Changes in the Carbon Footprint of Japanese ... - ACS Publications

May 6, 2014 - consumption-based GHG emissions derived from Japanese household consumption in. 2035 are estimated to be 1061 Mt-CO2eq (4.2% lower ...
6 downloads 0 Views 2MB Size
Download PDF
Policy Analysis pubs.acs.org/est

Changes in the Carbon Footprint of Japanese Households in an Aging Society Yosuke Shigetomi,*,†,‡,§ Keisuke Nansai,‡ Shigemi Kagawa,∥ and Susumu Tohno† †

Graduate School of Energy Science, Kyoto University, Yoshida-Honmachi, Sakyo-ku, Kyoto 606-8501, Japan Center for Material Cycles and Waste Management Research, National Institute for Environmental Studies, 16-2 Onogawa, Tsukuba, Ibaragi 305-8506, Japan § Japan Society for the Promotion of Science, Tokyo 102-0083, Japan ∥ Faculty of Economics, Kyushu University, 6-19-1 Hakozaki, Higashi-ku, Fukuoka 812-8581, Japan ‡

S Supporting Information *

ABSTRACT: As the aging and low birthrate trends continue in Japan, and as changes in the working population and consumption patterns occur, new factors are expected to have an impact on consumption-based greenhouse gas (GHG) emissions. We present the impacts of changes in the composition of Japanese households on GHG emission structures using current (2005) consumption-based accounting on the commodity sectors that are expected to require priority efforts for reducing emissions in 2035. This is done using the Global Link Input-Output model (GLIO) and domestic household consumption data and assuming that recent detailed consumption expenditures based on the Social Accounting Matrix (SAM) will continue into the future. The results show that consumption-based GHG emissions derived from Japanese household consumption in 2035 are estimated to be 1061 Mt-CO2eq (4.2% lower than in 2005). This study can be used to reveal more information and as a resource in developing policies to more meticulously and efficiently reduce emissions based on emission and import rates for each domestic and overseas commodity supply chain.

1. INTRODUCTION Controlling emissions of airborne greenhouse gases (GHG), which promote global warming, is the most pressing global environmental problem facing us today, and measures are being taken to address this challenge by countries all over the world, as reflected in the United Nations Framework Convention on Climate Change (UNFCCC).1 The Kyoto Protocol (COP3), adopted in 1997, instructed developed nations to reduce their GHG emissions, and the developed nations as a whole agreed to reduce their GHG emissions by at least 5% from their 1990 levels between 2008 and 2012.2 At the Toyako G8 Summit held in Hokkaido, Japan in 2008, the G8 countries agreed on a shared vision for all of the UNFCCC signatories of achieving at least a 50% worldwide reduction in GHG emissions by 2050. To achieve this goal, international efforts by developed countries and developing countries are essential. However, carbon leakage associated with the production of exports bound for developed nations has been pointed out by developing countries,3 and this is difficult to account for with a conventional production-based accounting framework in which producing countries and regions bear the responsibility for those emissions. Thus, in recent years, there has been discussion of the use of consumption-based accounting, in which the country or region that consumes a produced item is responsible for emissions.3−5 Consumption-based accounting covers more global emissions than production-based account© 2014 American Chemical Society

ing with limited countries, especially developed countries. Furthermore, it can be expected to increase the emissions mitigation options for the developed countries and will inevitably end up promoting the development of policies for clean production and international mitigation schemes such as clean development mechanisms (CDM).3,5 For signatories of the Kyoto protocol, for example, Kanemoto et al. (2014)6 reported that 28% of the 200 fastest-growing flows of CO2 originate in Annex B (most developed) countries, while 52% of the fastest growing flows are derived from transportation from non-Annex B (most developing) countries to Annex B countries. These results indicate that the jurisdiction of the protocol could grow from covering 28% of the fastest-growing flows to covering 80% of them if the same Kyoto signatories set targets based on consumption-based accounting in addition to production-based accounting. It is therefore essential to look to the future with an eye on promoting the management and reduction of emissions on a consumption basis, particularly in developed countries. Furthermore, by making it possible for carbon leakages to revert to consumer countries rather than to producers, this framework could help increase future Received: Revised: Accepted: Published: 6069

November 7, 2013 April 30, 2014 May 6, 2014 May 6, 2014 dx.doi.org/10.1021/es404939d | Environ. Sci. Technol. 2014, 48, 6069−6080

Environmental Science & Technology

Policy Analysis

opportunities for even newly developed countries like China and India, which have thus far been resistant to Kyoto Protocol negotiations because of their focus on economic growth, to participate in climate treaties.5 However, there is considerable room for discussion with regard to how consumption-based accounting should be used, and figuring out how to calculate the responsibility for emissions continues to present a challenge.5,7 Reports indicate that particularly in developed nations, household consumption (lifestyle) makes a significant contribution to consumption-based emissions.8,9 Japan is no exception when it comes to this trend, and Nansai et al. (2012a)10 showed that Japan’s consumption-based GHG emissions in 2005 were 1,675 Mt-CO2eq (Mt: 106 ton), 61% of which was attributable to household consumption. This value was 256 Mt-CO2eq larger than emissions calculated from production accounting. Many studies have been conducted in various countries to provide detailed analyses of the environmental load derived from household consumption from this perspective.11−20 It will be particularly meaningful to perform estimates and structural analysis of emissions focusing on household consumption. The few current studies that predict future scenarios using consumption-based accounting, to our knowledge, have examined the situation only in the UK.21,22 No such studies have been conducted in Japan. Also, as the aging and low birthrate trends continue in Japan, and its working population and consumption patterns change, new factors are expected to have an impact on consumption-based GHG emissions. From the perspective of consumption-based accounting, domestic household consumption in Japan, one of the world’s most prominent trading countries according to the UN statistics,23 is expected to contribute significantly to increased GHG emissions abroad. It is therefore growing increasingly important to quantitatively evaluate the future impacts of these changes on domestic and international GHG emissions. This study was designed to highlight the impact of changes in Japanese household composition on GHG emission structure using current consumption-based accounting. The results also show that there are commodity sectors in which technological improvements will be necessary if selective emission reductions are to be achieved.

siatt =

riatt =

(2)

