Ecological Indicators 169 (2024) 112898

Contents lists available at ScienceDirect

Ecological Indicators
journal homepage: [Link]/locate/ecolind

Original Articles

Advancing high-quality development in China: Unraveling the dynamics,

disparities, and determinants of inclusive green growth at the
prefecture level
Yufei Ren a, Xinna He a , Qunou Jiang a,* , Fan Zhang a, Bitian Zhang b,*
a
Department of Geography, Beijing Forestry University, Beijing 100083, PR China
b
Policy Research Center for Environment and Economy, Ministry of Ecology and Environment of the People’s Republic of China, Beijing 100029, PR China

A R T I C L E I N F O A B S T R A C T

Keywords: Inclusive green growth (IGG), which underscores the delicate balance between economic expansion, ecological
Inclusive green growth sustainability, and social well-being, has garnered substantial scholarly attention. Despite China’s remarkable
Two-stage SBM model economic growth in recent decades, this rapid development has also given rise to significant social and envi-
New-type urbanization
ronmental challenges. Thus, advancing IGG is pivotal for ensuring China’s transition toward high-quality, sus-
Panel quantile regression
tainable development. Here, a novel evaluation framework of IGG is established by incorporating the nexus
Chinese prefecture-level cities
between urban production and governance. Our approach combines a two-stage network SBM model with a
coupling coordination degree model, allowing for a comprehensive assessment of IGG across 281 Chinese
prefecture-level cities from 2006 to 2019. The spatiotemporal variations and regional disparities in IGG are
analyzed using the Dagum Gini model. Furthermore, we employ a panel quantile model and a spatial panel
regression model to explore the mechanisms through which new-type urbanization (NTU) and other socio-
economic factors influence IGG, with particular attention to individual heterogeneity, temporal dynamics, and
spatial spillover effects. The findings reveal that although urban IGG in China has generally trended upward,
substantial potential for further improvement remains. High-IGG cities are concentrated in the eastern region,
but the gap between cities is gradually narrowing. The panel quantile model demonstrates that NTU and its
components positively influence urban IGG, though the impact varies considerably across cities with different
IGG levels. Additionally, temporal heterogeneity and spatial econometric analyses indicate that the influence of
NTU on IGG is dynamic and exhibits significant spatial spillover effects. The policy implications highlight the
importance of tailored regional strategies to promote green technology adoption, industrial upgrading, and in-
clusive public services, particularly in underdeveloped regions.

1. Introduction 2020; Yuan et al., 2020). As the world’s second largest economy, China’s
undeniably high economic growth over the long term since the reform
Since entering the Anthropocene, social and economic systems and opening up has injected surging momentum into the global econ-
driven by unsustainable resource extraction and consumption could omy. At the same time, China’s extensive economic growth model, along
irreversibly destabilize the Earth system (Rockström et al., 2023). With with issues such as poor quality of development, inefficiency, and
over 80 % of global GDP being generated in cities, urbanization has the ecological degradation, have consistently been challenges the Chinese
potential to contribute significantly to sustainable growth through government has sought to address (Ren et al., 2022b; Kong et al., 2023).
enhanced productivity and innovation, provided it is effectively In particular, the proliferation of “black swan” events, such as the
managed (Shan et al., 2021). However, urbanization is also a double- outbreak of the COVID-19, could lead to increased inequality among
edged sword. High-intensity economic activities not only contribute to vulnerable social groups and a shock to their livelihood resilience (Qian
global warming, environmental pollution and ecological degradation, and Fan, 2020; Liang et al., 2022). In fact, the Chinese government has
but may also cause polarization in social development (Fenoglio et al., proposed high-quality development address these problems in 2017. It

* Corresponding authors at: Beijing forestry university, Policy Research Center for Environment and Economy, Ministry of Ecology and Environment of the People’s
Republic of China, PR China.
E-mail addresses: jiangqo@[Link] (Q. Jiang), [Link]@[Link] (B. Zhang).

[Link]
Received 2 October 2024; Received in revised form 28 October 2024; Accepted 22 November 2024
Available online 5 December 2024
1470-160X/© 2024 The Author(s). Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license ([Link]
nc-nd/4.0/).
Y. Ren et al. Ecological Indicators 169 (2024) 112898

emphasizes cultivating new growth drivers through innovation and on evaluation indicators and measurement methods of regional IGG. In
enhancing the quality and efficiency of economic development, moving this context, this work reviews the progress of related research, iden-
away from the focus on rapid growth rates of the past (Pan et al., 2021). tifies current challenges, and proposes improvement methods. The
Simultaneously, it seeks to enhance residents’ sense of happiness and current measure of green growth is based on statistical indicators. Ac-
fulfillment while improving the provision of public services. Addition- cording to the most widely used definition of IGG, existing research on
ally, it aims for the efficient use of resources, effective environmental evaluation indicators has shifted from a comprehensive indicator system
governance, and ecosystem restoration. Therefore, harmonizing China’s to an input–output framework. In the comprehensive indicator system,
green economic development, environmental governance and social the widely used indicator is the inequality-adjusted human development
governance is not only important for China to achieve high-quality index (Maji, 2019), but this indicator only portrays the imbalance of
development, but also has important positive implications for global social development and cannot comprehensively reflect the full picture
sustainable development. of green development. Kumar (2017) suggests evaluating IGG by
The literature on sustainable development suggested that attaining monitoring human capital, productive capital (manufacturing output or
strong sustainability—characterized by the absence of trade-offs be- GDP), and natural capital stock over time, which led to the proposal of
tween economic, social, and ecological objectives—is exceedingly rare; the Inclusive Wealth Index (IWI). Recently, some scholars have also
political decision-making frequently prioritizes economic interests, proposed a comprehensive index system that includes economic growth,
often to the detriment of social and ecological considerations (Gupta and income distribution, equity of opportunity, green development, and
Vegelin, 2016). Unfortunately, simply promoting economic growth environmental protection, aiming to realize a coordinated relationship
cannot effectively address the environment degradation, income dis- between the economy, society, and ecology (Gu et al., 2021; Fan et al.,
parities, and poverty that persist globally (Rammelt and Gupta, 2021). 2023). Data envelopment analysis (DEA) can maximize economic output
Meeting human needs and improving quality of life through the provi- while minimizing undesired output, laying the groundwork for inclusive
sion of competitively priced goods and services, while progressively green growth measurement (Sun et al., 2020; Ren et al., 2022a; Sun
reducing ecological impacts and resource intensity, are seen as the goals et al., 2022; Zhao et al., 2022a; Ma et al., 2024). While most conven-
for addressing these issues (Hickel, 2020; Ren et al., 2022b; Ofori et al., tional DEA models treat the operation of a Decision-Making Unit (DMU)
2023). This global concern is articulated in the UN’s 2030 Agenda, “The as a “black box” (Färe and Grosskopf, 2000; Kao, 2018), Certain research
Future We Want,” which seeks to mitigate the strain on natural re- efforts have aimed to unpack the “black box” by meticulously exploring
sources, address climate change, and foster greener, more inclusive the internal composition of DMUs. Notably, network DEA models have
growth. Consequently, the concept of Inclusive Green Growth (IGG) has made considerable strides in recent years, facilitating a more refined
emerged, emphasizing the equitable and deliberate reduction (de- analysis of the internal dynamics within a DMU, which frequently
growth) of over-consumption, over-accumulation, and expropriation to involve multiple interconnected sub-processes or stages (Tone & Tsut-
enhance human well-being, social justice, and environmental health. sui, 2009). Each stage is characterized by its unique inputs and outputs
Although there is no broad consensus on the definition of IGG, numerous and is linked to subsequent stages via intermediate flows, thereby
scholars have provided useful insights (Sun et al., 2020; Ren et al., capturing the complexity of the entire system (Kao, 2018). The Network
2022a; Wang et al., 2022; Ben Amara and Qiao, 2023; Fan et al., 2023; DEA model presents distinct advantages compared to traditional DEA
Ofori et al., 2023). For instance, Sun et al. (2020) proposed a definition approaches, which typically assume a single-stage process or indepen-
of IGG from three perspectives: social governance, economic develop- dent sectors. It more accurately reflects real-world scenarios by
ment, and the ecological environment; Aslam and Ghouse (2023) considering multi-stage, interrelated processes and treating intermedi-
pointed out that IGG is a complex concept encompassing economic ate products as both inputs and outputs. For example, Yang et al. (2024)
growth, environmental sustainability, institutions, and social digitiza- proposed a meta two-stage parallel entropy dynamic DDF-DEA model to
tion. In general, the core of relevant definitions is to achieve the uni- re-measure industrial eco-efficiency of 30 provinces in China from 2015
versal sharing of economic and ecological wealth under the premise of to 2021. Given that the city is a triadic and complex system encom-
economic growth. Notably, investment in public services that meet the passing economic production, environmental governance, and social
welfare needs of the population has long been viewed as economically development, harmonizing these three relationships is a critical aspect
inefficient (Tullock et al., 2002). of IGG’s connotation. To this end, a potential innovation of this study is
In China, balancing socio-economic development and environmental to combine the network DEA model and the coupled coordination de-
governance often presents a trade-off for policymakers, who typically gree model (CCDM) to calculate the IGG.
prioritize economic growth (Liu et al., 2020a). This focus can result in In recent years, a substantial body of literature has sought to uncover
government failures in the provision of public goods due to the con- the driving mechanisms behind the factors influencing IGG. These
centration of financial resources in these areas. Although existing studies predominantly utilize various models such as the Dual Machine
studies have pointed out the importance of increasing the supply of Learning model (Feng et al., 2024), Panel Threshold Model (Ma et al.,
public facilities to stimulate economic growth and improve environ- 2024), Spatial Difference in Difference model (SDIDs) (Ren et al.,
mental quality, there is a lack of attention to this in the relevant defi- 2022a), spatial econometric model (Fan et al., 2023) across different
nitions. Therefore, further research and analysis are needed to better spatial scales, particularly at the national and provincial levels. The
understand the concept of inclusive growth and its role in achieving empirical findings indicate that factors such as energy efficiency (Ofori
inclusive green growth. To fill this research gap, our potential contri- et al., 2022), economic policy uncertainty processes (Gu et al., 2021),
bution to the community is to enrich the conceptual framework of IGG government intervention (Aslam and Ghouse, 2023), innovation (Jia
based on an input–output framework, which refers to a development et al., 2023), digital economy agglomeration (Ren et al., 2022a), FDI
model that balances the relationship among green productivity, eco- (Ofori et al., 2023), and industrial upgrading (Fan et al., 2023) signifi-
nomic growth and social governance. Productive efficiency here refers cantly impact IGG. It is worth noting that the concept of IGG resonates
to the ability to efficiently allocate factor inputs to increase economic with the “new-type urbanization” strategy delineated in the National
output per unit and reduce environmental pollution; governance effi- New-type Urbanization Plan (2014–2020), promulgated by the Gov-
ciency refers to the ability to efficiently allocate financial resources to ernment of China in 2014. In contrast to traditional urbanization, new-
increase the supply of public services and facilities per unit and reduce type urbanization (NTU) underscores a people-oriented paradigm that
the income gap, thereby sustaining urban environmental management. integrates not only the economies of scale of cities but also their
This integration ensures that economic growth is not only quantitatively ecological quality and public services. Although the ecological ramifi-
robust but also qualitatively sustainable. cations of NTU have been extensively studied in the academic realm, the
IGG researchers have conducted numerous solid and fruitful studies impact of NTU on IGG has been sparsely examined, with the dynamics of

