Inequalities in carbon intensity in China: A multi-scalar and multi-mechanism analysis

Highlights

• A multi-scalar and multi-mechanism analysis framework is used.

• Carbon emission intensity data at multi-scale are estimated by remote sensing.

• We found a trend of decreasing inequality in carbon intensity in China.

• The emission of neighboring cities has a strong effect on a city’s emission dynamics.

• The disparity in China is sensitive to the regional level across multiple mechanisms.

Abstract

This study estimates disparities in carbon intensity in China using a multi-scalar and multi-mechanism analysis. In order to avoid the inconsistency between regional, provincial, and city-level data, city-level CO₂ emissions from energy consumption in China were estimated through Defense Meteorological Satellite Program/Operational Linescan System nighttime light imagery. Our results reveal a trend of decreasing inequality in carbon intensity in China, and the study also found that the contribution made by eastern China to that inequality decreased continuously, while the share of inequality in western China increased consistently during the study period. Spatial Markov chains were also applied to identify the spatiotemporal dynamics of inequalities in Chinese cities. The results show that there is a strong effect of the emission status of neighboring cities’ on a city’s emission dynamics and the effect of self-reinforcing agglomeration was significant. Based on a multi-level model, the study further revealed that the disparity in China’s carbon intensity levels was sensitive to the regional hierarchy across a variety of mechanisms acting as potential influencing factors. We found that technological progress and population density have a potential to mitigate the intensities driven by economic development, trade openness, road density, secondary industry proportion, and investment intensity. Through the present study, we argue that the policies targeting emissions mitigation in China have been restrained due to a lack of effective restraint in relation to the influencing factors that have promoted emissions levels, while mitigation factors have not been adequately exploited.

Introduction

Climate change, the main feature of which is global warming, is increasingly at the center of discussions within an international community that is alarmed by increasing sea levels, drought, receding glaciers, and rising temperatures [1]. According to the IPCC, CO₂ emissions now account for approximately 76.7% of total anthropogenic greenhouse gas emissions. As such, global warming can be largely attributed to the CO₂ emissions stemming from energy consumption aroused by anthropogenic activities [2], [3]. There are increasingly fierce International negotiations on the sharing of CO₂ emission reduction obligations. As largest developing country and second largest economy in the world, China’s CO₂ emissions have rapidly increased in recent years, accounting for 25.43% of the total emissions in world in 2009 [4]. As the largest CO₂ emitter in the world, currently there is a dual challenge in front of the Chinese government reducing CO₂ emissions while simultaneously ensuring economic growth [5]. Shouldering much responsibility, China has made efforts to reduce carbon emissions in a variety of ways, including the commitment made by the Chinese government at the climate change summit in Copenhagen to decrease China’s carbon intensity by 40–45% relative to the 2005 levels by 2020 [6], [7]. In the 13th Five-Year Plan (2016–2020), mitigation targets were formulated to reduce carbon intensity by 18% [8]. To meet this quota, the urgent issue of the way to effectively mitigate carbon intensity in China must be solved.

Carbon intensity, measured in terms of the ratio of total CO₂ emissions to economic scale, can reflect the energy and economic performance more accurately than per capita CO₂ emissions and aggregate CO₂ emissions measures [9]. Carbon intensity is an appropriate way of representing CO₂ emissions in relation to mitigation efforts in developing countries such as China, where sustainable economic development and emission reduction are both of considerable significance. The focus on carbon intensity began afterward the oil crisis in the late 1970s, and research has tended to have concentrated on developed countries, with relatively few studies focusing on developing countries [10], [11], [12]. Existing literature has also mainly been conducted at larger scale, with few attempts being made to address carbon intensity at finer spatial units. For example, Pretis and Roser [13] compared observational records and socioeconomic scenarios for carbon dioxide emission intensity at the global scale. Taking Mexico as their study area, González and Martínez [14] analyzed the effects of energy intensity, carbon coefficient, and carbon intensity structure from 1965 to 2010. In recent years, there have been many articles on the intensity of carbon emissions in China. For example, using structural decomposition analysis and quantile regression, Dong et al. [15] investigated the factors that drive changes in CEI in China. And studies on China’s carbon intensity have also focused on the national scale [15], [16], [17], the regional scale [18], or the provincial scale [19]. Notably, large regional differences exist between China’s coastal and inland areas in terms of resource endowment and economic development [20], [21]. Studies on the spatial distribution of CO₂ emission also shown that CO₂ emissions display commensurate regional characteristics [22], [23], [24]. The spatial differentiation in economic development and CO₂ emissions in China inevitably results in disparities in carbon intensity. It is crucial that the evolution path of carbon inequality is investigated with respect to the multi-scalar structure of China, if China is to achieve its international carbon intensity reduction commitments. However, current research addressing carbon intensity has only considered specific scales, neglecting regional inequality analyses. Only a handful of literature explored spatial inequality of carbon intensity. For example, employing Moran’s I index and a dynamic evolution model, Li et al. [25] estimated spatiotemporal heterogeneity of carbon intensity in China’s construction industry at province level. Yin et al. [26] divided China into eight regions and used the multiregional input-output model and structure decomposition analysis to explore the spatial characteristics of regional carbon intensity and found a difference in decline degree of regional carbon intensity across the country caused by greater differences among region. And above literature were only concerned with the inequality of carbon intensity at the regional level, lacking analysis at other levels. Due to the limitations of previous studies, this study evaluated inequalities in carbon intensity in China at city-level, provincial-level and regional-level, using a multi-scalar analysis.