Here, M (=6) is the number of household attributes. Eq 2 captures that, for example, households in their 60s have relatively higher medical expenses than households in their 20s. It should be noted that since commodity sector classification in the NSFIE differs from that in the JIOT, we mapped the expenditure categories in the NSFIE onto the commodity sectors i in the JIOT. In doing so, we subdivided the “petroleum refinery products (incl. greases)” sector in the JIOT, whose GHG emissions were high due to burning. The NSFIE is the public statistical survey that presents consumption expenditures by Japanese households. The NSFIE contains the expenditures per household per month by household attribute, such as income level or size of household, for 100 categories of expenditures. It is possible to use the NSFIE to quantitatively understand differences in consumption composition based on differences in household attributes, but in the NSFIE, no distinction is made between domestic and imported commodities. As Schreyer (2013)26 pointed out, there are major inconsistencies between the survey data on household consumption (e.g., NSFIE) for different countries and the Social Accounting Matrix27 (SAM) (e.g., JIOT household consumption expenditures), and eliminating these inconsistencies is an important issue. Even if annual consumption is calculated by multiplying the NSFIE consumption amount by the number of households, and then multiplying that by 12, there is a large difference between that result and the previously mentioned JIOT household consumption expenditures that are in the SAM. In this study, we formulated a quadratic programming att problem such that the distance function in terms of ratt i and si is minimized under the following constraints (3−7). 2 ⎛ riatt ̃ − riatt ⎞ min . ∑ ∑⎜ ⎟ + riatt ̃ , ∼siatt riatt ⎠ att = 1 i = 1 ⎝ M

N

2 ⎛ siatt ̃ − siatt ⎞ ∑ ∑ ⎜ att ⎟ si ⎠ att = 1 i = 1 ⎝ M

N

(3)

s.t. M

gi =

2. METHODS AND DATA 2.1. Estimates of Consumption Expenditures by Household Attributes. To focus on the effects of the aging and low birthrate trends on consumption-based emissions in this study, we defined household attributes (att = 1···6) following Kronenberg (2009)17 by the age of the head of household (1=20s: −29, 2=30s: 30−39, 4=40s: 40−49, 4=50s: 50−49, 5=60s: 60−69, 6=70s: 70-). We first calculated the expenditure ratio ratt i of commodity sector i per unit expenditure of household att as Piatt N ∑i = 1 Piatt

Piatt M ∑att = 1 Piatt

∑

riatt ̃ g att

att = 1

(4)

N

̃ =1 ∑ riatt i=1

(5)

riatt ̃ ≥0

(6)

siatt ̃ = riatt ̃ g att /gi

(7)

att

where gi and g represent the total expenditure for commodity sector i in the JIOT and the total expenditure of household att, respectively. Here, eq 4 indicates that JIOT household consumption expenditures of commodity sector i should coincide with the sum of the annual consumption of household attributes. Eqs 5 and 6 indicate that the total of the household expenditure ratios is equal to 1 and each ratio is non-negative, att respectively. Eq 7 expresses the relationship between ratt i and si . att We multiplied the optimal solution rĩ̂ by gatt, and att att ̂ determined the annual consumption gatt for sector i i = rĩ g

(1)

where Patt is the expenditure (million JPY/m) from the i Japanese Input-Output Table (JIOT)24 for commodity sector i taken from the National Survey of Family Income and Expenditure (NSFIE)25 and N (=409) is the number of JIOT sectors. We used eq 2 to calculate satt i , the share of household att accounted for by commodity sector i. 6070

dx.doi.org/10.1021/es404939d | Environ. Sci. Technol. 2014, 48, 6069−6080

Environmental Science & Technology

Policy Analysis

emissions associated with the disposing of these devices as “waste management services (private)”. Likewise, the indirect emissions, Satt, resulting from the domestic and international supply chains associated with household consumption were calculated based on eq 12 using the embodied emission intensities, called global GHG JI emission intensities,29 qJD i (t-CO2eq/million JPY) and qi (tJD CO2eq/million JPY). The element qi is the embodied emission intensity expressing the GHG volume caused by the global supply chain associated with the unit production of the Japanese domestic (JD) commodity sector i, while qJIi is the embodied emission intensity for the Japanese imported (JI) commodity sector i.

for each household (million JPY/y). However, for JIOT sectors with no corresponding NSFIE expenditure category (such as waste processing, wholesale, retail, etc.), we calculated gatt i by proportionally distributing the total expenditures according to the JIOT’s gi by the size of gatt. These expenditures account for consumer prices. Because the emission intensities used in calculating consumption-based emissions are based on producer prices, we converted giatt to f iatt, the annual consumption for sector i for each household at producer prices. Consumer prices consist of producer prices with retail trade and transportation margins added. The method for converting giatt into f iatt is provided in the Supporting Information (SI). Additionally, we included the condition that all expenditures related to education and health care are compensated by the Japanese government. Using eqs 8 and 9, we multiplied f att i by the ratio of imports mi obtained from the JIOT, and determined the consumption of domestic commodities f JD,att (million JPY/y) and the i consumption of imported commodities f JI,att (million JPY/y). i fi

JD,att

= (1 −

mi)f iatt

N

S att =

i=1

(9)

Thus, the total consumption figures for the six household attributes found in this study are consistent with the total expenditures on household consumption f in the JIOT (million JPY/y). 2.2. Calculating Consumption-Based GHG Emissions by Household. Consumption-based GHG emissions by household, Catt (t-CO2eq/y), is defined as shown in eq 10 as the sum of direct emissions derived from burning fuel through the use of private cars and home heaters, Datt (t-CO2eq/y), and indirect emissions generated by the supply chains of products and services consumed, Satt (t-CO2eq/y). The GHG considered are CO2, CH4, N2O, HFCs (hydro fluorocarbons), PFCs (per fluorocarbons), and SF6 (sulfur hexafluoride). C att = Datt + S att

(10)

att

D was calculated as shown in eq 11 by multiplying expenditures f JD,att and f JI,att by emissions coefficient qdirect (ti i i CO2eq/million JPY), which expresses the GHG directly produced by burning fuel associated with the unit consumption of sector i. N

Datt =

N

∑ qidirectfi JD,att + ∑ qidirectfi JI,att i=1

i=1

i=1

(12)

