Does the Participation Degree in Global Value Chains Influence Carbon Emission Transfer Through International Trade in Belt and Road Countries?

Abstract

Studying the impact of global value chains’ (GVCs) participation degrees on carbon emission transfer through international trade (CTIT) in Belt and Road Initiative (BRI) economies is of great significance because these economies are significant participants of GVCs and international trade. The current study, through the inter-regional Input-Output table, calculated the GVCs’ participation degrees (forward and backward participation) and CTIT (carbon emission transfer through export (ETET) and emission transfer through import [ETIT] trade) of 27 BRI economies from 2005 to 2018 and investigated the impact of the GVCs on CTIT. Several test results illustrated that endogenous issues did not affect the robustness of study discussions. The study articulates appropriate environmental governance policies that could realise emissions reduction goals. Significant results are (a) participation degree in GVCs increases the CTIT in BRI; (b) energy intensity, energy structure, final demand and secondary industry escalate CTIT; (c) the optimisations of participation degree in GVCs, energy intensity development, industrial structure optimisation and increased awareness of emission lessening among the BRI community could compensate for the growth in CTIT from the constant deepening of GVCs. This study delivers a comprehensive insight into understanding the driving forces that cause the changes in CTIT from the GVCs’ perspective.

JEL Codes: F02, F18, Q56

Keywords

Global value chains Belt and Road Initiative carbon emission transfer through international trade industrial structure energy intensity

Introduction

Devotedly reducing the emission of carbon to tickle the environmental degradation issues has become a fundamental goal for all economies across the globe. The global carbon mitigation agreements’ economic cost would increase a significant financial burden on several economies. More crucially, the close trading partnership has led to economic linkages among regions, though several regions might evade the task of their carbon reduction and circumvent emission reduction responsibilities (Zhangqi et al., 2021). Since the 1980s, multinational firms profit-seeking behaviour prompted them to allocate resources. As a result, the production process of enterprises has been divided into diverse ‘processes’, resulting in the creation of global value chains (GVCs) through fragmentations of the production process (Alfaro et al., 2019). According to Shankar et al. (2020), the comparative advantage concept states that with the growth of globalisation, products and goods are dispersed to several regions, from development, design and manufacturing, to marketing and services after the sale (Shankar et al., 2020). As a result, the developed regions tend to occupy the high ends of GVCs and carry high technology. At the same time, the case is reversed for developing and emerging economies. GVCs offer a channel for the flow of energy and technology resources between high-income, emerging and low-income economies and the rapid development of economic activities globally accompanied by increased foreign trade scale (Steingress et al., 2019). These fresh waves of globalisation illustrate the accelerated expansion of GVCs. This trend of GVCs’ expansions may carry opportunities to transfer resources and technology to various emerging and developing economies. But it can also transfer carbon emissions through international trade; with this, there could be a necessity to establish target strategies and guidelines on a regional and global basis to tickle adverse climate change caused by carbon emission transfer through international trade (CTIT).

In recent times, specific regional cooperation and especially the developing nations’ economic cooperation have become a vital driver in global environmental governance (Meng et al., 2018). In such cases, in 2013, China initiated the Belt and Road Initiative (BRI) project to expand the GVCs’ participation and encourage high efficiency in distributing resources. BRI nations make up over 68% of the world population and 30% of the total world GDP. Figure 1 shows that 73% of BRI’s economies are in developing and emerging countries and 27% are in developed countries (Nedopil, 2022). The BRI projects would accelerate energy use in upcoming years, and the energy structures of these economies are mainly composed of non-renewable energy sources, that is, oil, coal and gas (Rehman et al., 2019).

Along with that, several BRI economies have a system of imperfect pollution control, and their trade and financial operations are bound to display crucial challenges to the ecological system of the host country (Rehman et al., 2019). Moreover, the worldwide carbon emissions are linked with GVCs’ participation degree and international trade, as the consumption in several developed economies has escalated, with a significant portion of carbon originating from underdeveloped and developing economies (Shakib et al., 2021). The reason for selecting the Belt and Road countries for analysis is that the GVCs and international trade can benefit the BRI economies by increasing their participation in international trade and GVCs. But on the other end, participation in GVCs will increase the CTIT. With this, it is much needed to evaluate the role of GVCs’ participation degrees in carbon ETIT and ETET from the Belt and Road perspective.

With the development of the BRI project, the trade of exports and imports would declare the critical growth in CTIT of the economies along the Belt and Road. Moreover, the net trade volume along the BRI was USD 13.35 trillion in the year 2019 (UN COMTRADE), ‘accounting for about 40% of the total global trade volume, which makes Belt and Road countries a vital part of the global trade market’ (Changjian et al., 2021). Consequently, as the BRI region is an integral part of the current booming growth in international trade integration and GVCs, the emission transfer through international trade in the production process is also progressively evident. And most of the BRI economies, being at the lower end of GVCs, would also accelerate CTIT on an annual basis. As mentioned by Zhong et al. (2020) that ‘the global carbon leakage (including emissions transfer through exports trade, ETET for short and emissions transfer through imports trade, ETIT for short) keeps increasing with an annual growth of 5%’. Consequently, the massive economic benefits from GVCs’ participation and international trade could lead to adverse environmental impacts in the shape of carbon ETET and ETIT trade in Belt and Road countries.

