-3235. Luukkanen, J., Kaivo-oja, J., 2002a. ASEAN tigers and sustainability of energy use-- decomposition analysis of energy and CO2 efficiency dynamics. Energy Policy, 30 (4), 281-292. Luukkanen, J., Kaivo-oja, J., 2002b. A comparison of Nordic energy and CO2 intensity dynamics in the years 1960-1997. Energy, 27 (2), 135-150. Mashelkar, R. A. et al., 2002. Interim report of the expert committee on auto fuel policy. Government of India, New Delhi. Nag, B., Kulshrestha, M., 2000. Carbon emission intensity of power consumption in India: A detailed study of its indicators. Energy Sources, 22 (2), 157-166(10). National Bureau of Statistics of China, 2007. China Statistical Yearbook 2007. China Statistics Press. Available at: . Reuters UK, 2007. "Thailand's Bangchak sees biodiesel boom." January 16, 2007. Available at: . Reuters UK, 2008. "Thailand launches push to use more ethanol, gas." May 27, 2008. Available at: . 55 Rhee, H.C., Chung, H.S., 2006. Change in CO2 emission and its transmissions between Korea and Japan using international input-output analysis. Ecological Economics, 58 (4), 788-800. Saikku, L., Rautiainen, A., Kauppi, P.E., 2008. The sustainability challenge of meeting carbon dioxide targets in Europe by 2020. Energy Policy, 36 (2), 730-742. Schäfer, A., 2000. Regularities in travel demand: an international perspective. Journal of Transportation and Statistics, December, 1­31. Schipper, L., Marie-Lilliu, C., Gorham, R., 2000. Flexing the link between transport greenhouse gas emissions: a path for the World Bank. International Energy Agency, Paris, June. Schipper, L., Murtishaw, S., Khrushch, M., Ting, M., Karbuz, S., Unander, F., 2001. Carbon emissions from manufacturing energy use in 13 IEA countries: long-term trends through 1995. Energy Policy, 29 (9), 667-688. Schipper, L., Ng, W., 2004. Rapid motorization in china: environmental and social challenges. ADB-JBIC-World Bank East Asia and Pacific Infrastructure Flagship Study. Schipper, L., Scholl, L., Price, L., 1997. Energy use and carbon from freight in ten industrialized countries: an analysis of trends from 1973-1992. Transportation Research ­ Part D: Transport and Environment, 2 (1), 57-76. Scholl, L., Schipper, L., Kiang, N., 1996. CO2 emissions from passenger transport: a comparison of international trends from 1973 to 1992. Energy Policy, 24 (1), 17-30. 