Spatial and temporal heterogeneity of the impact of per capita income on household indirect carbon emissions in western China

ZHAO Chun-yan ,FU Wei ,* ,LUO Ming-can ,CHEN Jian-cheng

1 School of Economics and Management,Southwest Forestry University,Kunming 650224,CHINA

2 School of Economics and Management,Beijing Forestry University,Beijing 100083,CHINA

Abstract: With China entering the stage of high-quality development,the issue of carbon emission has become a hot research topic.This paper analyzes the different temporal and spatial effects of per capita income on household indirect carbon emissions in western China.Based on the data of Chinese Family Panel Studies (CFPS) in 2016 and 2018 in the western China,this paper uses Regression analysis and Bayesian correlation analysis to study the relationship between per capita income and household indirect carbon emissions.The results showed that the indirect carbon emissions generated by the expenditure on food,housing and household equipment in the household consumption structure in the western China were relatively high.In 2016-2018,the per capita income and per capita household consumption indirect carbon emissions in the western China showed an increasing trend.There was a positive correlation between per capita income and indirect carbon emissions of per capita household consumption,and its correlation was gradually enhanced in time dimension.In the spatial dimension,the household indirect carbon emissions in Yunnan,Qinghai,Guangxi Zhuang and Ningxia in the western China were greatly affected by per capita income,while the household indirect carbon emissions in Guizhou was least affected by per capita income.Finally,the paper puts forward some problems that we should consider in the process of facing the per capita income growth and climate change:the collection of carbon tax,the optimization of household consumption structure,the research and development of low-carbon products,and the differentiated carbon reduction.

Key words: per capita income;household indirect carbon emissions;spatial and temporal heterogeneity analysis

1 Introduction

The rise in global temperature caused by the increase in carbon emissions since the Industrial Revolution has attracted wide attention.Global temperatures are expected to rise by 1℃compared with pre-industrial temperatures,and 1.5℃between 2030 and 2052 with this rate continues(Liu et al.,2020).By 2100,the earth will have warmed by about 3.2℃compared to pre-industrial levels (Raftery et al.,2017).Studies have shown that the proportion of energy consumed and carbon emissions generated by households in some countries is still relatively high(Zhang et al.,2019),such as 80%in the United States(Bin et al.,2005),75%in India(Pachauri et al.,2002)and 40%in Japan(Long et al.,2018;Long et al.,2018)which is lower than 61%in 2012(Nansai et al.,2012)and 52%in South Korea(Park et al.,2007),China’s share of the category were between 30%and 40%(Wei et al.,2007;Liu et al.,2011).China’s national policy paid attention to the development of urbanization,and pointed out that speeding up the construction of domestic consumption policy with domestic big cycle as the main body (Xia et al.,2019),China’s carbon emissions were affected by the increase of household consumption(Dai et al.,2012).Although the total carbon emissions in the eastern and central regions of China were obviously higher than those in the western China,in 2000-2015 the growth rate of indirect carbon emissions from household consumption in the western China exceeded that in the central region by 207.1% (Cheng et al.,2021).Therefore,exploring the spatial and temporal heterogeneity analysis of household indirect carbon emissions in western China from the per capita income of consumers is helpful to reduce the carbon emissions caused by household energy consumption,and has certain practical significance for China to achieve the goal of “double carbon” and promote the green and high-quality development of society.

The carbon emissions of household consumption consists of direct carbon emission and indirect carbon emission.The carbon emissions generated by the consumption of electricity,coal and gasoline in household life become direct carbon emissions (Guo et al.,2018),and indirect carbon emissions refers to the carbon emission caused by the manufacturing process of non-energy products and services consumed in daily life,such as clothes and food (Tian et al.,2016).Some studies have concluded that the indirect carbon emissions generated by the use of non-energy products and services in households are much higher than the direct carbon emissions generated by energy consumption(Ottelin et al.,2018).At present,there are three main methods to calculate household indirect carbon emissions,namely,input-output method,consumer lifestyle method and life cycle evaluation method.Input-output method can calculate the carbon emissions generated by household products and services in the manufacturing process from macro and micro levels,and quantitatively reflect the relationship between input and output.Consumer Lifestyle Approach(CLA)is based on understanding consumers,and adds carbon emission coefficient to the input-output model to calculate indirect carbon emission.Life Cycle Assessment(LCA)is mainly to analyze the input and output data during the whole manufacturing cycle,and then calculate carbon emissions(Wang et al.,2019).

