Introduction

In the past few decades, profound changes have taken place in China’s economic system. The extensive mode of economic growth and backward technological level have resulted in severe environmental damage, and the contradiction between resource shortage and economic construction has increasingly intensified (Umar et al., 2020; Ren et al., 2023). In this context, the concept of green growth came into being, which has currently evolved into a trend in the global environmental and economic fields. On the involution, green growth refers to the way of minimizing the use of natural resources while mitigating environmental deterioration and climate change, which is a model innovation based on traditional economic development under the constraints of environmental capability and resource carry capacity and takes ecological protection as a considerable pillar to achieve sustainable advancement. Among them, environmental resources are the intrinsic factor, sustainable evolution is the goal, and the clean of the economic activities is the main content and approach (Dong et al., 2022; Zhao et al., 2022a). Since China’s previous economic development pattern is unreasonable, attaching great importance to speed and quantity, neglecting quality and efficiency, paying much attention to development and utilization, and overlooking protection and recovery. In other words, China’s economic advancement has been achieved at the cost of sacrificing environment to some extent. Thus, implementing the green sustainable evolution strategy and boosting green growth are the inevitable course for the new plan of China’s future evolution and modernization construction, as well as a long-term plan for the survival and development of the Chinese nation (Wang et al., 2022).

As an important part of the new development philosophy, green growth is the fundamental requirement of high-quality development and the only way to transform a large industrial country into an industrial power. To energetically strengthen China’s green growth, it is imperative to accelerate the adjustment of energy structure. Many scholars attributed the reason to the fundamental role of energy in residents’ life and the operation of enterprises (Shahbaz et al., 2022; Ren et al., 2021a). Currently, according to the statistics from former British Petroleum (BP), the primary energy use of China in 2021 was 157.65 exajoules, which accounted for 26.5% of total global energy consumption (BP, BP Statistical Review of World Energy, 2022). Besides, the data of China Statistical Yearbook (CSY) show that, the total use of coal and oil in China accounts for more than 70% of the country’s primary energy use (CSY, National Bureau of Statistics, China Statistical Yearbook, 2021). These statistical evidence states the necessity of boosting energy transformation. Among them, improving energy efficiency and boosting clean energy alternatives are two powerful weapons. On the one hand, an accelerated low carbonization transition of energy can effectively alleviate the high dependence of enterprises or residents on traditional fossil energy consumption and significantly improve energy utilization efficiency, thus curbing pollutant emissions and promoting green growth (Pingkuo & Huan, 2022; Dogan et al., 2022); on the other hand, optimizing the existing energy framework can contribute to the orderly evolution of clean energy industries, effectively replacing the use of high-carbon energy and facilitating the green growth of China (Dong et al., 2020; Yuan et al., 2018).

Although the underlying positive role of energy transition in accelerating green growth is widely acknowledged in political and academic circles, no evidence exists to substantiate this speculation. Whether reducing fossil energy use and actively boosting the renewable energy industry is conducive to green growth remains to be discussed and further explored. Assessing the green development consequences of transforming energy structure has important empirical value to help policymakers and stakeholders distinguish the effects and optimize the path of energy transition; however, very few scholars have conducted quantitative and systematic evaluation on the economic green promotion effect of transforming energy structure, especially an accurate measure of green growth. Under these circumstances, several vital questions pique our interest:

(1) How to assess green growth in a more comprehensive and efficient way in light of China’s reality? Although some scholars concern about the measurement of green growth in China, some deficiencies still exist. Further research on the evaluation of green growth is of great value to supplement existing literature and enrich relevant theories on green growth.

(2) Can China’s low-carbon energy transition indeed contribute to green growth? At present, some negative voices indicate that the clean energy transition will harm the ecological environment and restrict green growth while boosting the construction of renewable energy-related facilities. Thus, under the premise of vigorously advocating energy transition and green growth, evaluating the actual influence of energy transition on green growth provides valid evidence for rational adjustment of energy transition structure and strength.

(3) Is improved energy productivity an efficient route to accelerate green growth in the energy transition? It has been widely confirmed that improved energy efficiency is a powerful catalyst for saving energy and mitigating pollutants (Dogan et al., 2020; Taskin et al., 2022). Investigating whether energy efficiency is the transmission channel for energy transition to promote green growth not only facilitates the smooth transition of the current energy system, but also installs an effective accelerator for the timely realization of green growth goals.

To solve these questions, China’s green growth composite index, including three dimensions: economic growth, people’s livelihood, and ecological environment, is built, and the dynamic influence of transforming energy framework on green growth is detected by applying panel data of China’s 30 provinces from 2004 to 2018. Besides, four sensitivity checks to detect the sensitivity of the empirical findings are conducted and how energy productivity adjusts and affects the influence of low carbonization energy transition on green growth is tested. It can be found that the continuous progress of the energy transition can effectively facilitate the green evolution of the economy. It is a priority for the government to substantially boost the development of renewable energy sector and increase energy productivity.