From clarifying system boundaries, a multiregional inputoutput analysis (MRIO)30,31 is effective in identifying the environmental burden generated through the global supply chains associated with the unit production of goods and services. For this reason, identifying consumption-based environmental loads using MRIO has been promoted in recent years.32−40 In this study, the global GHG emission intensities were determined using the global link input-output model (GLIO).10,29,41−43 GLIO defines Japan’s input-output structure with 406 sectors of domestic commodities and 406 sectors of imported commodities, indicating that Japan has a high level of sector resolution. Also, by defining 230 countries and regions as international sectors, the GLIO is a simplified MRIO that guarantees system boundaries while achieving data management cost reductions. In the EORA44 as well, Japan’s sector resolution is carefully defined with about 400 sectors, but we used the GLIO model because it is easier to compare with previous studies10 and because it is highly consistent with the final demand amounts reported in the JIOT. The more detailed methods for calculating both models and the GLIO accounting system are elaborated in the SI. 2.3. Estimating Consumption-Based GHG Emissions Until 2035. Because the most recent JIOT is based on 2005 figures, we were able to calculate f JD,att , f JI,att , and Catt using the i i method noted above for 2005. Hereafter, to clarify the target year of the estimate y, we will denote these values as f (y)JD,att ,f i att , and C (y) , respectively. In this study, we focused on (y)JI,att i changes in household composition associated with Japan’s population decline, its aging and low birthrate trends, and estimated future consumption-based GHG emissions. Since the predicted number of households published by the National Institute of Population and Social Security Research (2013)45 was for 2035, we selected target years between 2005 and 2035, and calculated emissions for every five years, C (y)att, as follows: 2005 (y = 1), 2010 (y = 2), 2015 (y = 3), 2020 (y = 4), 2025 (y = 5), 2030 (y = 6), and 2035 (y = 7). We determined emissions through 2035 by hypothetically calculating the consumption of each household’s domestic commodities i and imported commodities i by comparing them to the number of households for each household attribute, Hatt (t), as shown in eqs 13 and 14. The element t reflects the target year: 2005 (t = 1), 2010 (t = 2), 2015 (t = 3), 2020 (t = 4), 2025 (t = 5), 2030 (t = 6), and 2035 (t = 7).

(8)

fi JI,att = mif iatt

N

∑ qi JDfi JD,att + ∑ qi JIfi JI,att

(11)

qdirect i

was calculated based on the direct emissions from the household consumption expenditures sector (2005 figures) reported in the Embodied Energy and Emission Intensity Data for Japan Using Input-Output Tables.28 CO2 is classified as an emission from the “gasoline”, “light oil”, “kerosene”, “LPG”, “coal products”, and “gas supply” sectors, so we found qdirect by i dividing the direct emissions by the consumption for each sector. Likewise, CH4 and N2O were classified into auto emissions, using “gasoline” and “light oil”, and household emissions, using “kerosene”, “LPG,” “coal products”, and “gas supply”. For the HFCs, we allocated emissions from the use of fixed air conditioners to “household air conditioners” and emissions from the use of household refrigerators as “household electric appliances”. We categorized emissions from the use of air conditioners in transport equipment as “passenger motor cars” and “trucks, buses, and other cars”, and classified 6071

dx.doi.org/10.1021/es404939d | Environ. Sci. Technol. 2014, 48, 6069−6080

Environmental Science & Technology

Policy Analysis

Figure 1. Composition of number of households (left) and population (right) by household attribute from 2005 to 2035.

Figure 2. (a) Household expenditures of 13 aggregated sectors by householder age group in 2005. (b) GHG emissions of 13 aggregated sectors by householder age group in 2005. (c) GHG emissions of three locational sectors by householder age group in 2005. “In Japan (Direct)”, direct emissions; “In Japan (Supply chain)”, indirect emissions from Japanese supply chains; “Overseas”, indirect emissions from foreign countries’ supply chains.

f (t )iJD,att = f (t − 1)iJD,att ×

H(t )att × θ(t )iatt H(t − 1)att

f (t )iJI,att = f (t − 1)iJI,att ×

(13)

6072

H(t )att × θ(t )iatt H(t − 1)att

(14)

dx.doi.org/10.1021/es404939d | Environ. Sci. Technol. 2014, 48, 6069−6080

Environmental Science & Technology

Policy Analysis

H (t)att refers to levels predicted by the National Institute of Population and Social Security Research. According to its predictions, the total number of households is estimated to rise from 2005 to 2020, then drop until 2035. On the other hand, it forecasts that the total population will decrease from 2005 mainly due to the declining number of children. Therefore, household size (persons per household) will also shrink with the population, which can influence per household expenditures. However, expenditures do not always decrease with shrinking household size. For example, most expenditures for food, like rice and bread, tend to be higher in larger households. On the other hand, there are some expenditures, such as eating out, that are higher for single person households than multifamily households.46 We explained such trends using θ(t)iatt coefficients that explain the influence on future household expenditures in sector i due to changes in household size (for detailed calculation methods see the SI). Since this study focuses on emissions that are impacted by changes in household composition, and since it is not easy to estimate future technological changes, including changes in global supply chains, the embodied emission intensities were fixed, regardless of the target year, and the estimated value for 2005 was used. We based the rigid factors for estimating the consumption-based GHG emissions in this study on the following considerations. • The shares of both domestic and import products for household expenditures are assumed to be constant as of 2005. • Assuming that there are no improvements in technologies and global supply chains since 2005, global GHG emission intensities and the coefficient matrix of GLIO2005 are rigid. • Future consumption patterns for 2010 to 2035 for each household keep being based on those of 2005.

40s, in particular, are estimated to drop by 9.19 million (−34.3%) persons from 2005 to 2035. 3.2. Characteristics of Consumption-Based GHG Emissions by Household in 2005. This section describes the characteristics of consumption-based GHG emissions by household attribute in 2005 from the following three perspectives. Figure 2 integrates the 409 sector types defined in this study into 13 aggregated sectors without distinguishing between domestic and imported commodities, and provides (a) a breakdown of household consumption expenditures per household by aggregated sector, (b) a breakdown of GHG emissions by consumption expenditure sector, and (c) a breakdown of GHG emissions by supply chain. Here note that “(7) transportation” mainly includes the commodity sectors associated with public transportation and cargo services. “Gasoline” and “light oil” associated with driving cars are in “(3) petroleum refinery and coal”, while commodity sectors related to purchasing cars such as “passenger motor car” are in “(4) transport machinery”. The corresponding relationships between the 409 sectors and the 13 aggregated sectors are shown in Table S1 in the SI. The supply chain categories reflecting the source of emissions are, as in previous studies,10 direct emissions from households (in Japan, direct), indirect emissions from the domestic supply chain (in-Japan supply chain), and indirect emissions from the overseas supply chain (overseas). Figure 2(a) shows that the largest household consumption expenditures per household occur among those in their 50s, followed closely by those in their 40s. Households whose heads are in their 40s to 50s are in periods of life when their household incomes are larger and, generally, when their children are growing older. Also, in households of this age, many people have purchased their own home, or have purchased or are trading up to better cars, and for this reason, they tend to have higher expenditures than other households. This trend is also evident in the fact that household consumption expenditures are larger among those in their 40s and 50s in such categories as “(9) education”, “(7) transportation”, and “(13) house rent, insurance, and others”. However, as shown in Figure 2(b), the consumption-based GHG emissions per household follow a different trend from the household consumption expenditures. The age group with the highest emissions per household is the 40s, at 25.3 t-CO2eq/ household, followed by the 60s and the 50s, at 24.7 t-CO2eq/ household and 24.6 t-CO2eq/household, respectively. One key reason that emissions are lower among households in their 50s, which have the highest consumption expenditures per household, is the particular characteristics of the household consumption expenditures pointed out above. Compared to the second place group (those in their 40s), for “house rent (imputed house rent)″, which is a category within “(13) house rent, insurance, and others” and that has an extremely low global emissions intensity, the impact of allocating over 400 000 JPY more per year of household consumption expenditures is particularly significant. These consumption trends for those in their 50s seem to be strongly related to their highest income. On the other hand, the emission contributions from “(10) medical and health care” and “(6) utility” are larger once people reach their 60s. That is, in many households in this age group and older, people spend more time at home due to retirement, which causes them to consume more household energy (costs for electricity and heating, for example). Also, their childrearing and rent burdens are greatly reduced, creating extra room for

