The Impact of Innovative and Low-Carbon Pilot Cities on Green Innovation

Abstract

Green innovation has emerged as a crucial strategy for reconciling economic development with environmental protection. While numerous policies target various aspects of innovation or green development individually, fewer policies address green innovation specifically. This raises the question of whether individual policies alone are sufficient to advance green innovation or whether a combination of policies is required. To address this, we analyze data from Chinese cities at the prefectural level or higher, focusing on the intersecting policies of innovative cities and low-carbon pilot cities as key explanatory variables. Using a longitudinal difference-in-differences model, our study examines the effects of these concurrent policies on green innovation and investigates the mechanisms underlying their effectiveness. Our findings reveal that the dual-pilot policy significantly promotes green innovation through four key pathways: regional economy, financial level, employment conditions, and education level. After confirming the efficacy of each policy individually, we rule out the impact of single or non-overlapping policies. This confirms that the observed enhancement in green innovation results from the combined effect of the two policies rather than from either policy in isolation. The study concludes with recommendations for further enhancing green innovation, based on the empirical evidence obtained.

1. Introduction

Innovation involves leveraging existing resources or knowledge to create new items or ideas that cater to the evolving needs of current residents. The advancement of innovation in urban settings markedly affects the efficiency of innovative activities on regional and national scales [1]. However, the efficiency of urban innovation is often impacted by the environmental regulations imposed by local city governments [2]. Specific environmental regulations, especially those focused on preserving the sustainable growth of urban areas, could potentially elevate the manufacturing and operational expenses for businesses within these localities. This could decrease the operating profit of these enterprises, negatively affecting the region’s economic growth and potentially hindering the progress of urban innovation [3]. The term “green innovation” has been introduced into urban development planning to address the issue of urban innovation and environmental sustainability. However, there are different interpretations of this term in the field of urban development research. Some scholars equate green innovation with environmental innovation, viewing it as a policy instrument for achieving economic and environmental sustainability in urban development using renewable and clean energy [4,5]. Others perceive green innovation as a process of green technology development driven by technological and market shifts, company factors, and external environmental elements [6], with green technology innovation encompassing three dimensions: economic benefits, innovation benefits, and environmental benefits [7]. Additionally, various scholars suggest that green innovation encompasses technologies, methods, and goods that conform to both environmental and economic standards [8]. This includes products, processes, and services designed to reduce environmental pollution and harm by employing novel technologies or strategies for managing emissions, recycling or repurposing waste, and discovering cleaner, more efficient sources of energy [9,10].

Green innovation is crucial for urban development in many respects. On a micro-level, it encourages urban businesses to actively adopt green innovation technologies, facilitating both “source reduction” and “end cleaning” during their development process. This accelerates the digital transformation of businesses and aligns their growth with the state’s advocated sustainable development direction [11]. On a macro-level, improving the caliber of urban green innovation can contribute to the optimization of the city’s industrial framework. This has a beneficial effect on the development of the city’s eco-friendly financial sector and accelerates China’s progress towards its goals of reaching a carbon peak and achieving carbon neutrality. In the end, this fosters simultaneous sustainable growth in both the nation’s economy and its environmental health [12]. Consequently, researchers are delving into the variables that affect green innovation. Recent studies highlight the beneficial impact of factors like financial spending [13] and environmental policy [14] on the promotion of green innovation. Yet, there is a predominant focus on diverse production variables in these studies, with a relative lack of attention to the implications of policy-related factors. While it is true that there are fewer policies specifically related to green innovation, they are being implemented in an orderly manner. Policies focused on specific aspects of innovation or green development may have individual or cumulative effects, thereby promoting the development of green innovation. In China, for example, the Implementation Plan on Further Improving a Market-Oriented Green Technology Innovation System was released by the National Development and Reform Commission and the Ministry of Science and Technology at the end of 2022, after policies focusing on innovation and green urban development were put into effect. In 2024, the CPC Central Committee and the State Council issued the Opinions on Accelerating Comprehensive Green Transformation of Economic and Social Development, which lays out systematic measures for accelerating comprehensive green transformation in economic and social development [15]. In addition, the State Intellectual Property Office has actively promoted the transformation and utilization of green technology patents by continuously improving the classification system of green patents and strengthening the statistical analysis of green patents [16]. Overseas, green innovation has shown a positive development trend, with countries promoting green transformation and economic growth through policy support and technological innovation [17]. For example, the European Union promotes green industry capacity development and technological innovation through policy planning, such as the Net Zero Industry Act; the Netherlands leads the offshore wind power industry; and the reduction of energy life cycle costs encourages government policy and financial support to address the decarbonization of electricity production [18], making a positive contribution to the realization of a clean and sustainable future.

The principle of decreasing marginal yields from conventional factor-driven gains distinctly illustrates that depending exclusively on traditional elements for a nation’s economic transition is insufficient to meet the demands of contemporary economic changes. Consequently, the adoption of a strategy focused on innovation-driven growth, coupled with expediting the shift of national economies towards sustainable urban development, stands as a vital goal in the economic progression of numerous countries. In this process of fostering innovation systems, the government’s role is indispensable [19]. Through the enactment of pertinent policies and measured intervention, the government can successfully counter the imbalance in innovation resources attributed to market inadequacies. This leads to the judicious distribution and advancement of innovative resources across regions. Consequently, it facilitates the systematic implementation of innovation initiatives at the regional level [20]. Over the past decade, the Chinese government has released a multitude of policy papers aimed at initiating pilot projects for innovative cities, aiming to drive urban economic improvement through technology [21]. The execution of these policies can be segmented into three distinct phases. The first phase (2006–2010) was a small-scale experimentation phase, with Shenzhen being chosen as China’s first innovative city pilot in 2008. The second phase (2010–2016) was a large-scale experimentation phase, during which the central government selected cities from various local pilots, propelling the construction of innovative city pilots to a policy diffusion stage. Concurrently, 60 national-level enterprises were established. The third phase (2016–present) is the policy diffusion stage, where the central policy reduces local empowerment and accelerates policy diffusion following interactive screening. These three stages of innovative city establishment have resulted in a national innovation network that spans the eastern, central, and western regions of mainland China [22].

In order to stimulate economic growth in China, the Chinese government has emphasized the advancement of the real economy, enacting various policies designed to promote the growth of the industrial sector. However, this growth has led to significant environmental issues. The industrial system’s development process emits a large volume of greenhouse gases, leading to ecological deterioration [23]. Therefore, achieving a balance between economic expansion and carbon emission control has become a critical challenge in the pursuit of sustainable urban development in China [24]. More critically, global warming, driven by excessive carbon dioxide emissions, poses a severe threat to human survival [25]. Furthermore, the growing population in developing countries is certain to escalate carbon emissions. Being the largest developing country globally, China carries a substantial duty in terms of energy saving and reducing emissions, which will greatly influence global carbon emissions [26,27]. As a result, the Chinese government launched the inaugural series of low-carbon pilot programs in five provinces and eight cities in 2010, setting specific local targets for carbon emissions [28]. Following this, the government introduced two more rounds of low-carbon pilot cities in 2012 and 2017, based on the initial pilot projects [29]. Up to the present, China has identified 36 cities and six provinces as low-carbon pilot cities [30]. Research shows that the establishment of a low-carbon pilot city has a favorable effect on improving urban air quality and ecological performance [31], but it also expedites the digital transformation of businesses [32].