The novelty and innovations of our study mainly include following three aspects. First, at present, scholars have not formed a unified standard for the measurement of green growth, especially in China. Building a more reasonable and scientific index is particularly essential to assess and understand the current situation of green economic evolution in China. To this end, a composite index by incorporating three dimensions by using the entropy method is efficiently calculated. This is of great value to help the political and academic circles more clearly recognize the green promotion of China’s economy and draw up more scientific and valid measures and strategies. Second, this discussion is one of the few that focuses on the green advancement effects of energy transition, and can efficiently assess the actual impact of optimizing the energy structure on green growth. Identifying and evaluating the energy transition-green growth nexus can contribute to lay down appropriate measures and regulations to boost green growth in China. Third, to explore how low-carbon energy transition affects green growth and their internal impact routes, the role of energy efficiency in influencing the dynamic linkage between energy transition and green growth are creatively investigated. This provides an effective theoretical reference and practical support for understanding the ways and channels through which low-carbon energy transition affects green growth.

The rest of this study is arranged in the following sections. The next section sorts out the relevant literature on the energy transition and green growth nexus, followed by the theoretical mechanism analyzed in Section 3. Section 4 builds the empirical model and sample data. Section 5 analyzes the specific estimated findings, while Section 6 further checks the role of energy efficiency. The last section concludes the entire paper.

Literature review

Research on the definition and measure of green growth

As a novel growth paradigm that pursues economic growth while inhibiting ecological deterioration and unsustainable use of natural resources, green growth draws increasing attention by the political and academic circles. With the country’s continuous emphasis on environmental protection and the gradual rise of the concept of conserving energy and mitigating pollutants, the discussion of green growth has been widely extensive. In this section, the related literatures on the definition and measures of green growth are reviewed, and the concrete content is presented in Table 10 in Appendix.

In 2005, the United Nations Economic and Social Commission for Asia and the Pacific (UNESCAP) underscored that green growth refers to replacing the traditional extensive economic growth model with low carbonization and sustainable economic growth model. With the gradual promotion of green growth and the deepening of the concept of energy saving and pollutant mitigation, the green growth paradigm was formalized at the UNESCAP conference in 2006 as a way to seek synergies between environmental sustainability and efficient economy. The Organization for Economic Cooperation and Development (OECD) defined green growth as maintaining rapid economic growth while ensuring that natural resources provide adequate resources and ecological services for people’s livelihood (OECD, 2011). This definition has been widely accepted by Baniya et al. (2021) and Jänicke (2012), amongst others. In addition, some scholars also aim to explore the concept of green growth. For example, Hickel and Kallis (2020) indicate that green growth means absolute decoupling of development from the resource application and adverse environmental consequences. In summary, green growth means ensuring both rapid economic growth and a sound ecological environment. In this regard, it is not feasible to limit measures to the control of pollution emissions only. The government needs to improve energy efficiency, enhance research and development (R&D) of environmental technologies, and optimize the current industrial structure to promote green economic growth.

How can green growth be measured? After understanding the concept of green growth, the way it is effectively measured becomes a crucial issue. Since green growth contains multiple dimensions, the current measurement of green growth mainly adopts a composite index across different countries (Tawiah et al., 2021; Wang & Shao, 2019). For instance, the OECD (2010) stresses that green growth contains five categories — environmental efficiency of consumption, environmental efficiency of production, environmental quality of life, stocks of natural capital and environmental quality, and economic actors’ responses. Based on this, Huang and Quibria (2013) expand these five types of indicators and put forward 22 more detailed indicators to measure green growth. These indicators are also employed by Ates and Derinkuyu (2021). Taking Bangladesh and Nepal as research objects, a set of six indicators are applied by Baniya et al. (2021) to assess green growth. Notably, at present, no unified measure standard exists to gauge green growth, particularly in China.

The nexus between energy transition and green growth

At present, massive energy consumption, especially traditional fossil energy, can provide a driving force and solid foundation for rapid economic development; however, it also creates a deteriorating ecological environment and serious resource shortage, a conclusion verified by a growing body of scholars (Dogan & Inglesi-Lotz, 2017; Inglesi-Lotz & Dogan, 2018). To this end, some scholars gradually extend their research on energy transition to the field of rapid economic growth (Balsalobre-Lorente & Leitão, 2020). More specifically, based on the data from 1850 to 1950 in Sweden, Kander and Stern (2014) empirically examine the gradual impact of the transformation of energy structure from traditional energy to modern energy on economic growth. They insist that traditional energy plays an important role in facilitating economic growth between 1850 and 1890, while from 1890 to 1950 the contribution of modern energy to economic growth is far greater than that of traditional energy. Then, based on the panel data of 25 developing Aisan countries between 2000 and 2016, Mohsin et al. (2021) analyze the impact of the transition from nonrenewable energy to renewable energy on economic growth. The empirical results suggest that increased renewable energy consumption is positively related to rapid economic growth. In addition, Murshed et al. (2021) employ data from Bangladesh for the period 1975-2016, and examine the environmental effects of energy use and other macroeconomic variables. They emphasize that increased non-fossil energy and hydroelectricity consumption are negatively associated with carbon footprint levels; in other words, energy transition in Bangladesh could be a panacea for solving the country’s growing environmental difficulties.

