Multiplicative Structural Decomposition Analysis Research Articles

Partner heterogeneity and driving factors of China's export embodied energy intensity

Although China's energy consumption is very high, foreign demand has become crucial factor affecting China's energy consumption due to exports. A detailed analysis of the embodied energy intensity of China's exports from the demand side could provide valuable insight into promoting China's energy-saving development. This paper explores disparities in the embodied energy intensity of China's exports to different trading partners and the energy supply structure behind these exports. Furthermore, it explores the driving factors of export embodied energy intensity using multiplicative structural decomposition analysis. The results show that China's export embodied energy intensity to developing countries, such as India, is higher than that of European and American countries; however, a convergence trend has emerged over time. Regarding the driving factors influencing changes in export embodied energy intensity over time, the sectoral energy intensity effect plays a pivotal role in promoting its decline, while the production structure and export structure effects change from inhibition to weak promotion. The adjustment of export structure has significant potential for reducing China's export embodied energy intensity, especially in emerging economies. Finally, this paper proposes policy directions for collaborative opening up, energy saving, and emission reduction goals.

Trade-induced carbon-economic inequality within China: Measurement, sources, and determinants

Carbon-economic inequality (CEI) occurs due to the unequal exchange between embodied CO2 emissions and value-added in interregional trade. How to reduce CEI has become an urgent task to realize the Sustainable Development Goals. In this study, we quantify the CEI in bilateral and regional trade in China using the embodied CO2 emission and value-added during 2012–2017. Then, we analyze the drivers of CEI changes using the multiplicative structural decomposition analysis. Finally, we measure national CEI using an extended Gini coefficient and further identify the sources of inequality using the Shapley decomposition approach. The results indicate that 1) inequalities in bilateral trade mainly occur between the three north regions (North, Northwest, and Northeast) and eastern coastal regions (Beijing–Tianjin, South coast, and Central coast), and the inequalities generally exhibit worsening trends during 2012–2017. In regional trade, the Northwest suffers the worst and widening inequalities, while the Southwest has obtained an expanded advantage. 2) the decomposition results show that coal efficiency and input structure disparities are the main factors that exacerbate CEI in bilateral trade between the three north and Beijing–Tianjin/South coast regions. Improved input structure is the leader in helping the Southwest reduce CEI in bilateral and regional trade. 3) the overall CEI continues to deteriorate, with the Gini coefficient increasing from 0.260 to 0.305 during 2012–2017, where regional gap and intra-regional inequality, coal-induced CO2 emission, and the electricity production industry are the leading causes. This exacerbated CEI calling for optimizing regional trade patterns and providing holistic solutions for sustainable development.

UNDERSTANDING THE CARBON INTENSITY OF SOUTH KOREA’S EXPORTS: A MULTIPLICATIVE STRUCTURAL DECOMPOSITION ANALYSIS

Despite the large and growing role of exports in Korea’s carbon emissions, they have been largely overlooked in Korea’s carbon neutrality strategy. With this in mind, we analyze Korea’s aggregate emission intensity of exports (AEIE), an indicator of the environmental efficiency of exports, which decreased from 1.37[Formula: see text]Kt/$M to 0.74[Formula: see text]Kt/$M during 2000–2014. Using the multi-regional input-output model and two-stage multiplicative structural decomposition analysis (MSDA), we uncover drivers of changes in the AEIE. Analysis of bilateral AEIE shows that exports to developed countries had a large impact on the decline while exports to developing countries have risen in importance. MSDA shows that the carbon intensity effect contributed most to the decline but that trade in intermediate goods and trade in final goods were also important. At the sectoral level, manufacturers of basic metals, electricity, gas, steam and air conditioning supply and transportation sectors were shown to be important drivers of the decline in the AEIE. As the first study to analyze the drivers of Korea’s AEIE, this paper suggests various abatement policies to help achieve the goal of carbon neutrality by 2050.

