**3. Results and Discussion**

The results and discussion are divided into the three sections of overall trends in emissions, changes within sectors and age groups, and the decomposition of the driving forces underpinning emissions.

### *3.1. Overall Trends of Total Direct and Indirect CO2 Emissions*

CO2 emissions induced by Japanese household consumption from 1990 to 2005 are shown in Table 1. Both direct and indirect CO2 emissions showed a significant growth trend during the analyzed period. In terms of overall emissions, indirect CO2 emissions levels remain nearly four times those of direct CO2 emissions. The annual average increase of direct CO2 emissions is relatively small, and emissions in 2005 are slightly lower than those in the year 2000. However, indirect CO2 emissions continuously increased, with an annual average increase of 6.6 Mt in 1990–2005. In terms of the growth rate of emissions, that of direct emissions was higher than that of indirect emissions during the studied period (see Table 1). These characteristics will be elaborated by observing the trends in direct and indirect CO2 emissions by sector, as presented in the following section.

**Table 1.** Compositions of Japanese household carbon footprints (CF) in 1990–2005


### *3.2. Changes in Direct and Indirect CO2 Emissions in Di*ff*erent Sectors and Age Groups*

The total direct and indirect CO2 emissions are disaggregated into sectors and household age groups to evaluate their impact on CO2 emissions. Fourteen sectors are considered: "food and non-alcoholic beverages," "alcoholic beverages and tobacco," "clothing and footwear," "housing," "furnishings," "medicals," "private vehicles," "public transport," "information and communication," "recreation and culture," "education," "restaurants and hotels," "consumable goods," and "margins, religions and other services." These sectors are determined based on the "Classification of Individual Consumption by Purpose."

Figure 1 presents the trends in total direct CO2 emissions by sector and average direct CO2 emissions per capita across age groups. To observe the impact of demographic factors on carbon emissions in an aging society in greater detail, we examined the per capita CO2 emissions as well as the per household emissions across di fferent age groups. Here, we focus on the per capita emissions (Figure 1b), because the per household emissions are a ffected by the average family size and composition of households attributed to each age group. The results of per household direct and indirect emissions are detailed in Appendix A.

Private vehicles and housing are the only two sectors that generate direct CO2 emissions from households, as fossil fuels are used for these activities. For instance, the private vehicles sector includes gasoline and light oil, while the housing sector includes kerosene, LPG, coal products, and city gas. As shown in Figure 1a, the private vehicles sector accounts for a large proportion of direct CO2 emissions throughout the analyzed period, impacting the high growth rate of direct emissions, as referred to above. From 1990 to 2005, Japanese car ownership rose from 57.99 million to 78.28 million vehicles—an increase of approximately 35% [42]. With this increase in car ownership, the growth rate of gasoline consumption was much larger than for other fuels. Consequently, the direct emissions for the private vehicles sector increased significantly compared to those for the housing sector.

Considering di fferent age groups, direct CO2 emissions (per capita) from the private vehicles sector are concentrated within two age groups—40s and 50s—as shown in Figure 1b. For those in their 40s, this may be due to work and family needs. With the increase of household savings in these age

groups, many families tend to own their own cars and use them frequently. The 50s age group has the highest direct CO2 emissions across all age groups, possibly because the annual income in the 50s is higher than that of other age groups [40]. In addition, with the increase in family members, the household size has also expanded to a certain extent when compared to others, potentially expanding the demand for private vehicles, particularly large-sized cars. As for the direct CO2 emissions from the housing sector, these are also concentrated in the 40s and 50s age groups. This may be because more people in their 40s and 50s are married and living with their children, and they tend to live in relatively large, energy-consuming houses. Furthermore, the changes in direct CO2 emissions in different age groups also showed certain peculiarities, such as CO2 emissions gradually increasing for householders in their 20s, reaching their peak in the 50s and then subsequently declining.