Overall, gradual energy transition is particularly useful for effectively alleviating the degeneration of the ecological environment, improving energy utilization efficiency, accelerating production efficiency, and jointly realizing a sound environment and rapid economic growth (Wang & Wang, 2020; He et al., 2021; Ren et al., 2021b); however, scholars have not directly assessed the underlying impact of energy transition on green growth, especially in China.

Literature gaps

Although plenty of researches have been conducted to measure green growth by a growing body of scholars, there are several study gaps in the current literature. First, although various scholars have applied multiple indicators to assess green growth in different economies (Wang & Shao, 2019; OECD, 2010; Huang & Quibria, 2013; Ates & Derinkuyu, 2021), there is no widely agreed unified index and few scholars have gauged a comprehensive and effective green growth composite index based on China’s specific reality. Second, it has been extensively acknowledged that remarkable results have been achieved in China’s energy transformation, few scholars have evaluated the potential impact of energy transition on green growth in China. Third, on the premise that very few scholars pay attention to the assessment of the green development effect of energy transition, it has become a question worthy of in-depth exploration how transforming energy structure affects the green promotion of economy. This issue offers empirical proof to help the government optimize the strategy of energy transformation and build the grand goal of socialist modernization and greening.

Theoretical framework and mechanism

Low-carbon energy transition indicates the gradual shift from an energy system dominated by high-carbon and high-polluting fossil energy consumption to an energy system highly dependent on clean and green renewable energy (Pingkuo & Huan, 2022; Dogan et al., 2022; Agyeman & Lin, 2022). At present, low carbonization and sustainable development has become the general consensus of the whole society. In the low carbonization transformation process of the current energy structure, the proportion of fossil energy has been greatly reduced, and the ratio of natural gas and clean energy use has gradually enhanced (CSY, National Bureau of Statistics, China Statistical Yearbook, 2021). Moreover, technological breakthroughs in energy efficiency and clean energy utilization have enormously contributed to the clean and low carbonization growth of China’s economy along with the transformation of the energy structure (Abbasi et al., 2022). Furthermore, in the process of vigorously advocating and supporting the improvement of energy technology and efficiency, boosting the low carbonization transition of the energy system can rapidly reduce the pollutant emissions from fossil energy, thus promoting the green evolution of economy in China. Therefore, the hypothesis is proposed as follows:

Hypothesis I: Energy efficiency can effectively moderate the impact of energy transition on green growth; put differently, the interaction between energy transition and energy efficiency can facilitate China’s green growth.

In addition, under the current background of carbon neutrality, energy structure transition aims to reduce dependence on fossil fuel and enhance the use of renewable energy (Balsalobre-Lorente & Leitão, 2020). Among them, to decrease the high dependence on traditional solid fossil energy, improving energy utilization technologies, expanding the surface area of fuel combustion, and widely applying energy-saving process facilities are imperative, which can help reduce pollutant emissions (Zhao et al., 2020). At the same time, due to the limitations of the current evolution of the clean energy industry, accelerated energy efficiency can greatly enrich the renewable energy technologies, thus making renewable energy a vital engine for the green evolution of China’s economy; therefore, the second hypothesis is proposed as follows:

Hypothesis II: Energy productivity is the transmission route through which low-carbon energy transition affects green growth.

Model and data

Model

As proposed by Grossman and Krueger (1991), the environmental consequences of any economic activity can be divided into scale, technical, and structural effects. Since ecological environment is an important component of green growth, this study successively introduces control variables from three dimensions of economy, technology, and structure according to the theory of three major effects, which can be assessed by economic growth, energy efficiency, and industrial upgrading, respectively. Furthermore, following the works of Shang and Liu (2021) and Vona and Patriarca (2011), urbanization and income inequality are also added into the model; thus, the basic framework consisting of a series of control variables is presented as follows:

$$GG{I}_{it}=f\left( EC{S}_{it},E{E}_{it}, Pgd{p}_{it}, IS{U}_{it}, Ur{b}_{it}, GA{P}_{it}\right)$$
(1)

where i and t in the above equation indicate the province and year, respectively. f(·) refers to the functional relationship. GGI represents green growth, ECS denotes the low carbonization transition of energy, which is gauged by the energy consumption structure, EE, Pgdp, ISU, Urb, and GAP are energy efficiency, economic growth, industrial upgrading, urbanization evolution, and income inequality of each province, respectively.

Out article performs natural logarithm processing on the above equation (i.e., Eq. (1)) to solve the problems of data volatility and heteroscedasticity. Thus, Eq. (1), with the natural logarithm form, can be rewritten as follows:

$$\ln GG{I}_{it}={\alpha}_0+{\alpha}_1\ln EC{S}_{it}+{\alpha}_2\ln E{E}_{it}+{\alpha}_3\ln Pgd{p}_{it}+{\alpha}_4\ln IS{U}_{it}+{\alpha}_5\ln Ur{b}_{it}+{\alpha}_6\ln GA{P}_{it}+{\eta}_i+{v}_t+{\varepsilon}_{it}$$
(2)

where α0 is the intercept term. ηi, vt, and εit denote the province-specific impact, year-specific impact, and error item, respectively. α1 − α6 in Eq. (2) denote the estimated parameters. Based on the theoretical analysis in Section 2, the parameter of energy transition (i.e. α1) is expected to be negative.