Structural decomposition analysis applied to energy and emissions: Frameworks for monthly data

Structural decomposition analysis (SDA) is a well-known approach to studying factors contributing to changes of an aggregate indicator in energy and emissions studies. Such studies normally rely on yearly data since input-output (I-O) tables are needed. With energy and economic transitions and seasonal factors, variations in renewable energy supply and in final demands of goods and services are becoming more prominent within a year in many countries. If monthly data are incorporated, some temporal dynamics within a year can be investigated in SDA application. In this paper, we propose an additive SDA framework and a multiplicative SDA framework that include monthly data to respectively reveal the drivers of temporal dynamics associated with energy/emissions embodiments and aggregate embodied intensity indicators. Based on China's 2018 and 2020 I-O tables, an empirical study is conducted using the proposed frameworks. The results obtained show that the increased granularity helps to reveal temporal dynamics mechanisms which will otherwise be overlooked. We discuss the findings and present areas for future research.

Two sides of the same coin: The economic and environmental effects of China's international trade from a global value chain perspective

China's accession to the World Trade Organization (WTO) has led to rapid economic growth and international trade development. However, China is also challenged with a heavy environmental burden due to the massive carbon emissions transferred through trade. By splitting production activities into traditional trade and global value chain (GVC) activities, this paper uses an intercountry input-output (ICIO) framework to study the imbalances of the economic and environmental effects between China's imports and exports at different levels. We define the indices value added per embodied emission in imports and exports (VPM and VPX, respectively). Additionally, we find a large gap between China's VPM and VPX, primarily because developed economies gain much higher value added per embodied emission than China gained through exports in GVC activities. Then, we study how to narrow the gap between China's VPM and VPX. The application of multiplicative structural decomposition analysis (SDA) and the logarithmic mean Divisia index I (LMDI-I) approach reveals the total and bilateral-sectoral contributions of the driving factor effects to the changes in China's VPM and VPX. The results provide tailored implications for promoting the comprehensive economic and environmental benefits of China's imports and exports.

Contributors and drivers of Chinese energy use and intensity from regional and demand perspectives, 2012-2015-2017

China has been experiencing significant energy use transition accompanied by emphasizing on dual control of energy (e.g. energy consumption and energy intensity). Energy use was expected to present obvious spatial characteristics and differentiated contributions across regions. A systematic energy analysis was carried out to 31 regions (including Tibet) in China for investigating the contributors to national achievements and key drivers of regional energy use. With the help of multi-region input-output model (MRIO) and structural decomposition analysis (SDA) techniques, we found that the aggregate energy use in China has been advanced effectively with a declined increasing total energy consumption and accelerated decreasing total aggregate energy intensity. From regional perspective, Eastern region contributed the most characterized by greater total embodied energy consumption and lower aggregate embodied energy intensity, such as Shandong, Hebei and Jiangsu. From demand perspective, regional energy use has been dramatically determined by investment, consumption and exports in sequence. The changes in regional energy use have been tremendously influenced by the sectoral energy intensity and domestic production structure with differentiated performance across regions. According to additive and multiplicative SDA results, two different regional energy conservation paths, i.e. efficiency- and structure-oriented improvements, were identified for China.

Improved granularity in input-output analysis of embodied energy and emissions: The use of monthly data

Input-output (I-O) analysis has been widely used in national energy and energy-related emission studies. These studies are generally conducted using annual data. In a growing number of countries, significant variations in renewable energy supply and in final demands of goods and services are observed over time within a year. These temporal variations cannot be captured in I-O analysis using annual data. To investigate such temporal dynamics, we propose an I-O analysis framework that uses monthly data. Further to that, the drivers in embodiments and aggregate embodied intensity (AEI) indicators are studied via Structural Decomposition Analysis (SDA). Additive SDA and multiplicative SDA are applied to reveal the temporal dynamics associated with energy and emission embodiments and AEI indicators, respectively. An application study using China's 2018 datasets show the importance of temporal dynamics in studying its embodiments and AEI indicators, with drivers of their changes show significant variations over months. It is shown that increased data granularity reveals useful information which would otherwise undetected if annual data are employed. Implications of the findings on future research are discussed.

