As the concept of CTI has deepened in recent years, many scholars have recognized its significant positive impact on reducing urban CEI. From a theoretical perspective, the Natural Resource-Based View suggests that GTI helps firms gain competitive advantages. It also makes their environmental strategies valuable, scarce, difficult to imitate, and irreplaceable strategic resources (Hu et al., 2023). Recent empirical research has confirmed that GTI promotes high-quality economic development and enhances environmental sustainability across various countries and regions (Wang et al., 2022). For instance, in terms of Wei and Ma (2024), R&D expenditures increase CO₂ emissions in APEC countries, while the number of patents reduces carbon emissions. After the discussion, it was found that GTI can effectively enhance a region’s carbon emission reduction capacity. Moreover, it also has significant spatial spillover effects on neighboring regions’ carbon emission reduction levels. At the international level, some scholars have used the CS-ARDL method to conclude that GTI positively impacts Thailand’s CEI and overall environmental improvement (Xuan et al., 2024). Research on developed countries, including the G7, also highlights that progress in green technology helps these countries achieve carbon emission reduction targets (Hu and Wu, 2022). This emphasizes the interdependence between technological innovation and environmental protection.

However, some research holds an opposing view on the emission reduction effects of GTI. They claim that GTP does not explicitly ensure CO₂ emission reductions. From the perspective of Rennings and Rammer (2011), the market cannot effectively promote GTI. Companies may require sufficient incentives or penalties to drive GTI. Lu and Lu (2024) observed that GTI exhibits an inverted U-shaped impact on the carbon intensity across Chinese provinces from the perspective of energy output growth. Furthermore, Mongo et al. (2021) proposed an indirect rebound effect of GTI. If this effect is significant enough, energy demand growth may outweigh supply-side efficiency gains. This could undermine GTI’s role in reducing CEI. On the international level, Suki et al. (2022) emphasized that while GTI improved the environmental performance of Italian cities, it did not obviously reduce CEI.

Additionally, a study in Malaysia employing the Bootstrapped ARDL method found that GTI significantly increases CEI in the short and long term (Li H et al., 2024). Other scholars contend that the reduction of carbon emissions by GTI occurs in distinct phases. Although GTI positively affects carbon reduction in the short term, its effectiveness diminishes over time and even becomes ineffective (Chen et al., 2023). Relevant research has shown that GTI can effectively reduce CEI, but it is established in developed countries. Besides, Lin and Ma (2022) pointed out a non-linear relationship between GTI and CEI. Significant reductions occur only when the GTI and human capital level reach a specific threshold. Zhu and Zhang (2023) assessed the single-threshold effect of income on GTI effectiveness using a threshold regression model based on data from 71 economies. The results showed that GTI led to significant CEI only when the income level exceeded a specific threshold. Moreover, Wang et al. (2023) found differences in the carbon reduction effects of GTI across sectors. In the construction sector, GTI likely increases CEI, while in the industrial sector, it produces relatively fewer emissions.

Porter (1991) contended that stringent, well-designed environmental regulations can stimulate GTI. This offsets compliance costs, increases firm productivity, and creates a win-win scenario. In this situation, both profitability and emission reduction targets are improved. Public Health Policy provides a new perspective on the impact of environmental regulations. It emphasizes how such regulations influence firms’ innovative behavior and contribute to emission reductions. Numerous studies have empirically tested this hypothesis (Li Q et al., 2024). The Innovation Offsets perspective suggests that under a well-functioning environmental regulatory system, the environmental benefits of resource efficiency improvements can outweigh the costs. These costs are associated with internalizing environmental impacts (Hart et al., 2022). Fragkiadakis and Paroussos (2021) confirmed a positive correlation between the intensity of environmental regulation and innovation output. Their study, which is based on data from the manufacturing sector of 17 European countries, supports this finding. Other research shows that stringent environmental policies, such as carbon taxes on excessive emissions, positively impact environmental sustainability. These policies also lead to increased environmental costs. Moreover, these policies significantly enhance environmental sustainability. It also increases investment in green technologies, further promoting sustainability in the G-7 economies (Aybudak et al., 2025). Chhabra et al. (2023) emphasized that implementing environmental laws, such as carbon pricing or taxation, is crucial for countries heavily reliant on importing non-renewable energy sources to meet consumption needs. Tang et al. (2025) found that the marginal impact of environmental policy synergies on per capita carbon emissions in China exhibits clear structural features at different levels. These synergies effectively lower per capita emissions and contribute to carbon neutrality targets.