At present,scholars have a novel research perspective on household carbon emissions,and have carried out research from many aspects.The rapid development of the Internet has not only boosted the growth of the global economy,but also changed the consumption behavior of people in China into a more enjoyable consumption mode.Chen et al.(2023)made an empirical analysis of the provincial panel data in China,and concluded that the development of the Internet increased carbon emissions by promoting consumption upgrading,and the effect was remarkable in eastern China.Qing et al.(2022) explored the relationship between digital finance and household carbon emissions in China,and reached the consumers that digital finance can increase household carbon emissions by increasing the scale of consumers,but also reduced household carbon emissions by promoting consumers’ green consumption behavior.Some scholars believed that the relationship between digital economy and carbon emissions was inverted U-shaped (Li et al.,2022).Some scholars have also discussed the relationship between population aging and household carbon emissions,and found that population aging will reduce carbon emissions to a certain extent (Kim et al.,2020;Fan et al.,2021),but some studies have also shown that population aging will increase carbon emissions (Zhang et al.,2016).Scholars’ researched on the relationship between income and carbon emissions was complicated (Wang et al.,2021),Hatziggeorgiou et al.(2011) through the research on the relationship between Greek GDP and carbon emissions from 1977 to 2007,it was concluded that the increase of per capita GDP will promote CO2emissions,and Liu et al.(2020)through the analysis of families in China,it was concluded that with the increase of per capita income of a family.Although the factor of household income inequality still has a positive impact on household carbon emissions,its impact intensity is gradually weakening.Cheng et al.(2021)obtained through empirical analysis that the income inequality of 30 provinces in China has a significant role in promoting direct carbon emissions,but the impact on indirect carbon emissions was not significant.Liu et al.(2019)obtained that the indirect carbon emissions generated by rich families in cities were higher than those of poor families,and more than 58%of the indirect carbon emissions in cities were from wealthy people.

With the ecological environment being increasingly damaged and carbon emissions increasing,it is of practical significance to explore the temporal and spatial heterogeneity of the impact of per capita income on household indirect carbon emissions in the western China of China,so as to slow down the growth of the total amount of indirect carbon emissions in the western China and promote the high-quality development of the western China.This paper mainly studied the relationship between per capita income and per capita household indirect carbon emissions in western China through regression analysis and correlation analysis,and understands the characteristics of indirect carbon emissions in western China and the differences of indirect carbon emissions in different provinces and regions through analysis.

2 Materials and methods

2.1 The research area

This paper studies 11 provinces in western China,including Yunnan,Sichuan,Shaanxi,Guizhou,Chongqing,Qinghai,Gansu,Ningxia,Inner Mongolia,Guangxi,and Xinjiang.The total area of the study area is 5639400 square kilometers,accounting for 58.74% of the total land area of China.The economic development of the western China has significantly improved,with its GDP rising from 1.65 trillion yuan in 2000 to 23.76 trillion yuan in 2021,and its share in the national GDP rising from 17.01 percent to 20.78 percent,and the population of the western China in 2021 is 379 million,accounting for 26.84%of the total population of the country.

2.2 Data source

The data source is mainly divided into two parts.Firstly,the data of eight types of consumption related to indirect carbon emissions of household consumption and the data of consumers’ per capita income are all from the database of Chinese Family Panel Studies(CFPS) in 2016 and 2018.The data collected by CFPS includes 31 provinces,autonomous regions and municipalities directly under the Central Government of China (excluding Hong Kong,Macao and Taiwan),which has a wide coverage,high credibility and high explanatory power.In this paper,the data content of 11 provinces in the western China was selected as the research objects,and 3725 urban and rural families were selected as the research objects by eliminating the samples with serious data deficiency.The specific data content is shown in Table 1.

Table 1 Number of Samples in western China of year 2016 and 2018

2.3 Research methods

In this paper,after screening,calculating and sorting out the corresponding data in the database of Chinese Family Panel Studies (CFPS) in 2016 and 2018,taking the annual per capita income in the western China as the independent variable,taking the per capita household indirect carbon emissions in the western China and the per capita indirect carbon emissions in eight types of consumption expenditure as the dependent variables,we explored the relationship among the variables by using the method of regression analysis and Bayesian correlation analysis,and analyzed the relationship between the variables in time perspective.