Contributors and drivers of Shanxi’s aggregate embodied carbon intensity (2002–2017) based on input–output and multiplicative structure decomposition analysis

Success stories of low carbon transition for energy enrichment areas in developing countries are scarce. Shanxi Province is a typical coal resource-based region with the highest carbon intensity in China. Determining the main contributors and drivers of high carbon intensities is a crucial issue for China to achieve carbon neutrality. This study constructed the environment extended input–output model and calculated Shanxi Province’s aggregate embodied carbon intensity (AECI) in 2002, 2007, 2012, and 2017. Then the contributors of AECI can be analyzed at the final demand level and sector level. A multiplicative structural decomposition analysis (SDA) was employed to investigate Shanxi’s AECI change drivers. The results show that outflow to other provinces replaced exports and became the primary contributor to AECI change after 2012. The four sectors, electricity, construction, mining, and metallurgy, contribute mostly to Shanxi’s AECI. The main driver of AECI change is gradually shifting from energy intensity to production structure and final demand factors. The framework and conclusion could be extended to the energy enrichment regions of developing countries to identify the critical action of low-carbon transition.

Structural path and decomposition analysis of aggregate embodied energy intensities in China, 2012-2017

The dependence of the economic development of a region on its energy consumption is a major challenge for developing countries. The aggregate embodied energy intensity (AEI) based on an input-output framework reflects the interaction between regional economy and energy consumption driven by final demand. Studying the change in AEI from the perspective of sector production and final demand can effectively control the source and trend of energy intensity change in order to control the growth of energy intensity. However, there is a relative lack of studies determining the causes of AEI changes from the perspective of sectors and final demands. This study attempted to identify the influencing factors of AEI changes and establish an AEI-based environmental policy in China. A multiplicative structural decomposition analysis and a structural path analysis were used to study the AEI under conditions of different final demands, sectors, and transmission layers, along with the factors influencing AEI changes in China from 2012 to 2017. Our results indicate that the final demand category of gross fixed capital formation made the largest contribution to AEI changes in 2012 and 2017, and the multiplicative intensity factor significantly inhibited the decrease in AEI. Among the sectors studied, only a few were found to significantly affect the change in AEI, with “S11-Processing of petroleum, coking, and nuclear fuel”, “S12-Chemical industry”, and “S14-Smelting and rolling of metals” being the key sectors that input high-energy-consumption products to other sectors.

Linkages-based indicators of production-source sectoral eco-efficiency with application to Polish data

The goal of this paper is twofold. First, a new approach to measuring production-source sectoral eco-efficiency in generalized input-output models is presented. The procedure may be used as a supplementary method to the traditional framework that is usually based on various decomposition techniques. Due to its simplicity and low data requirements, the approach provides a useful and flexible tool that allows a more general insight to be obtained into levels of emissions associated with production, which in turn is essential for constructing more sound indicators of eco-efficiency. Second, in order to trace the factors driving the dynamics of the proposed indicators of eco-efficiency, a two-step procedure based on combination of multiplicative structural decomposition analysis and a new variant of sensitivity analysis based on mathematical programing is outlined. In order to illustrate the potential applications of the new approach, an empirical analysis based on national input-output tables and environmental accounts for Poland is conducted. The empirical results provide solid evidence to claim that the sector Electricity, Gas and Water Supply plays a dominant role in shaping the structure of backward and forward linkages of all sectors of the Polish economy in the CO2-emmision-oriented generalized IO models. As a consequence, lowering the emission-output ratio in this particular sector has a particularly strong impact on improving production-source eco-efficiency of all sectors operating in Polish economy.