However, due to the inherent uncertainty and high capital requirements of innovative activities, some argue that environmental regulation is necessary to reduce CEI (Wang et al., 2025). Research based on Neoclassical Economics argues that ecological regulations increase compliance costs, such as pollution fees. This forces firms to redirect limited resources from innovation to basic emission reduction projects, possibly reducing overall innovation efficiency (Yin et al., 2022). Some research has examined the mechanism of environmental regulation. They conclude that regulation, acting as a driver, can mediate the role of GTI in promoting a low-carbon transition (Wang H. J et al., 2024). Furthermore, Wu and Yu (2025) argued that market-based environmental regulatory instruments, such as IERs, are most effective in promoting green innovation and emission reductions among Chinese firms.

Achieving carbon neutrality through GTI has become the consensus of most countries. The existing literature primarily focuses on two aspects of the relationship between GTI, energy consumption structure, and CEI. On the one hand, it examines the impact of GTI on the structure of energy consumption. Xu et al. (2022) emphasized a non-linear dynamic relationship between GTI and energy efficiency. When innovation levels exceed a critical threshold, CEI will be reduced significantly. From the perspective of Guo et al. (2024), stronger policy support is required when technological innovation levels are low. Once innovation levels improve, policy intensity should be reduced once innovation levels improve to promote green energy transformation. Yang et al. (2023) emphasized that GTI fosters the development of new energy sources. It will contribute to the reduction of fossil fuel consumption. Other scholars argue that GTI, as a key energy-saving mechanism, can guide resource flow from high-pollution, high-energy-consumption industries to low-energy, environmentally friendly industries. The process strongly promotes the energy transition (Cao and Xu, 2023).

Another aspect is that research has also focused on the impact of energy consumption structure on CEI. Zhang and Xu (2022) constructed an energy security index system and found that improving energy security helps decouple carbon emissions in the Yellow River Basin. Liu H. Z. et al. (2022) stated that an unreasonable energy consumption structure inhibits the low-carbon economic transformation regionally and in neighboring areas. Some scholars argue that reducing the cost of clean energy and promoting energy structure reform can eliminate dependence on traditional sources (Chen L et al., 2024). This paves the way for developing and using clean energy, such as solar energy. It also encourages the gradual replacement of high-carbon sources with low-carbon alternatives, thus reducing CEI.

In addition, some scholars have analyzed the impact of related policies and argued that energy quota trading has a dual green effect. It reduces carbon inequality within and between cities. The effect is more pronounced in national low-carbon pilot cities than in non-pilot cities (Haowei, 2025). Some scholars have examined the new energy city demonstration policy from constraint and incentive perspectives. They found that the carbon reduction effect of the NECDP becomes significant only after the second year, indicating a time lag (Che et al., 2023). Besides, some studies emphasize the potential complexity of GTI. Gu et al. (2019) argued that GTI improves energy efficiency, but its marginal effect gradually diminishes. Moreover, economic growth stimulates higher energy demand. This increases pollutant emissions in production processes and poses greater climate change risks. Some scholars have found that GTI may encourage residents to use energy-efficient products more frequently by reducing energy consumption costs. It will lead to increased CEI in turn (Pu et al., 2022).

In summary, existing research has examined the impact of CTI on CEI from different perspectives, providing a solid theoretical foundation for this study. However, there is still significant room for further exploration. Firstly, regarding the effectiveness of GTI in reducing carbon emissions, significant academic debate remains, and no consensus has been reached. Secondly, most existing research focuses on the effects of carbon emission reduction at the provincial level. However, as cities are the main drivers of economic development and the primary sources of carbon emissions, the carbon reduction mechanisms of GTI at the urban level remain underexplored. Thirdly, in terms of the mechanisms, although some scholars have highlighted the vital role of environmental regulation in the relationship between technological innovation and carbon emissions, most studies focus on its moderating effect. There has been relatively limited discussion on the mediation effect pathways. In addition, while existing research has examined the mediating roles of environmental regulation and energy consumption structure in the impact of GTI on CEI, few studies have integrated these factors into a unified framework. This limitation hinders the exploration of whether they form a chain mediation of GTI-environmental regulation-energy consumption structure-CEI.