In the above context, the BRI region’s CTIT, caused by the local and mainly foreign demand of individual country, has not fascinated the wide-ranging attention of scholars and government divisions. Consequently, the principal aim of conservation of energy and carbon emission reductions in a region and an economy might no longer be restricted to domestic consumption and manufacturing (Meng et al., 2018). Accordingly, from the perspective of CTIT influence on regional reduction obligations and emission accounting assignments in BRI projects, consolidating targeted governance and management in the CTIT is crucial in attaining the carbon reduction target in the BRI project. As a result, the study’s main objective is to study the impact of GVCs’ participation degrees on CTIT of BRI countries (27) from 2005 to 2016. And handsomely devise precise and beset ecological governance guidelines and policies from the BRI outlook to attain the goal of reducing CTIT along the GVCs participation.

Specifically, to enable the BRI project’s stakeholders to devise better environmental governance policies and measures, based on structuring a methodology, the current study calculates the GVCs’ participation degrees for Belt and Road countries based on the Input-Output database. Moreover, the present article fills the research gaps by organising a panel data methodology to vigorously inspect the influence of GVCs’ participation degrees (forward and backward participation degrees) on CTIT (export and import trade) in 27 BRI economies from 2005 to 2018, and also devise a policy concerning climate change and carbon reductions in the BRI region. From the academic perspective, the current study will deliver fresh insight to increase our understanding of the factors of the CTIT by devising long-term ecological governance measures at the Belt and Road level from the perspective of practicality.

The remaining study is arranged as follows: The second section comprises the methods and data. The third section holds the empirical results and discusses the results. The fourth section illustrates the summary contribution of the current study and advises policy recommendations.

Methods and Data

Total GVCs

Following the work of Koopman et al. (2014), the crucial independent regressor of GVCs’ participation degrees can be elaborated as below:

Forward participation degrees in GVCs it = \frac{DVA embodied in foreign exports it}{gross exports it} = \frac{DVA PETCit}{Goss exports it}

(1)

\begin{array}{l} Backward participation degress in GVCs it = \frac{FVAit}{Gross exports} \\ = \frac{(FVA final goods + FVA FCIC + FVA PEEC + FVA PETC) it}{Gross exports} \end{array}

(2)

\begin{array}{l} Participation degress in GVCit = Forward participation degrees in GVC it + \\ Backward participation degrees in GVCs it \\ = \frac{DVA final goods + FVA final goods + FVA (FCIC) + FVA (PEEC) + FVA (PETC) it}{Gross exports} \end{array}

(3)

Wang et al. (2019) mentioned that a country’s position in GVCs can be demonstrated by its backward and forward participation degrees in GVCs. As Equation (1) indicated, the former calculates the total domestic value added (DVA) that has been exported to the third economy as a quantity of total gross exports. Moreover, a high value expresses that more total DVA of a single economy has been carried down to the rest of the economies, illustrating a compact spot in the upper stream of GVCs. The latter indicates the portion of total foreign value added (FVA) in total gross exports (Equation (2)), a greater value of which suggests that FVA reflects a higher proportion of total gross exports. While FCIC is ‘finished and consumed goods in the importing country’, PEEC is ‘goods that are processed and exported back to the exporting country’ and PETC is ‘goods processed and exported to a third country’. In such scenario, an economy captures a powerful position in the lower stream of GVCs. Furthermore, the sum of backward and forward participation degrees in GVCs indicates the total participation degree in GVCs (Equation (3)).

CTIT

There are three different kinds of ‘Input-Output modelling’ methods: calculating regional pollutants emission and resources consumption transfer through trade (e.g., pollutant emission, energy resources and water resources) and relevant issues, namely ‘bilateral regional Input-Output analysis (BRIOA), single-regional Input-Output analysis (SRIOA) and multi-regional Input-Output analysis (MRIOA)’ Wiedmann et al. (2007). MRIOA better articulates heterogeneity, and the reason is it excels in discovering regional ecological influences and total energy use in exported and imported goods and capturing the associations among economic divisions. With this, we have employed MRIOA to calculate the CTIT, and based on the Input-Output table, we have

Y_{T}^{B} = \sum_{p = 1}^{N} Z_{t}^{bp} P + \sum_{p = 1}^{N} Y_{t}^{bp} .

(4)

So, $Y_{T}^{B}$ is a country B output in year T. Unit matric is P and matrix with final demand and intermediate distributions are shown by $A_{t}^{bp}$ and $F_{t}^{bp}$ . Current study suggested following settings, such as, $Z_{t}^{bp}$ shows the volume of intermediate inputs and $z_{t, j, t}^{bp}$ shows capacity of intermediate inputs from an economy’s s sectors to an economy’s s sector j in year T. In matrix of intermediate input, A_t comprising of $A_{t}^{bp}$ has similar construction with Z_t comprising of $Z_{t}^{bp}$ Consequently, II coefficient $a_{t, j, t}^{bp}$ = $z_{t, j, t}^{bp}$ / $Y_{t}^{bp}$ and $Y_{j, t}^{p}$ is sector J gross output of an economy’s R in year t, r, s ∈ N, i, j ∈ J, N = 27, J = 25. Hereby, Equation (1) could be shown as Y_t = Z_tP + F_tP. similarly, Y_t = A_tX_t + F_tP. So, we have: Z_tP = A_tX_t, and Y_t = (I – A_t) − 1 F_tP. Moreover, paper set M_t ≡ (I – A_t) – 1, I is unit matrix and M_t is Leontief inverse matrix. Accordingly, the matrix complete carbon emissions can be inferred as follows:

G_{t} = [\begin{array}{l} ({wa}_{t}^{1}) (M_{t}^{1, 1}) \dots ({wa}_{t}^{1}) (M_{t}^{1, n}) \dots ({wa}_{t}^{1, N}) (M_{t}^{1, N}) \\ \dots \dots \dots \\ ({wa}_{t}^{1}) (M_{t}^{1, 1}) \dots ({wa}_{t}^{1}) (M_{t}^{1, n}) \dots ({wa}_{t}^{1, N}) (M_{t}^{1, N}) \\ \dots \dots \dots \\ ({wa}_{t}^{1}) (M_{t}^{1, 1}) \dots ({wa}_{t}^{1}) (M_{t}^{1, n}) \dots ({wa}_{t}^{1, N}) (M_{t}^{1, N}) \end{array}]