3. RESULTS AND DISCUSSION 3.1. Estimates of The Number of Households and Population by Household Attributions from 2005 to 2035. Figure 1 depicts the breakdown of households and population by six household attributes from 2005 to 2035. The number of households (on the left) refers to national statistics45 and the population (on the right) is estimated by the method of this study. Total households are expected to increase from 49.1 million households in 2005 to 53.1 million households in 2020. After that, the number decreases to 49.6 million households by 2035. This number is almost the same as in 2005. On the other hand, the total population is estimated to be 126 million in 2010, and then fall, finally dropping to 109 million in 2035. This is 13.0% less than in 2005. The difference between the change in number of households and population will occur mainly due to a rapid increase in the number of older households and a decrease in the younger and middle (under 50s) population as Japan’s society ages and there are with fewer children. The number of households headed by those in their 70s in 2035 is estimated to be about 1.7 times larger than in 2005. In other words, the share of 70s households will climb steeply from 19.0% to a dominant 31.6% in those 30 years. Adding households in their 60s to the 70s makes up more than half of total households (51.6%) in 2035, reflecting the rapidly aging society. At the same time, the larger effect of fewer children is expected to cause the population of those in both their 40s and 50s to have smaller households. Those in their 6073

dx.doi.org/10.1021/es404939d | Environ. Sci. Technol. 2014, 48, 6069−6080

Environmental Science & Technology

Policy Analysis

Figure 3. Variations in the GHG emissions of the 13 aggregated sectors from 2005 to 2035. (a) Total emissions. (b)−(g) Emissions for each age group (20s to 70s and older).

and older. In all of the years in this period “(1) food”, “(3) petroleum refinery and coal”, “(6) utility”, and “(12) service” accounted for 70% or more of GHG emissions. This trend is generally consistent with the future estimates for the U.S. in 200416 and the UK in 2030.22 Additionally, the trends do not change much each year. It is therefore important to reduce emissions effectively to try to make way for technological improvements related to those sectors toward 2035. From 2005 to 2035, the total number of households is expected to be highest in 2020, but the consumption-based GHG emissions derived from Japanese household consumption looks to be highest in 2015, at 1150 Mt-CO2eq. This is 3.8% more than in 2005. After that, both figures decline, but ultimately, GHG emissions in 2035 are estimated to be 1,061 Mt-CO2eq (4.2% lower than in 2005). In other words, due to changes in family composition, GHG emissions are expected to increase by 42.6 Mt-CO2eq from 2005 to 2015, and then fall, decreasing to just 46.5 Mt-CO2eq below the 2005 level in 2035. Thus, some decreases in consumption-based GHG emissions can be expected to occur simply as a result of changes in household composition due to aging and the low birthrate trend. But it is going to be necessary to make further efforts to reduce emissions through technological means or trade structure strategies. The same has been said regarding GHG emissions derived from household consumption in Germany, where aging and the low birthrate trend are expected to progress much as they have in Japan.17 According to the trends in GHG emissions by head of household age group, the emissions of those in their 20s is expected to decline to 2035. Emissions of those in their 30s and 40s seem to be increasing to 2010 and 2015, respectively, then both decrease. Similar to those in their 20s, emissions for those in their 50s peaks in 2005, but after decreasing from then to 2015, increases to 2025, finally decreasing again to 2035. In contrast, the emissions for those in their 60s peaks in 2010,

returning those consumption expenditures to other commodity sectors. This is one reason that the overall emissions per household are greater among this group than among those in their 50s. Next, Figure 2(c) shows emission sources by supply chain, and reveals that differences in emissions in the “in-Japan supply chain” category significantly impact the overall differences. There is little difference between households in the “in-Japan direct” category, which reflects direct emissions, but a breakdown shows that the characteristic differ between the 20-to-40s, and the 50-to-70s. While the former reflects a larger consumption of “gasoline” and “light oil” than the latter due to the use of private cars, the trend for “kerosene”, which is consumed in household heaters, is the opposite. We believed this to be related to the fact that the amount of time spent at home, as mentioned above, differs among households. A similar trend has been seen in Germany.17 Finally, a comparison of households using the consumptionbased GHG emissions per unit expenditure (kg-CO2eq/1000 JPY) indicates that emissions are 1.5 times larger among those in their 70s, the age when emissions are highest, at 4.95 kgCO2eq/1000 JPY, than among those in their 50s, when emissions are lowest, at 3.40 kg-CO2eq/1000 JPY. While the emissions per unit expenditure by those in their 30s, 40s, and 50s are fewer than 4 kg-CO2eq/1000 JPY, those in their 70s and older and 60s whose emissions are 4.63 kg-CO2eq/1000 JPY are an important segment given that the number of senior households is expected to increase. 3.3. Estimating Consumption-Based GHG Emissions from 2005 to 2035. Figure 3 shows a breakdown of GHG emissions derived from household consumption by consumption expenditure sector from 2005 to 2035. Figure 3(a) shows the trends in total emissions, while the graphs in Figures 3(b) to (g) show the trends in emissions by the six age groups of the head of household, from those in their 20s to those in their 70s 6074

dx.doi.org/10.1021/es404939d | Environ. Sci. Technol. 2014, 48, 6069−6080

Environmental Science & Technology

Policy Analysis

Figure 4. Variations in the GHG emissions from 2005 to 2035 of the three locational sectors. (a) Total emissions. (b)−(g) Emissions for each age group (20s to 70s and older).

