1 Introduction

The energy consumption dominated by fossil fuel has been rapidly promoting economic development. Meanwhile, global CO2 emissions are rising, and the environmental issues are becoming worse, which is causing widespread international concern. In addition to natural factors, some scholars ascribe to the belief that human activities are the main cause of global warming; this belief has been supported by much research. Unlimited human activity accounted for 66 and 90 % of the factors leading to global warming, respectively, according to the third and fourth report of the Intergovernmental Panel on Climate Change (IPCC).

For a significant amount of time, energy consumption from industrial sectors was given more attention. Nevertheless, there has been a lack of attention given to household consumption as a driving force of production and the terminal unit of social consumption. Many studies indicate household CO2 emissions contribute more and more side effects to climate warming. Many researchers (Papathanasopoulou 2010; Feng et al. 2011a, b; Liu et al. 2011; Liu and Sweeney 2012; Hamamoto 2013; Jones and Kammen 2014) have confirmed that limiting household CO2 emissions is an effective measure to curb whole carbon emissions.

In 2009, the researchers at Michigan State University concluded that if all US families would take effective actions in energy conservation and emissions reduction, the total CO2 emissions in the USA would be reduced by approximately 7.4 % of the current level after a decade. According to the European Union (EU) statistics, the household energy demand in the EU in the 1990s was more than the industrial energy demand. Wang and Shi (2009) found that the proportion, which China’s household carbon emissions account for the total value, had increased from 19 to 30 % over 1995–2004; this proportion in the USA had already been up to 8 % in 2008 (Dietz et al. 2009). Therefore, there is enormous potential for households to conserve energy and reduce emissions, and future studies should investigate how to make the corresponding policy work toward these goals.

Household energy consumption can be divided into direct and indirect parts, and the latter refers to the carbon emissions that are caused by consuming non-energy commodities and services to satisfy people’s basic needs. However, the entire process consumes much energy. For quantifying indirect CO2 emissions from household consumption activities, input-output (IO) models, life cycle assessment (LCA) and consumer lifestyle approach (CLA) have been used to analyze environmental effects in different countries and regions (Zhang et al. 2015). Based on an IO model, Weber and Perrels (2000) estimated the carbon emissions from living consumption in the Netherlands, Germany, and France; Pachauri and Spreng (2002) made a comparative analysis to all kinds of consumption data from Indian dwellers; Kok et al. (2006) explored the calculation method of household carbon emissions; Lenzen (1998) assessed the effect on energy consumption and Greenhouse Gases (GHG) emissions of Australian dwellers; Munksgaard et al. (2000) estimated the CO2 emissions from 1966 to 1992 in Denmark; Druckman and Jackson (2009) also estimated the carbon footprint of the UK using a quasi-multiregional IO model. Wang et al. (2015) discussed the driving forces of indirect carbon emissions from household consumption in China; Dong and Geng (2012) performed a study of household carbon footprint. Applying hybrid LCA, Zhang et al. (2013) evaluated the carbon reduction potential of a typical household biogas system in rural China; Heinonen and Junnila (2011) proposed a hybrid LCA for calculating the consumption-based carbon footprints in metropolitan areas among 11 cities in Finland; Liu et al. (2007) calculated the electrical carbon emissions of citizens living in 1985 and 1998. Based on the application of a CLA, Wei et al. (2007) quantifies the direct and indirect impact of lifestyle of urban and rural residents on China’s energy use and the related CO2 emissions during the period 1999–2002; Feng et al. (2010) conducted a preliminary calculation for China’s rural and urban household carbon emissions.

China and the USA reflect similar land areas and infrastructure stock demands. In addition, during the past number of years, the sound national economic system and unbalanced regional development of the USA are parallel to China’s current economic situation Therefore, it is necessary to make a comparative analysis on the carbon emissions between the two countries. On the other hand, to clear the responsibility that the two countries should undertake on global carbon emissions reduction, it is necessary to perform a quantitative analysis on the two countries, which can provide a standard for a separation of the responsibility in the future.

Based on the above research results and analyses, there are barely any studies regarding household consumption between China and the USA; therefore, we decided to use an IO method to make a quantitative analysis on household emissions in China and the USA. We expect to provide some scientific support and policy proposals for domestic energy conservation and emissions reduction and future international negotiations in the field of climate change.