Variables and data

Dependent variable

To empirically check the low-carbon energy transition–green growth nexus, it is crucial to effectively quantify China’s green growth. Therefore, by referring to the research of Zhao et al. (2022b), the comprehensive indicator system of green growth includes three main components: economic growth, people’s livelihood, and ecological environment. Notably, the research of Zhao et al. (2022b) also posts the concrete measures corresponding to the three second-level indicators and their attributes. Based on the improved entropy technique discussed by Zhao et al. (2021a), this study evaluates the green growth composite index (denoted as GGI) in China for the period 2004-2018 in various provinces. This sample interval is determined by the availability of statistical data. In addition, the sub-indices of green growth –– economic growth index (EGI), people’s livelihood index (PLI), and ecological environment (EEI) –– are also calculated; the change trend graph of the annual average of the four indexes is presented in Fig. 1. Notably, the related statistics of these indicators are collected from the (CSY, National Bureau of Statistics, China Statistical Yearbook, 2021), the China Energy Statistical Yearbook (CESY) (CESY, 2021), and the China Rural Statistical Yearbook (CRSY) (CRSY, 2021).

Fig. 1
figure 1

Time trend graph of the averages of all indexes

Full size image

As this figure shows, China’s green growth during the sample period (i.e., 2004-2018) presents a U-shaped feature; to put it differently, China’s green growth initially declines due to environmental degradation caused by rapid economic growth; after reaching the turning point, economic growth slows down, environmental pollution is alleviated, and a strong economic foundation provides a driving force for promoting residents’ welfare. Thus, the green economy in China is increasingly improving.

Independent variables

Following the work of Dong et al. (2021), energy consumption structure (expressed as ECS) is calculated by the proportion of coal and oil use converted into standard coal by the conversion coefficient in primary energy use. Notably, CESY (2021) provides the data on energy consumption.

Control variables

(1) Energy efficiency (EE). Ulucak (2020) confirms that environmental technologies play a fundamental role in promoting green growth, and environment-related technologies are positively related to green growth; in other words, improving energy efficiency is particularly important for promoting green growth (Ringel et al., 2016). Hence, this study expects its coefficient to be positive.

(2) Economic development (Pgdp). This variable is characterized by the per capita gross domestic product (GDP) (Dong et al., 2018; Shahbaz et al., 2013). The increase in per capita income offers capital and economic support for achieving the Sustainable Development Goals (Adedoyin et al., 2020), and it is necessary to incorporate this variable into the model to control economic development’s impact on green growth.

(3) Industrial upgrading (ISU). This usually denotes the process of transformation from labor- and capital-intensive industry to knowledge- and technology-intensive industry and from high-carbon secondary industry to high value-added tertiary industry, which is conducive to promoting the expansion of economic scale and improving the deteriorating ecological environment (Guo et al., 2020). Thus, industrial upgrading is added into the estimation model, which is gauged by the ratio of the output value of tertiary industry to secondary industry (Wang et al., 2019; Yuan et al., 2020).

(4) Urbanization level (Urb). To effectively solve the dilemma of traditional urban development models, local governments will also focus on resource conservation and pollutant mitigation in the process of promoting urban construction (Qian et al., 2021). Hence, the impact of urbanization level on green growth cannot be ignored, and is assessed by the proportion of the population in urban regions of total population in this province.

(5) Income inequality (GAP). As some scholars have stressed, income inequality in a country not only presents a driver for renewable energy use, but can also facilitate the mitigation of carbon dioxide (CO2) and fossil energy (Baležentis et al., 2020; Uzar, 2020). Thus, we introduce income inequality in the equation and expect the parameter symbol to be negative.

A sample dataset of China’s 30 provinces covering the period 2004-2018 is applied to check the underlying impact of energy transition on green growth from the provincial perspective. Notably, Hong Kong, Macao, Tibet Autonomous Region, and Taiwan are not incorporated due to the unavailability of statistics. In summary, to identify the numerical characteristics of the variables, the specific name and descriptive statistics of the variables are posted in Table 1. In this table, the data of the descriptive statistics as a whole show no large fluctuations, and there is no extreme value.

Table 1 Definitions and descriptive statistics
Full size table

Findings and discussion

Technically, the regression procedures of conducting empirical analysis are plotted in Fig. 2. Clearly, preliminary inspections (see Section 5.1), baseline regression (see Section 5.2), robustness checks (see Section 5.3), and further discussion (see Section 6) are incorporated in the empirical process.

Fig. 2
figure 2

The specific estimation steps in this study

Full size image

Preliminary inspections

Prior to the baseline regression, the multicollinearity and correlation among used variables are first checked (see Table 2). In this table, the test values of the variance inflation factor (VIF) of the variables used are all less than the critical value of 10, which meets the empirical rule of multicollinearity. This implies that no serious multicollinearity is found between explanatory variables. Furthermore, the maximum and minimum of the absolute value in the correlation matrix are 0.8904 and 0.0874, respectively, and all correlation coefficients are all substantial. This confirms that no serious problems, such as uncorrelated or highly correlated problems, exist between the various explanatory variables (Yuan et al., 2020). Thus, multicollinearity between variables does not need to be considered in subsequent analysis.