2
Y. Ren et al. Ecological Indicators 169 (2024) 112898

changes in NTU largely overlooked. Moreover, many studies overlook studies focus on traditional urbanization metrics, this paper breaks new
the consideration of urban individual heterogeneity in econometric ground by analyzing how NTU’s emphasis on sustainability and inclu-
analysis, meaning that cities at different stages of development may sivity influences IGG outcomes, particularly in terms of green produc-
respond differently to influencing factors. This limitation can be tivity, governance efficiency, and social welfare. The research process of
addressed using panel quantile regression. Therefore, it is crucial to this study is structured into three key steps (Fig. 1).
develop a comprehensive NTU assessment index and employ both panel
quantile regression and spatial econometric models to examine whether 2. Methodology and data
improvements in NTU and its components contribute to enhancing IGG
in Chinese prefecture-level cities. 2.1. Two-stage network slack-based DEA model with undesirable output
Motivated by this context, based on panel data from 281 prefecture-
level cities in China from 2006 to 2019, this study first proposes a new Compared to traditional DEA methods, network DEA offers signifi-
conceptual framework for IGG, emphasizing the coordination between cant advantages in capturing the complexity of urban systems. Urban
the economic environment and government governance, and analyzes economic-social processes analyzed within the DEA framework are
its spatiotemporal evolution and regional disparities. Subsequently, the typically categorized into two approaches. The first treats urban pro-
nonlinear driving mechanisms and spatial spillover effects of influencing duction activities as a one-stage process within a black-box DEA
factors on IGG are empirically examined using panel quantile regression framework (Ren and Li, 2022; Zhao et al., 2022a). The second approach
and the spatial panel Durbin model. employs a separation DEA framework, disaggregating urban production
This study strives to make three main contributions to the IGG activities into multiple processes (Lo and Lu, 2009). However, these
research community: (i) This study develops a novel evaluation frame- processes are often treated as independent, failing to account for the
work for IGG by combining the two-stage network Slack-Based Measure interdependencies between urban processes. Network DEA, by contrast,
model with the Coupling Coordination Degree Model (CCDM). While not only recognizes the existence of urban multi-stage processes but also
previous research has often used static econometric models or simple effectively reflects the interdependence between them through the in-
input–output frameworks, this paper goes beyond that by introducing a clusion of intermediate products (inputs for one process and outputs for
dynamic, multi-stage approach that considers the interaction between another). As a result, this approach yields results that more accurately
production efficiency and governance efficiency. The innovative use of align with real-world dynamics (Tone and Tsutsui, 2009). In this study,
CCDM to measure the synergistic coordination between these sectors the city is modeled with two modules—production and governance
provides a more comprehensive evaluation of IGG, addressing both processes, represented as sectors k and ℎ—which cooperate and interact
economic and governance dimensions. (ii) Unlike conventional studies to drive urban operations. Consider N DMUs (i = 1,2,⋯,N), where each
that rely on average effects in regression analysis, this paper employs DMU consumes M inputs to produce E desirable (good) outputs and F
panel quantile regression to capture the heterogeneous impacts of NTU undesirable (bad) outputs. Let X =(x1, x2,…, xM) ∈ RM + , a G =(g1, g2, …,
across cities with varying levels of IGG. This is the first attempt to use gE) ∈ RE+, and B =(b1, b2, …, bM) ∈ RF+ represent the inputs, good outputs,
this method in the context of NTU and IGG, providing insights into how and bad outputs, respectively. Table 1 shows the production technolo-
cities at different stages of development respond to NTU interventions. gies under different DEA frameworks.
This approach allows for a deeper understanding of the nonlinear effects A comparison of the production technologies across different DEA
of NTU and its components on IGG, which previous research has often frameworks reveals that, unlike the separation DEA framework, the
overlooked. (iii) This study is among the first to systematically explore network DEA framework accounts for the linkage constraints of inter-
the role of New-Type Urbanization (NTU) in driving IGG across Chinese mediate products and integrates the stage processes of the two sectors
prefecture-level cities. It introduces a comprehensive NTU index system into a unified model (Tone and Tsutsui, 2009,2010). Here, Z(k,h) denotes
that incorporates green-oriented economic urbanization, inclusive so- the flow of intermediate products from sector k to sector ℎ. Various
cial urbanization, and ecological development. While many previous methods for handling these intermediate product linkages have been

Fig. 1. Research framework.

3
Y. Ren et al. Ecological Indicators 169 (2024) 112898

Table 1
Comparison of production technologies with different DEA frameworks.
DEA framework Production technology Production technology (constant returns to scale)
⎧ ⎫
T = {(X, G, B) : X can produce (G, B) } ⎪ ∑N ∑N ⎪
⎨ (X, G, B) : X ≥
⎪ λ X ,G ≤
i=1 i i
λG ⎪
i=1 i i
⎬
T = ∑
⎪ N ⎪
⎩ B≥
⎪ λ B , λ ≥ 0(i = 1, 2, ⋯, N) ⎪
i=1 i i i
⎭

{( ) ( )} ⎧ ⎫
Tk = Xk , Gk , Bk , Zk : Xk can produce Gk , Bk , Zk Th = ⎪ ( ) ∑N ∑N ⎪
{( h h h h ) ( ) } ⎪
⎨ Xk , Gk , Bk , Zk : Xk ≥ i=1
λk k
i Xi , G k
≤ i=1
λ k k
i Gi
⎪
⎬
X , Z , G , B : (Xh , Zh ) can produce Gh , Bh Tk = ∑ ∑ Th =
⎪ k N k k k N k k k ⎪
⎩Z ≤
⎪ λ Z ,B ≥
i=1 i i
λ B , λi ≥ 0(i = 1, 2, ⋯, N) ⎪
i=1 i i
⎭
⎧ ⎫
⎪ ( h h h k ) h ∑N h h h ∑N h h ⎪
⎪
⎨ X ,G ,B ,Z : X ≥ λ X ,G ≤
i=1 i i
λ G
i=1 i i
⎪
⎬
⎪ ∑N ∑N ⎪
h h h h h h h
⎩Z ≤
⎪ λ Z ,B ≥
i=1 i i
λ B , λi ≥ 0(i = 1, 2, ⋯, N) ⎪
i=1 i i
⎭

{( ) ( )} ⎧ ⎫
Tk = Xk , Gk , Bk , Zk : Xk can produce Gk , Bk , Zk Th = ⎪
⎪ ⎪
⎪
{( h h h h ) ( h h) } ⎪
⎪ ( ) ∑N ∑N ⎪
⎪
h h
X , Z , G , B : (X , Z ) can produce G , B Xk , Xh , Gk , Gh , Bk , Bh , Z(k,h) : Xk ≥
⎪
⎪
⎪
⎪ λk Xk , Xh ≥
i=1 i i
λh Xh ,
i=1 i i
⎪
⎪
⎪
⎪
⎪
⎨ ⎪
⎬
∑N ∑N ∑N ∑N
T = G ≤
k k k
λ G ,G ≤ h h h
λ G ,B ≥ k k k
λ B ,B ≥h
λk Bh ,
⎪ i=1 i i i=1 i i i=1 i i i=1 i i ⎪
⎪
⎪ ∑ ∑ ⎪
⎪
⎪ N N ⎪
λk Z λh Z
⎪ (k,h) k h ⎪
, λi , λi ≥ 0(i = 1, 2, ⋯, N)
⎪ (k,h) ⎪
⎪
⎪ i=1 i i
= i=1 i i
⎪
⎪
⎪
⎩ ⎪
⎭

Note: λi indicates the intensity vector. Zk and Zh are the intermediate outputs of sector k and the intermediate inputs of sector h, respectively.

proposed in the literature. Considering the practical situation, where governance efficiency (GE) and production efficiency (PE), as outlined
governments can adjust the supply of public services and products based in Eqs. (2) and (3), respectively (Tone and Tsutsui, 2009). GE quantifies
on budget allocations (Xiao et al., 2021), this study assumes that the effectiveness of the government sector in delivering public goods
∑N k (k,h) ∑
= Ni=1 λhi Zi , indicating that intermediate products are and services, calculated through Eq. (2). PE assesses the performance of
i=1 λi Zi
(k,h)

freely connected. In other words, the optimal value of the link activity the production sector, including profit generation and the creation of
may be equal to, less than, or greater than its observed value. environmental by-products, with the PE of the industrial sector derived
To evaluate both the overall and sectoral efficiency of DMUs, a from Eq. (3).
network DEA model based on slack is proposed—referred to as the ∑ k smk
1 − M1k M
network slack-based model—which incorporates intermediate goods. GE = ⎛
mk =1 xmk
⎞ (2)
The objective function is modified accordingly to address undesired ∑ ∑
1+ 1 ⎝ Ekk sgkk + Fkk gk ⎠
outputs simultaneously, allowing for a more comprehensive assessment Ek +F k e =1 gk f =1 bk

of efficiency. The network slack-based model with undesirable outputs

under the constraint of the free link case can be defined as follows:

( ) ( )
∑Mk k ∑M h h
wk 1 − 1
Mk
sm
mk =1 xmk + wh 1 − 1
Mh
sm
mh =1 xmk

Min ⎡ ⎛ ⎞⎤ ⎡ ⎛ ⎞⎤
∑ k ∑Fk k ∑ k ∑F k h
wk ⎣1 + 1 ⎝ Ekk sk
+ sk ⎠ ⎦
+ wh ⎣1 + 1 ⎝ Ehk sk
+ fh ⎠⎦
Ek +Fk e =1 gkk f k =1 bk Eh +Fh e =1 gkh f h =1 bk

∑ h smh
1 − M1h M
mh =1 xmh
∑N k
∑N h PE = ⎛ ⎞ (3)
s.t. Xk = λk Xk + SX Xh =
i=1 i i
λh Xh + SX
i=1 i i ∑ ∑
1+ 1 ⎝ Ehh ehh + Fhh fh ⎠
∑N k
∑N k Eh +F h e =1 ge f =1 bf h
Gk = λk Gk − SG Gh1 =
i=1 i i
λh Gh − SG
i=1 i i
∑N ∑N
Bk = λk Bk + SB Bh =
k
λh Bh + SB
6
(1)
i=1 i i i=1 i i
∑N ∑N 2.2. Measurement of inclusive green growth
λk Z λh Z
(k,h) (k,h)
i=1 i i
= i=1 i i
k
λki , λhi ≥ 0Sx , Sc , Ss ≥ 0
k k Enhancing urban Inclusive Green Growth (IGG) is a multifaceted
process that necessitates the coordination of multiple sectors. Specif-
where the slack variables for all inputs are expressed as SX = (s1, s2, …, ically, the city is conceptualized as comprising two primary components:
sM) ∈ RM+ . It is important to note that the efficiency derived from Eq. (1)
the government sector and the production sector. The government
reflects the overall efficiency, which can be decomposed into sector is tasked with delivering public goods and services and supporting
the production sector, which is tasked with economic activities. The

4
Y. Ren et al. Ecological Indicators 169 (2024) 112898

Dynamic equilibrium

Fig. 2. The conceptual framework of inclusive green growth.