Innovative cities and low-carbon pilot cities have effectively contributed to their respective fields by fostering innovation and reducing carbon emissions. However, it remains unclear whether these two policies, when combined, positively influence green innovation. This research aims to explore the impacts and underlying mechanisms of the ‘dual-pilot policy’ involving innovative cities and low-carbon pilot cities on green innovation, employing these policies as the explanatory factors. This research, utilizing data from Chinese urban centers at the prefectural level or higher, further seeks to ascertain if singular policies and their interplays are equally effective in fostering green innovation.

Compared with previous studies on innovative cities and low-carbon pilot city policies, the research in this study has the following three innovative points: First, in terms of the selection of explanatory variables, most scholars only choose a single policy as the explanatory variable in the research on the implementation effects of innovative cities and low-carbon pilot city policies to study the effects of these two policies on urban green development [33] and innovative development [34], respectively. Instead, this study chooses the dual-pilot policy as an explanatory variable to study the influence and mechanism of the joint implementation of the two policies on the development of urban green innovation. Second, in order to further explore and understand the influencing factors of urban green innovation, this study explores the impact mechanism of dual-pilot policy on urban green innovation from four dimensions, including regional economy, financial level, employment level, and education level, and studies the direct and indirect impact transmission mechanisms caused by the dual-pilot policy themselves. Third, in previous articles examining the effects of dual policies, scholars have typically neglected to exclude tests of the effects of policy overlap [35]. On the basis of establishing that innovative cities and low-carbon pilot city policies can have positive impacts on urban green development and innovative development, respectively, this study further excluded the impact of single pilot policy and single policy superposition to verify the net impact of dual-pilot policy on urban green innovation, ensuring the reliability and accuracy of this study. It can also provide a reference for other multi-policy research.

The rest of this document is structured in the following manner: First, we review related research findings, compile studies on innovative cities and low-carbon pilot city policies in their respective fields, and analyze the potential effects and mechanisms of these two overlapping policies on green innovation. In the third section, we describe the chosen models and data, and in the fourth section, we examine the empirical evidence surrounding the effects of the dual-pilot policy on green innovation and its mechanisms. We then investigate whether individual policies and their non-overlapping aspects can also influence green innovation, based on testing the effectiveness of innovative cities and low-carbon pilot city policies in their respective fields. Finally, based on our empirical findings, we put forth pertinent policy suggestions to lay the theoretical groundwork for the advancement of green innovation in China. These recommendations also serve as a useful guide for developing nations considering a green and synergistic approach to development.

2. Literature Review

2.1. Innovative Cities vs. Innovative Development

To deepen the scientific outlook on development, fully implement the strategy of independent innovation, and advance the strategy of innovation-driven growth, the Chinese government has actively initiated the pilot policy for innovative cities. This policy lays a robust foundation for China’s transformation into an innovative nation, with the strategy of developing innovative cities playing a crucial role in the economic and social progress of urban areas in China [36]. An early innovative city represents a novel form of urban society that integrates new elements, undergoes social and economic changes, and fosters innovation [37]. In contrast, modern innovative cities possess a strong capability for independent innovation, effectively harnessing various innovative components such as technology, talent, environment, culture, and regional resources to achieve both innovation and sustainable development [38]. In the context of China’s developmental efforts, the implementation of the innovative city policy has had positive implications. Economically, innovative cities have the potential to enhance the aggregation of scientific and technological expertise and to refine urban industrial structures during their development [39]. Additionally, these cities, due to their higher levels of economic development, can drive regional growth and further optimize their industrial structures by funding innovative technologies [40]. Regarding urban living standards, the development of innovative cities contributes to the growth of smart urban environments, improves the quality of life for residents, and, to some extent, enhances their sense of well-being [41].

Most scholars concentrate on the innovation effects produced by innovative cities. They typically approach this in two ways. The first approach involves focusing on innovative cities as the research scope and investigating the differences among them. As an illustration, Wang et al. examined the synergy between industry, academia, and research within innovative cities, as well as the spatial synergy among these cities [42]. The second approach considers innovative cities as a policy and assesses its effectiveness. For instance, Caragliu and Del Bo used the propensity score matching method to estimate the urban innovation effect of implementing the pilot policy of innovative cities [43]. He found that establishing innovative cities can foster urban innovation development, and the innovative technology produced in the process can have a spillover effect. Similarly, Fan et al. assessed the effect of the innovative cities pilot program using the difference-in-differences (DID) model [44]. He discovered that the policy could be effective by influencing government financial expenditure, urban industrial agglomeration, and the spatial synergy between innovative cities and the city’s industrial clustering. Moreover, several researchers have suggested that innovative cities could positively influence green development for two main reasons. First, implementing creative city pilot programs can reduce urban carbon emissions over the long run while simultaneously producing short-term advantages [45]. Second, by encouraging the expansion of human capital, knowledge stock, and research and development levels, innovative city development can raise a city’s green total factor productivity [46].

2.2. Low-Carbon Pilot Cities vs. Green Development

The notions of a low-carbon society and economy gave rise to the concept of a low-carbon city [47]. The notion of a low-carbon economy was initially presented by the UK government in 2003 in their Energy White Paper, in the context of global warming. They defined it as achieving high economic output and quality development while minimizing environmental pollution and damage. In 2007, the Japanese government put forth the idea of a low-carbon society. China is one of the world’s biggest carbon polluters, and as the globe struggles with severe issues brought on by climate change [48], China has implemented a low-carbon pilot city policy [49], which aims to reduce urban carbon emissions and promote the development of urban ecosystems. It is based on the idea that urban areas should have sustainable energy and ecological systems that are rooted in the city’s low-carbon production and consumption [50]. The development of low-carbon pilot cities may have a favorable impact on these cities’ industrial composition [51], resulting in enhanced rationalization and, eventually, sustainable growth. Furthermore, urban dwellers’ knowledge of low-carbon living can be increased via the establishment of low-carbon pilot cities, which can change their consumption habits and motivate them to make more environmentally friendly decisions [52].

The consequences of low-carbon pilot cities on the environment have been the main subject of research. These policies can either stimulate a low-carbon transformation and green production in businesses through digital technology, thereby promoting the city’s green development [53], or they can influence the city’s economic and technological structure, increasing green total factor productivity [54]. China’s low-carbon city pilot program can either directly or indirectly reduce urban carbon emissions by promoting industrial structure upgrades and technology innovation, given the country’s increasing environmental pollution and deteriorating effects of global climate change [55], thereby mitigating urban environmental pollution to some extent [56]. Moreover, low-carbon city pilot programs can have a favorable impact on the carbon emission efficiency of non-policy-affected cities [57], collectively advancing the green development of both the pilot cities and their surrounding areas. Green development depends on invention, and studies on the impact of low-carbon pilot cities on innovation are only just beginning. On a micro-scale, the implementation of low-carbon pilot city regulations can enhance resource allocation and foster technological innovation in Chinese listed enterprises [58]. It can also raise the demand for competent people in these organizations, which can subsequently enhance their profitability [59]. On a macro-scale, policies pertaining to low-carbon pilot cities have a substantial impact on technological innovation in both the pilot cities and the cities nearby. By easing financial limitations and streamlining the industrial structure, these measures can indirectly promote technical innovation [60].