(5)

while $M_{t}^{B, S}$ from an economy S to an economy B in year T shows the ‘Leontief inverse matrix’ with intermediate distributions. The coefficient of direct carbon emission is ${WA}_{J, t}^{B}$ country’ B sector j, and ${WA}_{J, t}^{B}$ = ${WR}_{J, t}^{B}$ / $Y_{J, t}^{B}$

Equation (8) shows that ${(G_{t}^{B, p})}^{'}$ = ${({WA}_{t}^{p})}^{'}$ ${WA}_{t}^{p}$ , and $G_{t}^{B, p}$ shows the direct and indirect carbon emission in economy B for single unit of final demand in economy P formed in economy B. So, the CTIT among economy P and B caused from global trade in year t $({CTIT}_{r - s, t})$ might be specified as below:

{CTIT}_{b - p, t} = [\sum_{k = 1}^{N} (G_{t}^{k, b})] f_{t}^{b, p} + G_{t}^{k, p} (\sum_{k = 1}^{N} f_{T}^{p, k}), p \neq r, s, k \in N .

(6)

In Equation (6), the first terms show the CTIT’s volume in final consumption because of export and import trade among an economy P and economy B. Numbers of BRI regions and countries are shown by k, k = N = 27. Furthermore, for an economy B, the final demand is delivered by another region (comprising economy B), and hereby, inter-regional trade in this course is meant to create emission of carbon. $\sum_{k = 1}^{N} (G_{t}^{k, b})^{'}$ indicates the row vectors of emission of carbons produced in BRI regions that are crucial for a single unit of final goods manufactured in economy B. The next term in Equation (6) shows the CTIT’s capacity in intermediate input because of regional trade among economies B and P. $\sum_{k = 1}^{N} f_{T}^{P , k}$ shows the final requirements created in economy S for final demand in rest of economies (k) (s = k, k = 27).

Additionally, for economy B, CTIT generated from international trade is made up of ETITs and ETETs. Equation (6) is given as

{ETIT}_{b, t} = \sum_{k = 1}^{N} [(\sum_{k = 1}^{N} G_{T}^{p, k})] f_{T}^{p, k} + \sum_{k = 1}^{N} [(G_{t}^{k, b})] \sum_{k = 1}^{N} f_{T}^{p, k}

(7)

{ETET}_{b, t} = [\sum_{k = 1}^{N} (G_{T}^{k, b})] (\sum_{k = 1}^{N} f_{T}^{p, k}) + \sum_{k = 1}^{N} [(G_{t}^{k, b}) (\sum_{k = 1}^{N} f_{T}^{p, k})] .

(8)

Variable Selection and Econometric Model

Herby, bases on above-mentioned context, an econometric model of panel data in the observation of time-based measurement has been further used. Besides, to evade the multi-collinearity matter among main regressors, the current study used the following specification:

\begin{matrix} E T E T_{i t} (E T I T_{i t}) = λ 0 + α 1 G V C_F o_{i t} + α 2 E N I N_{i t} + α 3 I N T I N_{i t} + \\ α 4 F I R E Q_{i t} + α 5 P R I N_{i t} + α 6 S E C I N_{i t} + α 7 E N S T_{i t} + \\ α 8 P G D P_{i t} + R & D_{i t} + δ i + ρ t + ξ i t, \end{matrix}

(9)

\begin{matrix} E T E T_{i t} (E T I T_{i t}) = λ 0 + β 1 G V C_b a k_{i t} + β 2 E N I N_{i t} + β 3 I N T I N_{i t} + \\ β 4 F I R E Q_{i t} + β 5 P R I N_{i t} + β 6 S E C I N_{i t} + β 7 E N S T_{i t} + \\ β 8 P G D P_{i t} + R & D_{i t} + δ i + ρ t + ξ i t, \end{matrix}

(10)

where t and i symbolise the year and country, respectively. E and 0 symbolises the error term and constant term, respectively. ETET_it and ETIT_it display the emissions transfer through import and export in the year for an economy r, respectively. GVC_bak and GVC_for illustrate the GVCs’ participation degrees in year t for a country r. Along with this, EN IN_it, INT IN_it, FI REQ_it, PR IN_it, SECIN_it, EN ST_it, PGDP_it and R&D_it, represent the dependent variables of the current mode, such as energy intensity, intermediate inputs, final requirements, primary industry (total share of primary industry), secondary industry (total share of secondary industry), total energy consumption structure, per capita gross domestic product and research and development, respectively. Moreover, to lessen the projected biasness triggered by the oversight of related variables, we considered the period fixed effects (ρt) and country fixed effects (δi) in our model. To evade the biasness, ρt and δi also be considered two random regressors, identically and independently distributed through variance and a zero mean in the model of the study. To enhance the estimated result consistency, the dependent regressors that would not only elaborate the understudy samples’ national attributes but also affect the interregional CTIT are chosen here.