essence of IGG lies in harmonizing the coupled relationship between the balanced and inclusive growth. The CCDM is particularly well-suited for
production and government sectors (Fig. 2). To this end, we have this purpose as it not only captures the dynamics between these two
innovatively introduced the CCDM to obtain IGG scores by capturing the systems but also quantitatively evaluates the degree to which they are
interaction between production sector and governance sector. There- coordinated. Traditional econometric models might overlook the
fore, the calculation of IGG is divided into two steps: complexity of interactions between governance efficiency and produc-
Step 1: The government sector utilizes public expenditure to provide tion efficiency. The CCDM, in contrast, allows for a nuanced under-
infrastructure, scientific and technological advancements, education, standing of how these two systems evolve over time, acknowledging that
and healthcare while striving to reduce income disparities among resi- improvements in one sector may influence, or be influenced by, changes
dents. The production sector leverages infrastructure and technological in the other (Lu et al.). This makes the model highly appropriate for
progress while consuming energy, natural resources, labor, and capital assessing a multifaceted process like IGG, where economic, environ-
to generate GDP as desired outputs and environmental pollutant emis- mental, and social systems are deeply interconnected. This interaction
sions as undesired outputs. Additionally, infrastructure and technology can be quantitatively assessed using the following formula.
are considered as intermediate products, playing a pivotal role in linking
{ /[ ( ) ] }1
the governance and production sectors. To thoroughly investigate the C = (U1 • U2 , ⋯, Um ) Π Ui + Uj m (4)
interrelationships between sectors within a city, this study employs a
two-stage slack-based network DEA model to evaluate production effi- where, U1 • U2 , ⋯, Um represent multiple sectors, C represents the
ciency (PE) and governance efficiency (GE). The detailed information of coupling degree, which elucidates the magnitude of intersectoral in-
the statistical summary of variables can be found in supplement material teractions from U1 to Um . In this work, the coupling degree encapsulates
(Table S1) (Table 2). synergistic interactions, mutual influences, and interdependent re-
Step 2: The concept of coupling, originally derived from physics, lationships between production efficiency and governance efficiency.
refers to the interaction between two or more systems. IGG requires the Therefore, the previous formula can be simplified to Eq. (1):
interaction of both production and governance sectors to achieve

{ / }1
2
C = (productionefficiency • governanceefficiency) [(production efficiency + governanceefficiency) ]2 (5)

5
Y. Ren et al. Ecological Indicators 169 (2024) 112898

Table 2 2.3. Core explanatory variables: new-type urbanization

Indicators used for measuring governance efficiency and production efficiency.
Stage Categories Variable Measurement Compared with traditional urbanization, which is centered on eco-
nomic development and relies on factor-input-driven and investment-
Governance Input Fiscal Total government fiscal
efficiency expenditure expenditures (Million driven growth, new-type urbanization emphasizes the promotion of
(government Yuan) people-centered urbanization (Chen et al., 2018). The core of new-type
sector) Land use Urban built-up area for urbanization is sustainable, high-quality, coordinated, and shared
Stage 1 each city (km2) development. While fewer studies have explored the interaction be-
Desirable Education Number of college
Output teachers per 10,000
tween NTU and IGG, but numerous studies have examined the rela-
people tionship between NTU and total factor productivity (Cheng and Wang,
Healthcare Number of doctors per 2023; Shao and Wang, 2023), air pollution (Huang et al., 2023), and
10,000 people greenhouse gas emissions (Xiao et al., 2023), providing us with useful
Resident Total retail sales of social
insights. Combining these insights, we assume that NTU impacts IGG
consumption consumer goods per
capacity capita (Million Yuan) through three aspects: the green-oriented economic urbanization pro-
Undesirable Income Ratio of urban to rural per cess, the inclusive social urbanization process, and ecological and
Output inequality capita income environmental development.
Intermediate Infrastructure Road miles per 10,000
people (Km/10,000
person)
(i) Green-oriented economic urbanization fosters industrial struc-
Technology Number of green patents ture transformation and upgrading, innovation factor agglomer-
per 10,000 people ation, green technology dissemination and diffusion, and
optimization of urban spatial structure (Liu et al., 2022). This
Production Input Labor Number of employees improves the level of economic development through structural
Efficiency (10,000 persons) effects, innovation effects, scale effects, and efficiency effects,
(production Capital Kt = Kt-1 *(1- a) + It/Pt. K while simultaneously reducing the emission of environmental
Sector) represents the fixed
pollutants.
Stage 2 capital for Yeart, while Kt-1
denotes the fixed capital (ii) Inclusive social urbanization enables more peasants to increase
from Yeart-1; The their income through employment transfer and enjoy better
parameter a is the capital
depreciation rate, set at
9.6 %. It refers to fixed Table 3
asset investment, and Pt is Indicators system for people-centric new-type urbanization.
the fixed asset deflator for
Primary indicator Secondary Measurement (Unit) Indicator
It. Unit: Million Yuan.
indicator properties
Electricity Total amount of
electricity consumption Green-oriented Intensity of GDP output per unit Positive
(10,000 tons of standard economic economic area of built-up area
coal) urbanization development (10,000 Yuan/km2)
Water supply Total water supply Household Proportion of total Positive
(Million cubic meters) consumption retail sales of consumer
Desirable GDP GDP at constant 2006 level goods in GDP (%)
Output prices (Million Yuan) Employment The ratio of tertiary Positive
Undesirable Sewerage Total wastewater volume structure industry employment
Output (100,000 tons) to total employment
Soot(dust) Total industrial soot (%)
emissions (tons) Investment by Fixed asset investment Negative
SO2 Total industrial SO2 asset as a share of GDP (%)
emissions (tons) Inclusive social Fiscal Per capita fiscal Positive
PM2.5 PM2.5 concentrations (μg/ urbanization expenditure expenditure (Yuan/
m3) person)
Income Ratio of urban to rural Negative
inequality per capita income (%)
Education Ratio of educational Positive
The IGG reflects the intensity of interaction between the production services expenditure to fiscal
expenditure (%)
sector and the management sector. Given that the coupling coordination
Public cultural Public library Positive
degree model illustrates the coherence and harmony of system devel- services collections (per book/
opment, we calculate IGG by revising the coupling coordination degree 10,000 persons)
model. The IGG is derived using the following formula: Medical services Number of beds in Positive
hospitals and health
√̅̅̅̅̅̅̅̅
IGG = C* T; (6) centers (per bed/
10,000 persons)
Environmental and Park green space Park green space per Positive
T = α*productionefficiency + β*governance efficiency ecological capita (m2/person)
development Municipal Sewage treatment rate Positive
where IGG represents the inclusive green growth (referred to as the urbanization sewage treatment (%)
degree of coordination), whereas α and β represent undetermined co- Cleaning Number of public Positive
efficients, which are set to 1/2 in this study. Both C and D range from facilities toilets per capita
(item/10,000 persons)
0 to 1, Both C and D range from 0 to 1, where a higher value signifies a
Environmental Frequency of words Positive
greater level of coherence among sectors. concern related to
environmental
governance in
government work
reports (%)

6
Y. Ren et al. Ecological Indicators 169 (2024) 112898

Table 4
Measurement of control variables and statistical summary.
Variable Definition Unit of Abbreviations Statistical Summary
measurement
Mean Max Min Std.

Population Population per square kilometer Person/km2 POPD 3852.73 20,093 248 2795.62
density
Government The share of fiscal expenditure over the total GDP % GOV 0.22 3.21 0.04 0.12
intervention
Foreign direct The percentage of the foreign direct investment in GDP % FDI 2.01 31.84 0.0002 2.34
investment
industrial industrial upgrading index is a dynamic process in which the industrial / IUI 6.44 7.83 5.51 0.35
upgrading structure evolves from low to high level. The detail information of the
index method can be found in previous studies (Wang et al., 2021; Zhang et al.,
2023).
Financial The ratio of Loan balance to deposit balance % FIN 0.91 6.43 0.06 0.58
development

public services by becoming urban residents, thus expanding the adopted quantile regression to address the limitations inherent in con-
urban consumer group and upgrading the consumption structure ventional approaches. Additionally, the quantile regression approach
(Jiang et al., 2022a). It also brings about significant investment proves beneficial in addressing issues that could significantly impact
demand for urban infrastructure and public service facilities, estimation accuracy, such as heteroscedasticity, outliers, and unob-
providing continuous impetus for economic development and served heterogeneity. The econometric model presented below is
raising the level of sustainable development (Zhao et al., 2022b). designed to address the conditional quantile function of panel data:
(iii) Ecological and environmental development emphasizes the co- Eq. (7) represents a quantile regression function at a τ quantile.
ordinated development of socio-economic and ecological systems QY (τ|Xit ) denotes the τth quantile of the dependent variable; Xit is the
in the urbanization process (Kang et al., 2021). With goals such as vector of explanatory variables. Z denotes for a k-vector of differentiable
resource recycling, urban green landscaping, and improving the transformations of the components of X. Machado and Silva (2019)
human environment, it fully considers the carrying capacity of named (αi + δi q(τ)) as a scalar coefficient, αi (τ), indicating the fixed
resources and the environment for urbanization development effect at a τ quantile for individual i. This encapsulates the effects of
(Kronenberg et al., 2020), thereby driving a green transformation time-invariant individual characteristics. Then, MM-QR estimates q(τ)
of the socio-economic development model and affecting IGG. by Eq. (8), where ρτ (A) = (τ − 1)AI{A ≤ 0} +τAI{A > 0} is the check-
function.
Therefore, this study built an indicator system for NTU based on
QY (τ|Xit ) = (αi + δi q(τ)) + Xʹit β + Zʹit γq(τ) (7)
references to existing studies (Lin and Zhu, 2021; Yu, 2021; Chen et al.,
2023). It contains three dimensions: green-oriented economic urbani- ∑∑ ( ( ) )
zation, inclusive social urbanization, Environmental and ecological min ρτ R
̂ it − ̂ δ i + Zʹit ̂γ q (8)
q
i t
development urbanization. Furthermore, we introduced the entropy
weight method and principal component analysis (PCA) to calculate a
2.5.2. Spatial panel Durbin model
comprehensive New-type urbanization score. The indicator system is
Elhorst (2014) identifies spatial Durbin model (SDM) to capture
presented in Table 3. The entropy weight method and PCA are explained
spatial correlation, which integrates features of both the spatial lag
in supplement material (Text S1-S2).
model (SLM) and spatial error model (SEM). Specifically, the SDM can
be represented as follows:
2.4. Control variables ( )
Yit = ρWYit + βXit + αWXit + λt + μi + εit , εit ∼ N 0, δ2 IN (9)
Referring to the existing literature (Sun et al., 2020; Gu et al., 2021; In this model, Yit is independent variables in ith city at time t. W
Ren and Li, 2022; Ma et al., 2024), other factors that affect urban IGG denotes the spatial weighting matrix. Xit represents observation values
were selected. These included population density (POPD), government of independent variables. ρ is the spatial regressive coefficient, indi-
intervention (GOV), foreign direct investment (FDI), industrial cating the impact of neighboring cities’ observation values on Yit . β
upgrading index (IUI), and financial development (FIN) (Table 4). A represents the coefficient of the independent variables. α is the coeffi-
correlation analysis was conducted in this study, and the findings reveal cient of the spatial lagged independent variable, capturing the effect of
that all correlation coefficients are below the threshold of 0.7, indicating neighboring cities’ independent variables on Yit . λt and μi denote a time-
an absence of multicollinearity concerns (Table S2). period effect and a spatial effect, respectively. εit is the random error
term. A pivotal differentiation between spatial econometric models and
2.5. Econometric models traditional econometric models is the integration of a spatial weighting
matrix in the former, which reflects interactions among spatial units.
2.5.1. Panel quantile regression The adjacent spatial weighting matrix was employed in this work.
Koenker and Bassett (1978) pioneered the introduction of quantile
regression, while Machado and Santos Silva (2019) advanced the field 2.6. Data sources
by proposing the method of moments quantile regression (MM-QR) with
fixed effects, thereby extending quantile regression to panel data con- This research assembled a panel dataset comprising 281 prefecture-
texts. Quantile regression estimates the impact of explanatory variables level cities in China over the period from 2006 to 2019. Socioeconomic
on a response variable across different quantiles and exhibits robustness data were mainly sourced from the China City Statistical Yearbook
to outliers. Furthermore, its coefficients can be interpreted as marginal (2007–2010), China City Construction Statistical Yearbook
effects, akin to linear regression. In this context, quantile regression (2007–2010), and municipal work reports (2007–2010), covering the
serves as a valuable complement to traditional methods for both poli- study periods. China’s annual Consumer Price Index (CPI) is applied to
cymakers and researchers. Consequently, several recent studies have deflate fiscal expenditure and household consumption, setting 2006 as