2.3. Dual-Pilot Policies vs. Green Innovation

Currently, as innovation resources become scarce and the ecological environment deteriorates, there is an increasing focus on green innovation. Through the optimization and transformation of urban industrial infrastructure, green innovation may economically support the growth of a green economy [61]. In terms of innovation benefits, green innovation can consolidate a city’s innovation resources to achieve innovative growth [62]. Simultaneously, it can also reduce urban air pollution, thus promoting the city’s residents’ low-carbon consumption from an environmental perspective [63]. As a result, developing urban green innovation is essential to stopping the deterioration of the urban environment, encouraging the growth of the urban green economy, and attaining a balanced development of the economy and ecology [64]. While most research on innovative cities centers on their influence on innovation, and studies on low-carbon pilot cities focus largely on their impact on green development, some study has begun to explore the beneficial implications of diverse policy pilots for green innovation. For example, Liu et al. examined the beneficial relationship between the degree of green innovation in enterprises located in the implementation area and the government’s low-carbon pilot city policy using the DID model [65]. He also looked at business investment in green innovation initiatives from a macro-viewpoint and found that government-led environmental policies had good benefits. Concurrently, Du et al. (2021) formulated a hybrid triangular envelope analysis model to evaluate a city’s eco-efficiency [66]. He also used a DID approach to assert that low-carbon pilot city can enhance eco-efficiency through green innovation mechanisms.

2.4. Inadequacies and Improvement of Existing Research

Through the collection and sorting of the existing literature, it can be seen that current scholars do not consider both policies when examining the impact of innovative cities and low-carbon pilot cities on green innovation. And since the two policies have a high degree of temporal overlap, it is impossible to determine whether the improvement in green innovation level is due to one of the policies or the superposition of the two policies at the same time. Therefore, this study addresses the shortcomings of the current study and examines the impact of innovative cities and low-carbon pilot cities on green innovation under the superposition of the two policies. In addition, this study extends the analysis of the impact mechanism of the superposition of policies of innovative cities and low-carbon pilot cities on green innovation.

The intersection of innovative cities and low-carbon pilot city policies can enhance the regional economy, financial status, employment, and education levels. First, the dual-pilot policy’s adoption eases corporate financing limits to hasten enterprises’ digital transformation by allowing the government to increase its financial investment in research and technology at the regional economic level. This ultimately leads to an increase in business revenue, driving the regional economy [67]. Second, at the financial level, the dual-pilot policy offers policy support and financial security for urban fintech development, promoting urban green savings and investment by reducing costs, thereby advancing urban green finance [68]. Third, regarding employment, the pilot cities’ governments will increase high-tech and service industries to improve the industrial structure [69] and provide more jobs, aiming for full employment of urban residents. Finally, implementing the dual-pilot policy at the education level can improve the city’s innovation environment, leading to increased government expenditure on education. This attracts innovative talents and fosters the city’s innovation development [70]. Enhancing the regional economy, financial level, employment conditions, and education level can further stimulate green innovation. First, urban green innovation is advanced, and green innovation research and development are encouraged when there is a stronger regional economy because it offers a solid material foundation for urban growth and long-term financial support for technological innovation [71]. Second, a better financial level might lessen the funding obstacles companies have when engaging in green innovation, hastening the growth of urban green innovation [72]. Third, more favorable work environments can encourage the expansion of green human resource management in urban areas. Green human resource management procedures have the power to impact workers’ environmental beliefs and behaviors [73], creating an inherent demand for green innovation technologies and furthering green innovation. Lastly, a higher education level signifies increased investment in human resource development, ensuring a pool of research talents for the city’s green innovation, thereby promoting the city’s green innovation [74].

3. Data and Modeling

3.1. Study Design

Based on existing research and related theories, this study investigates the impact of the “dual-pilot policy” on green innovation within innovative cities and low-carbon pilot cities, as illustrated in Figure 1. The research is structured into three main sections. Section 3 introduces and explains the selection of data and models, providing a foundation for the subsequent empirical analyses. Section 4 is divided into two subsections that evaluate the effectiveness of the dual-pilot policy. The first subsection assesses the impact on green innovation through benchmark regression, robustness testing, and parallel trend testing. The second subsection explores the pathways through which this impact occurs. Section 5 focuses on testing the effectiveness of individual policies, also divided into two parts. The first part examines the roles of innovative cities and low-carbon pilot cities in their respective areas to validate variable selection. The second part investigates the individual and non-overlapping effects of these cities on green innovation, aiming to discern whether the observed increase in green innovation is due to the combined effect of the two policies or attributable to a single policy.

3.2. Selection of Variables

3.2.1. Explained Variable: Green Innovation

This research directly measures each city’s capacity for green innovation by using the number of green patents as an index. All other uncontrolled external elements may be eliminated from the representation of each city’s excitement for green innovation through the use of green patent applications. In order to gauge each city’s capability for green innovation, the total number of green innovations and green utility model patents that each city applies for each year is chosen as the explanatory variable. The quantity of green patents is another common metric used in related research to assess green innovation. For example, Chen et al. investigated the connection between environmental sustainability and green innovation using the total number of environmental technology patents registered [75]. The green innovation data utilized by Bai and Lin in their investigation of the connection between green finance and green innovation came from the green invention patent applications of listed businesses [76]. In order to gauge green innovation in their study on the impact of the emissions trading policy on green innovation in the pilot and surrounding areas, Du et al. looked at the number of green invention patents and their authorizations, as well as the number of green utility model patents and their authorizations [77]. To quantify green innovation in their study on the connection between heterogeneous green innovation and cities’ carbon emission performance, Xu et al. employed the quantity of green patents and their authorizations [78]. To determine if environmental information sharing has an impact on green innovation, Lu and Li measured the amount of green innovation using the number of green patent applications [79].

3.2.2. Explanatory Variables: Innovative Cities and Low-Carbon Pilot City Policies

Cities are divided into two groups in this study: the treatment group and the control group. The dual-pilot policy is chosen as the explanatory variable. The cities in the treatment group are those that have executed the low-carbon pilot city policy as well as the innovative city policy, while the cities in the control group have executed one or neither of these policies. As Figure 2 illustrates, the policy was mostly implemented in 2011 during the 2008–2018 period. The majority of the cities adopting the creative city strategy are situated in China’s central and eastern coastal areas. The low-carbon pilot city policy, in contrast, was primarily implemented in 2010 during the 2010–2018 period and is concentrated in China’s central and southern regions. It is logical and scientifically appropriate to investigate the impact of the dual-pilot policy on green innovation, given the considerable time overlap between these two initiatives. Furthermore, compared with the number of cities implementing a single policy, fewer cities have implemented the dual-pilot policy, which was initially implemented in 2010 and is mostly implemented in the country’s east and center.

