*3.3. Decomposition Results*

### 3.3.1. Driving Forces of Total Direct and Indirect CO2 Emissions

Figure 3 shows the results of examining the factors affecting direct CO2 emissions during 1990–2005 using the IDA.

**Figure 3.** Driving forces of direct household CO2 emissions in Japan during 1990-2005. Δintensity is the intensity effect, Δpattern is the consumption pattern effect, Δvolume is the consumption volume effect, Δsize is the household size effect, Δdistribution is the household distribution effect, and Δhousehold is the household number effect.

Among factors, the intensity effect was the main driver in reducing direct CO2 emissions, while the consumption pattern was the main driver which increased emissions. The negative impact of the intensity effect on direct CO2 emissions has progressed over time, indicating that Japan has made substantial progress in the carbon reduction technology used in daily life since 1990.

Figure 4 describes the direct CO2 emissions by household energy item. The consumption of gasoline increased significantly prior to 2000, which is one of the reasons for the growth of direct CO2 emissions driven by the consumption pattern during this period. After 2000, along with the slowdown of the growth of gasoline consumption, the positive impact of the consumption pattern has also weakened. In addition, the total number of households in Japan increased from 40.67 million to 49.06 million between 1990 and 2005 [11], promoting a positive impact of the number of households on direct CO2 emissions. The positive effect of consumption volume on direct CO2 emissions weakened consistently between 1990 and 2005. This is probably related to the increase of small-scale households which could reduce their energy consumption to a certain extent. Moreover, the negative impact of household size and household distribution on direct CO2 emissions is gradually increasing, which is likely due to the influence of recent demographic trends such as an increase in one-person households and a reduction in household size because of an aging society with fewer children.

**Figure 4.** Trend in the direct CO2 emissions by energy item.

Considering the effects of household size, the number of households and household distribution in 1995, 2000 and 2005, we find that the sum of these effects gradually decreased direct CO2 emissions. This suggests that if Japan maintains an aging society with a low birth rate, direct CO2 emissions generated by the household sector will gradually decrease.

As for indirect CO2 emissions, both the total emissions and change in emission levels were higher than for direct CO2 emissions, and the sectoral composition of indirect CO2 emissions is more diverse. Further, the change in driving forces of indirect CO2 emissions are relatively complex. Figure 5 shows the SDA result for indirect CO2 emissions in 1990–2005. Indirect CO2 emissions grew rapidly from 1990 to 1995 with growth slowing down during 1995–2000 and slightly accelerating thereafter. This may be due to the post-bubble economy in which Japan adopted governmen<sup>t</sup> intervention policies to stimulate the recovery of the economy and increase residential consumption, resulting in an increase in indirect CO2 emissions between 1990 and 1995. With the change in policy direction from economic stimulus to economic constraint (including raising the consumption tax and increasing medical expenses in 1997) [43,44], the growth rate of indirect CO2 emissions slowed down in 1995–2000. After 2000, due to the effect of the internet bubble in the United States [45], Japan was forced to introduce looser monetary policies to stimulate economic development which increased its indirect emissions to some extent. With the increase in the number of households, the positive impact of the number of households on indirect CO2 emissions has gradually increased, becoming one of the main factors promoting indirect CO2 emissions post-2000. The positive impact of the supply chain structure on indirect CO2 emissions increased slightly before 2000, shifting to a negative effect after 2000. This may be caused by the transformation and maturing of enterprise, eliminating excess employment, equipment and debt through a severe restructuring process from the late 1990s to the early 2000s. From the 1990s, along with an emphasis on environmental protection and energy saving (e.g., the Kyoto Protocol adopted in 1997), the impact of the supply chain structure toward CO2 emissions became negative, which reflects the grea<sup>t</sup> development of low-carbon technology in the whole supply chain. The negative impact of the intensity effect on indirect CO2 emissions increased from 1990 to 2000 and weakened thereafter. After the bubble economy, economic recovery may be an important reason for the change in intensity effect. The impact of the consumption pattern on indirect CO2 emissions changed from negative to positive from 1995 to 2000, turning to negative once more after 2000, although this impact was relatively small. Because of the economic stimulus policy post-bubble economy, the choices of consumers tended to be toward high-quality and environment-friendly consumption, causing the consumption pattern to inhibit indirect CO2 emissions. However, during the period of economic constraint policies, consumers tended to choose goods with a high performance and low price, reducing the environmental awareness of consumption, resulting in a positive consumption pattern impact.

**Figure 5.** Driving forces of indirect CO2 emissions during 1990–2005. Δintensity is the intensity effect, <sup>Δ</sup>supply chain is the supply chain structure effect, Δpattern is the consumption pattern effect, Δvolume is the consumption volume effect, Δsize is the household size effect, Δdistribution is the household distribution effect, and Δhousehold is the household number effect.

Summarizing the effect of household size, the number of households and household distribution in 1995, 2000 and 2005 and observing the changes, it was identified that the effect is gradually changing toward the positive and increasing. This shows that if Japan maintains an aging society with a low birth rate, the indirect CO2 emissions generated by households will continue to grow.