56 Shrestha, R.M., Marpaung, C.O.P., 2006. Integrated resource planning in the power sector and economy-wide changes in environmental emissions. Energy Policy, 34 (18), 3801-3811. Shrestha, R.M., Timilsina, G.R., 1996. Factors affecting CO2 intensities of power sector in Asia: a Divisia decomposition analysis. Energy Economics, 18 (4), 283-293. Sperling, D., Salon, D., 2002. Transportation in developing countries: an overview of greenhouse gas reduction strategies. Pew Center on Global Climate Change. Arlington, Virginia. Times of India, 2007. "Soon, your car will come with a mileage tag." Nitin Sethi, October 5, 2007. Available at: . United Nations Statistics Division, 2008a. Key Global Indicators Database. Available at: . United Nations Statistics Division, 2008b. National Accounts Main Aggregates Database. Available at: . USDA, 2008. Global agriculture information network (GAIN) report: Thailand bio-fuels annual 2008. U.S. Department of Agriculture (USDA) Foreign Agricultural Service. Wang, C., Chen, J., Zou, J., 2005. Decomposition of energy-related CO2 emission in China: 1957-2000. Energy, 30 (1), 73-83. Webster, F.V., Bly, P.H., Johnson, R.H., Dasgupta, M., 1986a. Part 1: urbanization, household travel, and car ownership. Transport Reviews, 6 (1), 49-86. Webster, F.V., Bly, P.H., Johnson, R.H., Dasgupta, M., 1986b. Part 2: public transport and future patterns of travel. Transport Reviews, 6 (2), 129-172. 57 World Bank, 2008. World development indicators 2008. World Bank, Washington, DC. World Bank, 2006. World development indicators 2006. World Bank, Washington, DC. Wright, Lloyd, and Lewis Fulton. 2005. Climate change mitigation and transport in developing nations. Transport Reviews 25 (6):691-717. Wu, L., Kaneko, S., Matsuoka, S., 2005. Driving forces behind the stagnancy of China's energy-related CO2 emissions from 1996 to 1999: the relative importance of structural change, intensity change and scale change. Energy Policy, 33 (3), 319- 335. Yabe, N., 2004. An analysis of CO2 emissions of Japanese industries during the period between 1985 and 1995. Energy Policy, 32 (5), 595-610. Zhou, N., 2007.What do India's transport energy data tell us? A Bottom-up Assessment of Energy Demand in India Transportation Sector. Ernest Orlando Lawrence Berkeley National Laboratory. Presentation at 2007 8th ECEEE Summer Study, 4-9 June 2007.Available at: . 58 Appendix CO2 Emissions Change and Contributing Factors Bangladesh CO2 Emissions Factors Influencing the CO2 Emissions Change Year Change Fuel Modal Emission Transport Per Capita Influencing Factors (`000 tCO2) Population Mix Mix Coefficient Energy Intensity GDP 1981 -162 -1 0 0 -195 12 22 FM, EI 1982 151 2 0 0 128 -1 22 FM, EI, POP 1983 120 3 0 0 77 15 26 FM, EI, PC, POP 1984 -27 -3 0 0 -79 28 27 FM, EI 1985 273 3 0 0 233 9 29 FM, EI, PC, POP 1986 -27 -3 0 0 -79 23 32 FM, EI 1987 96 1 0 0 45 18 32 FM, EI, PC, POP 1988 86 -1 0 0 56 -3 34 EI, POP 1989 139 -3 0 0 101 4 36 EI, PC, POP 1990 25 1 -1 0 -70 57 38 FM, PC, POP 1991 166 0 1 0 108 18 40 MM, EI, PC, POP 