Data Sources

Following the methods mentioned above, GVCs data, ‘worldwide multi-regional Input-Output tables and sectoral emissions of carbon’ for each economy are measured from ‘the 2021 edition of OECD Inter-Country Input-Output (ICIO) tables’. The reasons behind the 27 countries selected for current analysis are: The OECD ICIO database has data for 36 industries and 64 regions for the period 2005–2018, and 27 Belt and Road economies data are available in the OECD ICIO database. Box 1 shows the 27 BRI countries selected for analysis of CTIT from 2005 to 2018. Furthermore, the Belt and Road countries chosen for study contribute over 43% of total global carbon emissions transferred through international trade (OECD) and make up 30% of global GDP (Han et al., 2020). Moreover, the data set from 2005 to 2018 is picked for the following reasons. First, this study focuses on addressing the association between CTIT and participation in GVCs and hereby delivers a theoretical foundation to explain the influence of GVCs’ participation degrees in the BRI region on the CTIT. The data set provided by these databases is the latest available. And current article principally efforts to extract the association between these two variables from an empirical perspective. With this, in the case of stakeholders, it is meaningfully significant to elaborate the influence of the BRI region’s GVCs’ participation on the CTIT to devise better targeted ecological governance and management of strategies and policies from the BRI trade partner perspective. Second, the current article required other socioeconomic and regional Input-Output data in a sequential period. In specific, all variables are also required to be matched in current econometric models.

Finally, the rest of the variables, such as industrial structure, energy consumption structure, PGDP and energy intensity, these datasets can be taken from the database of the World Bank (http://data.worldbank.org.cn) and variable’s descriptions are shown in Box 1. Additionally, to avoid the effect of the ‘Price factor’, all the dependent and crucial regressors are chosen in constant price (US dollars). Table 1 contains the descriptive statistics of employed variables.

Box 1.

List of Belt and Road Economies Selected for Study Analysis.

Bulgaria, Brunei Darussalam, China, South Korea, Cyprus, Czech Republic, Estonia, Greece, Croatia, Hungary, Indonesia, Kazakhstan, Cambodia, Latvia, Malaysia, Indonesia, Philippine, Romania, Poland, Russian Federation, Saudi Arabia, Singapore, Slovak Republic, Slovenia, Thailand, Turkey and Vietnam.

Table 1.

Variable Description.

Variables	Number of Countries		Sources	Mean	Median	SD	Maximum	Minimum
GVC_for	27	US dollar, millions	ICIO tables	0.83	0.83	0.70	2.25	1.64
GVC_bak	27	US dollar, millions	–	0.78	0.78	0.60	2.33	0.32
ETET	27	Kilotons	–	1.97	2.02	0.68	2.42	0.75
ETIT	27	Kilotons	–	5.12	2.01	0.78	3.81	0.81
EN IN	27	US dollar, millions	–	0.99	0.97	0.63	1.49	0.81
INT INP	27	US dollar, millions	–	5.33	5.25	0.74	7.01	4.11
FI REQ	27	US dollar, millions	–	5.12	5.18	0.68	6.87	3.81
PR IN	27	%	WBI	0.54	0.57	0.63	1.53	1.31
SEC IN	27	%	–	1.41	1.42	0.24	1.87	0.81
EN ST	27	%	–	1.82	1.88	0.21	2.00	1.01
PGDP	27	US dollar	–	3.97	4.12	0.51	4.86	2.67
R&D	27	R&D expenditure (% of GDP)	–	0.98	0.75	1.04	9.45	0.00

Source: OECD Inter-Country Input-Output (ICIO) tables and World bank Development Indicators (WBI).

Graphical Analysis

Figures 2 and 3 indicate the participation degree in GVCs of the major 27 BRI economies and carbon transfer through export and import in 2005–2018, respectively. These figures illustrate the influence of GVCs’ participation degrees on carbon transfer through export and import trade. The Belt and Road economies with high participation degrees in GVCs would transfer carbon emissions through export and import trade. To support this presumption, there has been indicated an increasing trend from 2005 to 2018 in both Figures 2 and 3. As GVCs rise, the emissions transferred through export and import trade rises.

Econometric Results and Discussions

Econometric Regression Outcomes

First, a series of tests such as ‘least square dummy variables’ and ‘F-value (p = 0.00) test’ were performed, and study regression results illustrated that the model has an individual effect. Consequently, ‘the confluent regression’ should not be used in the model. Second, the current article employed Hausman to discover if the random effect regressions cannot be employed. The result favoured the fixed-effect regression and was chosen as the better-fitted technique. Consequently, the appropriate results have been shown in Table 2.

Table 2.

Influences of BRI Region GVCs Participation Degree on the CTIT.

	GVC_for				GVS_bak
	ETIT (Import)		ETET (Export)		ETIT (Import)		ETET (Export)
	1	2	3	4	5	6	7	8
GVC_for	0.16* 0.01	0.14* 0.01	0.06* 0.01	0.01* 0.01
GVC_bak					0.07* (0.03)	0.01* (0.03)	0.10* (0.02)	0.04* (0.02)
EN IN		–0.42* (0.07)		–0.35* (0.06)		–0.39* (0.08)		–0.37* (0.07)
INT INP		0.66* (0.07)		–0.03* (0.06)		0.68* (0.07)		–0.04 (0.06)
FI REQ		–0.10 (0.08)		0.41 (0.06)		0.18* (0.08)		0.42* (0.06)
PR IN		(0.01) (0.05)		0.07* (0.04)		–0.03 (0.05)		0.07* (0.04)
SEC IN		0.71 (0.09)		0.43* (0.07)		0.85 (0.10)		0.55* (0.08)
EN ST		0.27* (0.08)		0.43* (0.07)		0.25* (0.10)		0.43* (0.07)
PGDP		0.16* (0.06)		0.29* (0.05)		0.10 (0.06)		0.28* (0.05
R&D		0.002 (0.26)		–0.00** (0.007)		–0.001 (0.0008)		0.00** (0.007)
Constant	1.51* 0.08	–1.52* (0.26)	–2.27 (0.19)	–2.26* (0.24)	1.59* (0.26)	–1.48* (0.29)	2.02 (0.20)	–2.23* (0.24)
Number of countries	27	27		27	27	27	27	27

Source: OECD Inter-Country Input-Output (ICIO) tables and World bank Development Indicators (WBI).