7
Y. Ren et al. Ecological Indicators 169 (2024) 112898

Fig. 3. The temporal variation of production efficiency and governance efficiency.

the base year. Atmospheric pollution from fine particulates (PM2.5) to smooth, indicating a narrowing gap between cities. However, the PE
metrics were derived from the global PM2.5 long-term remote sensing distribution remains steep, with most cities below the mean, high-
inversion dataset, which provided by the University of Washington (htt lighting significant polarization and low overall levels of PE. Histori-
ps://[Link]/acag/datasets/surface-pm2-5). To rectify any cally, China has adhered to the principle of prioritizing efficiency,
missing data, linear interpolation techniques were systematically directing production factors to concentrate in eastern coastal cities and
applied. cities with higher administrative levels and favorable locational condi-
tions (Deng et al., 2022). Over time, these cities have developed asym-
3. Results and discussion metric competitive advantages over other regions, leading to
polarization effects driven by development inertia. These findings sug-
3.1. Temporal variations in GE and PE gest considerable potential for improving green productivity in Chinese
cities.
This study employed a two-stage network SBM DEA model to eval- Theoretically, effective urban governance can enhance urban inno-
uate the governance efficiency (GE) and production efficiency (PE) of vation and labor productivity, reduce enterprise operating costs, and
281 Chinese cities from 2006 to 2019. Fig. 3(a and b) illustrates the contribute to improvements in urban green productivity. Thus, it is
temporal variation of GE and PE at the prefecture level. An analysis of essential to explore whether the enhancement of urban governance in
the annual mean trends reveals that GE consistently surpassed PE after Chinese cities is synergistically related to the development of a green
2016, reflecting a shift in urban development from a focus solely on economy. To this end, we first employed a t-test to determine whether
enhancing economic output and reducing environmental pollution to a there are significant differences between governance efficiency (GE) and
model that emphasizes governance efficiency. During this period, production efficiency (PE), and then applied the Pearson correlation
China’s urbanization rate surged from 43.9 % in 2006 to 60.6 % in 2019, coefficient to examine the relationship between GE and PE. As shown in
with an annual average of 17.85 million rural residents transitioning to Fig. 4(a–d), significant differences between PE and GE are observed
urban life, driving unprecedented demand for public services and across all time periods (tstudent = [5.65, 9.10], p-value < 0.01), con-
infrastructure. Since 2012, the Chinese government has prioritized so- firming that the research framework effectively distinguishes between
cial management and public services, strengthened the financial transfer the two concepts. Further analysis reveals that the correlation between
payment system, adjusted the redistribution of national resources, and GE and PE increased from 0.32 (Pearson’s r = 0.32, CI95%= [0.21,0.42])
provided targeted support to less-developed cities in central and western in 2006 to 0.44 (r = 0.44, CI95%= [0.34,0.53]) in 2010, a growth of 37.5
China through increased financial input and relevant policy measures, %. This upward trend continued, reaching 0.48 in 2014 (r = 0.48,
effectively fostering improvements in urban governance. Fig. 3 also CI95%= [0.38,0.56]), a further increase of 9.09 %. The results of the
depicts the probability distribution of GE and PE across Chinese cities primary function fitting (represented by the blue line in the figure)
each year. The GE probability distribution curve transitions from steep further validate the conclusions outlined above. These results suggest

8
Y. Ren et al. Ecological Indicators 169 (2024) 112898

Fig. 4. Pearson correlation scatter plot.

that green productivity has kept pace with improvements in urban having long depended on natural resources to develop pollution-
governance, indicating that a financial allocation system focused on intensive sectors such as heavy industry and chemicals. In particular,
public welfare can create a win–win situation for green productivity the path dependence created by this long-term reliance on resource-
growth. Despite a slight decline in PE in 2019, the correlation coefficient driven development has hindered the region’s pace of green economic
for 2019 remained at 0.32 (r = 0.32, I95%= [0.21,0.42]), suggesting that transformation, especially in the context of increasing resource scarcity.
China has established a stable public service investment mechanism that As a result, the average IGG value remains one of the lowest.
is resilient to macroeconomic fluctuations. Fig. 5(b–e) reveals a general positive correlation between IGG and
city size, suggesting that larger cities typically exhibit higher IGG.
However, this relationship does not extend to the central region, where
3.2. Spatiotemporal variation of IGG mega cities (population > 5 million) are increasingly burdened by urban
challenges. The research findings indicate that large cities (5 million >
The performance of IGG exhibited a fluctuating upward trajectory population > 1 million), with their relatively strong economic base,
throughout the study period, characterized by distinct periodicity advanced technology, and mature management practices, can deliver
(Fig. 5a). The performance of IGG can be delineated into three distinct more public services than less economically developed cities with the
phases: a decline from 2006 to 2011, an inverse U-shaped trajectory same level of financial input, thereby achieving higher IGG. However,
from 2012 to 2014, and a subsequent rapid ascent from 2015 to 2019. population concentration and industrial development in these cities also
The national average IGG during the 2006–2019 period stood at 0.55. give rise to “urban diseases,” which not only reduce urban green pro-
Among the sub-regions, East China recorded the highest IGG at 0.59, ductivity but also crowd out the provision of public services, thereby
whereas Central China exhibited the lowest at 0.53. Compared with the diminishing the effectiveness of urban IGG.
existing studies (Sun et al., 2020), the fluctuation status of IGG has a Fig. 6(a–d) illustrates the significant spatial differentiation in China’s
high similarity in the same time period. The reason may lie in the central urban inclusive green growth (IGG), which diverges notably from the
region’s historical role as a national coal energy base and industrial hub,

9
Y. Ren et al. Ecological Indicators 169 (2024) 112898

Fig. 5. Average value of IGG at prefecture-level in China (2006–2019).

economic development levels of Chinese cities. The figure reveals a clear their higher environmental costs (Gu et al., 2021; Fan et al., 2023). This
spatial pattern, with IGG decreasing progressively from coastal regions industrial disparity is a key theoretical determinant of regional IGG
to inland areas. From 2006 to 2019, the number of cities in the high differences, as cities that transition to greener, knowledge-intensive
coordination category increased from 54 to 70, raising their proportion industries are better positioned to balance economic growth with
from 19 % to 25 %. Spatially, in 2006, 2010, and 2014, high coordi- environmental sustainability. Additionally, policy implementation
nation clusters were concentrated around core cities in major urban timing and effectiveness contribute to temporal variations. The Chinese
agglomerations, including the Yangtze River Delta, Pearl River Delta, government has increasingly prioritized green growth, particularly since
Beijing-Tianjin, and Chengdu-Chongqing regions. By 2019, IGG in the early 2012, with policies focused on regional coordinated develop-
southern Hebei and central Sichuan cities transitioned from basic to high ment strategy, reducing pollution, and promoting industrial upgrading.
coordination. Moreover, cities with medium coordination experienced For example, IGG levels of cities surrounding core cities have signifi-
the most significant growth, rising from 139 in 2006 to 179 in 2019, cantly improved since 2014, particularly within the Beijing-Tianjin-
representing 64 % of the total. Their spatial pattern became more Hebei urban agglomeration. This finding may contradict most studies,
balanced, with a decline in range of dispersed distribution. New which suggest that Beijing’s siphoning effect has contributed to regional
medium-coordinated cities emerged mainly in central and western re- socioeconomic imbalances (Ren and Li, 2022). In fact, Beijing-Tianjin-
gions, such as Henan, Shanxi, Shaanxi, and the Hexi Corridor in Gansu, Hebei coordinated development strategy, which focuses on relieving
following the government’s strategy to foster collaborative economic Beijing’s non-capital functions, not only transfers industries to Hebei but
development across the eastern, central, and western regions. These also provides educational and medical resources. Furthermore, our re-
areas absorbed a substantial number of industries relocated from the sults show that as China’s economic development enters a period of
eastern coast, forming industrial chains that strengthened ties with growth speed adjustment, structural adjustment, and the digestion of
eastern regions. The proportion of cities with basic or moderately earlier stimulus policies, the production efficiency of marginal cities is
imbalanced coordination dropped significantly from 31 % in 2006 to 11 generally affected. This has led to a phenomenon where production ef-
% in 2019. These cities are primarily located in southern hilly areas, ficiency lags behind governance efficiency, gradually becoming
such as Gannan and Guizhou, and inland arid regions like Dunhuang and apparent after 2014. The finding indicates that the government needs to
Hami. With China’s new infrastructure initiatives, including the “Data in develop reasonable mechanisms and channels to fully leverage the role
the East, Computing in the West” strategy and solar photovoltaic power of social development in promoting green technology innovation and
stations, economically disadvantaged cities in challenging locations are the upgrading of regional industrial structures.
poised for future IGG improvement (Zhang et al., 2020; Peng and Dan,
2023). 3.3. Regional disparities analysis
Compared with related studies (Guo et al., 2020; Liu et al., 2020b;
Liu et al., 2021; Ren et al., 2022b), our research reveals that national Analysis of spatial pattern evolution reveals that cities situated near
center cities, primarily located in the eastern coastal region and central coastal and major inland economic zones demonstrate positive ag-
China, achieve the synergy between PE and GE. The relatively rigorous glomerations of IGG. However, it remains a critical question whether
IGG assessment framework effectively highlights the disparity in urban these agglomerations indicate a widening of IGG development dispar-
IGG development across different regions, avoiding the problem of ities between regions. This study employs the Dagum Gini coefficient
reduced differentiation caused by the overestimation of sustainable method, which accurately measures relative disparities across regions,
development improvements in most cities. The differences in IGG can identifies their sources, and effectively addresses data cross-overlapping
largely be explained by the variation in economic structures across re- issues (Dagum, 1998). As depicted in Fig. 7(a–d), the total Gini coeffi-
gions. Cities in the eastern region of China, particularly in the Yangtze cient for IGG initially increased and subsequently decreased, indicating
River Delta and Pearl River Delta, have a more diversified and advanced a reduction in regional disparities. Intra-regional analysis shows that
industrial base, with a greater emphasis on service-oriented and high- IGG disparities decreased most significantly in the western region, fol-
tech industries. In contrast, central and western cities are more depen- lowed by the northeastern and central regions. This outcome is not
dent on resource-extractive and pollution-intensive industries, such as unexpected and aligns with findings from previous research (Ke et al.,
mining and heavy manufacturing, which inherently lower IGG due to 2023). On the one hand, the western region has benefited from China’s

10
Y. Ren et al. Ecological Indicators 169 (2024) 112898

Fig. 6. Inclusive green growth performance at prefecture level in China from 2006 to 2019. Note: IGG scores between 0 and 0.4 indicate a moderate imbalance,
scores between 0.4 and 0.6 represent basic coordination, scores between 0.6 and 0.8 denote moderate coordination, and scores between 0.8 and 1 signify high
coordination.