highest are “JD137: gasoline” at 106 Mt-CO2eq and “JD306: retail trade” at 61.3 Mt-CO2eq. By aggregated sector, the largest expenditure category is “(3) petroleum refinery and coal”. Since emissions associated with the direct consumption of petrochemical products accounts for the vast majority of this, a first step will be to reduce the consumption of such products, in addition to electricity. This is also seen in Germany and the UK in 2030 by Kronenberg (2009)17 and Chitnis et al. (2012),22 respectively. On the other hand, the amounts of emissions in the commodity sectors related to diet, such as “JD394: general eating and drinking places” in category “(11) leisure”, and “JD36: slaughtering and meat processing” in category “(1) food”, were remarkable. Emissions in “JD394: general eating and drinking places” were higher than those produced by “JI140: LPG”, the third largest in “(3) petroleum refinery and coal”, and emissions in “JD36: slaughtering and meat processing” were very close to that level. Furthermore, since the vast majority of emissions resulting from “LPG” are accounted for by domestic emissions associated with direct consumption, “general eating and drinking place” emissions are 1.7 times higher than overseas emissions. This example demonstrates the size of our hidden emissions influenced by the use of everyday services and foods. Particularly remarkable in overseas emissions are the “JI84: woven fabric apparel” and “JI85: knitted apparel” categories in sector “(2) textiles”, whose production bases are expanding quickly in Southeast Asia thanks to its low costs. Therefore, commodity sectors like these are highlighted not in production-based accounting but in consumption-based accounting. Dietary habits reflect differences in living standards and household patterns, as well as differing preferences depending on the age of the head of household. For example, among younger households, the percentage of favorite foods accounted for by eating out, boil-in-the-bag foods, and juices is higher than in other households. Their staple tends to be bread rather than

then decreases to 2025, finally increasing again to 2035. Meanwhile, for those in their 70s, emissions are expected to increase steeply from 2005 to 2025, finally being 67% larger than their initial emissions. This increase is 122 Mt-CO2eq Remarkable reductions for those in their 20s and 30s of 75.1 Mt-CO2eq will occur, but the reductions achieved by the younger generations will not offset the increased emissions being produced by the older generations. The proportion of total emissions accounted for by households in their 60s and 70s will gradually rise from about 37% in 2005 to about 51% by 2035. For this reason, achieving technological improvements focused on commodity sectors strongly correlated with the consumption habits and lifestyles of middle-aged and olderaged households is likely to be effective in achieving emissions reductions. Figure 4 shows the trends in GHG emissions derived from household consumption by emission source for the same time period. “In-Japan supply chain” accounts for the most emissions at about 50%, but “overseas” also accounts for 34%. This “overseas” share is generally close to the “overseas” shares in the UK’s household consumption in 2004, which is estimated at about 40%.19 These indicate the importance of the overseas spillover effect of GHG emissions resulting from household consumption by developed countries like Japan and the UK. Also note that the emission sources of each household are almost the same. 3.4. Characteristics of Consumption-Based GHG Emissions Derived from Household Consumption in 2035. The major expenditures that will shape consumptionbased GHG emissions in 2035 are shown in Table 1. Table 1 shows the expenditure sectors that have the top five largest GHG emissions in 13 categories, differentiating between domestic products (JD) and imported ones (JI), and indicating the ratio of those expenditures attributed to the imported commodities. The expenditure category that yields the highest emissions is “JD296: electricity”, at 132 Mt-CO2eq The next 6075

dx.doi.org/10.1021/es404939d | Environ. Sci. Technol. 2014, 48, 6069−6080

Environmental Science & Technology

Policy Analysis

Table 1. Top five Household Domestic (JD) and Imported (JI) Commodity Sectors of Each Aggregated Sector, Their Highest Consumption-Based GHG Emissions for 2035, and the Ratios of Expenditures Attributed to the Imported Commodities. The Total Emissions for Japan and Overseas Are Also Shown GHG emissions ammount (Mt-CO2eq)

import ratio (%)

Japan (%)

overseas (%)

slaughtering and meat processing sushi, lunch boxes and other prepared dishes grain milling dairy farm products soft drinks

10.1 9.4

14.1 0.2

53.2 56.7

46.8 43.3

9.2 8.5 8.4

0.0 4.6 1.7

86.0 66.0 65.2

14.0 34.0 34.8

12.3 11.9 4.3

97.9 91.9 71.5

1.1 3.3 8.4

98.9 96.7 91.6

JI87 JI88

woven fabric apparel knitted apparel other wearing apparel and clothing accessories bedding other ready-made textile products

1.7 1.1

75.5 28.6

5.1 29.1

94.9 70.9

(3) petroleum refinery and coal 192.8 Mt-CO2eq

JD137 JD138 JI140 JD139 JD140

gasoline kerosene LPG (liquified petroleum gas) light oil LPG (liquified petroleum gas)

105.7 53.1 12.6 7.4 6.0

3.3 3.2 67.0 4.7 67.0

80.7 89.7 81.8 83.8 81.8

19.3 10.3 18.2 16.2 18.2

(4) transport machinery 60.1 MtCO2eq

JD252 JD253 JI252 JI266 JI254

passenger motor cars trucks, buses and other cars passenger motor cars bicycles two-wheel motor vehicles

15.8 3.1 2.0 1.0 0.4

12.7 0.0 12.7 76.4 58.9

61.4 68.3 61.4 6.5 16.1

38.6 31.7 38.6 93.5 83.9

(5) manufacturing product 54.9 MtCO2eq

JI279 JD130

5.9 5.6

82.8 14.1

3.7 56.1

96.3 43.9

JD235 JI151 JI243

jewelry and adornments cosmetics, toilet preparations and dentifrices household electric appliances miscellaneous leather products personal computers

5.1 4.3 3.4

16.9 90.1 55.5

41.1 3.1 9.9

58.9 96.9 90.1

(6) utility 176.3 Mt-CO2eq

JD296 JD298 JD302 JD304 JD300

electricity gas supply sewage disposal waste management services (private) water supply

131.8 30.5 9.7 2.1 1.6

0.0 0.0 0.0 0.0 0.1

91.5 83.3 88.8 96.3 77.4

8.5 16.7 11.2 3.7 22.6

(7) transportation 74.7 Mt-CO2eq

JD324 JD318

14.0 13.0

43.7 0.0

50.9 84.1

49.1 15.9

JD314 JI324 JI314

air transport road freight transport(except self-transport by private cars) railway transport (passengers) air transport railway transport (passengers)

11.6 9.5 5.4

4.0 43.7 4.0

57.6 50.9 57.6

42.4 49.1 42.4

(8) information and communication 13.9 Mt-CO2eq

JD337 JD346 JD336 JD347 JD343

mobile telecommunication newspapers fixed telecommunication publication information services

3.2 2.4 1.9 1.4 1.2

0.1 0.0 0.2 3.5 3.4

74.9 73.5 75.8 76.6 68.3

25.1 26.5 24.2 23.4 31.7

(9) education 4.8 Mt-CO2eq

JD352 JD356

school education (private) other educational and training institutions (profit-making) school education (public) social education (private, nonprofit) social education (public)

3.2 0.7

0.0 0.0

77.4 84.4

22.6 15.6

0.7 0.1 0.1

0.0 0.0 0.0

84.8 76.8 80.4

15.2 23.2 19.6

aggregated sector (sectoral emission) (1) food 149.6 Mt-CO2eq

commodity no. JD36 JD60 JD45 JD39 JD69

(2) textile 35.1 Mt-CO2eq

JI84 JI85 JI86

JD351 JD354 JD353

top five household commodity sectors in the aggregated sector

6076

dx.doi.org/10.1021/es404939d | Environ. Sci. Technol. 2014, 48, 6069−6080

Environmental Science & Technology

Policy Analysis

Table 1. continued aggregated sector (sectoral emission)

commodity no.