The remainder of the paper is organized as follows: Sect. 2 reviews the IO method. Section 3 follows with empirical results of the indirect energy consumption of households in China and the USA along with a related discussion. Conclusions and policy suggestions are presented in Sect. 4.

2 Methodology: Input–Output (IO) model

In order to get the precise results of carbon emissions, it is imperative to apply an effective analysis method, which is an IO model that correlates production and consumption by the total demand coefficient. The IO method that was initiated by Leontief (1936) is used to reflect the quantitative dependency relationship between IO in all of the departments of an economic system. The IO model is given in Eq. (1):

$$X = AX + Y,$$
(1)

where X is the gross output of the national economy, which is the addition of the intermediate utilization AX and the final utilization Y of the gross production value; A = a ij /X j ; a ij is the direct consumption coefficient, which represents the consumption input of sector i for the producing per unit sector j; Y is the final utilization, which includes consumption, investment, and import and export from citizens and government. We transform Eq. (1) into the following coefficient form:

$$X = \left( {I - A} \right)^{ - 1} Y,$$
(2)

where (I − A)−1 is the Leontief inverse matrix in which b ij represents the total demand of sector i of increasing per unit of the final utilization of sectors j. Assume E is the carbon emissions intensity matrix of the unit gross output, so the indirect carbon emissions is

$$C = E(I - A)^{ - 1} Y.$$
(3)

According to Eq. (3), the indirect carbon emissions from household consumption can be adjust into:

$$C_{\text{emb}} = E(I - A)^{ - 1} Y_{d}^{{\prime }} ,$$
(4)

where \(C_{\text{emb}}\) is the implicit carbon emissions from household consumption; E is the carbon emissions intensity of the eight consumption categories. However, due to unavailable energy consumption data of every consumption category, respectively, we adopt the data of additive energy consumption of every industry (ton of standard coal equivalent/tce). A is the 8 × 8 matrix of the direct consumption coefficient by merging the IO table; (I − A)−1 is the Leontief inverse matrix; \(Y_{d}^{{\prime }}\) is the amount of household consumer expenditure in the eight categories.

3 Data sources and processing

The data for the IO table of China’s dwellers were obtained from China Statistical Yearbook, and the US data were provided by the Energy Information Administration (EIA) Web site. According to the industry classification criteria of China Statistical Yearbook and Consumer Expenditure, we made an adjustment and a merger of the industries and sectors to divide household consumption into the following eight categories: Food; Clothing; Residence; Household Facilities and Articles; Education, Culture, and Recreation; Healthcare and Medical Services; Transport and Communications; and Miscellaneous Goods and Services. Each category has several corresponding industries (Tables 1, 2).

Table 1 China’s household consumption categories and merger criteria
Full size table
Table 2 US household consumption categories and merger criteria
Full size table

The IO table was introduced into China in the 1950s, and the first national vision was compiled in 1987. From then on, China had regulated that the IO table would be complied every 5 years. The basic IO table will be compiled in the years ending in 2 or 7, and the IO extended table will be complied in the years ending in 0 or 7. Hence, our paper only calculates the indirect carbon emissions from household consumption between China and the USA in 2002, 2005, 2007, and 2010, respectively.

4 Results and discussion

4.1 Gross household CO2 emissions between China and the USA

According to the above methods, we calculated the indirect carbon emissions from household consumption between China and the USA in 2002, 2005, 2007, and 2010 (Table 3).

Table 3 Indirect carbon emissions from household consumption between China and the USA over 2002–2010
Full size table

As shown in Table 3, USA indirect carbon emissions from household consumption were always higher than the emissions in China over 2002–2010. To be specific, USA carbon emissions were 1 242.15 million ton (Mt) in 2002, which is 3.5 times higher compared to China’s 358.41 Mt in same year. In 2010, the indirect carbon emissions from Chinese household consumption rose by 156.56 % to 919.54 Mt, while the USA indirect carbon emissions of 1 059.23 Mt were 1.5 times higher than those of China. All of the above facts led to China’s consumption level becoming lower than the USA.