1992 278 1 0 0 180 52 45 FM, EI, PC, POP 1993 56 -3 0 0 -37 48 48 PC, POP 1994 43 -2 0 0 -42 40 48 PC, POP 1995 458 0 0 0 342 65 52 EI, PC, POP 1996 270 -1 0 0 144 69 58 EI, PC, POP 1997 384 -7 -1 0 228 102 62 EI, PC, POP 1998 75 2 0 0 -99 105 66 FM, PC, POP 1999 -298 8 1 0 -462 91 63 EI 2000 -61 -13 -2 0 -223 118 59 FM, MM, EI 2001 748 12 2 0 559 108 66 FM, MM, EI, PC, POP 2002 62 2 0 0 -106 91 74 FM, PC, POP 2003 66 -7 -1 0 -125 125 74 PC, POP 2004 -68 4 1 0 -309 163 73 EI 2005 653 5 1 0 406 166 75 FM, MM, EI, PC, POP 59 Appendix (Cont'd) CO2 Emissions Change and Contributing Factors China CO2 Emissions Factors Influencing the CO2 Emissions Change Year Change Fuel Modal Emission Transport Energy Per Capita Influencing Factors (`000 tCO2) Population Mix Mix Coefficient Intensity GDP 1981 1,145 148 760 -85 -3,943 3,188 1,078 FM, MM, PC, POP 1982 4,438 -50 -97 -76 -2,911 6,292 1,280 PC, POP 1983 5,644 233 -603 -74 -3,425 8,185 1,329 FM, PC, POP 1984 6,927 198 -383 55 -6,843 12,612 1,289 FM, EC, PC, POP 1985 1,914 -463 -380 -366 -9,871 11,597 1,397 PC, POP 1986 7,081 -172 -583 -40 -1,157 7,440 1,593 PC, POP 1987 5,983 97 -777 26 -5,839 10,653 1,823 FM, EC, PC, POP 1988 7,152 229 -626 83 -5,408 10,938 1,936 FM, EC, PC, POP 1989 4,987 172 -276 68 -53 3,140 1,937 FM, EC, PC, POP 1990 -5,634 423 -533 83 -10,306 2,849 1,849 MM, EI, 1991 10,416 220 -1,758 59 611 9,535 1,749 FM, EC, EI, PC, POP 1992 8,359 452 -1,552 -74 -8,747 16,593 1,687 FM, PC, POP 1993 17,473 -181 -1,611 24 -471 17,982 1,729 EC, PC, POP 1994 -9,671 797 -247 -267 -28,974 17,273 1,747 MM, EC, EI, 1995 11,346 -326 -1,021 223 -3,590 14,372 1,686 EC, PC, POP 1996 6,827 510 -1,633 117 -7,838 13,948 1,723 FM, EC, PC, POP 1997 19,491 -320 1,326 -405 3,133 13,943 1,814 MM, EI, PC, POP 1998 18,378 -1,534 988 285 3,919 12,839 1,881 MM, EC, EI, PC, POP 1999 13,385 550 -893 -172 -1,649 13,542 2,007 FM, PC, POP 2000 12,410 -133 -457 -377 -4,786 16,568 1,595 PC, POP 2001 5,497 426 -307 -381 -12,927 16,985 1,701 FM, PC, POP 2002 13,906 417 -336 198 -7,625 19,617 1,635 FM, EC, PC, POP 2003 31,884 -909 414 201 6,773 23,745 1,660 MM, EC, EI, PC, POP 2004 26,606 4,102 -5,629 539 -745 26,569 1,769 FM, EC, PC, POP 2005 29,042 -555 -382 -309 -1,145 29,358 2,076 PC, POP 60 Appendix (Cont'd) CO2 Emissions Change and Contributing Factors India CO2 Emissions Factors Influencing the CO2 Emissions Change Year Change Fuel Modal Emission Transport Per Capita Influencing Factors (`000 tCO2) Population Mix Mix Coefficient Energy Intensity GDP 1981 847 50 -260 47 -2,595 2,302 1,303 FM, EC, PC, POP 1982 3,350 9 -423 61 1,529 844 1,330 FM, EC, EI, PC, POP 1983 2,513 -20 -299 83 -1,592 2,975 1,367 EC, PC, POP 1984 692 183 -739 -171 -1,209 1,260 1,369 FM, PC, POP 1985 4,270 8 -207 138 630 2,320 1,380 FM, EC, EI, PC, POP 1986 1,994 63 -792 61 -675 1,808 1,529 FM, EC, PC, POP 1987 4,991 127 -500 186 2,078 1,524 1,576 FM, EC, EI, PC, POP 1988 4,303 201 -585 -176 -2,540 5,754 1,650 FM, PC, POP 1989 4,315 -58 -647 -35 -133 3,479 1,709 PC, POP 1990 736 34 -293 -25 -3,819 3,108 1,730 FM, PC, POP 1991 4,543 251 -299 59 3,736 -953 1,751 FM, EC, EI, POP 1992 3,988 484 -616 140 -775 3,031 1,724 FM, EC, PC, POP 1993 1,530 744 -1,446 119 -2,407 2,759 1,761 FM, EC, PC, POP 1994 4,089 621 -1,199 -204 -2,176 5,282 1,766 FM, PC, POP 1995 11,688 312 -726 303 3,999 5,911 1,888 FM, EC, EI, PC, POP 1996 -23,543 849 110 287 -31,885 5,341 1,753 EI 1997 1,827 144 -67 -189 -1,971 2,356 1,555 FM, PC, POP 1998 3,450 -808 810 -151 -1,739 3,759 1,579 MM, PC, POP 1999 5,126 419 -200 -17 -1,698 4,988 1,635 FM, PC, POP 2000 1,832 -528 65 158 -1,808 2,274 1,672 MM, EC, PC, POP 2001 -614 -571 374 -36 -5,471 3,470 1,620 FM, EC, EI 2002 3,187 -111 140 -127 -429 2,137 1,578 MM, PC, POP 2003 1,622 133 -108 -143 -6,637 6,827 1,551 FM, PC, POP 2004 4,456 121 -217 363 -4,366 7,026 1,529 FM, EC, PC, POP 2005 -634 -656 841 14 -10,447 8,127 1,488 FM, EI 61 Appendix (Cont'd) CO2 Emissions Change and Contributing Factors Indonesia CO2 Emissions Factors Influencing the CO2 Emissions Change Year Change Fuel Modal Emission Transport Per Capita Influencing Factors (`000 tCO2) Population Mix Mix Coefficient Energy Intensity GDP 1981 1,786 11 -6 0 314 1,091 376 FM, EI, PC, POP 1982 1,172 44 -3 0 909 -167 389 FM, EI, POP 1983 -680 10 -1 0 -2,348 1,278 382 MM, EI 1984 355 -4 -5 0 -1,043 1,033 374 PC, POP 1985 633 6 16 0 -99 330 381 FM, MM, PC, POP 1986 1,248 -18 -15 0 24 864 393 EI, PC, POP 1987 2,092 1 -1 0 884 793 415 FM, EI, PC, POP 1988 2,049 10 -2 0 471 1,121 448 FM, EI, PC, POP 1989 2,318 -35 45 0 -83 1,904 486 MM, PC, POP 1990 3,371 17 9 0 720 2,078 546 FM, MM, EI, PC, POP 1991 3,653 12 8 0 731 2,318 583 FM, MM, EI, PC, POP 1992 2,063 -9 -5 0 -494 1,965 605 PC, POP 1993 2,575 17 9 0 -195 2,130 614 FM, MM, PC, POP 1994 4,865 1 1 0 1,746 2,478 639 FM, MM, EI, PC, POP 1995 3,903 -4 6 0 89 3,142 669 MM, EI, PC, POP 1996 5,341 22 5 0 1,493 3,096 724 FM, MM, EI, PC, POP 1997 6,055 -6 3 0 3,414 1,854 790 MM, EI, PC, POP 1998 -1,451 -51 14 0 7,010 -9,237 813 FM, PC 1999 -371 -4 -11 0 -821 -323 788 FM, MM, EI, PC 2000 5,122 10 -8 0 2,174 2,138 808 FM, EI, PC, POP 2001 3,436 19 11 0 939 1,600 866 FM, MM, EI, PC, POP 2002 892 -61 -5 0 -1,947 1,999 906 PC, POP 2003 1,064 -111 -15 0 -1,980 2,251 919 PC, POP 2004 6,448 -129 -26 0 3,032 2,591 979 EI, PC, POP 2005 -1,161 -108 -3 0 -5,144 3,074 1,019 FM, MM, EI 62 Appendix (Cont'd) CO2 Emissions