top five household commodity sectors in the aggregated sector

GHG emissions ammount (Mt-CO2eq)

import ratio (%)

Japan (%)

overseas (%)

(10) medical and health care 76.7 MtCO2eq

JD366 JD365

medical service (medical corporations, etc.) medical service (nonprofit foundations, etc.) medical service (public) social welfare (private, nonprofit) social welfare (public)

43.8 15.2

0.0 0.0

66.7 69.3

33.3 30.7

14.2 1.1 0.7

0.0 0.0 0.0

69.1 72.5 74.1

30.9 27.5 25.9

JD364 JD372 JD371 (11) leisure 83.4 Mt-CO2eq

JD394 JD397 JD390 JI397 JD396

general eating and drinking places hotels amusement and recreationfacilities hotels eating and drinking places with entertainment

30.6 13.4 12.5 6.3 4.6

6.2 25.6 1.0 25.6 4.4

55.1 48.6 81.6 48.6 59.0

44.9 51.4 18.4 51.4 41.0

(12) service 118.5 Mt-CO2eq

JD306 JD305 JD381 JD404 JD406

retail trade wholesale trade repair of motor vehicles ceremonial occasions supplementary tutorial schools, instruction services

61.3 22.2 6.5 6.3 4.1

0.0 0.0 0.0 0.2 0.0

84.3 71.2 67.5 80.0 81.3

15.7 28.8 32.5 20.0 18.7

(13) house rent, insurance and others 20.6 Mt-CO2eq

JD313 JD312 JD308 JD309 JI409

house rent (imputed house rent) house rent life insurance nonlife insurance activities not elsewhere classified

7.9 5.8 5.8 1.1 0.1

0.0 0.0 0.0 0.1 22.3

70.9 74.2 72.3 75.8 35.5

29.1 25.8 27.7 24.2 64.5

Table 2 shows the 10 sectors with the largest differences in consumption-based GHG emissions between 2035 and 2005 by

rice, and meat tends to account for a larger percentage of consumption than fish and seafood. On the other hand, among older households, fresh vegetables, fish and seafood, fruit, and rice account for the highest ratios of foods consumed, reflecting what would be considered a more traditional Japanese diet.47 Also, Japan has a low rate of food self-sufficiency and thus has to rely on imports for a large majority of its food, including its livestock feed. These necessarily make a sizable emissions contribution to the overseas supply chain, and constitute an important commodity cluster when considering the relationship between the trade structure and consumption-based GHG emissions. A review of transportation of such items is also important. For example, GHG emissions derived from the use of private cars is 117 Mt-CO2eq, which is the sum of emissions from both “gasoline” and “light oil”. This is 5.3 times the emissions derived from “railway transport (passenger)” and “bus transport service”, which are public means of transportation used on a daily basis. Of this, the emissions derived from the use of private cars were largest, at 7.5 times the domestic emissions and 2.4 times the overseas emissions of public transport. This suggests that expanding campaigns conducted by local governments aimed at promoting the use of public transportation, such as “no car days”, would be an effective way to reduce both domestic and overseas GHG emissions. In addition, the total emissions related to health care, for “JD364: medical service (public)”, “JD365: medical service (nonprofit foundation, etc.)”, and “JD366: medical service (medical corporation, etc.)”, is 73.2 Mt-CO2eq, making them the second highest overall. For a country that is experiencing rapid aging like Japan, controlling GHG emissions indirectly generated by the expansion of the health care system is going to become increasingly important.

Table 2. Top 10 Sectors with the Largest Differences and Change Ratios in Consumption-Based GHG Emissions Between 2035 and 2005

rank

sector no.

1 2

JD138 JD366

3

JD365

4 5 6 7 8

JD364 JD397 JI40 JD40 JI334

9 10

JI6 JD5

commodity sector kerosene medical service (medical corporations, etc.) medical service (nonprofit foundations, etc.) medical service (public) hotels frozen fish and shellfish frozen fish and shellfish travel agency and other services relating to transport fruits vegetables

difference

difference

[MtCO2eq]

[%]

3.95 3.49

8.0 8.7

1.28

9.1

1.20 0.66 0.51 0.44 0.39

9.2 5.2 8.1 8.1 15.4

0.37 0.33

14.5 5.4

total emissions. This gives a view of the situation from a different perspective than total emissions, and suggests the need to make controlling such increases a policy priority. The increases in emissions from “JD138: kerosene” are most remarkable. Since “kerosene” and “electricity”, which produce the largest emissions from 2005 to 2035 are due to direct use by households, consumers need to introduce energy-saving products and make greater efforts to adopt energy-saving strategies in their everyday lives. On the other hand, four of items in this table are accounted for by commodity sectors 6077

dx.doi.org/10.1021/es404939d | Environ. Sci. Technol. 2014, 48, 6069−6080

Environmental Science & Technology

Policy Analysis

related to diet, such as “JI40 (and JD40): frozen fish and shellfish”, “JI6: fruits,” and “JD5: vegetables”. Since these are in categories where emissions themselves are high, and where, in contrast to “kerosene” and “electricity”, it is difficult for consumers to restrict their consumption, the government and corporate sectors need to prioritize developing technological improvements to reduce emissions. Specifically, neither the technologies nor the supply chain associated with “frozen fish and shellfish” have seemed to do particularly well in reducing emissions. For Japan, which is also a great fisheries country, it will be important to pay attention to them. The three medical demand sectors are strongly influenced by the increase in middle- and older-aged households, which have higher ratios of medical expenditures, and are commodity sectors that must be paid close attention to as Japan’s society continues to age. The “hotels” sector is also expected to be impacted by trends among middle-aged and older households, which tend to enjoy more postretirement sightseeing and travel. Hertwich (2011)9 showed a graph that illustrates the consumption-based GHG emissions of per capita household consumption (t/capita/y) referred to in several articles. In it, the emission shares of “Health” in some countries like the UK, the US, The Netherlands and Denmark were estimated to rise remarkably from the 1990s to the 2000s. Although it is quite difficult to simply compare their results with ours due to the different methods used, consumption categories and the system boundaries, the results related to medical demands in this study seem generally consistent with such past trends in developed countries. 3.5. Further Perspectives on This Research. In this study we identified current consumption-based GHG emissions precisely by the age of the head of household in 2005 using the GLIO model and domestic household consumption data. Next, we estimated future consumption-based GHG emissions derived from Japanese household consumption due to changes in household composition. We also highlighted the commodity sectors expected to require priority efforts in order to reduce emissions in 2035. Kronenberg (2009)17 also estimated GHG emissions derived from household consumption, focusing on changes in household composition, but because that study looked at the domestic supply chain using German SIO tables, it differs from our analysis, which uses consumption-based accounting. Barrett and Scott (2012)21 and Chitnis et al. (2012),22 who made future estimates using consumption-based accounting, both presented results achieved through macro sector resolution based on a scenario analysis, while our study analyzed the impact on consumption-based emissions for each commodity sector in as much detail as possible. For example, the mere identification of large “food” emissions does little to show specifically what kind of “food” supply chain improvements or policies for consumers would be effective. The results presented here can not only be used to reveal more information, such as the future importance of foods such as “frozen fish and shellfish” and “fruits”, but also as a resource for developing policies to make more meticulous and efficient emissions reductions based on emission and import rates for each domestic and overseas commodity supply chain. With regard to the effectiveness of consumption-based accounting, Wiedmann (2009)31 argued that it is possible to make consumers aware of indirect GHG emissions derived from their lifestyles and consumption habits, and that future estimates of consumption-based GHG emissions based on