As shown in Fig. 1, although US indirect carbon emissions from household consumption have been always higher than that in China, the rapid growth of these emissions in China is apparent. The reason why the indirect emissions in China increased rapidly by 156.56 % over 2002–2010 is that the each expenditure of the consumption categories has a high growth rate. China’s per capita consumption expenditure was 4,144 RMB, while the value in 2010 reached 10,522 RMB, which was an increase by 151.91 %.

Fig. 1
figure 1

Indirect CO2 emissions from household consumption between China and USA over 2002–2010

Full size image

4.2 Comparison of industry structure of indirect household CO2 emissions between China and the USA

According to the accounting results of indirect household carbon emissions in the eight consumption categories between China and the USA over 2002–2010 (Figs. 2, 3), we find that the key industries of carbon emissions are different.

Fig. 2
figure 2

China’s indirect CO2 emissions from household consumption over 2002–2010

Full size image
Fig. 3
figure 3

US indirect CO2 emissions from household consumption over 2002–2010

Full size image

As shown in Fig. 2, most indirect carbon emissions in China over 2002–2010 are a result of Residence, which accounts for nearly 50 % of these emissions and is experiencing an increasing trend. As shown in Fig. 3, the allocation of US carbon emissions among each category is more balanced than the allocation in China. Furthermore, the carbon emissions from Household Facilities and Articles; Education, Culture, and Recreation; Transport and Communications; and Healthcare and Medical Services in USA are higher than those emissions in China.

The differences in the industry structure of indirect carbon emissions between the two countries are caused by the following reasons: (1) Due to the differences in economic development, the living standard of the USA was more than the living standard in China during the study period;

(2) Considering the differences in income and consumption concepts during the study period, China’s consumption mostly focused on satisfying people’s basic needs, especially Residence. For US residents, in ddition to satisfying people’s basic needs, the country was more willing to sharing the income on hedonic consumption, such as Education, Culture, and Recreation; and Transport and Communications.

In terms of Food, the USA maintained 9–10 %, while China decreased from 11.82 to 8.15 % over 2002–2010 (Tables 4, 5), which is consistent with the fact that an increasing consumption level leads to a decreasing Engel coefficient.

Table 4 China’s indirect carbon emissions from household consumption over 2002–2010 Mt
Full size table
Table 5 US indirect carbon emissions from household consumption over 2002–2010 Mt
Full size table

In terms of Household Facilities, although the usage had been rapidly growing over 2002–2010, carbon emissions remained at a relatively steady level. According to data from China Statistical Yearbook in 2002, washing machines, refrigerators, air conditioners, and televisions that every 100 dwellers possess in China’s towns were 92.9, 87.4, 51.1, and 126.4, respectively, while the number of rural dwellers was 31.8, 14.8, 2.3, and 108.6. In 2010, the number of these same items in towns was 96.9, 96.6, 112.1, and 137.4, respectively, and the number in rural areas was 57.5, 37.1, 12.2 and 118.2. The numbers indicate washing machines, refrigerators, air conditions, televisions have increased by 4.31, 10.53, 119.37, and 8.70 % in China’s towns and by 80.19, 205.41, 595.65, and 8.84 % in the rural areas. The above results were due to an energy conservation and emissions reduction policy during the “Eleventh-five-year Plan” in China, which was an energy conservation project, promoting 360 million efficient illuminating products, 30 million air conditioners, and 1 million fuel-efficient cars by fiscal subsidies.

4.3 Comparison of residence carbon emissions between China and the USA

By comparing the Residence carbon emissions between China and the USA, we recognized that the value in the USA remained steady at 400 Mt; however, the value in China increased from 150 Mt to 500 Mt over 2002–2010. US residential carbon emissions were 2.8 times the emissions in China in 2002; however, the value in China has exceeded that in the USA in 2010.

There may be two aspects leading to the difference:

First, the increasing housing demand in China gives rise to the increasing real estate investment and housing area. The per capita housing area in China’s counties increased by 28.68 % from 26.5 to 34.1 m2 over 2002–2010. Real estate investment in China’s towns increased tremendously by 448.54 % from 724.89 billion RMB to 3976.31 billion RMB over 2002–2010. In terms of housing area in China, the housing construction area was 1937.31 million square meters (Mm2) in 2002 (containing 732.09 Mm2 of commercial residential buildings), which increased by 148.17 % to 4807.73 Mm2 in 2010 (containing 3147.60 Mm2 of commercial residential buildings).