Change and Contributing Factors Korea CO2 Emissions Factors Influencing the CO2 Emissions Change Year Change Fuel Modal Emission Transport Per Capita Influencing Factors (`000 tCO2) Population Mix Mix Coefficient Energy Intensity GDP 1981 -3,354 7 -27 -2 -4,106 571 202 MM, EC, EI 1982 1,472 14 -47 -12 662 666 189 FM, EI, PC, POP 1983 3,535 33 -85 -14 2,114 1,273 213 FM, EI, PC, POP 1984 1,738 4 -13 55 347 1,130 215 FM, EC, EI, PC, POP 1985 1,519 38 -15 -39 291 1,057 186 FM, EI, PC, POP 1986 3,202 37 -30 -49 1,101 1,936 207 FM, EI, PC, POP 1987 4,424 -10 57 -90 1,837 2,383 247 MM, EI, PC, POP 1988 3,951 -11 -46 43 1,011 2,668 286 EC, EI, PC, POP 1989 4,279 -8 -55 -36 2,202 1,844 331 EI, PC, POP 1990 8,726 1,245 -1,046 39 5,134 2,976 377 FM, EC, EI, PC, POP 1991 5,351 43 -9 40 1,069 3,740 468 FM, EC, EI, PC, POP 1992 5,576 -6 25 26 2,548 2,443 540 MM, EC, EI, PC, POP 1993 6,973 -19 8 -37 3,534 2,888 600 MM, EI, PC, POP 1994 7,395 3 -62 -14 2,081 4,736 651 FM, EI, PC, POP 1995 9,060 57 -43 -14 2,569 5,749 742 FM, EI, PC, POP 1996 7,085 -24 22 -8 1,548 4,751 796 MM, EI, PC, POP 1997 1,057 -16 79 44 -2,963 3,123 790 MM, EC, PC, POP 1998 -11,666 -340 213 -92 -5,718 -6,323 595 FM, EC, EI, PC 1999 6,961 -83 -17 -21 -30 6,554 557 PC, POP 2000 6,479 27 -128 32 -388 6,227 709 FM, EC, PC, POP 2001 2,495 -4 -110 38 -807 2,712 666 EC, PC, POP 2002 5,538 34 -77 -147 -584 5,799 513 FM, PC, POP 2003 2,253 67 -12 49 -830 2,488 491 FM, EC, PC, POP 2004 -151 0 -117 24 -4,619 4,191 371 MM, EI 2005 -11,833 -295 -656 -68 -14,360 3,357 190 FM, MM, EC, EI 63 Appendix (Cont'd) CO2 Emissions Change and Contributing Factors Malaysia CO2 Emissions Factors Influencing the CO2 Emissions Change Year Change Fuel Modal Emission Transport Per Capita Influencing Factors (`000 tCO2) Population Mix Mix Coefficient Energy Intensity GDP 1981 483 -2 0 0 33 283 169 EI, PC, POP 1982 439 -3 0 0 26 234 182 EI, PC, POP 1983 1,172 4 0 0 683 278 207 FM, EI, PC, POP 1984 272 -21 0 0 -359 421 232 PC, POP 1985 471 -12 0 0 586 -356 253 EI, POP 1986 635 7 0 0 517 -163 273 FM, EI, POP 1987 602 3 0 0 59 238 301 FM, EI, PC, POP 1988 994 9 0 0 -64 727 322 FM, PC, POP 1989 1,133 32 0 0 53 699 348 FM, EI, PC, POP 1990 1,737 4 0 0 564 790 380 FM, EI, PC, POP 1991 1,090 -2 0 0 -274 958 408 PC, POP 1992 1,059 7 0 0 -315 949 418 FM, PC, POP 1993 313 -41 0 0 -1,225 1,155 424 PC, POP 1994 1,742 -23 0 0 202 1,116 447 EI, PC, POP 1995 1,121 -19 0 0 -660 1,319 482 PC, POP 1996 2,803 -7 0 0 795 1,468 547 EI, PC, POP 1997 3,485 77 2 0 1,693 1,102 612 FM, MM, EI, PC, POP 1998 -1,873 -150 0 1 188 -2,529 616 FM, PC 1999 5,406 68 8 0 3,734 952 643 FM, MM, EI, PC, POP 2000 1,690 150 -1 -1 -1,036 1,891 686 FM, PC, POP 2001 2,620 18 -4 