Note: *significant at 1%, **significant at 5% (standard error in parenthesis).

Table 2 reported the estimation result of the impact of GVCs’ participation degrees on CTIT in Belt and Road economies. Based on the scenario of individual regression and totalling all dependent regressors, the GVCs’ participation degrees estimated coefficient and GVCs forward and backward participation are significantly positive, illustrating the CTIT rise in Belt and Road countries with an increase in their participation degrees in GVCs and confirmed the work of Zhang et al. (2021) and Montalbano et al. (2018), and our result did not match the outcomes of Zheng et al. (2021). The estimation outcomes illustrate the continued integration of BRI economies into the GVCs, leading to the expansion of their industrial and economic activities. In such a scenario, GVCs accelerate the economy’s growth and increase the flow of high-carbon energy sources between diverse economies.

Robust Test

Previous methodologies illustrated that if the endogenous issues among variables are ignored, there might be inconsistency and bias in the estimation result (Bound et al., 1995; Klette & Griliches, 1996). Our article addresses the endogeneity issues of reverse causality, measuring error and omitted variables. From this perspective, to minimise the errors of outcome estimation triggered by the econometric model choice, this article introduces the ‘system generalised methods of moments’ (SYGMM).

As indicated in Table 3, the BRI participation degree in GVCs has a positive and significant impact on CTIT, and endogeneity issues do not hinder the results’ robustness. The BRI economies’ participation in GVCs both estimated coefficients are positive and significant in the outcomes from the two kinds of regressions. Tables 2 and 3 illustrate that the core explanatory variable approximation outcome’s reliability is confirmed. The results illustrated that the extension of GVCs and economic globalisation that pose opportunities for emerging and developing economies of Belt and Road would also lead to the continued rise of the CTIT. Because several of BRI’s economies are lying at the lower positions of GVCs, their increase in participation degree in GVCs could considerably influence tasks for reductions in CTIT. One plausible clarification is that the international market through production fragmentation has welcomed an era of GVCs. As part of this production fragmentation, BRI economies have a high demand for high-carbon-intensive goods from the upper-rank economies in GVCs. In this context, emerging and developing economies in BRI are subjected to industries with extravagant energy consumption and high pollution. While these BRI economies take advantage of their comparative advantage, they engross specific product segments, leading to the use of a mass of fossil fuel energy resources in the region. Though the high-income economies’ productions are mainly low and clean carbon, the desire for absolute necessities led to imports from the rest of the economies through international trade. As a result of this, the participation in GVCs could hasten the flow of CTIT across every economy. Accordingly, inside the context of closer interregional linkages, for the BRI economies, the perpetual deepening of their participation in GVCs would lead to an increase in the volume of CTIT, which would obtain widespread consideration from the several stakeholders.

In the case of rest the dependent regressors, the sign of several regressors in Table 3, which are secondary industrial structure, total energy intensity and the primary industrial structure, is not the same as in Table 2. The outcomes of the remaining regressors in Table 2 are constant with the results in Table 3, illustrating that some levels have confirmed the reliability of estimated results. Per Capita GDP coefficient is significant in Tables 2 and 3; however, the coefficient sign is negative different after endogeneity. The energy intensity variables in Table 3 are considered insignificant at a 10% significant level. However, its coefficient’s sign is identical to that in Table 2, demonstrating that it may result in insignificant outcomes after addressing endogeneity issues. However, Table 3 shows that the energy intensity, secondary industry structure, research and development (R&D) and share of the primary industrial structure estimated coefficients are not the same as in Table 2. And the energy consumption structure, intermediate inputs and input demand in Table 3 are constant with the results in Table 2, signifying that their impact on the CTIT may be uneven. As a matter of fact, Xu and Dietzenbacher (2014) illustrated that the home feature of economies primarily causes the influence of the fluctuations in ‘intermediate input demand’ on CTIT; however, Zhong et al. (2020) realised that the employed spital econometric regression models cause the unstable effect of these variables on CTIT. In short, based on Tables 2 and 3, the regression result approximation of crucial variables (i.e., GVCs participation degree) is consistent. Concerning the result discussions of rest of the critical variables, the effect of energy intensity on CTIT in the BRI region is significant and positive; the results coincide with Munir Ahmad et al. (2020) and did not match with works of Zheng et al. (2021). One of the probable reasons for this adverse effect of energy intensity on CTIT is the increase in consumption of non-renewable energy, which is widely accompanied by CTIT. Likewise, as for energy consumption structure, the calculated results are consistent with the outcomes of prior studies (Zhong et al., 2020). In the case of the energy consumption structure of BRI economies, one of the rational justifications is that these BRI economies are highly dependent on fossil fuels in their energy structure. Along with that, the size of CTIT among BRI regions is principally connected to the outflow and inflow of high carbon industrial production among different areas along the GVCs. So, the frequent energy consumption structure optimisations in the BRI region could diminish the urgency for the fossil fuel imported from the rest of the world (Zhong et al., 2020). However, it will also increase the domestic demand for industrial products from rest of economies. Accordingly, during country economic transition, these two forces may interact dynamically to undermine the impact of the total energy consumption structure. Lastly, the per capita GDP results are consistent with Xu and Dietzenbacher (2014) and oppose the work of Ali et al. (2021a, 2021b). The plausible explanation is that after endogeneity issues are measured, the results suggest that the BRI region’s economic growth level is not favourable enough to reduce the influence of CTIT on the distribution of obligations and assignments for the lessening the CTIT.