Second, residence expenditure items have lead to the enormous amounts of energy demand industries in China. In 2002 the housing energy demand was 262.89 million ton of coal equivalent (Mtce), which accounted for 16.50 % of all of the industries. The same data were 467.89 Mtce and 14.40 % in 2010, which was in the first place of all of the industries. In addition, the corresponding production and supply of electric power, heat power, and gas are in huge demand of coal, and the higher carbon emissions in those industries is an indispensable factor.

5 Conclusions and policy proposals

5.1 Conclusions

By calculating the indirect carbon emissions from household consumption between China and the USA in 2002, 2005, 2007, and 2010, we can make following conclusions:

(1) In terms of the gross indirect carbon emissions from household consumption, the USA has always been at a higher level than China. However, in recent years, China has observed an increasing trend year by year. In contrast, the USA appears to be in a downward trend. (2) US household indirect carbon emissions focus on Residence; Education, Culture, and Recreation; and Transport and Communications, which accounted for 39.5, 15.85, and 17.65 % of indirect carbon emissions, respectively, in 2010. In contrast, China’s Residence accounted for 50 % of indirect carbon emissions, while Education, Culture, and Recreation accounted for 2.28 % of these emissions, and Transport and Communications accounted for 2.48 %, which were lower percentages than the USA. (3) Although the number of household facilities in China is rapidly growing, carbon emissions remain relatively at a steady level. (4) In terms of the absolute value of indirect carbon emissions from housing, the USA remained steady at 400 million ton over 2002–2010, while China increased from 150 to 500 million ton over 2002–2010. US residential carbon emissions were 2.8 times the emissions in China in 2002, while China’s value has exceeded that in the USA in 2010.

China and the USA are in different periods of economic development. The USA had stepped into developed countries, where sound resource allocation and industrial structures existed. Through a long-term industrial society, USA accumulated emissions have been at a higher level. In contrast, China is still a developing country, which stands on a rapid urbanization and industrialization process. Therefore, the household requirements of basic needs; education, culture, and recreation; and healthcare need to be further improved. In addition, China has a larger population than the USA at present. Although, per capita indirect CO2 emissions are much less in China than those in the USA, China has an enormous potential for emissions in the future.

5.2 Policy and proposals

  1. 1.

    Estimate future potential carbon emissions in China, by the US current carbon emission condition: By comparison of the equivalent household consumption period between China and the USA, it is possible to forecast future carbon emissions in China, so the government should be encouraged to put forward solutions ahead of time or set measures to cater to the purpose of reducing emissions.

  2. 2.

    Clear the responsibility for carbon reduction between China and the USA in future climate negotiations: Different economic developments led to different historic accumulated emissions, so it is unreasonable to divide countries’ responsibilities by the same standards. Therefore, potential carbon emissions in developed counties and rigid household demands in developing counties should be taken into consideration.

  3. 3.

    Offer government guidance to residents in choosing a rational housing area: At present, it is a common phenomenon that China’s residents blindly pursue bigger per capita housing areas, which lead to more and more residential carbon emissions. The government should prolong the service life of building and improve the energy-saving design standard of new residences. No doubt that “short-lived building” is mainly caused by external factors (e.g., excessive demolition and construction) and internal factors (e.g., poor property). Therefore, on the basis of guidance in choosing rational housing areas, the government should prolong the housing lifespan for the sake of reducing residential indirect energy consumption.

  4. 4.

    Encourage the development of new technologies: The government should provide support to energy-saving household facilities and green residences, such as financial support for efficient illuminating products and energy-saving air conditioners and televisions.

  5. 5.

    Promote energy-saving and emission-reducing projects in special fields, especially for the energy-intensive production and supply industries: For instance, the government could implement industrial boiler reforms, motor system energy conservation, and surplus heating and pressure usage. For heavy energy-consuming enterprises, the government could reinforce energy-saving management and improve energy efficiency.