Table 2 VIF test and correlation matrix
Full size table

In Fig. 3, this study further presents the scatter plots between energy transition and green growth. Obviously, both energy consumption structure and green growth as well their sub-indexes show a significant negative correlation. This has provided preliminary evidence for the discussion of the energy transition-green growth nexus in China.

Fig. 3
figure 3

Correlation trend graph

Full size image

Baseline findings

Table 3 reports the empirical outcomes of random effect (RE), feasible generalized least squares (FGLS), the differential generalized method of moments (Diff-GMM), and the system generalized method of moments (Sys-GMM) methods. Since a potential issue of endogeneity may exist in the sample data, the Sys-GMM method is selected as the baseline estimated strategy, which can solve the endogeneity problem to some extent (Huang, 2010). In addition, there are still two reasons to support the Sys-GMM approach as the baseline estimation — one reason for choosing this method to assess the energy transition-green growth nexus is that the potential impact of energy transition on green growth in which hysteresis features may exist, and the other reason is due to the sample data characterized by large N (30 provinces) and small T (15 years).

Table 3 Benchmark regression outcomes
Full size table

As this table shows, in the dynamic panel estimation, the values of AR (1), AR (2), and Sargan reveal that the estimated results have passed the Arellano-Bond (A-B) and Sargan tests. The coefficients of the energy consumption structure (i.e., ECS) in the above four techniques (i.e., RE, FGLS, Diff-GMM, and Sys-GMM) are substantially negative; to put it differently, the gradual advancement of energy transition can boost the rapid growth of China’s economy. The results of the Sys-GMM show that a 1% reduction in the proportion of coal and oil use can effectively enhance China’s green economy by 0.209%.

On the one hand, the gradual transformation of energy structure plays a crucial role in addressing pollutant emissions; this view is also obtained by Yuan et al. (2022). Specifically, first, energy transition is a gradual transformation to a modern and clean energy system, which can significantly reduce the proportion of fossil energy with high-polluting features and increase the ratio of clean energy. This can effectively alleviate the deteriorating ecological environment (Murshed et al., 2021). Second, the transition of energy structure can greatly accelerate the high-speed evolution of the natural gas industry and provide perfect conditions and excellent opportunities for the prosperity of China’s renewable energy industry; in other words, clean energy has been vigorously developed in the current context (Yuan et al., 2018; Hao et al., 2021). Notably, Wang and Wang (2015) have ever obtained a conclusion: the growing use of clean and renewable energy can obviously mitigate ecological damage.

On the other hand, in addition to optimizing the ecological environment, energy transition is equally critical in improving residents’ well-being and boosting economic growth. More specifically, continuously improving technologies in the energy transition process provides a strong guarantee for economic stability and sustainable development (Mohsin et al., 2021). In addition, environmental improvement in the context of energy transition can significantly increase residents’ quality of life and promote household welfare. It is also confirmed by Bohlmann et al. (2019) that the low-carbon upgrade of the energy industry can help eliminate sunset industries and accelerate the derivation of emerging industries, thus providing residents with more job opportunities, increasing income levels, and expanding the scale of the country’s economy.

Regarding the control variables, it is perceptible that improved energy productivity and urbanization evolution are positively associated with green growth in China, while the high-speed evolution of the economy at this stage and the gradual expansion of the income gap are not conducive to the accelerated promotion of a green economy. To be specific, strengthening energy efficiency can effectively expand the fuel combustion area, reduce total energy use, promote the development of clean energy industry, and accelerate green economic growth. This conclusion is confirmed by Ringel et al. (2016). At this stage, China is committed to the construction of smart and green cities. The growing rise of an energy-conserving construction industry, the wide use of energy-saving materials, and the large-scale popularization of natural gas have laid the foundation of green development (Qian et al., 2021; Viitanen & Kingston, 2014). Economic evolution driven by massive fossil fuel consumption and the widening income gap are not conducive to the green evolution of economy and society. Hence, strengthening the country’s energy technology reforms and promoting the construction of smart, low-carbon cities are essential for green growth.

Robustness checks

Applying the proxy variable of energy transition

To quantitatively examine the sensibility of the above empirical finding –– low-carbon energy transition negatively affects green growth in China, we first conducts a robustness check by applying the alternative measures of energy transition. In other words, this study replaces ECS in Eq. (2) with the ratio of fossil energy use (ECS_1; the proportion of coal, oil, and natural gas converted into standard coal to primary energy use) and the ratio of natural gas use in primary energy use (ECS_2) to re-estimate the econometric model by using the various methods.