this kind of detailed sector resolution will play an important role in taking advantage of this approach. Because this study focused on how changes in household composition will affect consumption-based GHG emissions, as noted in our methods and data, the production technologies (emission intensities), global supply chain structures (GLIO coefficients), prices and household consumption patterns in this study are fixed on 2005 data, except for the numbers of households and populations. For example, when today’s 20-year olds enter their 50s in 30 years, they will not be able to have the same consumption patterns as they do now. Their incomes and expenditures will increase as they get married and have children. Also, what patterns to expect of future 20-year-olds being born now is difficult. Thus, we assumed that today’s young households will adopt the consumption patterns of today’s older households as they grow older. Since the Fukushima Daiichi Nuclear plant disaster of 2011, the need to review Japan’s energy mix has come to the forefront. After the disaster, the territorial GHG emissions for 2011 were reported to be 1.31 Gt-CO2eq (Gt: 109 ton), about 4% larger than the 1.26 Gt-CO2eq of 2010. Moreover emissions continued to increase in 2012 to 1.34 Gt-CO2eq.48 Now the trend is not toward increasing generation by nuclear plants and resuming the operation of those that have been stopped, so the amount of petroleum fuels imported is expected to continue to rise, at least for a while. Therefore, the prices and emission intensities of energy sectors in 2005 are potentially much higher since 2011. On the other hand, the results of this study indicate that Japanese consumption-based GHG emissions derived from household consumption are estimated to drop naturally because of an aging society with low birthrate without interventions of any new technologies and new policies. In other words, the data presented here might be considered a base scenario for 2035. In the future, incorporating future trends in technology levels or changes in the international trade that incorporate the international supply chain into the scenario will improve the accuracy of the estimates so they can be better used in the management of consumption-based GHG emissions.

■

ASSOCIATED CONTENT

S Supporting Information *

Additional descriptions of the methodologies and tables are provided in the Supporting Information. This material is available free of charge via the Internet at http://pubs.acs.org.

■

AUTHOR INFORMATION

Corresponding Author

*Phone: +81 75-753-5618; fax: +81 75-753-5619; e-mail: [email protected]. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This research was supported in part by the Environmental Research & Technology Development Fund (K122024) of the Ministry of Environment, Japan, and a Grant-in-Aid for Research (No. 25281065) from the Ministry of Education, Culture, Sports, Science and Technology, Japan. 6078

dx.doi.org/10.1021/es404939d | Environ. Sci. Technol. 2014, 48, 6069−6080

Environmental Science & Technology

■

Policy Analysis

(25) National Survey of Family Income and Expenditure; Ministry of Internal Affairs and Communications Japan: Tokyo, 2004. (26) Schreyer, P. Social accounting matrix and microdata: New areas of research. In 21st International Input-Output Conference & The Third ed. of the International School of Input-Output Analysis; Kitakyushu, Japan, 2013. (27) Miller, E. R.; Blair, D. P. Input-Output Analysis Foundations and Extensions, 2nd ed.; . Cambridge University Press, New York, 2009. (28) Nansai, K.; Moriguchi, Y.: Embodied energy and emission intensity data for Japan using input−output tables (3EID): For 2005 IO table. CGER, National Institute for Environmental Studies, Japan, 2013; http://www.cger.nies.go.jp/publications/report/d031/index. html. (29) Nansai, K.; Kondo, Y.; Kagawa, S.; Suh, S.; Nakajima, K.; Inaba, R.; Tohno, S. Estimates of embodied global energy and air-emission intensities of Japanese products for building a Japanese input−output life cycle assessment database with a global system boundary. Environ. Sci. Technol. 2012b, 46, 9146−9154. (30) Lenzen, M.; Pade, L. L.; Munksgaard, J. CO2 multipliers in multi-region input-output models. Econ. Syst. Res. 2004, 16 (4), 391− 412. (31) Wiedmann, T.; Lenzen, M.; Turner, K.; Barrett, J. Examining the global environmental impact of regional consumption activities  Part 2: Review of input−output models for the assessment of environmental impacts embodied in trade. Ecol. Econ. 2007, 61, 15− 26. (32) Hertwich, E. G.; Peters, G. P. Carbon footprint of nations: A global, trade-linked analysis. Environ. Sci. Technol. 2009, 43, 6414− 6420. (33) Wiedmann, T. A review of recent multi-region input−output models used for consumption-based emission and resource accounting. Ecol. Econ. 2009, 69, 211−222. (34) Baiocchi, G.; Minx, J. Understanding changes in the UK’s CO2 emissions: A global perspective. Environ. Sci. Technol. 2010, 44, 1177− 1184. (35) Davis, S. J.; Caldeira, K. Consumption-based accounting CO2 emissions. Proc. Natl. Acad. Sci. U.S.A. 2010, 107 (12), 5687−5692. (36) Peters, G. P.; Minx, J. C.; Weber, C. L.; Edenhofer, O. Growth in emission transfers via international trade from 1990 to 2008. Proc. Natl. Acad. Sci. U.S.A. 2011, 108 (21), 8903−8908. (37) Lenzen, M.; Moran, D.; Kanemoto, K.; Foran, B.; Lobefaro, L.; Geschke, A. International trade drives biodiversity threats in developing nations. Nature 2012, 486, 109−112. (38) Barrett, J.; Peters, G.; Wiedmann, T.; Scott, K.; Lenzen, M.; Roelich, K.; Qur, C. L. Consumption-based GHG emission accounting: A UK case study. Clim. Policy 2013, 13 (4), 451−470. (39) Lenzen, M.; Moran, D.; Bhaduri, A.; Kanemoto, K.; Bekchanov, M.; Geschke, A.; Foran, B. International trade of scarce water. Ecol. Econ. 2013, 94, 78−85. (40) Wiedmann, T. O.; Schandl, H.; Lenzen, M.; Moran, D.; Suh, S.; West, J.; Kanemoto, K. The material footprint of nations. 2013. Proc. Natl. Acad. Sci. U.S.A. 2013, DOI: /10.1073/pnas.1220362110. (41) Nansai, K.; Kagawa, S.; Kondo, Y.; Suh, S.; Inaba, R.; Nakajima, K. Improving the completeness of product carbon footprints using a global link input-output model. The case of Japan. Econ. Syst. Res. 2009, 21 (3), 267−290. (42) Nansai, K.; Kagawa, S.; Kondo, Y.; Suh, S. Chapter 8: Simplification of multi-regional input-output structure with a global system boundary. Global link input-output model (GLIO), In The Sustainability Practitioner’s Guide to Multiregional Input-Output Analysis. Common Ground; Murray, J, Lenzen, M., Eds.; IL, 2013a. (43) Nansai, K.; Kagawa, S.; Kondo, Y.; Tohno, S.; Suh, S. Chapter 19: Estimating global environmental impacts of goods and services produced in Japan using a global link input-output model (GLIO). In The Sustainability Practitioner’s Guide to Multiregional Input-Output Analysis. Common Ground; Murray, J, Lenzen, M., Eds.; IL, 2013b. (44) Lenzen, M.; Kanemoto, K.; Moran, D.; Geschke, A. Mapping the structure of the world economy. Environ. Sci. Technol. 2012, 46 (15), 8374−8381.