### 3.3.2. Driving Forces of Indirect CO2 Emissions of Key Sectors

By observing the changes in indirect CO2 emissions in different sectors, this study identified four sectors with significant growth (amount) in indirect CO2 emissions from 1990–2005. These are the housing, medical, private vehicle and public transport sectors. We emphasize discussion about the decomposition results for these four sectors, as shown in Figure 6, because we consider it essential with regard to the relationship between an aging, shrinking population and increasing household CO2; those for the other 10 sectors and additional insights for the two most remarkable sectors showing the highest and lowest growth rate (i.e., the information and communication and clothes and footwear sectors, respectively) are detailed in Appendix A.

**Figure 6.** *Cont*.

**Figure 6.** Structural decomposition analysis (SDA) results for indirect CO2 emissions between 1990 and 2005 for four selected sectors. Δintensity is the intensity effect, <sup>Δ</sup>supply chain is the supply chain structure effect, Δpattern is the consumption pattern effect, Δvolume is the consumption volume effect, Δsize is the household size effect, Δdistribution is the household distribution effect, and Δhousehold is the household number effect. (**a**) Housing sector, (**b**) public transport, (**c**) private vehicles, and (**d**) medical sector.

First, the housing sector (Figure 6a), which produced the largest indirect CO2 emissions at any time between 1990 and 2005, increased consistently between 1990 and 2005. The technology and size effects are the main drivers which reduce indirect CO2 emissions. The rapid development of energy-related technologies has greatly reduced the indirect CO2 emissions from housing. At the same time, with the growth of single-member families, the proportion of small-scale households has gradually increased [46]. Compared with average-sized households, small-scale households utilize a lower number of consumables and appliances (e.g., air-conditioners, etc.), leading to a lower level of indirect CO2 emissions. The number of households, consumption volume, and consumption pattern were the main drivers of CO2 emissions growth. From 1990 to 2005, the number of households in Japan expanded. At the same time, the increase in single-member households has increased the demand for housing, further expanding consumption volumes. Therefore, the positive impact of the number of households and consumption volume on indirect CO2 emissions increased. As for the consumption pattern, residents seeking a better quality of life tend to invest in quality of life outcomes, leading to an increase in indirect CO2 emissions.

The public transport sector is the second-largest indirect CO2 emitter among the four sectors. As shown in Figure 6b, changes in consumption volume and the number of households were the main drivers which promoted the growth of indirect CO2 emissions in this sector. The number of households in Japan continued to grow from 1990 to 2005, increasing the positive effect, while, due to economic situation changes, residents might choose public transport in order to reduce living costs. An increase in environmental awareness has also prompted people to use more public transport. In addition, indirect CO2 emissions have been restricted to a large extent by the household size and household distribution effects. The reason for this may be that small-scale households may prefer to travel by lower-cost public transport when compared to private transport, with the same trend shown by the elderly. Moreover, the development of low-carbon technologies has also greatly reduced the CO2 emissions associated with the public transport sector.

The private vehicle sector is one of the main sources of direct CO2 emissions from households (see Figure 1), while this sector also has a strong impact on indirect CO2 emissions. With the increasing number of households in Japan [46], the demand for private cars is also expanding [42], causing the number of households and consumption volume effects to become important drivers in promoting indirect CO2 emissions in this sector, as shown in Figure 6c. The supply chain effect has increased indirect CO2 emissions, which is likely due to economic globalization, causing the production and manufacture of automobiles to be regionally diversified, whereby technological di fferences in production and transport between regions may place upward pressure on indirect CO2 emissions. The consumption pattern has become the main driver restricting indirect CO2 emissions, perhaps because people have become more likely to use public transport due to the abovementioned reasons, although many households still have a need for a private vehicle. With increasing awareness of environmental protection, households are willing to consider the purchase of environmentally friendly automobile models. Meanwhile, with the development of automobile manufacturing technology, consumers are more willing to buy fuel-e fficient vehicles. Household size and household distribution effects had a stable, inhibitory e ffect on indirect CO2 emissions, mainly because an aging society and the increase in households with less members reduced the consumption of private vehicles.

Although the medical sector has the lowest indirect CO2 emissions among the four sectors analyzed, it has the highest growth rate, at 73% (Figure 6d). Considering the current situation of Japan's aging society, the medical sector has a grea<sup>t</sup> impact on the lives of elderly people. Therefore, it is of grea<sup>t</sup> practical significance to analyze the influencing factors of indirect CO2 emissions in the medical sector. Consumption patterns and consumption volume have strongly contributed to the increase in indirect CO2 emissions as elderly households increase. In the context of an aging society, Japan's elderly population continues to grow, leading to the continued expansion of national medical expenditure. Furthermore, the proportion of imported drugs has been gradually increasing [42], which is also the main reason for the decreasing impact of supply chain structure on indirect CO2 emissions. Though the impact of technology on indirect CO2 emissions changed during 1990–2005, it has always been negative. This is probably due to the continuous development of medical manufacturing technologies which are more environmentally conscious. Finally, the household size also plays an important role in inhibiting indirect CO2 emissions. The decrease in household size (i.e., a relative increase in the share of small households in the total households) has greatly reduced household consumption of medicine and consequently reduced indirect CO2 emissions to a certain extent.