The estimated outcomes of the above four techniques are listed in Table 4. Notably, regardless of the approach and indicator, the positive stimulus of clean energy transition on green economic growth is reliable. Furthermore, the coefficients of ECS_1 and ECS_2 are -0.325 and -0.041, respectively, and provide an interesting conclusion: the decline in the share of fossil energy use has been significantly greater than the combined effect of coal and oil in strengthening China’s green growth. As a relatively green transitional fuel, decreasing the application and dependence on natural gas will also support and stimulate the acceleration of green growth. In addition, the coefficient sign and significance of the control variables in Tables 3 and 4 are basically consistent; this strongly proves the reliability of the estimated outcomes.

Table 4 Outcomes of applying the proxy variable of energy transition
Full size table

Eliminating the particular values in the sample

In this subsection, the statistical data of Beijing, Tianjin, Shanghai, and Chongqing are excluded due to the prominent data in these municipalities to further test whether the benchmark regression outcomes are robust (see Table 5). The coefficients of lnECS in the four methods are substantially negative at the 1% level. This finding show that the estimation outcomes after excluding particular values still support the conclusion that transforming energy structure can boost the green evolution of economy.

Table 5 Outcomes of eliminating the particular values
Full size table

Addressing the endogenous issue

In this section, this study tends to conduct the robust check by using the alternative estimated approaches which can efficiently solve endogeneity — instrumental variable (IV) technique built under the condition of the heteroscedasticity put forward by Lewbel (2012) and two stage least squares (2SLS) method. Notably, the latter mainly applies the trade openness (denoted as Tra; gauged by the ratio of import and export trade converted into RMB to total output value of each province) (Pan et al., 2021; Stevens et al., 2013).

Table 6 lists the corresponding test findings. In the 2SLS estimates, the p-value of the underidentification test (i.e., K-P rk LM) is significant at the 1% level, while the statistical value of the weak identification test (i.e., K-P rk Wald F) is significantly larger than the critical value (i.e., 16.38). Furthermore, the p-value of the Durbin-Wu-Hausman (DWH) test verifies the existence of endogeneity. The results of these three tests also verify the validity of using 2SLS method to deal with endogeneous problems. In addition, as posted in Table 6, the estimated parameters of energy structure and energy efficiency are negative and positive, respectively, which again verifies the main conclusions of this study — accelerating low-carbon energy transition and improving energy efficiency can facilitate green growth.

Table 6 Outcomes of alternative estimated techniques
Full size table

Dividing the composite index into three sub-indexes

Except for applying the alternative measures and regression methods, this study further examines the sensibility of the regression outcomes by investigating the impacts of energy transition on the sub-indexes of green growth; the test findings are illustrated in Table 7.Obviously, whether the proportion of coal and oil consumption (ECS) or fossil energy consumption (ECS_1) is used, low carbonization transition of energy has an obvious effect on boosting economic evolution, deepening household social welfare and treatment, and strengthening the construction of ecological civilization. In other words, the optimization of the current energy structure can help accelerate green growth, which significantly verifies the empirical findings.

Table 7 Outcomes of dividing the composite index into three sub-indexes
Full size table

Further discussion

Moderating role of energy efficiency

After investigating the causal low-carbon energy transition–green growth nexus in China, we further check how energy efficiency moderates the process of optimizing energy framework in boosting green economic growth. To this end, the interaction term of ECS and EE is introduced into Eq. (2) for further estimation. Table 8 provides the empirical results by applying ECS and ECS_1 simultaneously. In this table, (1) and (3) indicate the regression outcomes without adding the interactive term, while (2) and (4) represent the empirical results by introducing the interaction term.

Table 8 Regression outcomes of the moderating analysis
Full size table

In (1), it is perceptible that the coefficients of lnECS and lnEE are negative and positive, respectively, while that of the interactive item is obviously positive. This implies that improved energy efficiency can effectively adjust the dynamic energy transition-green growth nexus; specifically, the lower the share of solid fuel use, the greater the role of energy efficiency in accelerating green growth. In other words, as energy efficiency increases, the lower the ratio of solid fuel use is, the more conducive it is to promoting green growth. Notably, low carbonization transition of the overall energy system refers to the gradual shift from an energy framework in which highly polluting solid fuels occupy a large share to an energy system in which clean energy is the main fuel (Dogan et al., 2022). In this process, energy efficiency has increased the output value created by unit energy input, reduced primary energy consumption, especially fossil energy, and improved the level of environmental protection technology. This is of great value in boosting low-carbon energy transition, thus accelerating the process of green growth. Furthermore, by comparing and analyzing the regression results without and with interaction term, it can be found that when introducing the interaction term, the absolute values of the parameters of lnECS and lnEE increase significantly, which emphasizes that the interactive effect between the clean transition of the energy framework and the substantial improvement of energy productivity can effectively strengthen their role in promoting green growth.