CLA method is used to calculate the indirect carbon emission of household consumption,that is,the product of each consumption expenditure and its corresponding carbon emission coefficient is summed up.The specific calculation formula is shown in Formula (1),and the indirect carbon emission coefficient refers to the data obtained by Liu et al.(2012)in the research process,and the specific data is shown in Table 2.

Table 2 Household indirect carbon emission coefficient

Among them,EiCis the indirect carbon emission of households,IiCis the amount corresponding to each kind of household consumption expenditure in yuan,andCiCis the carbon emission coefficient corresponding to various consumption expenditures in kg CO2/yuan.

3 Results

3.1 Analysis of indirect carbon emission structure of household consumption

In this paper,the proportion of indirect carbon emissions generated by eight kinds of consumer expenditures to the total indirect carbon emissions of households was integrated.The specific data results are shown in Table 3,and Figure 1 is derived according to the data in Table 3.As can be seen from Table 3,among the indirect carbon emissions from household consumption in western China,the proportion of carbon emissions from food,housing and household equipment is relatively high,accounting for more than 18%,followed by indirect carbon emissions from cultural,educational,entertainment,medical care and transportation and communication,accounting for more than 10%,in the end,the indirect carbon emissions of household consumption expenditure account for a small proportion of clothing and other expenditure types,accounting for less than 5%.Compared with 2016,the proportion of indirect carbon emissions from food,clothing,transportation,communication,culture education and entertainment consumption expenditure increased in 2018,while the proportion of indirect carbon emissions from consumption expenditure of housing and household equipment decreased in 2018 compared with 2016,while the proportion of indirect carbon emissions from medical care and other consumption expenditures remained unchanged.

Figure 1 Indirect carbon emission structure of household consumption in western China

Table 3 Indirect carbon emission structure of household consumption in western China,%

3.2 Temporal difference of indirect carbon emissions per capita income and per capita household consumption

With the implementation of the national western development policy and the development of the“the belt and road initiative”,the economic development of the whole western China has been promoted.From 2016 to 2018,the per capita income in western China showed an overall rising trend,the annual per capita income increased from 14577 yuan to 18840 yuan.With the rapid economic development in the western China,the total amount of carbon emissions in the western China has also increased,and the corresponding indirect carbon emissions of household consumption in the western China are also increasing.The indirect carbon emissions of household consumption per capita in the western China are in a state of growth from 2002 kg CO2to 2239 kg CO2in 2016-2018,and the growth rate is relatively slow.After 2016,with the increase of per capita income,the growth rate of per capita household indirect carbon emissions in the western China has slowed down,and the annual growth rate fluctuates around 1.1.

3.3 Temporal perspective of the impact of per capita income on household indirect carbon emissions in the western China

Correlation analysis of the variables studied can better explore the influence relationship between variables.In this paper,Pearson correlation coefficient is obtained by Bayesian correlation analysis,and the change of influence intensity between two variables is analyzed from the time perspective through the change of Pearson correlation coefficient.In this paper,we first explore the overall impact of the change of per capita income on the total indirect carbon emissions of per capita household consumption from 2016 to 2018,and then study the relationship between per capita income and the indirect carbon emissions of household consumption corresponding to eight different consumption expenditures.

As can be seen from Table 4,on the whole,there is a significant correlation between the per capita income and the per capita total indirect carbon emissions of household consumption in the western China from 2016 to 2018,and there is a positive correlation between them,indicating that the higher the per capita income,the higher the direct carbon emissions of households.Although there is always a positive correlation between the two variables,the correlation intensity has changed from a time perspective.In 2016,the correlation intensity between the variables was 0.209,and in 2018,the correlation intensity between the two variables was 0.379,indicating that the increase of per capita income promoted the increase of household indirect carbon emissions to a certain extent,and the impact of this per capita income on household indirect carbon emissions was gradually enhanced.