Grand Western Development Initiative (Jia et al., 2020), which has 2019, the development gap between the eastern and western regions
significantly enhanced infrastructure development and facilitated the (0.092) and between the eastern and central regions (0.09) was more
growth of emerging green energy sectors, such as photovoltaic panel pronounced than in other regional pairings. One plausible theoretical
production and solar photovoltaic generation (Qiu et al., 2022). On the explanation stems from path dependence, as cities in the eastern region
other hand, the Chinese government’s ongoing policy of opening-up, have historically benefited from early policy support, including foreign
particularly through the Belt and Road Initiative, has encouraged capi- direct investment (FDI), innovation clusters, and economic reforms.
tal inflows and the exchange and dissemination of technological These advantages have enabled eastern cities to adopt green technolo-
knowledge and resources, thereby promoting industrial upgrading and gies and establish sustainable infrastructure ahead of their counterparts
fostering a more inclusive economic transformation (Zhu et al., 2024). in central and western China (Ren et al., 2022). In contrast, central and
By the end of the study period, regional disparity rankings were as fol- western cities continue to grapple with the legacy of resource-dependent
lows: northeast (0.078), west (0.086), east (0.089), and central (0.092). economic models, which complicates the transition to greener alterna-
Interestingly, the within-region disparity is more pronounced in the tives due to sunk costs and institutional inertia (Guan et al., 2018).
central region compared to other regions. While the central region is Table 5 presents the sources and contribution rates of regional dis-
typically viewed as the core intersection of east–west and north–south, parities in IGG. This study systematically decomposes regional differ-
with dense transportation networks that support the growth of the ences into intra-regional, inter-regional, and super-variable density
regional industrial economy, a different dynamic emerges when disparities, providing deeper insights into the spatial dynamics of IGG.
excluding the core cities. Edge cities in this region, located closer to the The results indicate that intra-regional disparities have remained rela-
Yangtze River Delta and Pearl River Delta—China’s most developed tively stable over time, with contribution rates fluctuating slightly be-
areas—are subject to the siphoning effect, experiencing a drain in talent, tween 27.99 % and 28.212 %. This suggests that disparity within
capital, and resources. Inter-regional Gini coefficients demonstrate that individual regions has not undergone significant change during the
IGG gaps between regions have consistently narrowed over time. By study period. In contrast, inter-regional disparities and super-variable

11
Y. Ren et al. Ecological Indicators 169 (2024) 112898

Fig. 7. The results of Dagum gini coefficients.

density exhibit an inverse relationship. While inter-regional disparities, needed to reduce these differences and promote more equitable green
which measure the differences in IGG between various regions, fluctuate growth across all regions.
between 22.242 % and 17.511 %, super-variable density—a metric that
captures the overlap or convergence of IGG across regions—ranges from
49.768 % to 54.276 %. This inverse pattern implies that as regional 3.4. Econometric analysis
disparities decrease, the convergence of IGG across regions increases,
leading to more uniform development outcomes. On average, intra- 3.4.1. Panel quantiles regression results
The potential distortion of regression results due to non-normal data
regional, inter-regional, and super-variable density disparities account
for 28.084 %, 21.219 %, and 50.696 % of total disparities, respectively. distribution necessitated a rigorous assessment of normality through
five tests. Table 6 reveals that, although skewness coefficients were
The consistently higher contribution of super-variable density compared
to intra- and inter-regional disparities suggests a growing overlap in IGG below 3 and kurtosis values did not exceed 10, the variables only
approximated normality. Furthermore, the Kolmogorov-Smirnov and
performance across regions. This reflects a trend toward regional
convergence in IGG, signaling progress toward a more balanced distri- Shapiro-Wilk tests, despite yielding values close to 1, rejected the null
hypothesis of normal distribution based on p-values. The Jarque-Bera
bution of green growth among China’s urban areas. These findings
highlight a positive shift toward regional equilibrium in IGG, where test corroborated these findings, confirming that the data deviates
from normality. Consequently, quantile regression was deemed the most
cities across different regions are gradually aligning in their green
development efforts. However, the persistence of both intra- and inter- appropriate and robust method for this study.
We investigated the results of the fixed effect panel quantile
regional disparities indicates that further policy interventions are
regression, which explicitly describes the heterogeneous impacts of

12
Y. Ren et al. Ecological Indicators 169 (2024) 112898

Table 5
Decomposition Results of the Dagum Gini Coefficient for City-Level IGG in China from 2006 to 2019.
Year Total Intra-regional Inter-regional Super variable density

Source Contribution Source Contribution Source Contribution

rate (%) rate (%) rate (%)

2006 0.111 0.031 27.99 0.025 22.242 0.055 49.768

2007 0.115 0.031 27.359 0.027 23.34 0.057 49.301
2008 0.11 0.031 28.473 0.02 18.141 0.059 53.386
2009 0.11 0.031 28.15 0.023 21.131 0.056 50.719
2010 0.112 0.031 27.846 0.026 22.965 0.055 49.189
2011 0.105 0.03 28.142 0.023 21.501 0.053 50.357
2012 0.103 0.029 28.309 0.018 17.4 0.056 54.291
2013 0.09 0.025 28.081 0.016 17.187 0.049 54.732
2014 0.091 0.026 28.199 0.022 24.339 0.043 47.462
2015 0.086 0.024 27.819 0.021 23.959 0.041 48.222
2016 0.081 0.023 28.087 0.019 23.687 0.039 48.226
2017 0.088 0.024 27.785 0.021 23.344 0.043 48.87
2018 0.086 0.025 28.733 0.017 20.318 0.044 50.949
2019 0.089 0.025 28.212 0.016 17.511 0.048 54.276
Mean 0.098 0.028 28.084 0.021 21.219 0.0499 50.696

Table 6
Tests of normal distribution.
Variable Skewness Kurtosis Shapiro-Wilk test Kolmogorov-Smirnov(K) test Jarque–Bera test Obs.
*** ***
IGG − 0.109 2.898 0.997 0.033* 9.516 3934
NTU 0.007 2.938 0.997*** 0.823*** 0.668 3934
GD 0.727 3.752 0.971*** 0.0563*** 439.46*** 3934
SUR 0.329 2.982 0.992*** 0.019* 71.21*** 3934
EED − 1.816 8.567 0.855*** 0.095*** 7245.147*** 3934
POPD 0.459 1.694 0.872*** 0.136*** 417.946*** 3934
GOV 0.149 2.692 0.996*** 0.026*** 30.257*** 3934
FDI − 0.889 4.883 0.953*** 0.055*** 1100.513*** 3934
IUI 0.171 2.830 0.997*** 0.021** 24.083*** 3934
FIN 0.750 3.698 0.945*** 0.261*** 448.658*** 3934
**
Note: The symbols *, , and *** denote the significance at 10%, 5%, and 1%, respectively.

Table 7
The results of panel quantile regression.
Variables Model 1 Model 2

25th 50th 75th 25th 50th 75th

lnNTU 0.491*** 0.409*** 0.333*** ​ ​ ​

(0.037) (0.027) (0.034)
lnGD ​ ​ ​ 0.273*** 0.298*** 0.319***
(0.4) (0.028) (0.036)
lnSUR ​ ​ ​ 0.423*** 0.341*** 0.269***
(0.036) (0.026) (0.033)
lnEED ​ ​ ​ 0.039* 0.025 0.012
(0.023) (0.017) (0.021)
lnPOPD 0.001(0.002) 0.002 0.002 0.001 0.002 0.003
(0.002) (0.002) (0.002) (0.002) (0.002)
lnGOV − 0.145*** − 0.149*** − 0.153*** − 0.083*** − 0.109*** − 0.132***
(0.022) (0.016) (0.021) (0.022) (0.016) (0.02)
lnFDI 0.007* 0.003** 0.001 0.007* 0.004* 0.001
(0.004) (0.003) (0.004) (0.004) (0.003) (0.004)
lnIUI 1.997*** 2.163*** 2.319*** 1.556*** 1.694*** 1.815***
(0.223) (0.161) (0.205) (0.221) (0.158) (0.201)
lnFIN 0.023** 0.038*** 0.052*** 0.018** 0.033*** 0.046***
(0.009) (0.006) (0.008) (0.009) (0.006) (0.008)
Estimator FE FE FE FE FE FE

NTU, GD, SUR, EED, and control variables on local IGG. Three specific significant across all quantiles. The coefficients of lnNTU are 0.491,
quantiles were selected to capture a comprehensive view across different 0.409, and 0.333 at the 25th, 50th, and 75th quantile, respectively. This
IGG levels: the 25th quantile model elucidates the influence of explan- demonstrates a decreasing trend across all selected quantiles but shows a
atory variables in low-IGG cities, the 50th quantile model highlights stronger effect at the lower quantiles. Our study also demonstrates that
their impact on medium-IGG cities, and the 75th quantile model focuses the positive effect of NTU is not uniform across cities with different IGG
on high-IGG cities. The results presented in Table 7 and Fig. 8 indicate levels. As industrial layout adjustments move market capital to periph-
that the effects of all selected variables on IGG are heterogeneous. eral areas, lower-IGG cities are undergoing a new round of industrial
The effect of lnNTU on IGG is heterogeneous and statistically restructuring (Tang, 2021). In the context of globalized markets and a

13
Y. Ren et al. Ecological Indicators 169 (2024) 112898

Fig. 8. Estimation results of the panel quantile regression. Note: The shaded areas denote 95% confidence band for the panel quantile regression estimates. Two
dotted lines represent the OLS regression result.

unified national market policy, lower-IGG cities benefit from compara- the development of compact, energy-efficient cities that reduce
tive advantages in factor costs, such as labor and land, compared to core transportation-related pollution and increase the effectiveness of
cities, which face rising costs and stricter industrial policies. This public services. Furthermore, NTU facilitates the creation of more
advantage amplifies the positive effect of NTU. The results of Model 2 quality jobs within green industries, elevates residents’ incomes,
reveal three potential mechanistic channels through which NTU in- and fosters public environmental awareness and green con-
fluences IGG: sumption practices (Dell’Anna, 2021). Increased green economic
output further empowers the government to allocate more re-
(i) From a theoretical perspective, the impact of different urbani- sources toward environmental protection and social governance,
zation processes on IGG reveals distinct dynamics. Both lnGD thus strengthening environmental regulations and enhancing the
(green development-oriented urbanization) exhibits positive social security system. The findings also suggest that for low-IGG
correlations with IGG at the 1 % significance level across quan- cities, which face systemic constraints such as limited talent
tiles. The influence of lnGD intensifies at higher quantiles, pools, underdeveloped technological capabilities, and insuffi-
ranging from 0.273 to 0.319, indicating that cities with higher cient infrastructure in sectors like education, healthcare, and
IGG levels benefit more from green development initiatives. NTU transportation, these factors hinder their ability to attract capital
promotes the transformation and upgrading of industrial struc- and industry, thereby limiting their capacity to improve IGG.
tures by encouraging the development of high-tech, low-pollu- (ii) The coefficients for lnSUR diminish slightly at upper quantiles,
tion industries, and promoting green technology innovation with values of 0.423 and 0.341 at the 25th and 50th quantiles,
(Huang et al., 2023). This transition leads to more efficient use of respectively. These values, which surpass those of lnGD, suggest
resources and a reduction in environmental pollution, contrib- that for cities with lower IGG, fostering inclusive social urbani-
uting positively to IGG. Additionally, NTU fosters the optimiza- zation plays a more critical role in enhancing IGG outcomes. NTU
tion of urban spatial structures (Cheng et al., 2023), encouraging seeks to expand access to public services and improve living