The treatment impact of the dual-pilot policy is indicated in this study through the interaction term between the dummy variables for innovative cities and low-carbon pilot cities (Treat_it). Specifically, we categorize the cities within the research scope into two groups based on whether and when they have implemented the dual-pilot policy. This implies that cities that have implemented the dual-pilot policy constitute the experimental group. When a city i implements the innovative cities and low-carbon pilot city policies at the same time in year t, the meaning of Treat_it in that year, and the following years, is taken as one. Conversely, cities that have not implemented the dual-pilot policy at the same time are the control group, and the value of Treat_it is taken as zero. As Figure 1 illustrates, the dual-pilot policy’s implementation is focused in the years 2010–2018. As a consequence, the values of Treat_it in the various cities under investigation vary over time.

3.2.3. Control Variables

Since a city’s degree of green innovation may also be influenced by other factors, the study included the following four control variables to account for this factor:

(1)

Industrial structure: share of secondary industry

The industrial structure of a city can reveal differences in resource allocation and factor inputs among cities [80], while changes in this structure can indirectly suggest the city’s future development direction [81]. The National Development and Reform Commission of China, in its draft for public comment on the Guidance Catalogue for Industrial Structure Adjustment (2023 edition), suggests that China’s industrial structure policy aims to enhance the manufacturing industry’s environmental sustainability and to actively encourage the growth of emerging technology industries to boost innovation. As a result, the fraction of secondary industry is used in this study to quantify industrial structure.

(2)

Market environment: total retail sales of consumer goods

The white paper, “China’s Green Development in the New Era”, emphasizes that China has widely adopted green production and lifestyle practices, leading to a growing consumer demand for green products. A robust market environment, which is fundamental for innovation and development [82], can generate substantial market demand [83]. The development of new goods and technology is subsequently fueled by this demand. Additionally, the rise in market demand stimulates the enhancement and innovation of supply-side products [84]. Consequently, the market environment is chosen as the control variable in this study, and the growth of the market is gauged by the total retail sales of consumer products.

(3)

Land use: urban construction land area

The report from the 20th Party Congress explicitly asserts that China’s modernization is one where humans coexist harmoniously with nature, implying that China should actively foster the development of an ecological civilization system. Urban land use planning plays a crucial role in building China’s ecological civilization [85], and sensible planning of urban land use, coupled with the assessment of land ecological value, can further stimulate urban green innovation [86]. Consequently, this study selects the area of urban construction land, which reflects the city’s land use situation, as its control variable.

(4)

Standard of living: average wage of employed workers

The living standards of urban residents can serve as an indicator of urban consumption levels [87]. Currently, in China, urban consumption levels are continually improving, and the concept of green, low-carbon living is increasingly integrating with residents’ daily consumption habits [88]. As consumer groups become more aware of green consumption, this will further drive the transformation towards green, low-carbon industries and the innovative development of green technology products [89]. Therefore, this study uses the average wage of employed workers as a measure of the living standards of urban residents, serving as the control variable.

3.2.4. Intermediate Variables

As proxies for the four mediating variables related to regional economy, financial level, employment, and education level, we use regional GDP, loan balances of financial institutions, the amount of urban employees at the end of the period, and the quantity of students enrolled in general colleges and universities, respectively. The selection is based on a theoretical analysis of how dual-pilot policy influences green innovation. The aim is to empirically test and validate these impact mechanisms.

3.3. Research Modeling

3.3.1. DID Modeling

The purpose of this study is to examine the effect and mechanism of the dual-pilot policy on urban green innovation. Given that the implementation of these policies is exogenous to individual entities and there is no issue of reverse causality, the DID (difference-in-differences) model has been selected to address potential endogeneity concerns. Currently, scholars commonly use the DID model to study the effects of policy implementation. For instance, Wang et al. [90] employed a PSM-DID model to investigate the impact of China’s carbon trading pilot system on the transition to a low-carbon economy. Similarly, Zhao et al. [91] utilized the DID model to address endogeneity issues when examining the effects of the extended producer responsibility system on enterprise green technology innovation. Considering that each city studied in the sample data had different implementation times for their pilot policy, the study used the heterogeneous timing DID model, which is suitable for situations where individuals in the processing group were exposed to different policy impacts at different times. In Ruixue Jia [92]’s study, she took the establishment of modern treaty ports as a quasi-natural experiment. Since different regions in China set up treaty ports at different times, the study adopted the heterogeneous DID timing model to explore the influence of treaty ports on the population and economic development of modern China. In addition, Fauver et al. [93] used the heterogeneous timing DID model to evaluate the effect of reform in 41 countries with different reform times in the study of the impact of board reform in different countries on enterprise value. The following is the related multi-period DID model:

$$Y_{it}={\beta }_0+{\beta }_{1}Trea{t}_{it}+{\beta }_{2}Contro{l}_{it}+{\gamma }_i+{\mu }_{it}$$

(1)

In model (1), Y_it is the level of green innovation growth of a city, and Treat_it is a dummy variable for the dual-pilot policy. The value of this variable is determined by whether the dual-pilot policy was implemented in city i during year t. If the policy was in effect, the indicators for that year and subsequent years are assigned a value of one; otherwise, they are given a value of zero. Subsequently, β₁ is the Treat_it coefficient estimate, whose sign and size reflect the influence and direction of the dual-pilot policy on the growth of green innovation in the city. If β₁ is positive and significant, this suggests that the dual-pilot policy significantly promotes the green innovation development of cities, thereby demonstrating the policy’s effectiveness. Furthermore, The collection of control variables is represented by Control_it, which means the set of variables that could influence the level of green innovation development in cities and vary across individual cities. This removes the possibility that the experiment’s outcomes will be influenced by other variables. Finally, μ_it is the error term, and γ_i stands for each unique city fixed effect.

3.3.2. Modeling Intermediary Effects

To examine the impact mechanism of the dual-pilot policy on urban green innovation, this study employs a step-by-step testing method to analyze the mediating effects. This approach not only clarifies the direct impact of the dual-pilot policy on urban green innovation but also allows for the comparison of the direct effect with the indirect effects mediated by other variables. This provides quantifiable data that can inform policy formulation and development directions for urban green innovation. This methodology has also been used in other studies. For example, Cao et al. [94] utilized stepwise tests to explore the path through which environmental regulations affect employment, and Chen et al. [95] used the mediation effect model to estimate the mechanism by which large-scale implementation of green credits in China influences provincial economic growth. Their analyses followed a three-step process. First, we use model (1) to construct a regression of the dummy variable Treat_it of the dual-pilot policy on the degree of urban green innovation Y_it. If the regression coefficient β₁ is significant, we then build model (2) so that we can perform a regression analysis of Treat_it on Med_it, the mediating variable. A significant influence coefficient a₁ suggests a significant effect of the dual-pilot policy on the mediating variable. Finally, we construct model (3) to perform a regression of Treat_it and Med_it on Y_it. If the coefficient a₁ of Treat_it in model (2) is significant, and the regression coefficients a₂ of Treatit and b₂ of Med_it in model (3) are also significant, this suggests a partial mediation effect. This indicates that the dual-pilot policy has the ability to both directly and indirectly advance urban green innovation, with the indirect effect having an a₁b₂ magnitude due to the involvement of mediator factors. A full mediation effect is shown if the regression coefficient a₂ of Treat_it in model (2) is not significant, but only the regression coefficient b₂ of Med_it in model (3) is, and the coefficient a₁ of Treat_it in model (2) is. Stated differently, the dual-pilot policy’s ability to impact the growth of urban green innovation is limited to its impact on the mediator factors.