Building on the research gaps mentioned above, this study focuses on whether GTI can reduce urban CEI at the city level. It also examines whether environmental regulation and energy consumption structure can play a chain mediation role in achieving urban carbon reduction goals. The potential contributions of this paper are as follows. Firstly, in terms of research methodology, this study differs from traditional single mediation analysis. It reveals that GTI affects CEI through a chain path. This path involves environmental regulation and energy structure transformation, enriching the research perspective. Secondly, existing research has mainly focused on the provincial level, while cities have not been adequately explored as the main battleground for carbon reduction. In the context of increasing resource and environmental constraints, the urban carbon reduction effects of CTI urgently need to be studied. Thirdly, this paper differs from previous research that focuses on the economic impact of environmental regulation. It treats environmental regulation as a policy incentive variable and explores its mediation paths. It suggests that GTI only provides the direction for enterprises. The innovation compensation effect and environmental compliance cost of environmental regulation policies drive this transformation. This study aims to provide references for regulatory institutions to develop low-carbon development strategies suited to China’s actual conditions.

GTI is an innovation seeking to mitigate the conflict between economic expansion and ecological degradation. It is crucial to promote green, low-carbon economic development and achieve the dual-carbon goal (He et al., 2025). This view is supported by the theory of technology spillover, suggesting that green technology produces significant positive spillover effects as a special public good (Rong et al., 2023). On the one hand, applying key technologies such as renewable energy and energy storage can transform urban energy supply from centralized fossil energy systems to distributed clean energy. This shift breaks the traditional dependence on high-carbon pathways. Combined with e-mobility technology, this creates a chain system of emission reduction at the source and control at the end. Consequently, production costs for enterprises will decrease. This will stimulate production motivation, attract broader participation in GTI activities, and generate economies of scale (Colombelli et al., 2025). On the other hand, the potential of GTI to reduce emissions is also supported by Externality Theory. This theory argues that the environmental costs of corporate carbon emissions and the social benefits of green innovation are inadequately reflected in market pricing mechanisms (Hu and Tian, 2024). Similarly, using R&D subsidies, tax incentives, and other measures will encourage enterprises to adopt cleaner and more efficient green technologies. As a result, both capital and human investments in GTI will be increased. This will enhance individual enterprises’ ability to reduce emissions. In addition, a significant spillover effect will be generated, ultimately promoting the industry’s emission reduction process (Zhang et al., 2024).

The impact of GTI on carbon emissions remains controversial at the theoretical level. The Rebound Effect theory suggests that by reducing the cost of green products, GTI may stimulate larger-scale consumption. This may lead to an increase in overall carbon emissions instead of a decrease (Wang Q et al., 2024). According to the time-lag effect theory, it is a lengthy process for GTI to move from research and development to industrial application. This delay makes it challenging to achieve significant emission reductions in the short term (Yang Y. et al., 2023). Additionally, under resource constraints, large-scale investments in GTI may crowd out the improvement of traditional emission reduction technologies. This includes energy efficiency enhancements and other proven emission reduction paths. As a result, overall efficiency may decline (Li C et al., 2024). What is the actual emission reduction effect of GTI in the context of China’s rapid urbanization? Due to theoretical controversies and situational complexity, the exact impact of GTI on urban carbon emission reduction requires empirical testing. Therefore, this paper proposes the following hypothesis:

Hypothesis 1

GTI has a significant effect on urban carbon emissions

Environmental regulation refers to the administrative and legal measures governments use to guide firms in following specific policies. According to Porter’s hypothesis, technological progress reduces the costs of environmental monitoring and pollution control. This makes previously unaffordable or technologically infeasible environmental standards achievable (Song et al., 2022). The application of GTI prompts the government to set carbon emission caps. Consequently, mandatory constraints are imposed on high-emission industries, forcing them to adopt emission reduction measures (Cui et al., 2022). In addition, GTI has expanded the scope of environmental standards. Using tools such as carbon tax systems and green financial subsidies has shifted the regulatory approach towards multi-incentive mechanisms. This has accelerated the internalization of price mechanisms, directed resources toward low-carbon sectors, and reduced per-unit carbon emissions in urban areas (Ai et al., 2023). By utilizing institutional constraints and incentives, environmental regulation motivates market participants to reduce emissions, enhancing carbon reduction effectiveness. From the perspective of resource allocation, strict environmental regulations increase the costs and risks for high-carbon industries. Thus, their marginal returns decline. At the same time, the green technology sector benefits from high returns due to policy support (Zhang et al., 2024). This drives the concentration of factors such as capital, talent, and land in low-carbon industries, accelerating the translation of innovations into the marketplace.