14
Y. Ren et al. Ecological Indicators 169 (2024) 112898

conditions by focusing on equitable urban development (Zhao innovation. It facilitates the transition toward a green economy by
et al., 2022b). In the “Key Tasks for Promoting New-type Ur- providing financing for projects aimed at reducing emissions, improving
banization Construction”, released annually by the Chinese gov- energy efficiency, and fostering sustainable urban development (Lv
ernment since 2014, the primary objective has consistently been et al., 2021). The consistent positive correlation across quantiles in-
the acceleration of the citizenization of the migrant agricultural dicates that financial development plays a crucial role in fostering IGG
population. This policy focuses on optimizing settlement condi- across cities, irrespective of their level of development, by channeling
tions in cities and improving basic public services for the resident resources into more productive and environmentally friendly
population, thereby helping to bridge the urban–rural divide and investments.
reduce income inequality. These efforts not only enhance human Government intervention exerts obvious negative effect on IGG at
capital but also stimulate domestic consumption, fostering more selected quantiles and passed the 1 % confidence level test. Excessive or
inclusive and sustainable economic growth. By enhancing social inefficient government intervention, especially in low-IGG cities, can
equity, NTU directly supports IGG by ensuring that the benefits of create distortions in the market. Overregulation may stifle innovation by
growth are more evenly distributed across the population, which imposing rigid compliance costs on industries, discouraging investment
in turn improves social stability and overall development resil- in green technologies, and delaying necessary economic transformations
ience. Importantly, the regression results indicate that the posi- (Cheng et al., 2020). Furthermore, government policies that are not
tive impact of inclusive social urbanization on IGG surpasses that well-targeted or are influenced by interest groups may inadvertently
of green development-oriented urbanization across quantiles. protect inefficient industries, thus impeding progress towards greener
Such a shift may indicate an increasing awareness that a socially production models. This explains the observed negative correlation at
equitable environment, rather than industrial capacity alone, specific quantiles, where government intervention may hinder rather
serves as the key driver for attracting talent and capital, ulti- than enhance IGG.
mately enhancing green innovation and advancing IGG. These Fig. 7 illustrates the variations in coefficients across major quantiles.
findings are consistent with the theoretical insights of Brock In terms of positive effects, the marginal effects of lnGD, lnIUI, lnPOPD,
(2020), emphasizing the critical role of social inclusivity in sus- and lnFIN on IGG increase with rising quantiles, indicating that as the
tainable urban development. levels of IGG grow, their positive impact on inclusive green growth
(iii) The empirical results indicate that the coefficients for lnEED becomes more pronounced. Conversely, the marginal effects of lnNTU,
(environmental and ecological development urbanization) are lnSUR, lnEED, and lnFDI on IGG decrease with increasing quantiles,
positive at the 25th quantile with a 1 % significance level, suggesting diminishing returns as IGG rises. Meanwhile, the marginal
underscoring the critical role of environmental and ecological effects of lnGOV show a slow decreasing trend across different quantiles.
urbanization in enhancing IGG. NTU places significant emphasis In summary, different variables influence IGG differently across various
on ecological conservation and environmental management. levels, emphasizing the need for tailored policy approaches to effec-
Cities adhering to NTU principles prioritize green spaces, pro- tively promote inclusive green growth.
mote the recycling of resources, and implement stricter envi-
ronmental regulations, all of which contribute to the 3.5. Further analysis
enhancement of IGG. By incorporating ecological redline and
basic farmland protection boundary considerations into urban 3.5.1. Temporal heterogeneity analysis
spatial planning (Chen et al., 2021; Guo et al., 2023), NTU en- Referring to “National New-type Urbanization Planning
ables cities to strike a balance between economic development (2014–2020)”, NTU construction was delineated into two phases: early
and environmental preservation, ensuring that growth does not stage (2006–2013) and later stage (2014–2019). This section was un-
compromise long-term ecological sustainability. In regions where dertaken to investigate the temporal heterogeneity in NTU’s impact on
NTU is effectively implemented, we observe improvements in air IGG. The results, as shown in Table 8, reveal that NTU has a more
and water quality, reduced waste generation, and more sustain- pronounced impact on IGG during the later phase compared to the early
able urban ecosystems, all of which bolster IGG performance phase. A similar trend is observed with lnGD. Notably, the positive effect
(Jiang et al., 2022b; Li et al., 2023). Besides, as emphasized by of lnSUR diminishes over time, indicating that the benefits of promoting
the Chinese government’s assertion that “Lucid waters and lush inclusive social urbanization gradually wane. In contrast, lnEED does
mountains are invaluable assets,” natural capital becomes a not show a significant effect on IGG in the early stages but progressively
cornerstone for fostering IGG, particularly in less developed cit- demonstrates positive effects in the later stages, particularly in cities
ies. Industrialization of ecological assets represents an innovative with lower IGG. Our findings confirm that the growing impact of NTU on
practice model, demonstrating how ecological assets can be enhancing IGG through green-oriented development and environmental
leveraged to drive sustainable urban growth and bridge devel- governance channels is attributable to the continuous improvement of
opmental gaps. policy frameworks and institutional support. The decline in the urban-
ization coefficient for inclusive society may be attributed to inefficient
In terms of control variables, industrial upgrading is positively and allocation of financial resources, despite the annual increase in public
significantly correlated with IGG, and the regression coefficients are service expenditure (Shi et al., 2023). This inefficiency has led to
relatively stable across quantiles. Industrial upgrading enhances pro- redundancy and resource wastage in public services, thereby con-
ductivity by shifting from traditional, resource-intensive sectors to high- straining human capital development.
tech and service-oriented industries, which typically have lower envi-
ronmental footprints (Xie and Teo, 2022). This shift leads to more effi- 3.5.2. Spatial spillover effect
cient use of resources and fosters innovation, contributing to green Significant spatial autocorrelation in the independent variables can
economic growth. The stability of the coefficients across quantiles sug- result in biased estimates when using traditional econometric methods.
gests that industrial upgrading consistently supports IGG, regardless of a This study employed the global Moran’s I index to assess the spatial
city’s development level, by enabling more sustainable industrial prac- autocorrelation of IGG. The findings revealed that Moran’s I index
tices and reducing reliance on pollution intensive industries. values were statistically significant at the 1 % level and exhibited a
The regression results also suggest that financial development is gradual increase from 2006 to 2019, indicating an enhanced positive
positively correlated with IGG across all quantiles. An advanced finan- spatial autocorrelation of IGG (Table S3). Consequently, the spatial
cial sector improves the allocation of capital, making it easier for busi- econometric model was deemed appropriate for this analysis. The
nesses to access funds for investment in green technologies and findings from the SDM models incorporating spatial and period fixed

15
Y. Ren et al. Ecological Indicators 169 (2024) 112898

Table 8
Heterogeneity panel quantile regression results for the two time periods.
Model 3 Early stage Later Stage

25th 50th 75th 25th 50th 75th

lnNTU 0.419*** 0.371*** 0.334*** 0.687*** 0.615*** 0.552***

(0.059) (0.041) (0.052) (0.047) (0.053) (0.068)

Model 4 Early stage Later Stage

lnGD 0.195** 0.183** 0.173*** 0.541*** 0.503*** 0.470***

(0.095) (0.082) (0.117)) (0.068) (0.048) (0.061)
lnSUR 0.456*** 0.376*** 0.315*** 0.258*** 0.214*** 0.178***
(0.057) (0.050) (0.070)) (0.065) (0.046) (0.059)
lnEED 0.015 0.025 0.032 0.133*** 0.123*** 0.115***
(0.030) (0.026) (0.037) (0.044) (0.031) (0.040)
Control variables YES YES YES YES YES YES
Estimator FE FE FE FE FE FE

Table 9 Table 10
The results of Spatial Durbin Model. Decomposition of spatial effects.
Model 5 Model 6 Direct effect Model 7 Model 8

lnNTU 0.729*** ​ lnNTU 0.732*** ​

(26.788) (17.873)
lnGD ​ 0.409*** lnGD ​ 0.414***
(18.691) (19.571)
lnSUR ​ 0.307*** lnSUR ​ 0.306***
(14.253) (14.254)
lnEED ​ 0.101*** lnEED ​ 0.104***
(6.735) (7.034)
W*lnNTU − 0.229*** ​
(− 5.101)
Indirect effect
− 0.104**
​ ​
W*lnGD ​
(− 3.13) lnNTU 0.060*** ​
W*lnSUR ​ − 0.119** (2.518)
(− 3.646) lnGD ​ 0.061
W*lnEED ​ 0.005 (1.419)
(0.235) lnSUR ​ − 0.013
ρ (spatial lag) 0.369*** 0.356*** (− 0.309)
(19.367) (18.502) lnEED ​ 0.059*
σ2 0.029 0.029 (1.851)
2
Adjusted R 0.549 0.565
Log-likelihood 1437.410 1495.302
Wald spatial lag 141.588*** 135.167*** surrounding cities. Theoretically, the population concentration driven
Wald spatial error 23.405*** 32.039***
by NTU enhances cross-regional industrial chain integration and
LR spatial lag 144.295*** 140.942***
LR spatial error 25.453*** 34.636*** specialization (Wu et al., 2022), which helps reduce the marginal costs
Hausman test 58.185*** 62.055*** for both local and neighboring enterprises. Simultaneously, NTU pro-
Control variables YES YES motes the clustering of individuals with diverse educational back-
Estimator Two-way Two-way grounds and skill levels, facilitating cross-regional exchange of
FE FE
technology and knowledge. However, our findings reveal a persistent
“beggar-thy-neighbor” phenomenon in IGG cooperation among regions.
effects (Table 9) reveal that the estimated coefficients of ρ were uni- Another possible explanation is the demonstration effect, that is, under
formly positive and statistically significant at the 1 % level, substanti- the influence of the “promotion tournament” and the “one-vote veto”
ating the existence of a spatial spillover effect of IGG among prefecture- mechanism of environmental protection, local leaders are motivated to
level cities. This implies that an increase in IGG within one city catalyzes adopt the environmental regulations, environmental standards and
corresponding enhancements in neighboring cities. other systems of neighboring cities to strengthen IGG.
In contrast to nonspatial econometric models, the marginal effects of
explanatory variables on the dependent variable are not directly 4. Conclusions and policy implications
captured by spatial models. Consequently, the conclusions are drawn
from the results presented in Table 10, where the direct and indirect 4.1. Main findings
effects (spatial spillover effects) have been derived through decompo-
sition analysis. It was found in Model 7 that lnNTU exerts statistically A vast array of scholarly work emphasizes the necessity to transcend
significant positive direct and indirect effects on IGG at the 1 % signif- mere economic growth to enhance environmental quality, generate
icance level. Specifically, a 1 % increase in lnNTU was associated with a employment opportunities, and reduce inequalities (Sun et al., 2020;
0.732 improvement in IGG within the city and a 0.060 increase in IGG in Rammelt and Gupta, 2021; Wang et al., 2022; Ben Amara and Qiao,
neighboring cities. The empirical results reveal that the direct co- 2023; Qi et al., 2023; Xing et al., 2024). This necessitates an agenda for
efficients of lnGD, lnSUR, and lnEED were positive at the 1 % signifi- sustainable and inclusive green growth (Aslam and Ghouse, 2023). We
cance level. Additionally, the indirect effect of lnEED passed the 10 % posit that a key conception of IGG is to attain a synergistic balance be-
significance test, indicating that ecological environment governance has tween urban production efficiency and governance efficiency under the
a spatial spillover effect, which will drive the IGG increase in the constraints of limited capital, materials, environmental capacity, and