Mediation effect analysis

Model construction

Considering the significant moderating effect of energy productivity in the low-carbon energy transition–green growth link, another question is also worth discussing: whether the continuous improvement of energy productivity is a powerful influence path to boost green economic growth in the process of evolving energy structure? To this end, following the research of Yuan et al. (2020), Chen and Lee (2020) and Zhao et al. (2021b), the classical mediation effect model is employed to clarify how low-carbon energy transition affects green growth in China through energy efficiency. The conventional equations are presented as follows:

$$y=\beta x+{\mu}_1$$
(3)
$$m=\alpha x+{\mu}_2$$
(4)
$$y={\beta}^{\prime }x+\eta m+{\mu}_3$$
(5)

The specific procedures for checking the mediation effect between variables are posted as follows: (1) Testing the total impact of x on y in Eq. (3), if the parameter of x (i.e., β) is significant, a total effect exists; (2) checking the indirect effect between x and y; in other words, examining the parameter of x in the second equation (i.e., α) and the parameter of m in Eq. (5) (i.e., η). If both are substantial, we can check the direct effect; otherwise, the Sobel technique is used to check whether the product of a and η is prominent, if the null hypothesis (H0: aη=0) is rejected, a direct effect exists; and (3) checking the parameter of x in Eq. (5), if β′ is not significant, only a mediation effect exists; on the contrary, if β′ is significant, there is a partial mediation effect.

In this study, x is low-carbon energy transition, m is energy efficiency, and y is green growth. By incorporating the lagging term of the explained variable and control variables, the concrete mediation effect model is built as follows:

$$\ln GG{I}_{it}={\varphi}_0+{\varphi}_1\ln GG{I}_{i,t-1}+{\varphi}_2\ln EC{S}_{it}+\sum_{k=3}^6{\varphi}_k\ln Ctr{l}_{it}+{\eta}_i+{v}_t+{\varepsilon}_{it}$$
(6)
$$\ln E{E}_{it}={\xi}_0+{\xi}_1\ln E{E}_{i,t-1}+{\xi}_2\ln EC{S}_{it}+\sum_{k=3}^6{\xi}_k\ln Ctr{l}_{it}+{\eta}_i+{v}_t+{\varepsilon}_{it}$$
(7)
$$\ln GG{I}_{it}={\alpha}_0+{\alpha}_1\ln GG{I}_{i,t-1}+{\alpha}_2\ln EC{S}_{it}+{\alpha}_3\ln E{E}_{it}+\sum_{k=4}^7{\alpha}_k\ln Ctr{l}_{it}+{\eta}_i+{v}_t+{\varepsilon}_{it}$$
(8)

where φ0, ξ0, and α0 refer to the constant term, and φ1 − φ6, ξ1 − ξ6, and α1 − α7 are the estimated parameters. Ctrlit includes Pgdp, ISU, Urb, and GAP.

Empirical results

This study examines the mediation effect model through the stepwise regression method and lists the corresponding findings in Table 9. Furthermore, the robust results are listed in Table 9 by applying ECS_1 as the independent variable. The statistical data of (1)-(3) and (4)-(6) in the table are the estimated outcomes of Eq. (6)-(8), respectively.

Table 9 Regression outcomes of the mediation analysis
Full size table

(1) in Table 9 lists the empirical findings of estimating Eq. (6). Obviously, a decrease in the proportion of coal and oil use of total energy use by 1% can promote China’s green growth by 0.226%; to put it differently, the accelerated transition of the current energy consumption structure can enhance the green evolution of the economy. The corresponding reasons can refer to the interpretation of the benchmark regression results discussed in Section 4.2.

The empirical findings of estimating Eq. (7) by applying the Sys-GMM approach are illustrated in (2) of Table 9. As this table shows, a 1% decrease in the ratio of coal and oil consumption can improve energy efficiency by 0.147%. This stresses that low-carbon energy transition positively facilitates China’s energy utilization efficiency. Specifically, to realize the Sustainable Development Goals and ensure the long-term sustainability of the economic system, which relies on a favorable ecological environment, vigorously optimizing the current energy structure and encouraging its fast transition are imperative. On the one hand, local governments should vigorously advocate decreasing dependency on the total consuming of nonrenewable energy dominated by coal and petroleum (Pingkuo & Huan, 2022); on the other hand, under the premise that it is difficult to reduce total energy use and widely apply clean energy in the short term, actively encouraging relevant energy enterprises to accelerate pollution reduction technology reforms, greatly increasing the degree of solid fuel combustion, optimizing highly polluting fuel stoves, and strengthening energy productivity have become the corresponding measures that the nation needs to take urgently to promote energy system reform and transition (Balsalobre-Lorente et al., 2021). In addition, the government will take measures to stimulate the rapid advancement of renewable energy industries and greatly strengthen related infrastructure to accelerate the improvement of energy efficiency (Hao et al., 2021).

Regarding (3) in Table 9, the regression parameter of lnECS is substantially negative. This underscores the existence of direct effect in the ECS–GGI nexus. Besides, the parameter of lnEE in (3) shows that, if energy efficiency increases by 1%, China’s green growth will improve by 0.097%. This empirically verifies that improved energy productivity is beneficial for enhancing green advancement of economy. As Ringel et al. (2016) stressed, relevant policies and regulations of ameliorating energy productivity play the crucial roles in stimulating the transition to a harmonious, green, clean, and sustainable economy. This finding is also supported by Ulucak (2020) and Dell’Anna (2021): improved energy efficiency is positively associated with the gradual development of a green economy.