Table 3.

Study Outcomes After Checking the Endogeneity Issues for BRI Project Economies.

	GVC_for				GVC_ bak
	ETIT		ETET		ETIT		ETET
	1	SYGMM	2	SYGMM	3	SYGMM	4	SYGMM
GVC_for	0.15* (0.02)	0.04* (0.12)	0.02* (0.01)	0.004* 0.02
GVC_bak					0.23* (0.03)	0.07* 0.15	0.18* (0.02)	0.03* (0.15)
ETIT		0.05 (0.12)		–0.09 (0.04)
ETET						–0.24* (0.10)		–0.01* (0.05)
EN IN	–0.83* (0.07)	0.12 (0.21)	0.19* (0.04)	0.03 (0.12)	0.69* (0.07)	0.06 (0.17)	0.03 (0.05)	0.03 (0.14)
INT INP	–0.57* (0.06)	0.58* (0.18)	0.18* (0.04)	–0.16* (0.12)	0.49* (0.05)	0.69* (3.39)	0.15* (0.03)	–0.23* (0.17
FI REQ	0.30* (0.07)	0.06 (0.24)	0.72* (0.04)	0.48* (0.17)	0.30* (0.06)	0.28* (0.20)	0.62* (0.04)	0.56* (0.19)
PR IN	–0.19* (0.03)	–0.07 (0.20)	–0.12* (0.02)	0.01 (0.13)	–0.20* (0.03)	–0.16* (0.13)	–0.11* (0.02)	0.03 (0.15)
SEC IN	–0.03 (0.07)	1.19* (0.45)	–0.18* (0.04)	0.92* (0.20)	–0.08 (0.05)	1.24* (0.28)	–0.17* (0.03)	0.95* (0.27)
EN ST	0.32* (0.07)	0.46 (0.61)	0.07* (0.04)	0.78* (0.48)	0.30* (0.07)	0.29 (0.42)	0.13* (0.04)	0.79* (0.55)
PGDP	–0.40 (0.03)	–0.32** (0.18)	–0.26* (0.03)	0.29* (0.13)	–0.47* (0.03)	–0.19** (0.14)	–0.29* (0.02)	0.29* (0.18)
R&D	–0.02** (0.01)	0.03 (0.19)	0.00 (0.00)	0.01 (0.02)	0.03 (0.00)	0.03 (0.00	0.00** (0.001)	0.02 (0.04)
Constant	–0.71 (0.22)	–0.03 (0.31)	–1.29* (0.15)	–0.004 (0.12)	–0.13* (0.43)	–0.231 (–1.87)	–0.87* (0.17)	-0.21 (0.31)
A, R (2)		0.073		0.321		0.051		0.311
Hansen test		1.00		1.0		1.0		1.0
F	513.54 (0.00)		587.3 (0.00)		509.0 (0.00)		595.8 (0.00)
Number of countries	27	27	27	27	27	27	27	27

Source: OECD Inter-Country Input-Output (ICIO) tables and World bank Development Indicators (WBI).

Note: *significant at 1%, **significant at 5% (T-statistics in parenthesis).

Heterogenous Analysis

In international trade and participation in GVCs, the economic division among economies is progressively apparent. Consequently, the study conducts an experimental analysis to find what influence GVCs’ participation may have on CTIT in BRI economies with different economic development levels.

Based on the role in international trade, GVCs, carbon emission and World Bank categorisation, the BRI economies are separated into two sample groups of developed and emerging and developing economies (Nedopil, 2022). The impact of GVCs participation on the CTIT in both groups of the BRI project was significantly positive and conferred with the study’s key conclusion, as shown in Tables 4 and 5, that the CTIT in the BRI regions increases with the increase in their GVCs participation. So, rapid economic globalisation leads to fast growth in GVCs and international trade, which results in an upsurge of CTIT in BRI regions and might influence policies and measures for emission reduction. For the BRI region, energy management, conservation and focusing on emission reduction are vital for tackling CTIT. Furthermore, the estimation outcomes for dependent variables in Table 4 coincide with Tables 2 and 3 results, authorising the consistency of experimental outcomes. Likewise, the positive impact of energy intensity is similar in both groups of BRI regions. The final demand coefficient is more significant in high-income countries than in the emerging and developing countries of Belt and Road. The coefficient of secondary industries is positive and significant. Emerging and developing economies’ coefficient of energy structure is comparatively high. The estimated coefficient of PGDP is positive in both groups. The technologies and energy sources used by the emerging and developing economies of BRI in their industries and manufacturers are immeasurably out-dated and have less R&D activities. As a result, the developed countries transfer their carbon-intensive industries to these economies through emission trading systems; consequently, the lack of clean technologies and proper R&D activities, out-dated technologies and non-renewable energy consumption escalate the level of CTIT in the BRI economies.

Table 4.

Impacts of GVCs on the CTIT in Case of Emerging and Developing Countries of BRI.