However, the mediating mechanism of environmental regulation is not always effective. As regulators, information asymmetry theory suggests governments often lack sufficient technical and cost information. This makes it difficult to accurately assess firms’ abatement potential and compliance costs, leading to indiscriminate implementation of regulations (Tian et al., 2025). Additionally, environmental regulatory policies encounter practical constraints during implementation, including high costs and resistance from interest groups (Alexander et al., 2024). Such constraints significantly weaken the mediating effect. In the early stage of technological innovation, the lack of maturity and limited market acceptance of green technologies may result in overly advanced policies that are disconnected from reality. Therefore, compliance costs for enterprises increase, creating a compliance burden effect and inhibiting short-term incentives to reduce emissions (Han and Shao, 2022). More critically, behavioral distortion theory suggests that stringent environmental regulations may cause enterprises to develop a regulatory avoidance mindset. This leads them to circumvent regulations through industrial shifts, data falsification, and exploiting policy loopholes, rather than invest in GTI (Yu et al., 2021). Given the academic disagreement on the mediating role of environmental regulation, it is essential to verify whether GTI can reduce urban CEI through this pathway. Consequently, this paper proposes the following hypothesis:

Hypothesis 2

Environmental regulation plays a mediating role in GTI and reducing urban CEI, though its direction remains uncertain.

Resource allocation theory holds that the market serves as a fundamental mechanism for distributing resources. It guides the flow of resources to more efficient and lower-cost areas through price signals (Struben et al., 2020). Based on this theory, GTI can reduce carbon emissions by optimizing the energy consumption structure. On the supply side, GTI adjusts the relative prices of energy sources and reduces the production cost of renewable energy. With the rise of emerging energy supply modes, market choices are becoming more diversified, and competition among enterprises is intensifying. Under the market price mechanism, the cost advantage of clean energy will be transmitted to various industries (Liu et al., 2021). As a result, the development of the energy system towards decarbonization is accelerated. From the demand management perspective, this advantage will be passed to consumers as clean energy costs decrease. It will incentivize them to reduce the use of high-carbon energy. Carbon pricing and other tools will amplify price signals, further guiding consumers to make low-carbon choices (Li et al., 2024). From the perspective of energy efficiency improvement, the energy substitution theory suggests that enterprises can adjust their energy structure. The optimization of energy consumption structure promotes the concentration of energy use in high-efficiency sources, reduces energy losses, improves energy use efficiency, and enhances emission reduction effectiveness (Zhang et al., 2023).

However, significant challenges have been encountered in the practical application of this theoretical expectation. The effectiveness of the mediating role of the energy consumption structure remains a subject of intense debate in existing research. Path-dependence theory emphasizes high sunk costs, technological lock-in, and interest groups’ resistance to the established energy system. Accordingly, the energy consumption structure becomes highly rigid. Breaking free from the established path remains difficult for GTI, even when driven by technological innovation, particularly in the short term (Mi et al., 2021). Traditional high-carbon energy sources maintain a dominant position in the energy consumption structure (Yong et al., 2023). Market failure theory further suggests that information asymmetry, monopoly power, and externalities in the energy market imply that the market mechanism alone may not achieve effective energy structure transformation. In addition, the capacity of technological innovation to optimize the structure may be weakened by market failures (Wang et al., 2020). More importantly, some scholars contend that adjusting the energy consumption structure frequently lags behind GTI. In some cases, this delay may render the mediating role insignificant and even hinder carbon emission reduction (Huang et al., 2022). Given the divergence in existing research on the mediating mechanism of energy consumption structure, it remains to be tested whether GTI can reduce urban carbon emissions through this pathway. Thus, this paper proposes the following hypothesis:

Hypothesis 3

Energy consumption structure plays a mediating role in GTI and reducing urban CEI, though its direction remains uncertain.