We note the existence of a body of previous literature addressing the ways in which energy intensity [10], [27], industry structure [14], energy consumption [11], economic growth [28], [29], technology advance [30], [31], and foreign direct investment (FDI) [32], [33] affect carbon intensity. For example, Tsai [34] found that the carbon intensity was mainly influenced by energy intensity, economic growth, proportion of the secondary industry, and fiscal expenditure. However, few of these studies have considered the underlying mechanisms behind inequalities in carbon intensity in relation to carbon intensity reduction targets in China. Better understanding the multiple mechanisms at work in carbon intensity inequality is thus of considerable importance to China’s sustainable economic development and emissions reduction strategies. Knowledge about the ways in which carbon intensity varies in terms of multiple scales and multiple mechanisms is, in other words, critical. This paper therefore employs multi-scalar and multi-mechanism analysis (using Theil’s index, spatial Markov chains, and a multi-level model), in hope of exposing the disparities and underlying mechanisms at work in carbon intensity. To do this, we first explored the evolution path taken by carbon inequality, paying attention to its multi-scalar (inter-regional, provincial, and city-level) structure, rather than the specific scales studied in previous literature. Second, the underlying mechanisms of uneven regional carbon intensity values were examined, providing targeted policy recommendations for low-carbon development. Explicitly, the empirical findings of our study may be helpful for the government in sketching appropriate and directed policies, in order to meet reduction strategy targets and achieve sustainable economic development.

To sum up, numerous studies have measured the pattern characteristics of carbon intensity. However, current research addressing carbon intensity has mainly been conducted at specific scales, neglecting regional inequality analyses. Only a handful of literature explored spatial inequality of carbon intensity but those literatures were only concerned at the regional level, lacking analysis at other levels. Meanwhile, little attention has been paid to the underlying mechanisms behind inequalities in carbon intensity. Therefore, the purpose of this study is to focus on the above gaps in the research field and the contribution of this paper mainly include the following aspects: first, this study is investigated evolution path of carbon inequality with respect to the multi-scalar structure of China rather than the specific scales studied in previous literature, and the empirical findings may be helpful for China to achieve its international carbon intensity reduction commitments. Second, since previous studies has investigated the driving forces of carbon intensity but little attention has been paid to the underlying mechanisms behind inequalities in carbon intensity. This study also examined the underlying mechanisms of uneven regional carbon intensity values. Therefore, this paper will contribute to studies related to the inequalities in carbon intensity.

In this study, three geographical scales are considered—namely, inter-regional, provincial, and city scales. The Theil index and its decomposition techniques were employed to outline the inequalities present in carbon intensity in China. Spatial and non-spatial Markov Chain was then implemented in order to unfold the spatiotemporal dimensions of carbon intensity inequalities in China at the city level. Markov chains are widely used to analyze spatiotemporal dynamics in the distribution pattern of regional inequality [35]. By comparing the corresponding elements in spatial and non-spatial probability transition matrixes, we were able to understand the relationship between the probability of the carbon intensity of a city moving up or down, as well as that of its surrounding neighbors, and explore whether there a spatial spillover effect existed. Further, a multi-mechanism analysis was conducted, whereby seven potential influencing factors were explored in relation to their roles as underlying mechanisms of uneven carbon intensity in China. Using a multi-level spatial model, this study was able to reveal the relative importance of these determinants over space and time. The remainder of this study is organized as follows. The study areas were briefly introduced in Section 2. Then, the methodology and data were explained in section 3, which consists of Theil’s index, spatial Markov chains, and multi-level model. Section 4 presents the findings and interpretation. Conclusions are given in Section 5.