$$Me{d}_{it}={\alpha }_1+a_{1}Trea{t}_{it}+c_{1}Contro{l}_{it}+{\lambda }_{1i}+{\epsilon }_{1it}$$

(2)

$$Y_{it}={\alpha }_2+a_{2}Trea{t}_{it}+b_{2}Me{d}_{it}+c_{2}Contro{l}_{it}+{\lambda }_{2i}+{\epsilon }_{2it}$$

(3)

In models (2) and (3), Med_it denotes the mediating variable, and this study chooses the regional economy, financial level, employment situation, and education level as the mediating variables; a₁ and a₂ represent the estimated coefficients of the effect of Treat_it in models (2) and (3), respectively; c₁ and c₂ denote the regression coefficients of the control variable Control_it; b₂ denotes the estimated coefficients of the effect of the mediating variable Med_it on the explanatory variable Y_it; λ_1i and λ_2i denote city individual fixed effects; and ε_1it and ε_2it denote the error terms.

3.4. Data Sources

This research, which covers 2003–2020, focuses on 284 mainland Chinese cities that are larger than prefectures. The green innovation measurement indicators are sourced from the China Research Data Service Platform, and the remaining data are derived from the “China Urban Statistical Yearbook”. For cities with data lacking in particular years, we utilize the interpolation approach to fill in the gaps.

Table 1. Descriptive statistical analysis of variables.

Variable	Obs	Mean	StD.Dev.	Min	Max
Y	5112	4.917 × 102	1.764 × 103	0	3.467 × 104
X	5112	6.7 × 10−2	2.5 × 10−1	0	1
C1	5112	4.697 × 10	1.134 × 10	1.93	90.97
C2	5112	7.249 × 106	1.222 × 107	238	1.593 × 108
C3	5112	2.249 × 102	2.958 × 103	1	1.011 × 105
C4	5112	4.144 × 104	2.487 × 104	9.81	3.206 × 105
M1	5112	1.820 × 107	2.991 × 107	3.177 × 105	3.870 × 108
M2	5112	2.306 × 107	5.689 × 107	2.815 × 105	8.104 × 108
M3	5112	5.099 × 10	7.790 × 10	4.05	1.143 × 103
M4	5112	8.149 × 104	1.611 × 10	8.055 × 10	1.307 × 106

displays the descriptive statistics for the chosen data.

4. Analysis of the Effects of the Dual-Policy Pilot

4.1. Parallel Trend Test

The idea behind the dual-policy pilot is to take away the impact of variables other than the policy. This is accomplished by matching the control group and the treatment group to the policy shocks that transpired prior to the shift in the green innovation trend. The panel data selected for this study spans from 2003 to 2020, with the first dual-pilot policy implemented in 2010. This means that some pilot cities lack sample data for the first seven periods of the policy. Therefore, it is necessary to include the time before the first seven policy periods in each city’s panel data and exclude its time dummy variables to prevent multicollinearity. The parallel trend is then selected from the first seven periods to the second seven periods of the policy. Figure 3 presents the parallel trend test results, where −i (i = 2, 3, …, 7) denotes the year i before the policy shock, and i (i = 2, 3, …, 7) denotes the year i after the policy shock. The results demonstrate that prior to the policy shock, the temporal dummy variable coefficients failed the significance test. The effects of the dual-pilot policy on green innovation did not level out for three years following the policy shock. Three years after the dual-pilot policy was implemented in the dual-pilot cities, the time dummy variable coefficients passed the significance test. The overall development of green innovation showed a year-over-year increase, indicating that the parallel trend test was successful.

4.2. Benchmark Regression

Based on the findings of the Hausman test, this study employs individual fixed effects for the benchmark regression;

Table 2. Benchmark regression results for dual-pilot policy.

Variables	Coefficients	Variables	Coefficients
X	3.757 × 102 ***	C3	−2.581 × 10−2 ***
	(0.000)		(0.000)
C1	−4.735 ***	C4	−6.867 × 10−3 ***
	(0.003)		(0.000)
C2	1.412 × 10−4 ***	cons_	−4.397 × 10
	(0.000)		(0.615)
sigma_u	469.4031	sigma_e	742.4067

Note: p-statistics are in parentheses; *** indicates significance at the 1 percent level.

shows the outcomes of the benchmark regression analyzing the influence of the dual-pilot policy on green innovation. In order to account for city fixed effects, this regression incorporates the explanatory variables (green innovation), the explanatory variable (dual-pilot policy), and the control variables (industrial structure, market environment, land use, the standard of living). The findings of the regression analysis indicate that, at the 1% significance level, the dual-pilot policy’s regression coefficient on green innovation is positive, indicating that the policy encourages green innovation in every city. This is because the implementation of the dual-pilot policy facilitates a more strategic and optimal allocation of resources across the city. It enhances support for the research and development of innovative green technologies through increased government financial input and human resources. Concurrently, the policy encourages enterprises to adopt cleaner production methods and technologies, thereby expanding the demand for innovative green technologies. This increased demand subsequently drives the supply of innovative green technologies. The control variables of land use, living standard, and industrial structure all pass the 1% significance level test; however, they all have negative effect coefficients, meaning that they all impede the city’s green innovation. This is due to the following: (1) A rise in the industrial structure indicator, which shows the share of the secondary industry, which causes the share of the tertiary industry where research input is relatively low to fall, limiting the growth of green innovation; (2) An increase in land use area, which results in a decrease in grassland, wetland, lakes, and so on, in the same area, reducing the demand for green innovation due to the decrease in green space; (3) An improvement in living standards, which means an increase in urban residents’ affluence level. According to the “inverted U-shape” characteristics of the environmental Kuznets curve (EKC), as affluence level increases, the growth rate of carbon emissions decreases, leading to a reduction in green innovation investment and ultimately inhibiting green innovation development. (4) The estimated coefficient of the market environment’s impact on green innovation is positive at the 1% significance level, indicating that it aids in the city’s green innovation development. This is because a well-developed market environment signifies an increase in each city’s residents’ consumption demand, and the continuous upgrading of consumption demand can provide innovative impetus for social production, further promoting green innovation in products.

4.3. Robustness Tests

(1)

Addition of Control Variables

Given the potential of urban green spaces to contribute to a city’s green development and provide social utility, this study includes the control variable “green area” to further assess the impact of the dual-pilot policy on green innovation.

Table 3. Robustness test results in the dual-pilot policy.

Variables	The Addition of Control Variables’ Coefficient	Adding Replacement Variables’ Coefficient	Placebo Test Results’ Coefficient	Time Fixed Effects Test
X	3.758 × 102 ***	3.806 × 102 ***	6.904 × 101	3.757 × 102 ***
cons_	−4.461 × 101	−2.830 × 102 ***	−4.767 × 101	−4.397 × 101
control variable	Yes	Yes	Yes	Yes
fixed effect	Yes	Yes	Yes	Yes

*** indicates significance at the 1 percent level.

presents the regression results after incorporating the green area control. The dual-pilot policy passes the 1% significance level test with a positive coefficient, demonstrating the reliability and robustness of the initial regression findings.