1 2,502 -603 706 FM, EC, EI, POP 2002 810 8 3 2 -597 709 686 FM, MM, EC, PC, POP 2003 2,246 7 2 -3 260 1,294 687 FM, MM, EI, PC, POP 2004 2,662 11 0 0 -22 1,968 705 FM, PC, POP 2005 -51 -73 -1 2 -1,970 1,275 716 FM, MM, EI 64 Appendix (Cont'd) CO2 Emissions Change and Contributing Factors Mongolia CO2 Emissions Factors Influencing the CO2 Emissions Change Year Change Fuel Modal Emission Transport Per Capita Influencing Factors (`000 tCO2) Population Mix Mix Coefficient Energy Intensity GDP 1986 54 -2 0 0 -59 88 27 PC, POP 1987 28 -4 0 -1 -13 18 27 PC, POP 1988 766 392 -119 0 423 42 28 FM, EI, PC, POP 1989 -604 -107 -27 -1 -541 38 34 FM, MM, EC, EI 1990 47 4 1 1 -191 202 29 FM, MM, EC, PC, POP 1991 -556 -8 0 -1 -205 -360 18 FM, EC, EI, PC 1992 -101 4 -2 -1 -9 -111 17 MM, EC, EI, PC 1993 97 23 16 2 85 -41 13 FM, MM, EC, EI, POP 1994 -317 -31 -17 -1 -287 7 11 FM, MM, EC, EI 1995 157 25 2 0 -46 164 13 FM, MM, PC, POP 1996 46 6 3 0 18 12 7 FM, MM, EI, PC, POP 1997 -67 -10 0 0 -91 22 11 FM, EC, EI 1998 -21 -40 4 0 -12 17 10 FM, EI 1999 36 24 -1 0 -12 18 7 FM, PC, POP 2000 153 2 -1 0 142 2 8 FM, EI, PC, POP 2001 -44 -9 0 0 -46 2 8 FM, EI 2002 112 9 1 0 63 26 12 FM, MM, EI, PC, POP 2003 -43 -39 5 0 -62 42 12 FM, EI 2004 162 34 -4 0 23 91 17 FM, EI, PC, POP 2005 17 12 4 0 -70 57 14 FM, MM, PC, POP 65 Appendix (Cont'd) CO2 Emissions Change and Contributing Factors Pakistan CO2 Emissions Factors Influencing the CO2 Emissions Change Year Change Fuel Modal Emission Transport Per Capita Influencing Factors (`000 tCO2) Population Mix Mix Coefficient Energy Intensity GDP 1981 714 2 1 1 167 341 201 FM, MM, EC, EI, PC, POP 1982 963 -4 2 0 462 288 215 MM, EI, PC, POP 1983 47 17 4 1 -529 327 228 FM, MM, EC, PC, POP 1984 863 -16 -3 -1 442 200 240 EI, PC, POP 1985 531 3 0 1 -176 445 258 FM, EC, PC, POP 1986 523 -1 -5 -1 -14 273 270 PC, POP 1987 1,652 17 5 3 927 405 295 FM, MM, EC, EI, PC, POP 1988 883 -2 -6 0 -27 588 329 PC, POP 1989 531 -3 -1 2 -106 302 337 EC, PC, POP 1990 335 40 2 0 -301 248 346 FM, MM, PC, POP 1991 508 3 -1 0 -188 338 356 FM, PC, POP 1992 1,812 5 -5 -1 685 747 381 FM, EI, PC, POP 1993 2,303 7 -3 0 1,998 -131 432 FM, EI, POP 1994 1,053 10 0 0 348 220 476 FM, EI, PC, POP 1995 766 10 -1 0 -205 472 489 FM, PC, POP 1996 2,082 -28 -1 1 1,104 486 520 EC, EI, PC, POP 1997 545 -1 0 0 318 -315 543 EI, POP 1998 1,075 22 0 -1 464 28 562 FM, EI, PC, POP 1999 1,373 7 0 1 480 292 594 FM, EC, EI, PC, POP 2000 -320 26 1 0 -1,396 443 607 EI 2001 -211 -1 0 0 -668 -141 600 FM, EI, PC 2002 89 -2 0 0 -695 188 598 PC, POP 2003 1,062 -20 0 -1 -144 614 613 PC, POP 2004 1,633 -1 0 0 -23 1,011 644 MM, PC, POP 2005 -2,829 3 0 0 -4,790 1,329 630 EI 66 Appendix (Cont'd) CO2 Emissions Change