	GVC_for				GVC_bak
	ETIT		ETET		ETIT		ETET
	1	2	3	4	5	6	7	8
GVC_for	0.35* (0.04)	0.37* (0.43)	0.10 (0.06)	0.26* (0.03)
GVC_bak					0.12* (0.04)	0.08** (0.04)	0.13* (0.03)	0.10* (0.04)
EN IN		0.50* (0.10)		0.17 (0.09)		0.68* (0.12)		0.30* (0.10)
INT INP		0.68* (0.10)		–0.02 (0.09)		1.04* (0.11)		0.24** (0.09)
FI REQ		–0.33* (0.10)		0.14** (0.08)		–0.48* (0.11)		0.06 (0.09)
PR IN		–0.01 (0.07)		–0.05 (0.06)		0.01 (0.01)		–0.02 (0.07)
SEC IN		0.50* (0.14)		0.39* (0.12)		0.63* (0.16)		–0.50* (0.14)
EN ST		0.43* (0.08)		0.44* (0.07)		0.43* (0.09)		0.44* (0.07)
PGDP		0.002 (0.07)		0.41* (0.06)		–0.15 (0.001)		0.29* (0.07)
R&D		0.03 (0.00)		0.002 (0.002)		–0.001 (0.004)		–0.004 (0.003)
CONSTANT	–1.35* (0.04)	–1.08* (0.37)	–1.89* 0.07	–1.64* (0.34)	–1.64* (0.37)	–1.42* (0.43)	1.83* 0.21	–1.90* (0.36)
Number of observation	196	196	196	196	196	196		196

Source: OECD Inter-Country Input-Output (ICIO) tables and World bank Development Indicators (WBI).

Note: *significant at 1%, **significant at 5% (T-statistics in parenthesis).

Table 5.

Impacts of Participation Degree in GVCs on the CTIT in Developed Countries of BRI.

	GVC_for				GVC_bak
	ETIT		ETET		ETIT		ETET	ETIT
	1	2	3	4	5	6	7	8
GVC_for	0.12* (0.02)	0.08* (0.02)	–0.02** (0.01)	–0.02** (0.01)
GVC_bak					0.01** (0.05)	–0.09* (0.05)	0.12 (0.03)	0.01 (0.04)
EN IN		0.35* (0.13)		0.17 (0.11)		0.17 (0.13)		0.22* (0.10)
INT INP		0.41* (0.10)		–0.27* (0.08)		0.39* (0.10)		–0.26* (0.08)
FI REQ		0.06** (0.12)		0.65* (0.10)		0.04 (0.12)		0.69* (0.10)
PR IN		–0.05 (0.07)		–0.04 (0.06)		–0.10 (0.08)		-0.03 (0.06)
SEC IN		0.94* (0.11)		0.43* (0.09)		1.04* (0.12)		0.41* (0.09)
E ST		–0.41* (0.34)		1.25* (0.28		–0.47** (0.36)		1.21* (0.28)
PGDP		–0.10 (0.12)		0.41* (0.00)		0.02 (0.11)		0.37* (0.09)
R&D		0.01 (0.01)		0.01 (0.01)		–0.34 (0.77)		–0.01 (0.01)
CONSTANT	–1.57* 0.64	–0.47 (0.74)	–4.59* 0.03	–4.52* (0.61)	–1.47* (0.68)	–0.34 (0.77)	–3.35 (0.39)	–4.47* (0.61)
Number of observation	216	216	216	216	216	216	216	216

Source: OECD Inter-Country Input-Output (ICIO) tables and World bank Development Indicators (WBI).

Note: *significant at 1%, **significant at 5% (T-statistics in parenthesis).

Conclusion and Policy Recommendations

Conclusion

The current study aimed to deliver an improved understanding of the features of the CTIT variations from the Belt and Road perspective. And to experimentally elaborate, what impact would the GVCs’ participation degrees of BRI have on the carbon ETET and ETIT trade? In the context of the BRI region, the current article creates a panel methodology to inspect the influence of GVCs’ participation degree on the CTIT on the foundation of calculating the GVCs’ participation degree within an Input-Output framework for 27 Belt and Road countries from 2005 to 2018. On this basis, we expressed the appropriate environmental governance strategies and policy inferences for CTIT regarding carbon mitigation and climate change. The core conclusions and recommendations for policy implications are as below.

Concerning empirical results, they have two sides: the first one is that the estimated coefficient of GVCs (forward and backward participation) is positive and significant, illustrating that the CTIT raises as Belt and Road increase participation in GVCs. Additionally, potential endogenous issues were not interfering with the conclusion’s robustness which was displayed by a series of tests. Second, regarding remaining regressors, our estimated outcomes propose that lessening the energy intensity of the BRI region will improve corresponding energy efficiency because energy intensity is increasing CTIT in the BRI region. The estimated coefficient results of secondary industry and final demand of BRI regions are discovered to influence CTIT positively. At the same time, the conclusions displayed that potential endogenous risk did not affect the result and showed strong robustness. The energy consumption structure was discovered to be significant, have a positive value, and have the same value after considering the endogeneity issue. The result implies that energy structure has a stable impact on CTIT in the BRI region. Furthermore, the per capita GDP of the BRI region is positive and significant, illustrating that the level of the BRI region’s economic development level contributes to the rise of CTIT and should be considered seriously by the stakeholder.

Policy Implications

Concentrating on policy implications, the fast development of economic globalisation poses many opportunities for the BRI region. There is an accelerating global integration of production chains and cross-border investments, eventually encouraging GVCs’ participation. The GVCs participation degree of several BRI emerging economies, such as China, Russia, Indonesia and Bangladesh would continue to promote. Consequently, the international trade in the BRI region would also swiftly rise, resulting in the rise of CTIT. So, regarding environmental governance policy, the policymaker and advisor for reduction of carbon transfer through international trade would not automatically set policies to a single or some regions in reflection of fragmentation of production process. In its place, from the GVCs perspective, the CTIT reduction responsibilities can be divided. This procedure could deliver a way to proficiently allocate responsibility for emissions reduction and assist the developing economies of the BRI project in coping with their CTIT. Along with this, the rise in participation in GVCs of the BRI region would enhance the flow of factors of production and various resources between economies and could significantly increase the CTIT. Optimising industrial structure, increasing energy intensity and improving awareness regarding carbon reduction and energy conservation could minimise the CTIT. Furthermore, the infrastructure of the BRI project should be energy efficient and consume less energy. Belt and Road economies need to switch to sustainable energy sources such as wind or solar power to minimise the CTIT.