Table 4 Analysis results of correlation between per capita income and total indirect carbon emissions of per capita household consumption

As can be seen from Table 5,the per capita income in the western China has a positive effect on the carbon emissions generated by eight kinds of consumption expenditures,such as food,clothing,housing,household equipment,medical care,transportation and communication,culture education and entertainment,and others,but the promotion intensity of carbon emissions corresponding to different consumption expenditure is different.On the whole,from the perspective of time,the indirect carbon emissions corresponding to the three types of household consumption expenditures of food,clothing,transportation and communication between 2016 and 2018 are greatly influenced by per capita income,and regression coefficient Beta has been ranked in the top three.Indirect carbon emissions from household consumption expenditure of housing,household equipment,medical care,culture and education and entertainment,and other five types are also affected by per capita income,but the impact is relatively small compares with that of food,clothing,transportation and communication,especially the indirect carbon emissions corresponding to the consumption expenditure of medical care and other this two aspects are very little affected by the change of per capita income.Between 2016 and 2018,the regression coefficients corresponding to these two kinds of consumption expenditures are all below 0.16.

Table 5 Regression analysis results of per capita income and indirect carbon emissions of 8 types of per capita household consumption

From the time point of view,the positive promotion effects of the increase of per capita income on the indirect carbon emissions of food,clothing,housing,household equipment,medical care,transportation and communication,culture education and entertainment,and other eight kinds of consumption expenditures all show increasing trend,which shows that the increase of per capita income in the western China will still bring the price of increasing household indirect carbon emissions and increasing environmental burden.However,from promotion degree of per capita income to the indirect carbon emission of household equipment has increased,the change of its promotion degree is relatively small among the eight categories of consumption expenditure.The reason is that after 2014,China began to lay out special vehicles for new energy,and its purchase price is more affordable than that of ordinary cars while reducing the consumption of petroleum energy,and the research and development of low-carbon and environmentally-friendly green household appliances in China has never stopped.The indirect carbon emissions generated by food,clothing,culture education and entertainment,transportation and communication are still greatly affected by per capita income,which shows that we need to strengthen low-carbon propaganda and make everyone realize that low-carbon behavior is not only to buy green energy cars and green household appliances,but also it save food and recycle waste clothes.

3.4 Spatial perspective: the impact of per capita income on household indirect carbon emissions in the western China

The economic development in western China is rapid,but the economic development among difference provinces is uneven,and the relationship between per capita income and household indirect carbon emissions in different provinces will be different.Therefore,it is necessary to discuss the influence of per capita income on household indirect carbon emissions according to provinces.The relevant results of the discussion by province are shown in Table 6.The results show that the impact of per capita income on household indirect carbon emissions has obvious differences among provinces.The per capita income in the western China has a promoting effect on household indirect carbon emissions,and the promotion degree is different in difference provinces.In Guizhou,the correlation coefficient between per capita income and household indirect carbon emissions is 0.176,and the increase of per capita income has shown the effect of increasing household indirect carbon emissions,but the force is relatively small,indicating that the province pays more attention to low-carbon living consumption while increasing per capita income;The regression coefficients between the two variables in Sichuan,Chongqing,Shaanxi,Gansu,Inner Mongolia and Xinjiang are all between 0.4 and 0.6,indicating that the household indirect carbon emissions in these provinces are greatly affected by the increase of per capita income;The regression coefficient of Yunnan,Qinghai,Guangxi and Ningxia all exceeds 0.6,which means that household indirect carbon emissions in these provinces are extremely vulnerable to per capita income.

Table 6 Regression analysis results of per capita income and per capita household indirect carbon emissions in various provinces

Based on the above analysis of the impact of per capita income on household indirect carbon emissions in different provinces,the classification results of household indirect carbon emissions in western provinces in 2016 and 2018 are drawn.The per capita household indirect carbon emissions are divided into five levels: below 1500 is a lower level,1500-2000 is a low level,2000-2500 is a intermediate level,2500-3000 is a high level,and greater than 3000 is a higher level.As can be seen from Table 7,the per capita household indirect carbon emissions in Guangxi have been at a relatively low level,below 1500 kg CO2/yuan.The per capita household indirect carbon emissions in Qinghai province and Xinjiang are above 3,000 kg CO2/yuan,which is at a relatively high level.The per capita household indirect carbon emissions in Inner Mongolia decreased from above 3000 kg CO2/yuan in 2016 to between 2000-2500kg CO2/yuan in 2018,which means that the per capita household indirect carbon emissions in Inner Mongolia have a remarkable low-carbon environmental protection effect,while those in other provinces are between 1500-3500.