Upgrading or downgrading: China's regional carbon emission intensity evolution and its determinants

One of China's major national targets is to environmentally upgrade its economy. In this paper, we define environmental upgrading as lowering the carbon intensity. The disparities among China's regions suggest to examine China's carbon emission performance at the regional level. For this purpose, we use inter-regional input-output tables (for 2002, 2007, and 2012) that distinguish processing exports from ordinary exports. The regional emission intensities (EIs) show environmental downgrading in the period 2002–2007 and upgrading during 2007–2012. To identify the determinants of the evolution of regional EIs, we have employed a multiplicative structural decomposition analysis. Changes in direct emission coefficients and changes in production technology are found to be the major determinants. However, next to these standard determinants, we also evaluate the effects on the changes in regional EIs of changes in inter-regional trade and changes in inter-regional spillovers. Changing inter-regional trade is found to have increased the EI significantly in western and central regions. This suggests that more “dirty” production was shifted from coastal to inland regions. Our study yields clear policy recommendations for achieving China's transformation to a low-carbon economy. • China's regional carbon emission performance experienced “downgrading” first and then “upgrading” during 2002–2012. • There are large regional disparities in the emission intensity (EI) evolution. • China's inter-regional input-output tables that distinguish processing exports are adopted. • Multiplicative structural decomposition analysis (SDA) is used to identify the determinants of China's regional EI changes. • Shifts of “dirty” production from coastal to inland regions have increased the EIs in inland regions.

Diversified carbon intensity under global value chains: A measurement and decomposition analysis

The rise and gradually developing of global value chains (GVC) have changed the essence of international trade, and further complicated the issue of carbon emission. It is of great significance to identify and study the practical carbon intensity in a country under the condition of GVC. Based on Inter-country Input-Output (ICIO) model, this paper establishes a framework for measuring the carbon intensities embodied in aggregated demand and distinguishing pure domestic demand, traditional trade, simple GVC route, complex GVC route at the global level and for 41 economies from 2010 to 2014 according to demand heterogeneity, and uses multiplicative structural decomposition analysis (SDA) method to decompose the driving factors affecting their changes. The results indicate that, in most developed countries, the carbon intensities embodied in international trade were significantly greater than pure domestic carbon intensity, whereas it was opposite in most developing countries. The carbon intensity embodied in traditional trade was generally lower than that in GVC routes, and in low carbon intensity (high carbon intensity) sub-group, the carbon intensity embodied in simple GVC route was lower (higher) than that in complex GVC route. Moreover, the emission coefficient effect was the main driver of the decline in almost all carbon intensities, especially embodied carbon intensity in GVC routes. On the contrary, the intermediate import effect was the main inhibitor, and this effect would cause more rise to embodied carbon intensity in simple GVC route, compared to embodied carbon intensity in complex GVC route.

Multiplicative Structural Decomposition Analysis of Spatial Differences in Energy Intensity among G20 Countries

Comparing the spatial differences in the energy intensity of the Group of Twenty (G20) countries and identifying the factors that influence these differences can help the G20 countries formulate targeted policies to achieve energy conservation goals. This study analyzes the spatial differences in the G20 countries’ energy intensity at the aggregate and sectoral levels based on an input–output framework and reveals its driving factors by employing multiplicative structural decomposition analysis, obtaining the sectoral energy intensity, input structure, and final demand structure effects. The results show that: (1) the gap in aggregate energy intensity among the G20 countries tended to converge from 2000 to 2014 with the reducing energy intensity in Russia, India, China, and South Korea having great potential to reduce global energy consumption and improve global energy efficiency; (2) in 2014, the main driving forces for above-average energy intensity was the sectoral energy intensity effect in India, South Korea, and Canada, the input structure effect in Russia and China, and the final demand structure effect in Indonesia; (3) using the average of the G20 countries as a reference, the energy reduction potential of China, Russia, India, South Korea, Indonesia, and Canada is 62.75, 31.94, 21.24, 7.67, 1.47, and 0.81 exajoules (EJ), respectively. The embodied energy consumption decline in these countries was equivalent to 21.78% of the G20’s total energy consumption in 2014; and (4) the most important factor of the high embodied energy intensity of key sectors in India and South Korea is the sectoral energy intensity effect, while for Russia and China, it is the input structure effect.