REFERENCES

(1) Intergovernmental Panel on Climate Change. Climate Change 2007: Synthesis Report; Contribution of Working Groups I, II and III to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change, 2007. (2) United Nations Framework Convention on Climate Change. The Kyoto Protocol, United Nations Framework Convention on Climate Change, 2005; http://unfccc.int/kyoto_protocol/items/2830.php. (3) Peters, G. P.; Hertwich, E. G. Post-Kyoto greenhouse inventories: Production versus consumption. Clim. Change 2008, 86, 51−66. (4) Munksgaard, J.; Pedersen, A. K. CO2 accounts for open economies: Producer or consumer responsibility. Energy Policy 2001, 29, 327−334. (5) Peters, G. P. From production-based to consumption-based national emission inventories. Ecol. Econ. 2008, 65, 13−23. (6) Kanemoto, K.; Moran, D.; Lenzen, M.; Geschke, A. International trade undermines national emission reduction targets: New evidence from air pollution. Global Environ. Change 2014, 24, 52−59. (7) Lenzen, M.; Murray, J.; Sack, F.; Wiedmann, T. Shared producer and consumer responsibility − Theory and practice. Ecol. Econ. 2007, 61, 27−42. (8) Hertwich, E. G. Life cycle approaches to sustainable consumption: A critical review. Environ. Sci. Technol. 2005, 39, 4673−4684. (9) Hertwich, E. G. The life cycle environmental impacts of consumption. Econ. Syst. Res. 2011, 23 (1), 27−47. (10) Nansai, K.; Kagawa, S.; Kondo, Y.; Suh, S.; Nakajima, K.; Inaba, R.; Oshita, Y.; Morimoto, T.; Kawashima, K.; Terakawa, T.; Tohno, S. Characterization of economic requirements for a “carbon-debt-free country. Environ. Sci. Technol. 2012a, 46, 155−163. (11) Munksgaard, J.; Pedersen, K. A. Impact of household consumption on CO2 emissions. Energy Econ. 2000, 22, 423−440. (12) Pachauri, S.; Spreng, D. Direct and indirect energy requirements of households in India. Energy Policy 2002, 30, 511−523. (13) Lenzen, M.; Dey, C.; Foran, B. Energy requirements of Sydney households. Ecol. Econ. 2004, 49, 375−399. (14) Mäenpäa,̈ I.; Siikavirta, H. Greenhouse gases embodied in the international trade and final consumption of Finland: An input− output analysis. Energy Policy 2007, 35, 128−143. (15) Park, H.-C.; Heo, E. The direct and indirect household energy requirements in the Republic of Korea from 1980 to 2000An input−output analysis. Energy Policy 2007, 35, 2839−2851. (16) Webber, C. L.; Matthews, H. S. Quantifying the global and distributional aspects of American household carbon footprint. Ecol. Econ. 2008, 66, 379−391. (17) Kronenberg, T. The impact of demographic change on energy use and greenhouse gas emissions in Germany. Ecol. Econ. 2009, 68, 2637−2645. (18) Liu, H.-T.; Guo, J.-E.; Qian, D.; Xi, Y.-M. Comprehensive evaluation of household indirect energy consumption and impacts of alternative energy policies in China by input−output analysis. Energy Policy 2009, 37, 3194−3204. (19) Druckman, A.; Jackson, T. The carbon footprint of UK households 1990−2004: A socio-economically disaggregated, quasimulti-regional input−output model. Ecol. Econ. 2009, 68, 2066−2077. (20) Druckman, A.; Buck, I.; Hayward, B.; Jackson, T. Time, gender and carbon: A study of the carbon implications of British adults’ use of time. Ecol. Econ. 2012, 84, 153−163. (21) Barrett, J.; Scott, K. Link between climate change mitigation and resource efficiency: A UK case study. Global Environ. Change 2012, 22, 299−307. (22) Chitnis, M.; Druckman, A.; Hunt, C. L.; Jackson, T.; Milne, S. Forecasting scenarios for UK household expenditure and associated GHG emissions: Outlook to 2030. Ecol. Econ. 2012, 84, 129−141. (23) UN, Monthly Bulletin of Statistics, 2012; http://unstats.un.org. (24) JIOT. Input-Output Tables; National Federation of Statistical Associations, Ministry of Internal Affairs and Communications Japan, Tokyo, 2005. 6079

dx.doi.org/10.1021/es404939d | Environ. Sci. Technol. 2014, 48, 6069−6080

Environmental Science & Technology

Policy Analysis

(45) National Institute of Population Social Security Research, Population Statistics of Japan. http://www.ipss.go.jp/p-info/e/ psj2013/PSJ2013.asp. (46) Family Income and Expenditure Survey, 2005; Ministry of Internal Affairs and Communications Japan: Tokyo. (47) National Health and Nutrition Survey, 2005; National Institute of Health and Nutrition (in Japanese). (48) National Institute for Environmental Studies, Greenhouse Gas Inventory Office of Japan. http://www-gio.nies.go.jp/aboutghg/nir/ nir-j.html.

6080

dx.doi.org/10.1021/es404939d | Environ. Sci. Technol. 2014, 48, 6069−6080