The mechanism through which GTI influences urban carbon emission reduction is viewed differently in academia. The supporting view is that GTI can optimize the energy consumption structure and achieve the carbon reduction target. This is achieved through dual regulation: policy guidance and the demonstration effect. Regarding policy guidance, the government provides signals to market players by formulating green innovation development plans and related policies (Chen F et al., 2024). This reduces the cost of technology application and accelerates the market penetration of low-carbon technologies. At the level of the demonstration effect, industrial ecology theory suggests that technological innovation networks and industrial ecosystems incentivize upstream enterprises to provide cleaner raw materials. They also encourage downstream enterprises to offer more environmentally friendly products and services (Chen and Jiang, 2025). Environmentally friendly products and services promote the efficient utilization of energy. Furthermore, the synergistic development of upstream and downstream enterprises is facilitated, forming carbon synergy and emission reduction effects within the industrial chain. According to Porter’s hypothesis, clean energy gains policy advantages as environmental standards gradually rise with technological support. In contrast, high-carbon energy faces elimination pressure. As a result, key market players are forced to invest in high-efficiency, low-carbon clean energy technologies (Gu et al., 2022). As the proportion of clean energy increases, the policy resistance of traditional high-pollution industries weakens. This creates conditions for the implementation of stricter environmental standards. With the support of technology, a positive feedback loop of regulatory reinforcement and structural optimization forms, ultimately achieving carbon reduction.

However, skeptical views challenge the effectiveness of such chained mediating mechanisms. Resource curse theory suggests that local governments may rely more on traditional taxes and employment in resource-rich regions. This can lead to insufficient enforcement of environmental regulations. Therefore, even when available, advanced green technologies may not be effectively translated into strict environmental constraints (Rammer et al., 2017). Neoclassical economics further suggests that excessive environmental regulations may create a crowding-out effect. This effect diverts resources away from technological innovation and weakens firms’ capacity for structural adjustment (Zhang and Fan, 2024). In addition, some scholars using cross-country panel data have found that, in many developing countries, environmental regulations are often reduced to policy instruments. The intensity of these regulations is driven more by political cycles and economic interests than by GTI. More critically, institutional economics theory emphasizes that the effectiveness of the chain mediation mechanism depends heavily on institutional quality. In regions with lower institutional quality, a broken chain may exist among GTI, environmental regulation, and energy restructuring (Zeng et al., 2022). Most existing research focuses on single-path analysis and lacks an in-depth examination of the multiple mediation effect (Liu et al., 2025). As a result, consensus has not been reached regarding whether GTI can upgrade the energy consumption structure and achieve carbon emission reduction through strengthened environmental regulation. This controversy requires clarification through more comprehensive empirical research. Accordingly, this paper proposes the following hypothesis:

Hypothesis 4

(H4). Whether GTI can reduce CEI by strengthening environmental regulation and thus promoting energy structure upgrades remains to be verified.

In summary, the mechanism of GTI on urban CEI is shown in Figure 1.

Research focuses on Chinese cities at the prefecture level and above. However, due to numerous missing values in cities from Qinghai and Xinjiang, only provincial capitals were included in the data analysis. The final research sample consists of 272 cities at the prefecture level in China, excluding alliances and states. The GTI data primarily comes from the China National Intellectual Property Administration (CNIPA). The remaining variables are sourced from the China City Statistical Yearbook, China Energy Statistical Yearbook, China Regional Economic Statistical Yearbook, China Industrial Statistical Yearbook, China Labor Statistical Yearbook, China Environmental Statistical Yearbook, the National Bureau of Statistics of China, and the EPS database.

Before the analysis, the data for certain variables (GTI, DOP, and HE) were log-transformed. Simultaneously, selected variables were winsorized at the 1st and 99th percentiles (CEI and GTI). The risk of systematic bias was minimized due to the low proportion of missing values in the raw data (<1%). Missing values mainly occurred in isolation within a single year. To ensure data completeness, missing values were supplemented using interpolation. Additionally, to assess potential correlation issues between the variables, a VIF test was conducted. The descriptive statistics of the variables are presented in Table 1.

Table 1 shows that the maximum VIF value is 3.61, and the average is 2.24. These values are relatively stable, indicating no multicollinearity in the data. Additionally, the sample data used in this paper is short panel data. A Unit Root Test was conducted on the panel data to prevent pseudo-regression issues caused by a unit root in the data. The results are shown in Table 2. The results show that each variable rejects the unit root hypothesis at the 1% significance level. Therefore, the data is stationary and does not require further processing. Simultaneously, robust standard errors clustered at the city level address the potential heteroskedasticity issue.