(2)

Replacement of Control Variables

Changes in the share of a specific industry can reflect shifts in the shares of other industries. To further investigate the impact of the dual-pilot policy on green innovation, this study replaces the “share of the secondary industry” with the “share of the tertiary industry” as a measure of each city’s industrial structure. After substituting the control variables, the results presented in Table 3 show that the regression coefficients for the dual-pilot policy remain positive and significant, confirming the validity and reliability of the initial regression results.

(3)

Placebo Test

To further minimize the impact of other policies and unmeasurable factors beyond the dual-pilot policy, this study conducts a placebo test by adjusting the policy’s implementation period. Specifically, the time dummy variable for the dual-pilot policy in the treatment group is modified to reflect an earlier start date. Since the dual-pilot policy was first introduced in 2010, following the single-pilot policy by two years (with Shenzhen becoming an innovative pilot city in 2008), the policy period is adjusted to begin one year earlier, effectively setting it three years in advance. Table 3 displays the regression results for the dual-pilot strategy’s effect on green innovation after this adjustment. The findings indicate that the dual-pilot policy does not pass the 10% significance level test, confirming that changes in urban green innovation are indeed attributable to the dual-pilot policy. This result suggests that confounding factors other than the dual-pilot policy are unlikely to foster urban green innovation, underscoring the robustness and reliability of the initial findings.

(4)

Controlling for time fixed effects

Given that each city in the study area may be influenced by external factors such as macroeconomic conditions and technological progress in different years, this study employs time fixed effects to account for the impact of unobservable exogenous shocks due to temporal changes. The regression results presented in Table 3 show a positive and significant coefficient for the dual-pilot policy, confirming that the original regression findings are robust and reliable.

4.4. Impact Mechanism Test

This study begins with a stepwise technique test to evaluate the impact of the dual-pilot policy on green innovation. The findings are shown in

Table 4. Results of the stepwise test of the impact mechanism of the dual-pilot policy on green innovation.

	Regional Economy		Financial Level		Employment		Education Level
	(1)	(2)	(3)	(4)	(5)	(6)	(7)	(8)
M		3.41 × 10−5 ***		2.570 × 10−5 ***		2.169 ***		−5.079 × 10−3 ***
		(0.000)		(0.000)		(0.000)		(0.000)
X	3.399 × 106 ***	2.407 × 102 ***	1.17 × 107 ***	5.563 × 10	2.229 × 10 ***	3.084 × 102 ***	2.427 × 104 ***	4.800 × 102 ***
	(0.000)	(0.000)	(0.000)	(0.360)	(0.000)	(0.000)	(0.000)	(0.000)
C1	1.211 × 105 ***	−9.217 ***	−2.133 × 105 ***	4.070 × 10−1	2.485 × 10−1 ***	−5.623 ***	1.276 × 102 *	−4.436 **
	(0.000)	(0.000)	(0.000)	(0.798)	(0.000)	(0.002)	(0.099)	(0.011)
C2	2.128 ***	7.8 × 10−5 ***	4.176 ***	4.310 × 10−5 ***	2.490 × 10−6 ***	1.452 × 10−4 ***	3.196 × 10−3 ***	1.668 × 10−4 ***
	(0.000)	(0.000)	(0.000)	(0.000)	(0.000)	(0.000)	(0.000)	(0.000)
C3	1.641 × 10	−2.248 × 10−2 ***	1.482 × 102 *	−2.573 × 10−2 ***	1.194 × 10−3 ***	−2.451 × 10−2 ***	1.710 × 10−2	−2.183 × 10−2 ***
	(0.619)	(0.000)	(0.061)	(0.000)	(0.000)	(0.000)	(0.928)	(0.000)
C4	1.357 × 10 ***	−9.187 × 10−3 ***	−1.047 × 102 ***	−6.030 × 10−3 ***	9.950 × 10−5 ***	−8.508 × 10−3 ***	1.851 × 10−1 ***	−7.784 × 10−3 ***
	(0.003)	(0.000)	(0.000)	(0.000)	(0.000)	(0.000)	(0.000)	(0.000)
cons_	−3.710 × 106 ***	9.084 × 10	6.330 × 106 ***	−1.816 × 102 **	2.365 × 10 ***	−6.999 × 10	4.302 × 104 ***	1.998 × 102 **
	(0.000)	(0.235)	(0.000)	(0.029)	(0.000)	(0.456)	(0.000)	(0.031)
sigma_u	5,599,941.9	655.54362	13,368,655	544.80772	46.294316	662.83906	118,871.67	605.05967
sigma_e	5,619,074.9	717.27847	13,472,870	656.51616	26.555194	740.24603	32,083.142	724.37623

Note: p-statistics are in parentheses; *, ** and *** indicate significance at the 10 per cent, 5 per cent and 1 per cent levels, respectively.

. The regression findings of the policy’s influence on the local economy are shown in Table 4’s Columns 1 and 2. The results show that the estimated coefficient of the dual-pilot policy on the regional economy is positive and passes the 1% significance level. Additionally, at the 1% significance level, the calculated coefficients of the dual-pilot policy and the regional economy on the impact of green innovation are positive. This implies that the dual-pilot policy has a direct impact and that it can promote the growth of green innovation by raising the level of the regional economy. The regression findings of the policy’s effect on the financial level are shown in Table 4’s columns 3 and 4. The findings demonstrate that the financial level regression coefficient for the dual-pilot policy is positive at the 1% significance level, as is the financial level regression coefficient for green innovation. Nevertheless, the effect mechanism at the financial level is valid since the regression coefficient of the dual-pilot policy is not significant. By raising the financial bar, it appears that the dual-pilot policy can encourage the growth of green innovation. Columns 5 and 6 of Table 4 provide the regression findings that describe the employment effect mechanism. The calculated employment coefficient of the dual-pilot policy passes the 1% significance test and is positive. This shows that the dual-pilot policy has a direct impact on employment and can raise the city’s level of green innovation. The influence mechanism of education level’s regression findings is shown in Table 4, columns 7 and 8. At the 1% significance level, the dual-pilot policy’s regression coefficient on the degree of schooling is positive. Nonetheless, the education level’s regression coefficients are negative, suggesting a lag in the adoption of green innovation through educational initiatives. As a result, by raising educational standards, the dual-pilot policy promotes green innovation both directly and indirectly. The dual-pilot policy’s direct effect is 442.322, whereas the education level’s indirect effect is −152.38, indicating that the policy’s direct effect outweighs the education level’s indirect effect.

4.5. Analysis of Regional Heterogeneity

Given the broad range of sample sizes, the variability in other policies enacted within each city, and differences in resource endowments, industrial structures, and talent intensity, analyzing regional heterogeneity is crucial for a comprehensive understanding of how the dual-pilot policy impacts green innovation across different regions. This study conducted a sub-sample regression of 284 cities above the prefecture level in China, categorized into eastern, central, and western regions, to explore the regional characteristics of the dual-pilot policy’s effects on urban green innovation.

As shown in

Table 5. Analysis of regional heterogeneity in dual-pilot policy.