In summary, it is concluded that the mediation effect between low carbonization transition of energy and green growth exists. In other words, the gradual transition of the energy use framework dominated by high-polluting fossil fuel can promote green and sustainable development by improving energy efficiency while directly accelerate green growth in China. Through the moderating and mediation analysis, the theoretical mechanism of energy transition in affecting green growth through energy productivity is plotted in Fig. 4.

Fig. 4
figure 4

The theoretical mechanism chart between energy transition and green growth

Full size image

Conclusions and policy implications

Our discussion on the relationship between energy transition and green growth effectively supplements the current literature. In this study, the causal energy transition–green growth nexus is examined by using multiple green growth indicators and a dataset covering the period 2004-2018. Additionally, whether reducing energy intensify can effectively regulate and serve as a valid transmission pathway for energy transition and green growth has become a question worthy of exploration. The main findings are stressed as follows:

  1. (1)

    The outcomes of the benchmark regression insist that a 1% decrease in the ratio of fossil energy use can greatly boost the speed of green evolution of China’s economy by 0.209%. This emphasizes the positive energy transition-green growth nexus, which implies that the continuous evolution of low carbonization transition of energy industry can help accelerate the process of green growth in China. This general conclusion is also confirmed by a series of sensitivity checks.

  2. (2)

    Another interesting finding is related to the crucial role of energy efficiency in moderating the energy transition-green growth nexus; more specifically, the product coefficient of ECS and EE is substantially positive, implying that the interactive actions of transforming energy structure and enhancing energy productivity can vigorously promote the green evolution of China’s economy and accelerate the effects of energy structure adjustment and energy efficiency in facilitating green growth.

  3. (3)

    We also discover that improved energy productivity can serve as a favorable and stable transmission route in the process of energy transition impacting green growth. By implications, actively giving impetus to the low carbonization transition of the energy system will further boost the green development of the economy by increasing the energy productivity while directly expediting China’s green economic evolution.

Following the above empirical outcomes, a series of policy implications are put forward to achieve the goal of green growth. First, the basic finding drawn from multiple provinces in China effectively stresses the importance of accelerating clean transition of the existing energy structure, namely from a high-carbon energy structure dominated by fossil energy to a low-carbon energy system centered on clean energy. Notably, low-carbon energy transition is a long and arduous task that requires the efforts of both local governments and the market to effectively achieve the goals, especially in the post COVID-19 epidemic era. Under the guidance of the current carbon neutrality target, it is imperative for government to create policy environment by promulgating corresponding policies to facilitate the rapid evolution of clean energy industries and provide financial support to enterprises that actively improve energy combustion facilities and innovate clean technologies. In addition, the market should also actively play the role of its “invisible hand” to regulate and standardize enterprises’ production and R&D activities. To this end, enterprises should strengthen technological innovation and cooperation in accordance with the national policy direction, accelerate the course of energy productivity reform, change the traditional concept of interests, and increase the publicity of reducing pollutants and conserving energy.

Second, the empirical findings show that improved energy efficiency is conducive to regulating energy transition and green growth nexus, and is also an important channel in the ECS–GGI nexus; therefore, local governments and enterprises should do their utmost to enhance energy efficiency innovation activities. COVID-19 pandemic, which has lasted for three years, has severely damaged the country’s economic aggregate and hindered the rapid pace of economic advancement. In this background, optimizing and upgrading the existing industrial framework, the rapid evolution of urbanization, and the strengthening of independent innovation ability are all valid measures to improve energy productivity. Accordingly, the development of an industrial structure dominated by the service industry and the establishment of a complex, functional, and energy-conserving urbanization development model are crucial guarantees for the rapid transition of the future economic development mode. In addition, local governments should increase scientific and technical input, encourage the R&D of energy-saving technologies in the form of subsidies or tax breaks, or use financial support to improve national independent innovation capacity in the energy industry, so as to accelerate energy efficiency and relieve the situation of increasingly tight resources.

Third, the outcome of narrowing income gap in quickening green growth implies that narrowing income inequality between urban and rural regions and achieving common prosperity are effective channels to boost the China’s sustainable economic development. Specifically, the government should further increase its support of township enterprises, improve the distribution system, and increase subsidies to farmers. At the same time, it is crucial for provincial policy makers and relevant stakeholders to take measures to improve the urban-rural household registration regulations, actively enhance the supply of social security facilities, promote the deep integration of urban and rural areas, and narrow the household income gap. In addition, the country should gradually strengthen the construction of urbanization and accelerate its development and promote the gradual transformation of the rural population into urban residents. Under these measures, farmers’ income will gradually increase, and the urban-rural income gap will narrow, thereby helping to speed up the green growth of China’s economy.

Our article only offers an initial piece of evidence for exploring the energy transition-green growth nexus, there are still some understudies. One involves the issue of sample data. This article mainly detects the green growth effect of energy transition from the dimension of provincial panel data. In future studies, we will further analyze and discuss the relationship between variables from city-level or industrial dimensions. Another is associated with determinants of green growth. Our study only checks the effect of transforming energy structure on green growth. By referring to the works of Naeem et al. (2022) and Dutta et al. (2023), empirically discussing whether green finance and green investment are valid measures to promote green growth is imperative in future studies.