Before conducting the benchmark regression, model selection is performed. The results of the Hausman test, shown in Table 3, reveal that the P-value is less than 0.01, rejecting the null hypothesis. Therefore, the panel fixed-effect model is selected for regression. Columns (2) and (3) show that when control variables are added sequentially without fixed effects, the regression coefficients for carbon emission levels are significantly negative at the 1%. This indicates that GTI substantially reduces urban CEI. Columns (4) and (5) include individual and year fixed effects, respectively. The impact of GTI on CEI remains significantly negative and passes the 1% significance level. This finding provides preliminary support for hypothesis H1. This finding aligns with multiple studies (Zhang et al., 2024; Liu et al., 2022), which confirm the significant emission reduction effect of GTI. This effect is realized through recycling, reuse, and waste reduction, which lower resource consumption and carbon emissions during production. Furthermore, this finding confirms Porter’s theoretical expectation, which suggests that technological innovation can create a win-win situation for both the environment and economy (Porter and van der Linde, 1995). However, unlike Miao et al. (2024), which resulted from a short period and limitations in indicator selection, this study uses a more extended time window, effectively avoiding the discrepancies due to differing time methods. This approach effectively mitigates discrepancies in results arising from variations in measurement methodology (Li et al., 2023). It provides more robust empirical evidence for the carbon reduction effect of GTI.

Columns (1) and (2) in Table 4 validate the role of environmental regulation in urban carbon emissions. The results show that the regression coefficient for environmental regulation is 2.213, significantly positive at the 1% level. Meanwhile, the regression coefficient for carbon emissions remains significantly negative at the 1% level. Furthermore, after performing the Bootstrap mediation effect test, the estimated coefficient is −0.663, and the 95% confidence interval excludes 0. This suggests that GTI reduces carbon emissions by enhancing environmental regulation, thereby verifying hypothesis H2. This result aligns with Hu and Xiong (2024), which indicates that GTI can improve local government enforcement of environmental regulations and significantly reduce carbon emissions. It also confirms how environmental regulations incentivize enterprises to reduce emissions by lowering technology costs. However, unlike Ma et al. (2022) and Lu et al. (2023), which focus primarily on the direct impact of environmental regulations on GTI, this paper reverses the trend. It explores how GTI can strengthen the enforcement of environmental regulations, revealing the dynamic role of technological progress in institutional improvement.

The test results on the mediating role of the energy consumption structure are presented in columns (3) and (4) of Table 4. Table 4 shows that the coefficient for GTI’s influence on the energy consumption structure is 0.023, significantly positive at the 1% level. The absolute value of its negative influence on carbon emissions is smaller than the benchmark coefficient of 0.159. This indicates that the energy consumption structure plays a partial mediating role. The estimated coefficient in Bootstrap’s 1000 tests is −0.187, and the 95% confidence interval excludes 0. This indicates that GTI can reduce carbon emissions by optimizing the energy consumption structure, thereby verifying hypothesis H3. These findings align with the study by Guo and Tan. (2024), which suggests that new technologies reduce the cost of clean energy development and improve its economic feasibility and competitiveness. However, unlike Zhao et al. (2025), who examined the impact of green innovation on the energy mix from the perspective of corporate investment, this paper systematically verifies the mediating mechanism at the city level. It quantifies the magnitude of the mediating effect through a Bootstrap test, providing more rigorous empirical evidence for GTI’s indirect emission reduction pathway.

This paper examines the direct effect of GTI on CEI using Model 6 from the SPSS plug-in PROCESS. We conducted a Bootstrap test at the 95% confidence level, with 5000 sampling repetitions. The results are shown in Table 5. The coefficient of environmental regulation on energy consumption structure is 2.272, with a positive significance level. This indicates a significant chain mediation effect between environmental regulation and energy consumption structure. The indirect impact is greater than 0. Moreover, the Bootstrap test results in Table 5 show that the estimated coefficient of GTI is −0.017. The 95% confidence interval excludes 0, indicating that GTI can optimize the energy consumption structure by improving the environmental regulation level, thus reducing CEI. Hypothesis H4 is verified. The chain mediation aligns with Li et al. (2024), which argues that GTI enhances the technical feasibility of environmental regulation. Strengthened environmental regulation, in turn, guides market players to shift to cleaner energy options through the price mechanism (Liu et al., 2020). Unlike Zhang et al. (2024), who focus on the direct emission reduction effect of and Esperon-Rodriguez et al. (2022), who emphasize the independent impacts of environmental regulation and energy structure. This paper demonstrates that GTI achieves emission reductions and efficiency gains through two key pathways: regulatory reinforcement and structural optimization.