**Figure 1.** Sectoral composition of direct CO2 emissions from 1990–2005. (**a**) Total (Mt-CO2/y) and (**b**) per capita by age group (t-CO2/y). Inverted triangles denote the noteworthy household age groups as detailed in the main text.

The change in indirect CO2 emissions in different age groups is similar to that of direct CO2 emissions (see Figures 1b and 2b). Considering the order of growth in sectoral indirect emissions, we selected food and non-alcoholic beverages, housing, and public transport sectors for discussion here, as shown in Figure 2a. First, the indirect CO2 emissions generated by the housing sector were mainly concentrated in the three age groups of the 40s, 50s, and 60s, as demonstrated by Figure 2b. The reason behind the similarity between direct and indirect CO2 emissions of those in their 40s and 50s is that residents modify their houses to meet the needs of family life and child-rearing. For the 60s age group, a consideration of the living environment and living conditions for old-age life could increase the cost of housing and lead to the production of more indirect CO2 emissions.

**Figure 2.** Sectoral composition of indirect CO2 emissions from 1990–2005. (**a**) Total (Mt-CO2/y) and (**b**) per capita by age group (t-CO2/y). Inverted triangles denote the noteworthy household age groups as detailed in the main text.

Indirect CO2 emissions from the food and non-alcoholic beverages sector were also concentrated in the 40s, 50s and 60s age groups. In the 40s and 50s age groups, the expansion of household size may lead to an increase in food consumption, which could increase indirect CO2 emissions. As for the 60s age group, elderly people tend to spend more money on high-quality, expensive food compared to the other age groups. In addition, although the household size is smaller than the 40s and 50s age groups, it is still larger than the age groups of the 20s, 30s, and 70s. These factors combined make the 60s age group the third-largest indirect CO2 emitter on a per capita basis.

Indirect CO2 emissions from the public transport sector were concentrated in the 20s and 50s age groups. Compared with other age groups, the proportion of private vehicle possession in the 20s age group is relatively low, leading to an increased use of public transport, which could increase indirect CO2 emissions from this sector. People in the households with in the 50s age group produced the highest CO2 emissions from both private vehicles and public transport sectors. Generally, people in the 50s age group have the highest annual income, and with the expansion of household size, families are more likely to use private vehicles alongside the children of these families using public transport for attending school.

While the overall CO2 emissions of the public transport sector are increasing, the growth rate is not as fast as that of the private vehicles sector. In addition, there are different trends in indirect emissions from the private vehicles and public transport sectors by age group. In 2005, indirect emissions for the private vehicles sector for those in their 20s decreased by 21%, while those in other age groups, particularly the 30s and 60s, increased by 13–35% compared to 1990. The reasoning for such a decline in the 20s age group is that they were more likely to purchase smaller, less expensive vehicles such as lightweight automobiles (known as kei-cars in Japan) due to financial aspects. On the other hand, because age groups from the 30s to 50s are more likely to be involved in child-raising activities, family life has led to an increased demand for private transport. Furthermore, due to the demographic shift related to an aging, shrinking population leading to a postponement of childbearing age, those in their 60s also have experienced an increase in their use of private vehicles.

For public transport, growth trends were evident across all age groups. Among them, the growth is particularly significant for those in their 20s and 40s. In the 20s age group, the reduction in the use of private vehicles led to an increase in public transport demand. The use of public transport in the 40s

age group also increased, which is likely due to the expansion of travel needs for work and school as well as an increase in the use of private vehicles, although at a lower rate than for the 30s, 50s, and 60s.

On average, annual income increases from the 20s to the 50s, subsequently declining and leading to an increase in consumption and savings, creating upward pressure on CO2 emissions [40]. Meanwhile, because of the decline in income after the 50s, indirect CO2 emissions also tend to decline. Further, the household size expands from the 20s to the 50s, while post 50s, the household begins to shrink as children become independent and establish their own households, which is likely linked to the peak in CO2 emissions observed for the 50s age group.