16
Y. Ren et al. Ecological Indicators 169 (2024) 112898

social input. Consequently, a potential contribution of our research to productivity (Chen et al., 2023). Additionally, high-IGG cities should
the IGG community is the deconstruction of the urban socio-economic implement stringent environmental regulations to ensure that industrial
“black-box” from a systemic perspective, establishing a nexus frame- growth does not come at the expense of ecological degradation.
work between the socio-economic system and the economic-
environmental system. Such an approach effectively addresses the lim- 4.2.2. Strengthen inclusive social urbanization in medium and low-IGG
itations of existing research, which often incorporates environmental, cities
social, and economic factors within a single framework, thus failing to For cities with medium and low IGG levels, the priority should be
capture the nuanced intrinsic links among economic, environment and enhancing social inclusivity by improving public services and infra-
social development. Given the Chinese government’s ambitious new- structure. Our findings suggest that inclusive social urbanization,
type urbanization strategy, it is imperative to analyze whether this including better access to education, healthcare, and public trans-
new-type urbanization can effectively foster IGG, thereby contributing portation, has a strong positive impact on IGG, particularly in less-
to regional high-quality development. This study draws the following developed cities. Policymakers should focus on narrowing the urban-
conclusions: –rural income gap by increasing investments in social services, which
can boost human capital and support long-term sustainable growth.
(i) The key factor constraining the growth of IGG has shifted from Additionally, expanding affordable housing and enhancing urban
governance efficiency in the early stages to production efficiency. employment opportunities will attract talent and drive consumption,
This shift indicates that the Chinese government’s progressive leading to more balanced and inclusive urban development (Li et al.,
increase in the supply of public services and infrastructure has 2020).
substantially improved urban green growth. On the production
side, there is an urgent need to vigorously promote industrial 4.2.3. Enhance ecological urbanization and support green infrastructure
green transformation and upgrading, as well as to advance clean Environmental and ecological urbanization is essential for sustaining
production technology and circular economy practices to miti- long-term IGG growth, particularly in cities facing environmental
gate environmental impacts. challenges. Policymakers should promote the development of green
(ii) Urban IGG in China exhibits a fluctuating upward trend; how- infrastructure, such as urban parks, sustainable water management
ever, the overall level remains suboptimal, highlighting the systems, and pollution control technologies, to improve the ecological
considerable potential for further improvement. Moreover, IGG quality of urban areas (Kang et al., 2021; Huang et al., 2023). Moreover,
demonstrates pronounced spatial–temporal differentiation char- implementing policies that encourage resource efficiency and recycling
acteristics, with particular emphasis on enhancing the national can help cities reduce waste and environmental degradation while
IGG level in the central mega cities. High-IGG cities are concen- enhancing their capacity to support future population growth.
trated in the eastern region, but the gap between cities is grad-
ually narrowing. 4.2.4. Promote regional coordination and knowledge spillover
(iii) The panel quantile model substantiates that new-type urbaniza- The results of this study highlight the importance of regional coor-
tion, and its components play a crucial role in improving IGG. dination mechanisms in enhancing IGG, particularly through the spill-
Notably, the impact of green-oriented economic urbanization on over effects of NTU. Cities with high levels of NTU can act as hubs of
IGG intensifies as the urban IGG level increases. Conversely, the innovation, sharing best practices and technological advancements with
positive impact of inclusive social urbanization and environ- neighboring cities. To harness these benefits, regional governments
mental and ecological development urbanization diminishes with should develop inter-city collaboration frameworks that facilitate the
increasing urban IGG levels, suggesting the possible presence of transfer of green technologies, labor, and capital across city boundaries
the Matthew effect in regions with social and economic advan- (Yao et al., 2020). This could involve joint infrastructure projects,
tages. Less developed cities need to attract the agglomeration of regional economic zones, or coordinated environmental policies that
production factors by enhancing human capital and expanding leverage the strengths of more developed cities to boost IGG in less-
the supply of basic public services and facilities, thereby facili- developed areas.
tating IGG progress.
(iv) The analysis of the time heterogeneity model and spatial panel 4.3. Limitations and future work
model reveals that the impact of new-type urbanization and its
components on IGG exhibits dynamic change characteristics and Although this paper constructs a multi-process IGG assessment
a positive spatial spillover effect over time. This underscores that framework and provides a systematic analysis of spatio-temporal evo-
the continuous promotion of the new-type urbanization strategy, lution and influencing factors, it still has several limitations. Firstly, due
especially green development-oriented urbanization, will signif- to the “black swan” event of the COVID-19 pandemic that began in 2020
icantly enhance urban IGG. Additionally, the establishment of a and significantly impacted social and economic development, we
pilot mechanism for IGG construction demonstration cities serves shortened the research time series to avoid the influence of abnormal
as a catalyst for regional IGG progress. data fluctuations on the results, which may limit the reflection of the
latest developments. Secondly, data availability constraints prevented
4.2. Policy implications us from including the environmental governance sector as part of the
city’s operational processes, and ecological functions were not consid-
Drawing from the empirical results, the following policy recom- ered in the selection of indicators. Finally, the impact of new urbani-
mendations may be provided: zation on IGG involves complex mechanisms; however, this study only
explores the three broad areas of green development, inclusive gover-
4.2.1. Prioritize green-oriented economic urbanization in high-IGG cities nance, and environmental protection, without delving into more
High-IGG cities, particularly in the eastern region, should focus on detailed mechanisms, thus limiting the granularity of the results. Future
consolidating and advancing green-oriented economic urbanization by research could enhance this approach by collecting socioeconomic and
promoting industrial upgrading and green technology innovation. Given ecological data over extended periods, for instance, using remote
that these cities already have the necessary infrastructure and financial sensing to gather data on ecosystem services. Additionally, future
resources, policymakers should incentivize investments in digital tech- studies should consider the relationship between urban socioeconomic
nologies, cleaner production technologies and support research and development and eco-environmental quality as a multi-stage process,
development in sectors that promote energy efficiency and green allowing urban IGG to be assessed in terms of urban production

17
Y. Ren et al. Ecological Indicators 169 (2024) 112898

efficiency, social governance efficiency, and environmental efficiency. Fenoglio, M.S., Rossetti, M.R., Videla, M., 2020. Negative effects of urbanization on
terrestrial arthropod communities: a meta-analysis[J]. Glob. Ecol. Biogeogr. 29,
1412–1429.
CRediT authorship contribution statement Gu, K., Dong, F., Sun, H., et al., 2021. How economic policy uncertainty processes impact
on inclusive green growth in emerging industrialized countries: a case study of China
[J]. J. Clean. Prod. 322, 128963.
Yufei Ren: Writing – review & editing, Writing – original draft, Guan, X., Wei, H., Lu, S., et al., 2018. Assessment on the urbanization strategy in China:
Visualization, Methodology, Investigation, Funding acquisition, achievements, challenges and reflections[J]. Habitat Int. 71, 97–109.
Guo, Y., Tong, L., Mei, L., 2020. The effect of industrial agglomeration on green
Conceptualization. Xinna He: Writing – original draft, Investigation.
development efficiency in Northeast China since the revitalization[J]. J. Clean. Prod.
Qunou Jiang: Writing – review & editing, Writing – original draft, 258, 120584.
Supervision, Resources, Project administration. Fan Zhang: Writing – Guo, X., Zhang, Y., Guo, D., et al., 2023. How does ecological protection redline policy
review & editing, Writing – original draft, Investigation. Bitian Zhang: affect regional land use and ecosystem services?[J]. Environ. Impact Assess. Rev.
100, 107062.
Writing – review & editing, Writing – original draft, Supervision, Re- Gupta, J., Vegelin, C., 2016. Sustainable development goals and inclusive development
sources, Project administration. [J]. Int. Environ. Agreem.: Politics Law Econ. 16, 433–448.
Hickel, J., 2020. The sustainable development index: measuring the ecological efficiency
of human development in the anthropocene[J]. Ecol. Econ. 167, 106331.
Declaration of competing interest Huang, Z., An, X., Cai, X., et al., 2023. The impact of new urbanization on PM2.5
concentration based on spatial spillover effects: evidence from 283 cities in China[J].
Sustain. Cities Soc. 90, 104386.
The authors declare that they have no known competing financial Jia, J., Ma, G., Qin, C., et al., 2020. Place-based policies, state-led industrialisation, and
interests or personal relationships that could have appeared to influence regional development: evidence from China’s Great Western Development
the work reported in this paper. Programme[J]. Eur. Econ. Rev. 123, 103398.
Jia, L., Xu, R., Shen, Z.Y., et al., 2023. Which type of innovation is more conducive to
inclusive green growth: independent innovation or imitation innovation?[J].
Acknowledgments J. Clean. Prod. 406, 137026.
Jiang, N., Jiang, W., Zhang, J., et al., 2022b. Can national urban agglomeration
construction reduce PM2.5 pollution? Evidence from a quasi-natural experiment in
This work was supported by The National Natural Science Founda- China[J]. Urban Clim. 46, 101302.
tion of China (Grant NO. 42101190, Fundamental Research Funds for Jiang, C., Li, J., Liu, J., 2022a. Does urbanization affect the gap between urban and rural
the Central Universities (BLX202213), and China Postdoctoral Science areas? Evidence from China[J]. Socioecon. Plann. Sci. 82, 101271.
Kang, J., Zhang, X., Zhu, X., et al., 2021. Ecological security pattern: a new idea for
Foundation (Grant NO. 2020M680018). balancing regional development and ecological protection. A case study of the
Jiaodong Peninsula, China[j]. Glob. Ecol. Conserv. 26, e01472.
Appendix A. Supplementary data Kao, C., 2018. A classification of slacks-based efficiency measures in network data
envelopment analysis with an analysis of the properties possessed[J]. Eur. J. Oper.
Res. 270, 1109–1121.
Supplementary data to this article can be found online at [Link] Ke, N., Lu, X., Kuang, B., et al., 2023. Regional disparities and evolution trend of city-
org/10.1016/[Link].2024.112898. level carbon emission intensity in China[J]. Sustain. Cities Soc. 88, 104288.
Koenker, R., Bassett, G., 1978. Regression quantiles[J]. Econometrica 46, 33–50.
Kong, L., Wu, T., Xiao, Y., et al., 2023. Natural capital investments in China undermined
Data availability by reclamation for cropland[J]. Nat. Ecol. Evol. 7, 1771–1777.
Kronenberg, J., Haase, A., Łaszkiewicz, E., et al., 2020. Environmental justice in the
context of urban green space availability, accessibility, and attractiveness in
Data will be made available on request. postsocialist cities[J]. Cities 106, 102862.
Li, S., Liu, Y., Elahi, E., et al., 2023. A new type of urbanization policy and transition of
References low-carbon society: a “local- neighborhood” perspective[J]. Land Use Policy 131,
106709.
Li, K., Qin, Y., Wu, J., 2020. Recent housing affordability in urban China: a
Aslam, A., Ghouse, G., 2023. Targeting the new sustainable inclusive green growth: a
comprehensive overview[J]. China Econ. Rev. 59, 101362.
review[J]. Clean. Respons. Consumpt. 11.
Liang, X., Rozelle, S., Yi, H., 2022. The impact of COVID-19 on employment and income
Ben Amara, D., Qiao, J., 2023. From economic growth to inclusive green growth: how do
of vocational graduates in China: evidence from surveys in January and July 2020
carbon emissions, eco-innovation and international collaboration develop economic
[J]. China Econ. Rev. 75, 101832.
growth and tackle climate change?[J]. J. Clean. Prod. 425.
Lin, B., Zhu, J., 2021. Impact of China’s new-type urbanization on energy intensity: a
Brock, G., 2020. Justice for People on the Move: Migration in Challenging Times[C].
city-level analysis[J]. Energy Econ. 99, 105292.
Cambridge University Press.
Liu, H., Cui, W., Zhang, M., 2022. Exploring the causal relationship between
Chen, M., Liu, W., Lu, D., et al., 2018. Progress of China’s new-type urbanization
urbanization and air pollution: evidence from China[J]. Sustain. Cities Soc. 80,
construction since 2014: a preliminary assessment[J]. Cities 78, 180–193.
103783.
Chen, D., Pan, Y., Jin, X., et al., 2021. The delineation of ecological redline area for
Liu, Q., Wang, S., Li, B., et al., 2020b. Dynamics, differences, influencing factors of eco-
catchment sustainable management from the perspective of ecosystem services and
efficiency in China: a spatiotemporal perspective analysis[J]. J. Environ. Manage.
social needs: a case study of the Xiangjiang watershed, China[J]. Ecol. Ind. 121,
264, 110442.
107130.
Liu, D., Xu, C., Yu, Y., et al., 2020a. Economic growth target, distortion of public
Chen, W., Wang, G., Xu, N., et al., 2023. Promoting or inhibiting? New-type urbanization
expenditure and business cycle in China[J]. China Econ. Rev. 63, 101373.
and urban carbon emissions efficiency in China[J]. Cities 140, 104429.
Liu, W., Zhan, J., Zhao, F., et al., 2021. Exploring the coupling relationship between
Cheng, Z., Li, L., Liu, J., 2020. Natural resource abundance, resource industry
urbanization and energy eco-efficiency: a case study of 281 prefecture-level cities in
dependence and economic green growth in China[J]. Resour. Policy 68, 101734.
China[J]. Sustain. Cities Soc. 64, 102563.
Cheng, Z., Li, X., Zhang, Q., 2023. Can new-type urbanization promote the green
Lo, S.-F., Lu, W.-M., 2009. An integrated performance evaluation of financial holding
intensive use of land?[J]. J. Environ. Manage. 342, 118150.
companies in Taiwan[J]. Eur. J. Oper. Res. 198, 341–350.
Cheng, Z., Wang, L., 2023. Can new urbanization improve urban total-factor energy
Lu, Y., Zhang, Y., Cao, X., et al., Forty years of reform and opening up: China’s progress
efficiency in China?[J]. Energy 266, 126494.
toward a sustainable path[J]. Sci. Adv. 5: eaau9413.
Dagum, C., 1998. A New Approach to the Decomposition of the Gini Income Inequality
Lv, C., Bian, B., Lee, C.-C., et al., 2021. Regional gap and the trend of green finance
Ratio[C], Springer.
development in China[J]. Energy Econ. 102, 105476.
Dell’anna, F., 2021. Green jobs and energy efficiency as strategies for economic growth
Ma, R., Liu, H., Li, Z., et al., 2024. Promoting sustainable development: revisiting digital
and the reduction of environmental impacts[J]. Energy Policy 149, 112031.
economy agglomeration and inclusive green growth through two-tier stochastic
Deng, X., Liang, L., Wu, F., et al., 2022. A review of the balance of regional development
frontier model[J]. J. Environ. Manage. 355, 120491.
in China from the perspective of development geography[J]. J. Geog. Sci. 32, 3–22.
Machado, J.A.F., Santos Silva, J.M.C., 2019. Quantiles via moments[J]. J. Econ. 213,
Elhorst, J.P., 2014. Spatial panel data models[C]. In: Spatial Econometrics: From Cross-
145–173.
Sectional Data to Spatial Panels. Springer Berlin Heidelberg, Berlin, Heidelberg, pp.
Maji, I.K., 2019. Impact of clean energy and inclusive development on CO2 emissions in
37–93. doi: 10.1007/978-3-642-40340-8_3.
sub-Saharan Africa[J]. J. Clean. Prod. 240, 118186.
Fan, S., Huang, H., Mbanyele, W., et al., 2023. Inclusive green growth for sustainable
Ofori, I.K., Gbolonyo, E.Y., Ojong, N., 2022. Towards inclusive green growth in Africa:
development of cities in China: spatiotemporal differences and influencing factors
critical energy efficiency synergies and governance thresholds[J]. J. Clean. Prod.
[J]. Environ. Sci. Pollut. Res. 30, 11025–11045.
369, 132917.
Färe, R., Grosskopf, S., 2000. Network DEA[J]. Socioecon. Plann. Sci. 34, 35–49.
Ofori, I.K., Gbolonyo, E.Y., Ojong, N., 2023. Foreign direct investment and inclusive
Feng, Y., Sun, M., Pan, Y., et al., 2024. Fostering inclusive green growth in China:
green growth in Africa: energy efficiency contingencies and thresholds[J]. Energy
identifying the impact of the regional integration strategy of Yangtze River
Econ. 117, 106414.
Economic Belt[J]. J. Environ. Manage. 358, 120952.