Variables	Eastern Region	Central Region	Western Region
X	3.409 × 102 **	4.461 × 102 ***	2.636 × 102 ***
	(0.016)	(0.000)	(0.000)
_cons	1.579 × 10	1.531 *	8.536 × 10−1 ***
	(0.962)	(0.982)	(0.990)
control variables	Yes	Yes	Yes
fixed effect	Yes	Yes	Yes

Note: p-statistics are in parentheses; *, ** and *** indicate significance at the 10 per cent, 5 per cent and 1 per cent levels, respectively.

, the regression results indicate that the dual-pilot policy positively and significantly influences urban green innovation in all three regions—east, central, and west. However, the impact varies: the central region experiences the greatest effect, followed by the eastern region, with the western region showing the smallest effect. This variation may be attributed to the central region’s robust industrial foundation and its industrial structure being more conducive to the dual-pilot policy’s implementation. The central region’s economic development goals include upgrading innovation capacity, aligning with the dual-pilot policy’s objectives. Conversely, the western region, with a weaker economic base and less advanced industrial and technological infrastructure, experiences a smaller impact. This region’s industrial structure is less suited to the dual-pilot policy’s requirements compared to the eastern and central regions.

5. Individual Policy Tests

5.1. Tests of the Impact of Individual Policy in Their Respective Dimensions

Prior empirical testing has already shown us that the dual-pilot policy greatly increases urban green innovation. We separately investigate the influence of the low-carbon pilot cities on urban green development and the impact of the innovative city policy on urban innovation development in order to further support the appropriateness of choosing both policies.

(1)

Testing the impact of innovative city policy on innovation development

The findings of the innovative city policy’s regression outcomes and parallel trend test results on each city’s innovation development are shown accordingly in Figure 4 and

Table 6. Benchmark regression results of innovative city policies on innovation development.

Variables	Coefficients	Variables	Coefficients
X	8.696 ***	C3	−1.337 × 10−4
	(0.000)		(0.114)
C1	−1.415 × 10−1 ***	C4	1.254 × 10−4 ***
	(0.000)		(0.000)
C2	1.08 × 10−6 ***	cons_	3.625 *
	(0.000)		(0.070)
sigma_u	16.490122	sigma_e	14.468223

Note: p-statistics are in parentheses; * and *** indicate significance at the 10 per cent and 1 per cent levels, respectively.

. The effect coefficients generally exhibit an upward trend, and the regression coefficients are positive and significant following the adoption of innovative city policies, according to the test findings presented in Figure 4. The regression results from Table 6 reveal that the regression coefficient of innovative city policy on innovation development is positive and significant at the 1% level. These findings indicate that innovative city policy positively influences city innovation development, thereby promoting each city’s innovation development.

(2)

Testing the impact of a low-carbon pilot city on green development

The findings of the regression analysis and the parallel trend test of the low-carbon pilot city policy’s effects on each city’s green development are shown in Figure 5 and

Table 7. Benchmark regression results of low-carbon pilot city policies for green development.

Variables	Coefficients	Variables	Coefficients
X	1.360 × 105 ***	C3	−2.045
	(0.000)		(0.610)
C1	8.472 × 103 ***	C4	9.055 ***
	(0.000)		(0.000)
C2	1.025 × 10−2 ***	cons_	−8.246 × 104
	(0.000)		(0.457)
sigma_u	1,191,965.1	sigma_e	686,156.11

Note: p-statistics are in parentheses; *** indicate significance at the 1 per cent level.

, respectively. The low-carbon pilot city policy did not have a substantial impact on green development in the year that followed its introduction, according to the test findings shown in Figure 5. Nonetheless, the low-carbon pilot city policy’s regression coefficient is noticeably positive and rising four years after it was implemented. The regression analysis in Table 7 demonstrates a positive and significant regression coefficient of the low-carbon pilot city strategy on green development at a 1% level. These results imply that each city’s green growth may be further enhanced by a low-carbon pilot city policy.

5.2. Non-Superposition Test for Individual Policy Implementation

Given that cities are not simultaneously influenced by the policies of innovative cities and low-carbon pilot cities, a city may be affected by only one of these policies at a time, resulting in a change in green innovation. To eliminate this effect and prove that green innovation can only significantly change when affected by both innovative city and low-carbon pilot city policies, this study adjusts the timing of policy shocks for the cities in the study to when they were first affected by either the innovative city policy or the low-carbon pilot city policy. This modification permits a test that does not overlap. The findings of the single pilot policy’s parallel trend test on the growth of urban green innovation are shown in Figure 6. The results from the figure show that the parallel trend test fails, indicating that green innovation is significantly enhanced only when it is affected by both policies simultaneously.

5.3. Testing Individual Policies for Green Innovation

To confirm that green innovation results from the combined influence of both innovative city and low-carbon pilot city policy, rather than either policy in isolation, this section investigates the individual impacts of innovative cities and low-carbon pilot city policy on urban green innovation:

(1)

Innovative cities’ test for green innovation

The cities sampled in the study were divided into two categories: those affected by innovative city policies and those that were not. These categories were then tested for parallel trends. The results of this parallel trend test are shown in Figure 7, examining the impact of innovative city policies on each city’s green innovation development. The results show that, with a confidence interval that includes 0, the estimated coefficients of innovative city policy on urban green innovation first decline and subsequently rise when these policies are put into practice. The failure of the parallel trend test indicates that the innovative city policy initiatives do not substantially advance the growth of green innovation in urban areas.

(2)

Low-carbon pilot cities as a test for green innovation

In order to ascertain if the low-carbon pilot city policy has a beneficial impact on cities’ development of green innovation, this study divides the sample cities into two groups: those that are impacted by the policy and those that are not. The findings of the parallel trend test of the policy’s influence on the growth of green innovation in each city are shown in Figure 8. The test findings show that there is no consistent pattern of rise in the estimated coefficients of the low-carbon pilot city policy on urban green innovation after the policies are put into place. Moreover, the parallel trend test is not fulfilled since their confidence intervals encompass zero. As a result, the low-carbon pilot city policy has little effect on the growth of green innovation in the city.

6. Conclusions

6.1. Main Findings

Using the number of green patents as the explanatory variable, this study selects panel data from 284 prefecture-level cities in China from 2003 to 2020 to examine the effects of these two policies on green innovation. Initially, we evaluate and assess the impact of the dual-pilot policy on the growth of urban green innovation using a multi-period DID model. Secondly, we employ a systematic approach to examine the mechanism by which the dual-pilot policy influences green innovation. Finally, in order to further improve the dependability, we examine the effects of a single-pilot policy on innovation and green development, the non-overlapping effect of a single policy on green innovation, and a single policy on green innovation. Following this method of empirical investigation, we specifically conclude as follows:

The dual-pilot policy effectively promoted the development of green innovation in the pilot cities, a finding supported by several tests. Among the control variables, the improvement of industrial structure, land use, and living standards hindered green innovation, while the improvement of the market environment had a facilitating effect.

(1)

In terms of impact mechanisms, the stepwise test reveals that the dual-pilot policy can foster green innovation in cities via four channels: enhancing the regional economy, financial level, employment level, and educational level in the city.