and Contributing Factors The Philippines CO2 Emissions Factors Influencing the CO2 Emissions Change Year Change Fuel Modal Emission Transport Per Capita Influencing Factors (`000 tCO2) Population Mix Mix Coefficient Energy Intensity GDP 1981 -588 3 4 0 -760 42 124 EI 1982 321 4 6 0 142 52 118 FM, MM, EI, PC, POP 1983 -86 0 0 0 -178 -27 118 EI, PC 1984 -304 -2 -2 0 59 -473 114 FM, MM, PC 1985 -196 -2 -2 0 148 -447 108 FM, MM, PC 1986 290 -1 -2 0 140 43 109 EI, PC, POP 1987 541 0 1 0 331 89 119 MM, EI, PC, POP 1988 446 0 0 0 91 225 130 EI, PC, POP 1989 721 1 1 0 358 218 143 FM, MM, EI, PC, POP 1990 340 2 2 0 140 41 155 FM, MM, EI, PC, POP 1991 -591 7 8 0 -569 -186 149 EI, PC 1992 950 4 5 0 919 -129 152 FM, MM, EI, POP 1993 969 4 3 0 803 -14 172 FM, MM, EI, POP 1994 8,646 960 121 0 7,106 219 240 FM, MM, EI, PC, POP 1995 1,681 0 9 0 871 416 385 MM, EI, PC, POP 1996 2,265 -7 3 0 1,162 688 419 MM, EI, PC, POP 1997 2,246 -1 9 2 1,137 638 460 MM, EC, EI, PC, POP 1998 -53 -12 4 1 86 -609 477 FM, PC 1999 1,282 -13 -14 -3 529 294 489 EI, PC, POP 2000 -276 30 -15 0 -1,679 886 502 MM, EI 2001 979 17 -13 3 549 -88 511 FM, EC, EI, POP 2002 -348 -31 -12 0 -1,356 541 509 FM, MM, EI 2003 691 1 7 -1 -510 675 519 FM, MM, PC, POP 2004 557 -43 54 0 -970 993 524 MM, PC, POP 2005 -1,447 7 -17 3 -2,649 703 506 MM, EI 67 Appendix (Cont'd) CO2 Emissions Change and Contributing Factors Sri Lanka CO2 Emissions Factors Influencing the CO2 Emissions Change Year Change Fuel Modal Emission Transport Per Capita Influencing Factors (`000 tCO2) Population Mix Mix Coefficient Energy Intensity GDP 1981 -3 -1 -1 0 -118 79 36 FM, MM, EI 1982 118 1 1 0 30 60 27 FM, MM, EI, PC, POP 1983 -112 0 2 0 -214 70 31 EI 1984 249 1 -2 0 140 84 26 FM, EI, PC, POP 1985 102 1 0 0 -15 80 36 FM, PC, POP 1986 12 -1 0 0 -91 60 44 PC, POP 1987 60 -1 0 0 19 6 37 EI, PC, POP 1988 24 -3 1 0 -36 26 35 MM, PC, POP 1989 -129 -3 2 0 -184 22 34 FM, EI 1990 93 1 -1 0 -59 124 27 FM, PC, POP 1991 18 4 -1 0 -98 76 36 FM, PC, POP 1992 285 2 0 0 168 90 24 FM, EI, PC, POP 1993 195 1 1 0 0 157 36 FM, MM, PC, POP 1994 342 1 -1 0 171 129 43 FM, EI, PC, POP 1995 192 1 -1 0 8 136 48 FM, EI, PC, POP 1996 1,349 17 10 0 1,168 109 45 FM, MM, EI, PC, POP 1997 -418 -4 -2 0 -701 231 58 FM, MM, EI 1998 215 -41 41 0 7 160 49 MM, EI, PC, POP 1999 484 2 -1 0 276 137 70 FM, EI, PC, POP 2000 251 0 -1 0 -56 220 88 PC, POP 2001 -193 -8 4 0 -107 93 -175 FM, EI, POP 2002 78 -6 2 0 -123 126 78 MM, PC, POP 2003 364 -7 -1 0 52 251 69 EI, PC, POP 2004 386 -20 0 0 102 242 63 EI, PC, POP 2005 -524 39 -53 0 -807 235 62 MM, EI 68 Appendix (Cont'd) CO2 Emissions Change and Contributing Factors Thailand CO2 Emissions Factors Influencing the CO2 Emissions Change