18
Y. Ren et al. Ecological Indicators 169 (2024) 112898

Pan, W., Wang, J., Lu, Z., et al., 2021. High-quality development in China: measurement Wang, D., Hou, Y., Li, X., et al., 2022. Developing a functional index to dynamically
system, spatial pattern, and improvement paths[J]. Habitat Int. 118, 102458. examine the spatio-temporal disparities of China’s inclusive green growth[J]. Ecol.
Peng, Z., Dan, T., 2023. Digital dividend or digital divide? Digital economy and urban- Ind. 139.
rural income inequality in China[J]. Telecommun. Policy 47, 102616. Wang, X., Song, J., Duan, H., et al., 2021. Coupling between energy efficiency and
Qi, X., Wang, B., Gao, Q., 2023. Environment, social and governance research of industrial structure: an urban agglomeration case[J]. Energy 234, 121304.
infrastructure investment: a literature review[J]. J. Clean. Prod. 425. Wu, X., Huang, Y., Gao, J., 2022. Impact of industrial agglomeration on new-type
Qian, Y., Fan, W., 2020. Who loses income during the COVID-19 outbreak? Evidence urbanization: evidence from Pearl River Delta urban agglomeration of China[J]. Int.
from China[J]. Res. Soc. Stratif. Mobil. 68, 100522. Rev. Econ. Financ. 77, 312–325.
Qiu, T., Wang, L., Lu, Y., et al., 2022. Potential assessment of photovoltaic power Xiao, Y., Huang, H., Qian, X.-M., et al., 2023. Can new-type urbanization reduce urban
generation in China[J]. Renew. Sustain. Energy Rev. 154, 111900. building carbon emissions? New evidence from China[J]. Sustain. Cities Soc. 90,
Rammelt, C.F., Gupta, J., 2021. Inclusive is not an adjective, it transforms development: 104410.
a post-growth interpretation of Inclusive Development[J]. Environ. Sci. Policy 124, Xiao, H., Wang, D., Qi, Y., et al., 2021. The governance-production nexus of eco-
144–155. efficiency in Chinese resource-based cities: a two-stage network DEA approach[J].
Ren, Y., Li, Z., 2022. Unraveling the dynamics, heterogeneity, determinants of eco- Energy Econ. 101, 105408.
efficiency in Beijing-Tianjin-Hebei urban agglomeration, China[J]. J. Environ. Xie, R., Teo, T.S.H., 2022. Green technology innovation, environmental externality, and
Manage. 317, 115407. the cleaner upgrading of industrial structure in China — Considering the moderating
Ren, S., Li, L., Han, Y., et al., 2022a. The emerging driving force of inclusive green effect of environmental regulation[J]. Technol. Forecast. Soc. Chang. 184, 122020.
growth: does digital economy agglomeration work?[J]. Bus. Strateg. Environ. 31, Xing, Q., Wu, C., Chen, F., et al., 2024. Intranational synergies and trade-offs reveal
1656–1678. common and differentiated priorities of sustainable development goals in China[J].
Ren, Y., Yuan, W., Zhang, B., et al., 2022b. Does improvement of environmental Nat. Commun. 15, 2251.
efficiency matter in reducing carbon emission intensity? Fresh evidence from 283 Yang, L., Chen, S., Chiu, Y.-h., et al., 2024. Reassessment of industrial eco-efficiency in
prefecture-level cities in China[J]. J. Clean. Prod. 373, 133878. China under the sustainable development goals: a meta two-stage parallel entropy
Rockström, J., Gupta, J., Qin, D., et al., 2023. Safe and just Earth system boundaries[J]. dynamic DDF-DEA model[J]. J. Clean. Prod. 447, 141275.
Nature 619, 102–111. Yao, L., Li, J., Li, J., 2020. Urban innovation and intercity patent collaboration: a
Shan, Y., Fang, S., Cai, B., et al., 2021. Chinese cities exhibit varying degrees of network analysis of China’s national innovation system[J]. Technol. Forecast. Soc.
decoupling of economic growth and CO₂ emissions between 2005 Chang. 160, 120185.
and 2015[J]. One Earth 4, 124–134. Yu, B., 2021. Ecological effects of new-type urbanization in China[J]. Renew. Sustain.
Shao, J., Wang, L., 2023. Can new-type urbanization improve the green total factor Energy Rev. 135, 110239.
energy efficiency? Evidence from China[J]. Energy 262, 125499. Yuan, Y., Wang, M., Zhu, Y., et al., 2020. Urbanization’s effects on the urban-rural
Shi, J., Dai, X., Duan, K., et al., 2023. Exploring the performances and determinants of income gap in China: a meta-regression analysis[J]. Land Use Policy 99, 104995.
public service provision in 35 major cities in China from the perspectives of Zhang, Q.-F., Tang, X., Xiao, Y., et al., 2023. Coordination of industrial structure and eco-
efficiency and effectiveness[J]. Socioecon. Plann. Sci. 85, 101441. efficiency in ecologically fragile areas: a case study of the Loess Plateau, China[J].
Sun, Y., Ding, W., Yang, Z., et al., 2020. Measuring China’s regional inclusive green J. Environ. Manage. 331, 117237.
growth[J]. Sci. Total Environ. 713, 136367. Zhang, H., Wu, K., Qiu, Y., et al., 1969. Solar photovoltaic interventions have reduced
Sun, H., Mao, W., Dang, Y., et al., 2022. What inhibits regional inclusive green growth? rural poverty in China[J]. Nat. Commun. 2020, 11.
Empirical evidence from China[J]. Environ. Sci. Pollut. Res. Int. 29, 39790–39806. Zhao, X., Ma, X., Shang, Y., et al., 2022a. Green economic growth and its inherent driving
Tang, W., 2021. Decentralization and development of small cites: evidence from county- factors in Chinese cities: based on the Metafrontier-global-SBM super-efficiency DEA
to-city upgrading in China[J]. China Econ. Quart. Int. 1, 191–207. model[J]. Gondw. Res. 106, 315–328.
Tone, K., Tsutsui, M., 2009. Network DEA: a slacks-based measure approach[J]. Eur. J. Zhao, Z., Pan, Y., Zhu, J., et al., 2022b. The Impact of Urbanization on the Delivery of
Oper. Res. 197, 243–252. Public Service-Related SDGs in China[J]. Sustain. Cities Soc. 80, 103776.
Tone, K., Tsutsui, M., 2010. Dynamic DEA: a slacks-based measure approach[J]. Omega Zhu, H., Chen, S., Irfan, M., et al., 2024. Exploring the role of the belt and road initiative
38, 145–156. in promoting sustainable and inclusive development[J]. Sustain. Dev. 32, 712–723.
Tullock, G., Brady, G.L., Seldon, A., 2002. Government Failure: A Primer in Public Choice
[C]. Cato Institute.

19