(2)

A supplementary test of individual pilot policies proved that innovative cities and low-carbon pilot city policies, correspondingly, successfully encouraged green growth and innovation in cities, supporting the validity of the policy selection process. Further evidence that only the execution of dual-pilot policy may considerably advance the development of urban green innovation was obtained by evaluating non-overlapping and single-pilot policy on the subject of urban green innovation.

6.2. Discussion

Recent research on policy pilots and green innovation is progressively expanding. Beyond examining how different policy pilots impact green innovation, some studies have found that ESG reports can enhance green innovation practices by fostering competitive advantages within the industry [96]. The effectiveness of this practice is closely related to technological orientation and innovation capabilities [97]. Additionally, ongoing improvements in the green innovation chain contribute to solving environmental issues and promoting sustainable development [98]. When integrated with green human resource management, these improvements can positively influence environmental performance, ultimately supporting the goal of carbon neutrality [99].

In three respects, this study adds to the body of literature.

First, unlike previous studies that concentrated on the impact of a single pilot policy on urban green innovation, this study simultaneously examines both the low-carbon pilot city policy and the innovative city pilot policy as dual pilot initiatives. This approach overcomes the limitations of single-policy analysis typically found in traditional policy effect studies.

Second, in terms of research design, this study goes a step further to ensure that the observed improvement in urban green innovation is indeed the result of the combined effect of the dual pilot policies. To this end, the study not only tests the impact of each policy individually on green innovation but also includes a non-superposition test of the policies. This approach enhances the reliability and accuracy of the findings and can serve as a reference for future research on the effects of multiple policies.

Third, extending the analysis of the dual-pilot policy’s impact on urban green innovation, this study delves into the underlying mechanisms across four dimensions: regional economy, financial level, employment level, and education level. This comprehensive approach enhances the theoretical understanding of green innovation and contributes to the broader literature on the subject.

The findings of this study offer valuable insights for other developing countries, particularly regarding the promotion of a green transition through policy design. First, when formulating green development policies, it is crucial to adopt a holistic approach that considers multiple socioeconomic factors rather than focusing on a single policy objective. Second, developing countries can benefit from the experiences of pilot policies to understand how to effectively foster innovation within complex socio-economic contexts. Moreover, research indicates that various factors—including regional economies, financial level, and employment and education level—play a significant role in green innovation. Therefore, in their pursuit of green development, developing countries should integrate these socio-economic factors to construct a robust policy framework that supports green innovation.

6.3. Policy Recommendations

Our empirical findings and conclusions suggest several development proposals to effectively foster green innovation in China’s cities. These proposals, intended to guide policymakers, are derived from three key areas: the nature of green innovation itself, the control variables, and the transmission mechanisms.

First, the advancement of green innovation in cities necessitates dependable policy support and financial contributions. The government should actively encourage innovative and green development in cities, while also ensuring that both aspects develop in harmony. Since the start of its economic reforms, ecological and environmental circumstances have suffered as a result of China’s economic expansion, leading directly to more severe environmental and ecological issues. Given this environmental context, the government should implement relevant green development policies, broaden the application of the low-carbon pilot city policy, enhance environmental regulation, and subsequently improve the efficiency of green development as a means to foster green innovation. The first stage in promoting creative growth in the city is to aggressively expand the application of the innovative city policy, with the goal of promoting it widely. Next, it is crucial to increase financial investment in talent cultivation and technological R&D support. Concurrently, we should enhance talent attraction in the innovative city while nurturing innovative talents, leading to the aggregation of other elements that collectively drive the city’s innovative development. Ultimately, we need to expedite the digital development of the city, strengthen the internet infrastructure, and provide a strong basis for creative expansion. In summary, it is recommended that the government coordinate the execution of the dual-pilot strategy, promote technological innovation in urban areas via the lens of green development, and amplify green development with the energy of urban innovation. This approach will ultimately enable the mutual enhancement and influence of urban innovation and green development, jointly facilitating the city’s innovative green development.

Second, it is imperative to rationally manage land use, optimize the industrial structure, improve the market environment, and hasten the establishment of a low-carbon, green economy development system. First, because China’s economic development is in a pivotal transition and upgrading phase, acquiring the advantage of industrial transformation is crucial in order to promote the streamlining and modernization of the industrial framework. This can be achieved by integrating industrialization with manufacturing and service industries and by vigorously advancing China’s information technology with the aid of high-quality and new technology. Second, with the national innovation-driven and sustainable development plan in place, the complexity of urban land use issues is increasingly evident. In response, the government should expand the green use of urban land, refine the land use legal system, and establish a comprehensive green city construction system. Third, a regulatory service system for the market environment should be established, relevant institutional reform measures implemented, and green consumption demand stimulated to invigorate green innovation and development. Fourth, the construction of China’s low-carbon economic development system can be promoted in two ways. First, the government should encourage the green transformation, upgrading, and development of enterprises across various industries to foster the green upgrading of China’s industry and establish a green supply chain. Second, the government should enhance the green standard system and intensify law enforcement supervision to ensure the implementation of green development-related policies, thereby creating a conducive atmosphere for the development of a low-carbon economy. Concurrently, it should also stimulate the innovation and development of social green technology and provide robust financial support.

Finally, we should promote the coordinated development of regional economies, raise the level of green financial development and social employment, strengthen ideological and political education regarding green and low-carbon development, and cultivate relevant professionals. (1) Developing countries should support and guide the clustering development of green industries through rational regional economic layout. The government can set up green industrial parks to focus on the development of low-carbon technology and environmental protection industries and form a green innovation ecosystem in the region. At the same time, the government can improve the regional policy system and actively implement it, making full use of the comparative advantages and factor endowment conditions of each city region and promoting various types of factor flows to maximize the use of resources. (2) They should also increase financial support for green innovation, establish a green financial system, and encourage banks, funds, and other financial institutions to increase investment in green technology research, development, and application. The government can reduce the financing cost of green innovation projects through green bonds, green loans and tax incentives to attract more social capital into the green field. (3) The government can carry out ecological infrastructure construction and create more jobs to raise the level of social employment. In addition, it can also cultivate technical talents adapted to the needs of the green industry through vocational training and skills upgrading programs, especially encouraging young people and the unemployed to enter the field of green technology and environmental protection, so as to provide sufficient human resources for green innovation. (4) Additionally, the concept of green development should be integrated into the education system and improvements should be made in levels of education. The government should encourage universities to cooperate with enterprises and scientific research institutions to establish an industry-university-research cooperation mechanism for green innovation, and universities should popularize the knowledge and education of green development, cultivate scientific research talents with the concept of green development, and strengthen the reserve of talents for the development of green innovation in China.

6.4. Limitations

This study has two limitations. First, while it examines the influence of innovative city policies and low-carbon pilot city policies on urban green innovation, it does not account for other policies or events that could impact urban development. These include city cluster policies, smart city initiatives, and the introduction of high-speed rail, among other external factors that contribute to urban growth. Future research should consider expanding the range of factors or policies that influence urban development to provide a more comprehensive analysis. Second, the research focuses on cities as a whole to assess the impact of the dual-pilot policy, meaning the findings are generalized at the city level. This approach overlooks the specific effects of the dual-pilot policy on individual cities and the varying degrees of impact. Future studies should concentrate on typical cities to explore potential differences between individual city outcomes and aggregate results.