Year Change Fuel Modal Emission Transport Per Capita Influencing Factors (`000 tCO2) Population Mix Mix Coefficient Energy Intensity GDP 1981 472 30 0 0 -123 388 177 FM, PC, POP 1982 -339 385 -434 0 -793 336 166 MM, EI 1983 1,919 25 1 0 1,312 406 175 FM, MM, EI, PC, POP 1984 2,027 30 2 0 1,287 509 199 FM, MM, EI, PC, POP 1985 757 -6 8 0 116 430 209 MM, EI, PC, POP 1986 803 13 8 0 -18 588 212 FM, MM, PC, POP 1987 2,398 38 -11 0 880 1,268 224 FM, EI, PC, POP 1988 2,786 68 -1 0 365 2,107 248 FM, EI, PC, POP 1989 3,860 33 0 0 1,263 2,280 284 FM, EI, PC, POP 1990 3,096 11 -2 0 353 2,418 316 FM, EI, PC, POP 1991 1,014 -35 -1 0 -1,241 1,954 337 PC, POP 1992 1,667 -24 -7 0 -572 1,927 344 PC, POP 1993 5,310 53 0 0 2,667 2,215 374 FM, EI, PC, POP 1994 4,190 -4 -1 0 973 2,799 424 EI, PC, POP 1995 7,290 78 -4 0 3,407 3,327 482 FM, EI, PC, POP 1996 4,305 11 0 0 1,488 2,248 558 FM, EI, PC, POP 1997 -306 -80 -1 0 480 -1,271 567 FM, MM, PC 1998 -6,893 -126 0 0 -1,513 -5,774 520 FM, EI, PC 1999 1,344 55 -2 0 -652 1,487 455 FM, PC, POP 2000 -1,094 -14 0 0 -3,156 1,647 430 FM, EI 2001 1,179 -34 0 0 253 578 382 EI, PC, POP 2002 2,541 1 2 -1 122 2,045 372 FM, MM, EI, PC, POP 2003 3,293 44 -3 0 -118 3,009 360 FM, PC, POP 2004 4,047 69 -3 0 793 2,812 376 FM, EI, PC, POP 2005 1,194 42 0 0 -1,288 2,057 383 FM, PC, POP 69 Appendix (Cont'd) CO2 Emissions Change and Contributing Factors Vietnam CO2 Emissions Factors Influencing the CO2 Emissions Change Year Change Fuel Modal Emission Transport Per Capita Influencing Factors (`000 tCO2) Population Mix Mix Coefficient Energy Intensity GDP 1981 -357 4 -2 0 -460 67 34 MM, EI 1982 77 -2 12 0 -64 101 29 MM, PC, POP 1983 196 1 8 0 64 92 31 FM, MM, EI, PC, POP 1984 -50 1 2 0 -204 117 34 EI 1985 1,090 77 -34 0 964 38 45 FM, EI, PC, POP 1986 339 4 -14 0 264 13 72 FM, EI, PC, POP 1987 526 0 7 0 395 37 87 MM, EI, PC, POP 1988 287 16 -20 0 95 101 95 FM, EI, PC, POP 1989 -352 29 -34 -18 -606 185 92 MM, EC, EI 1990 547 -5 -11 -8 371 111 87 EI, PC, POP 1991 -223 2 14 -9 -472 153 88 EC, EI 1992 489 -10 -4 -5 153 269 87 EI, PC, POP 1993 2,518 34 5 -4 2,044 329 110 FM, MM, EI, PC, POP 1994 726 -11 -3 3 108 489 139 EC, EI, PC, POP 1995 -263 11 2 0 -974 559 137 EI 1996 735 13 -2 3 17 563 141 FM, EC, EI, PC, POP 1997 751 14 -2 11 51 543 134 FM, EC, EI, PC, POP 1998 536 -18 10 8 16 390 129 MM, EC, EI, PC, POP 1999 704 -11 9 -12 257 333 128 MM, EI, PC, POP 2000 880 -13 12 5 174 685 17 MM, EC, EI, PC, POP 2001 791 -12 8 -7 34 613 155 MM, EI, PC, POP 2002 2,608 -25 -13 9 1,735 729 173 EC, EI, PC, POP 2003 2,406 36 163 -37 1,135 881 228 FM, MM, EI, PC, POP 2004 3,985 4 -25 55 2,555 1,135 262 FM, EC, EI, PC, POP 2005 -284 -2 -198 -2 -1,757 1,404 271 FM, MM, EC, EI 70