China’s aggregate embodied CO2 emission intensity from 2007 to 2012: A multi-region multiplicative structural decomposition analysis

The striking disparities among China’s provinces suggest the need for regionalized carbon emission intensity mitigation measures. This study extends the aggregate embodied energy/emission intensity measures proposed by Su and Ang (2017) to a national multi-region setting, defines the aggregate embodied CO2 emission intensity (AECI) indicators at the aggregate, provincial, and final demand category levels from the demand perspective, and adopts multiplicative structural decomposition analysis (SDA) to identify the driving factors of the historical changes over the 2007–2012 period in China. On the basis of China’s latest multiregion input-output tables, we find that (a) the AECIs show heterogeneities across the provinces and final demand categories, with the AECIs of inland provinces higher than those of coastal provinces and the AECIs of investment and export higher than those of consumption; (b) the generally downward trends of the AECIs hold at different levels, and the improvements in carbon emission performances are more pronounced in inland provinces, presenting a convergence toward the levels of developed coastal provinces; (c) according to the multiplicative SDA results, the input structure effect and the emission intensity effect generally contribute to the decline of AECIs. By providing a deeper understanding of carbon emission performance and its driving factors at the subnational level, our study finally proposes some policy recommendations.

Cross-country analysis of relationship between material input structures and consumption-based CO2 emissions

This study focuses on the effects of changes in material and energy input structure on the consumption-based CO2 emissions. Previous studies demonstrated the connection between domestic structural changes including the shift toward a service economy and the increase in greenhouse gas emissions embodied in consumptions of a specific country. This study is an important follow-up research that examines the environmental effects across countries and evaluates whether or not the development levels of countries can explain those environmental effects. Specifically, we employed a multiplicative structural decomposition analysis based on the World Input–Output Database during 1995–2008 and decomposed consumption-based CO2 emissions of 40 nations into the following four inducement sources: (1) inputs from material goods (including energy) to material goods, (2) inputs from material goods to services, (3) inputs from services to material goods, and (4) inputs from services to services. The results from this paper is helpful for policymakers to identify the target segment of supply chain to reduce the domestic CO2 emissions without lowering the level of the country’s production activities.

Driving forces of national and regional carbon intensity changes in China: Temporal and spatial multiplicative structural decomposition analysis

With the increasing pressure on reducing CO2 emissions, China promised to reduce carbon intensity by 60–65% by 2030 from 2005 levels. This study aims at identifying the driving forces of national and regional carbon intensity changes in China at multiple levels by a newly extended multiplicative structural decomposition analysis. Attribution analysis is further adopted to identify sectors with large intensity-reduction potential. National and regional carbon intensity changes during 2007–2012 are decomposed into three determinants: intensity (or efficiency) effect, input structure effect and final demand effect. Temporal decomposition results suggest that 29.0% decline of national carbon intensity is mainly due to intensity effect, while input structure and final demand effect drive the increment of national carbon intensity. Eight regions are divided into two groups: carbon intensity in Northwest, South Coast and Northeast increased due to input structure and final demand effect; carbon intensity in other regions decreased due to intensity effect and final demand effect. Investment and export are the dominant final demand categories to carbon intensity decline in most regions. Spatial decomposition results reveal the huge contribution discrepancy of driving forces among 30 provinces, and 30 provinces are accordingly classified into four groups. For most regions, simultaneously optimizing input structure and final demand are preferred in sectors with large intensity-reduction potential like Mining, Manufacture, Metals and metal productions and Production and supply of electricity, gas and water. Targeted intensity-reduction strategies at multiple levels are suggested.