Footnotes

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship and/or publication of this article.

Funding

The authors received no financial support for the research, authorship and/or publication of this article.

References

Ahmed

, Kousar

, Pervaiz

, & Ramos-Requena

J. P.

(2020). Financial development, institutional quality, and environmental degradation nexus: New evidence from asymmetric ARDL co-integration approach. Sustainability, 12(18), 7812. 10.3390/su12187812

Alfaro

, Chor

, Antras

, & Conconi

(2019). Internalizing global value chains: A firm-level analysis. Journal of Political Economy, 127(2), 508–559.

Ali

M. U.

, Gong

, Ali

M. U.

, Asmi

, & Muhammad

(2020). CO₂ emission, economic development, fossil fuel consumption and population density in India, Pakistan, and Bangladesh: A panel investigation. International Journal of Finance & Economics, 1–14. 10.1002/ijfe.2134

Ali

M. U.

, Gong

, Ali

M. U.

, Wu

, & Yao

(2021a). Fossil energy consumption, economic development, inward FDI impact on CO₂ emissions in Pakistan: Testing EKC hypothesis through ARDL model. International Journal of Finance & Economics, 26, 3210–3221. 10.1002/ijfe.1958

Ali

M. U.

, Zhimin

, Asmi

, Xue

, & Muhammad

(2021b). The nexus between environmental degradation and industrial development in Pakistan and roles of financial development and fossil fuel. Environmental Progress and Sustainable Energy, 40, e13621. 10.1002/ep.13621

Bound

, Jaeger

D. A.

, & Baker

R. M.

(1995). Problems with instrumental variables estimation when the correlation between the instruments and the endogenous explanatory variable is weak. Journal of the American Statistical Association, 90(430), 443–450.

Casella

, Bolwijn

, Moran

, & Kanemoto

(2019). Improving the analysis of global value chains: The UNCTAD-Eora Database. Transnational Corporations, 26(3). http://worldmrio.com/unctadgvc/

Changjian

, Zhuang

, Xiao

D. C.

, & Yu

(2021). Factors affecting changes of greenhouse gas emissions in Belt and Road countries. Renewable and Sustainable Energy Reviews, 147, 111–220.

Han

, Lao

, Yao

, Zhang

, & Meng

(2020). Carbon inequality and economic development across the Belt and Road regions. Journal Environmental Management, 262, 110–250.

10.

Klette

T. J.

, & Griliches

(1996). The inconsistency of common scale estimators when output prices are unobserved and endogenous. Journal of Applied Econometrics, 11(4), 343–361.

11.

Koopman

, Wang

, & Wei

S. J.

(2014). Tracing value-added & double counting in gross exports. American Economic Review, 104(2), 459–494.

12.

Meng

, Peters

G. P.

, Wang

, & Li

(2018). Tracing CO₂ emissions in global value chains. Energy Economics, 73, 24–42.

13.

Montalbano

, Nenci

, & Pietrobelli

(2018). Opening and linking up: Firms, GVCs, and productivity in Latin America. Small Business Economics, 50(4), 917–935.

14.

Nedopil

(2022). Countries of the Belt and Road Initiative. Green Finance & Development Center, FISF Fudan University.

15.

Rehman

, Rauf

, Ahmad

, Chandio

A. A.

, & Deyuan

(2019). The effect of carbon dioxide emission and the consumption of electrical energy, fossil fuel energy, and renewable energy, on economic performance: Evidence from Pakistan. Environmental Science and Pollution Research, 26(21), 21760–21773.

16.

Shakib

, Yumei

, Rauf

, Alam

, Murshed

, & Mahmood

(2021). Revisiting the energy-economy-environment relationships for attaining environmental sustainability: Evidence from belt and road initiative countries. Environmental Science and Pollution Research, 29, 3808–3825. 10.1007/s11356-021-15860-9

17.

Shankar

(2020). Role of global economic policy uncertainty on firms participation in innovation and new product introductions: An empirical study in African SMEs. Transnational Corporations Review, 12(4), 360–378.

18.

Steingress

(2019). Market size and entry in international trade: Product versus firm fixed costs. Review of International Economics, 27(5), 1351–1370.

19.

Wang

, Wan

, & Wang

(2019). Participation in GVCs and CO₂ emissions. Energy Economics, 84, 104561.

20.

Wiedmann

, Lenzen

, Turner

, & Barrett

(2007). 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. Ecological Economics, 61(1), 15–26.

21.

, & Dietzenbacher

(2014). A structural decomposition analysis of the emissions embodied in trade. Ecological Economics, 101(5), 10–20.

22.

Zhang

D. Y.

, Wang

, Löschel

& Zhou

(2021). The changing role of global value chains in CO₂ emission intensity in 2000–2014. Energy Economics, 93, 105053.

23.

Zhangqi

, Zhifang

, & Jianwu

(2021). Does the participation in global value chains promote interregional carbon emissions transferring via trade? Evidence from 39 major economies. Technological Forecasting and Social Change, 169, 120–806.

24.

Zheng

, Zhang

, Wei

, Song

, Dietzenbacher

, Wang

, & Guan

(2021). Regional determinants of China’s consumption-based emissions in the economic transition. Environmental Research Letters, 15(7), 074001.

25.

Zhong

, Thill

J. C.

, Sun

, & Wang

(2020). Carbon emissions embodied in trade and urban regional climate policy-making in the Shanghai mega-region. In J. C. Thill (Ed.), Innovations in urban and regional systems (pp. 385–416). Springer.