Table 7 The per capita household indirect carbon emissions level of provinces

4 Discussion

The implementation of the large-scale development policy in the western China has promoted the rapid economic development,but it has also brought a series of environmental changes.Based on the relevant data of CFPS in 2016 and 2018,this paper analyzes the proportion of household indirect carbon emissions in the western China,and studies the temporal and spatial heterogeneity of the relationship between per capita income and household indirect carbon emissions.The proportion of indirect carbon emissions in households is the same as that of Liu et al.(2019),that is,food and housing are the two main sources of indirect carbon emissions in households,while the proportion of cultural,educational,entertainment,medical care and transportation and communication is higher than that of Liu's research,which verifies that Liu mentioned in the article that"people are more willing to spend money on these projects to improve their living standards"and obtained"the indirectness of families with different incomes.”

With regard to the exploration of the influencing factors of household indirect carbon emissions,scholars have studied the carbon intensity,population size and consumption structure,and concluded that the expansion of carbon intensity and population size will increase carbon emissions,and the change of consumption structure will lead to the decline of household indirect carbon emissions,with the conclusion that the decline of the proportion of food and housing has the greatest contribution to the reduction of carbon emissions (Liu et al.,2019) and Cheng et al.(2021) used the method of quantile regression to explore the impact of income inequality on household indirect carbon emissions.In this paper,the per capita income is not classified in the research process,and the influence of family development cycle,education level and other factors is not considered,so we will consider the influence of more factors on household indirect carbon emissions in future research.

5 Conclusions and recommendations

This paper explores the temporal and spatial heterogeneity of the impact of per capita income on indirect carbon emissions of household consumption in western China from 2016 to 2018,and explores the relationship between the two variables by regression and Bayesian correlation analysis.The findings are as follows: (1) From the analysis of the structure of household indirect carbon emissions,the proportion of carbon emissions corresponding to food,housing and household equipment is relatively high.(2) During 2016-2018,the per capita income in the western China showed a state of rapid growth,and the indirect carbon emissions of household consumption in the western China also increased year by year.(3) There is a positive correlation between per capita income and per capita household indirect carbon emissions in western China,and from the time dimension analysis,with the increase of per capita income in the western China,household indirect carbon emissions also increase.Moreover,the indirect carbon emissions generated by the eight types of consumption expenditure are gradually affected by the per capita income,but the change of the increase intensity is different,and the change of the increase intensity of carbon emissions from household equipment is the smallest.(4) From the spatial dimension,there are differences in the impact of per capita income on household indirect carbon emissions in different provinces in the western China,among which Yunnan,Qinghai,Guangxi and Ningxia are the most affected by per capita income,while Guizhou is the least affected by per capita income.

Therefore,in the face of per capita income growth and climate change,we should consider the following issues:

First,it is necessary to levy a carbon tax.As China's economy enters the stage of high-quality development,our government attaches great importance to the environmental pollution caused by development and actively adopts carbon emission reduction measures.Effective carbon emission reduction requires not only effective actions of the government,but also the joint efforts of the whole society.As the higher the income,the more indirect carbon emissions are generated.Therefore,it is suggested that a certain proportion of carbon tax can be levied on different income groups to reduce carbon emissions to a certain extent.

Second,to improve the household consumption structure.The improvement of income level ensures the improvement of people's quality of life,but at the same time,family consumption habits should be adjusted appropriately to achieve a green and low-carbon transformation in life and achieve a win-win situation in life and low carbon.For example,to consider buying green and energy-saving household equipment,choosing the way of green travel,giving priority to the purchase of new energy products,and reducing the purchase and use of products or services that bring high carbon emissions.

Third,to strengthen the research and development of low-carbon emission products.The indirect carbon emissions generated by residential consumption expenditure are growing rapidly,so the development of low-carbon emissions and energy-saving buildings should be encouraged.To support the development and use of new energy sources such as wind energy,geothermal energy and solar energy to reduce the emission of greenhouse gases such as carbon dioxide.

Fourth,differential reduction of carbon is necessary.Due to the differences of regional economic development and regional cultural differences,the carbon emissions of different provinces are affected by these factors.Therefore,effective low-carbon emission reduction measures should be taken according to regional characteristics to improve the efficiency of low-carbon development.

Acknowledgments

This research is supported by the National Natural Science Foundation of China (Grant No.72264035).