Multiplicative structural decomposition and attribution analysis of carbon emission intensity in China, 2002–2012

Abstract Energy/emission intensity indicators measure the relationship between economic development and climate change. These intensity indicators are preferentially used in assessment criteria for mitigation goals in places such as China. This paper investigates the driving factors of changes in the national CO2 emission intensity and their contributions to reductions at the sector level in China from 2002 to 2012 using multiplicative structural decomposition analysis (SDA) and attribution analysis. Both Leontief and Ghosh input-output models are used in the study. The empirical results indicate: 1) the energy intensity effect is the main driver that decreases the aggregate emission intensity from both the demand and supply sides; 2) structural effects, such as the energy structure effect and domestic Ghosh structure effect, promote the increase in aggregate emission intensity principally; and 3) to a large extent, sector “smelting and pressing of metals”, “manufacture of non-metallic mineral products” and “chemical industry” are the top three sectors that contribute to the negative energy intensity effect. Ultimately, several discussions and conclusions associated with the empirical results, result stability and research extensibility are presented.

Multiplicative structural decomposition analysis of aggregate embodied energy and emission intensities

Aggregate intensity indicators, such as the ratio of a country's energy and emissions to its GDP, are often used by researchers and policymakers to study energy and environmental performance. This paper analyzes the relationship between energy (or emissions) and value added (or GDP) from a different viewpoint, namely from the demand rather than the production perspective, using the input–output (I–O) framework. The aggregate embodied intensity (AEI), defined as the ratio of embodied energy (or emissions) to embodied value added, can be defined at the aggregate, final demand category and sectoral levels. The total aggregate intensity can be presented as a weighted sum of the AEIs at the final demand category or sectoral level. Changes of the AEI at different levels can be decomposed to identify the driving forces using multiplicative SDA. A study using the latest 2007 and 2012 datasets of China indicates that (a) its aggregate intensity of CO2 emissions was mainly determined by the AEI in investment and (b) the emission intensity effect generally contributed the most to the AEI ratio changes at different levels. The proposed framework can be applied to other aggregate intensity indicators and extended to multi-country/region analysis.

Multiplicative structural decomposition analysis of energy and emission intensities: Some methodological issues

Structural decomposition analysis (SDA) has been a popular tool for studying a country's energy or emission performance. At the same time, there has been an increasing use of intensity indicators, such as energy consumption and carbon emissions per unit of economic output, in energy and emission performance reporting. The ratio change of an energy or emission intensity indicator can be more conveniently handled in the multiplicative form. In the context of multiplicative SDA of intensity indicators, this study investigates three specific methodological issues. The first is about sub-aggregate decomposition which provides detailed results at the sectoral/regional level to explain the observed aggregate intensity change. The second is the possible linkages between multiplicative SDA and additive SDA when studying changes in an intensity indicator over time. Arising from the convergence between SDA and index decomposition analysis (IDA) in application, the third issue is about the conceptual and empirical linkages between these two decomposition analysis techniques in the multiplicative form. A better understanding of these three issues will help to promote the use of multiplicative SDA of intensity indicators. A case study that looks into China's energy intensity change from 2007 to 2012 is presented.

Multi-region comparisons of emission performance: The structural decomposition analysis approach

Structural decomposition analysis (SDA) has been widely used by researchers to study changes in carbon emissions or aggregate emission intensity over time in a country. These studies may be called temporal-SDA analysis. Similarly, SDA analysis can be conducted by studying variations in carbon emissions or aggregate emission intensity between countries or between regions in a country, i.e. a decomposition analysis conducted spatially. In spatial-SDA analysis, the objective is often to understand the contributions of factors such as emission intensity, Leontief structure, and final demand in explaining the difference in total carbon emissions or aggregate emission intensities between two countries or regions. We review the literature of spatial-SDA analysis and propose a spatial-SDA framework for multi-region comparisons. Both the additive and multiplicative SDA forms are presented in the framework. Using the framework, 30 geographical regions in China are compared and ranked based on their emission performance. This proposed framework can also be used to evaluate other performance indicators in multi-region comparisons.