Regional variations in GTI capacity arise due to differences in geographic environment, resource type, and economic development. Therefore, the heterogeneity effect of GTI on urban CEI is analyzed based on geographic location, resource type, and urban scale.

With the deepening of the dual-carbon goal and the innovation-driven development strategy, GTI has become a key driving force for urban low-carbon transformation. This paper examines 272 prefecture-level cities in China from 2006 to 2022 based on resource allocation and externality theories. The study applies a panel fixed-effect model and a chain mediation model to investigate the effect and path of CEI in cities empowered by GTI. The following key findings are obtained:

1. GTI can significantly reduce CEI. The conclusion holds even after replacing the explained variable, lagging the core explanatory variable by one period, and using the instrumental variable method.

Based on the above conclusions, the following recommendations are proposed:

Firstly, establishing a mechanism for green innovation entities to enhance the effectiveness of carbon reduction through technology. To achieve the dual-carbon goal, it is necessary to deepen the integration among industry, academia, and research. This includes creating a green technology transfer platform, reducing information barriers to technology diffusion, and promoting the adequate flow of green technologies among various innovation entities. Simultaneously, policy tools such as consumption subsidies and tax reductions should be used to raise consumer awareness and acceptance of green technologies. This would incentivize consumers to choose green products and services, thus creating stable market demand. Furthermore, mechanisms such as carbon trading and tax subsidies should be improved to internalize the positive externalities of GTI, thus enhancing corporate innovation incentives and achieving dual benefits for the environment and society. Additionally, an innovative insurance system needs to be established to address the negative externalities resulting from market failures. By improving risk-sharing mechanisms, the investment risks for innovation entities can be reduced, creating a favorable environment for GTI.

Secondly, environmental regulation policies should be strengthened to facilitate the energy structure and foster new urban carbon reduction activities. The study has shown that environmental regulation ensures the complete realization of GTI’s carbon reduction potential. Therefore, while developing green technologies, the government should strengthen the top-level design of environmental regulations. Based on the economic development level of each city, targeted environmental standards should be designed and effectively implemented. Furthermore, the energy consumption structure is a critical pathway for carbon reduction. Governments at all levels should leverage regional comparative advantages to accelerate energy-saving technology upgrades. They should guide the integration of green technologies into high-energy, high-pollution traditional sectors and drive the energy industry toward green and low-carbon development. Last but not least, the decisive role of the market in resource allocation must be emphasized. For high-carbon industries, strict emission standards and total volume control should be implemented. Market-based tools such as carbon trading and green finance should guide capital flow toward clean energy and low-carbon technologies, achieving synergies between environmental regulation and energy structure optimization and improving emission reduction efficiency.

Thirdly, implementing a region-specific development strategy and shaping a diverse environment for GTI. Given the different stages and characteristics of development in various cities, a new development paradigm should be established. It is characterized by clear features, complementary advantages, and efficient collaboration. Eastern cities should focus on accelerating the aggregation of green technology resources, optimizing energy supply and demand structures, and maximizing the diffusion of innovation outcomes to create regional demonstration effects. For central cities, it is essential to strengthen environmental regulations and leverage geographic proximity to diffuse eastern technologies. Meanwhile, the conversion rate of clean energy technologies should be improved, and cooperation with eastern cities should be enhanced. Although the GTI system in Western cities is relatively underdeveloped, the region is abundant in resources. Hence, the development strategy should focus on exploring renewable energy sources like solar and wind. Moreover, support for green finance, tax incentives, and other related policies should be intensified. The study also found that the carbon reduction effect of GTI is not significant in resource-based, small-scale cities. This indicates a path dependence in fossil-fuel-rich regions with high transformation costs. Therefore, a gradual environmental regulation strategy should be implemented, with transitional periods and phased targets. This approach would avoid the economic impact of one-size-fits-all environmental standards and enable green, low-carbon development.