The sample interval was 2006–2016, and the paper covered 30 provinces (including municipalities directly under the Central Government and autonomous regions) (excluding data from Hong Kong, Macao, Taiwan, and Tibet Autonomous Region). Definition of Main Variables and their data sources were related to the statistical yearbooks of previous years (Table 1).

(1)GTI. Since the number of green invention patent applications (GIPA) can be time-sensitive to directly examine green technology innovation activities of enterprises, GIPA was adopted as the main characterization indicator of GTI (Bai et al., 2019). Specifically, according to the “Green List of International Patent Classifications” launched in 2010, we used patent classification numbers to search for GIPAs and calculate the total number of green patent applications per year by region.

Drawing from the research of Peng and Yuan (2018), this study employed the min-max standardization method to calculate the composite index of ERs and set Model 5) for calculation. E R s i j = E R S M i j − min ( E R S M i ) max ( E R S M i ) − min ( E R S M i ) + E R R I i j − min ( E R R I i ) max ( E R R I i ) − min ( E R R I i ) ,   i = 1,2 , … , 11 ;   j = 1,2 , …   3 (5) where ERs _ij is the composite index of ERs in the jth province in year i; ERSM _ij and ERRI _ij are the proportion of ERs investment amount and the average income of ERs in the jth province in year i, respectively. max(ERSM _i ) and min(ERSM _i ) denote the maximum and minimum values of the proportion of ERs investment amount in each province of the country in year i, respectively; max(ERRI _i ) and min(ERRI _i ) denote the maximum and minimum values of the average ERs income of each province in the country in year i, respectively.

Here, the SDM is selected as the optimal choice by combining the Lagrange Multiplier (LM) and Wald tests, and the results of the Hausman test are used to select the fixed effects estimation results and set the time fixed model. As shown in Table 2, Models (1)–3) are the results of the maximum likelihood estimation of the SDM with the inclusion of the cross terms of GC, ERs, and GC in turn.

As can be seen from Table 2, ERs hinder GTI in China’s manufacturing industry. The analysis of the coefficient estimates for the explanatory variables shows that the regression coefficient of ERs is significantly negative under the three models, indicating that there is a significant negative relationship between the intensity of ERs and GTI in manufacturing in the region. This implies that all provinces in the country are in the primary stage of cost saving, with weak innovation support, such as technological improvement and optimization of management models. Innovation compensation can hardly compensate for the high production costs caused by ERs, and supporting funds and policies also lead to the effect of industrial GTI transformation being difficult to show. Thus, it is difficult to offset the negative impact of ERs on the crowding-out effect of innovation input.

Second, Table 2 shows that GC policy could play the “Porter effect” and positively promote GTI in China’s manufacturing industry. From the estimated coefficients of the variables in Model (2), the incentive effect of GC for technological innovation can be effectively brought into play in the policy context of green finance booming. The reason may be that, by the type of enterprise, high energy-consuming enterprises such as “two high and one leftover” face financing constraints, which need to be eliminated, or they are prompted to improve the efficiency of GTI and adjust their industrial structure to reduce undesirable output (Hsu et al., 2014). Simultaneously, enterprises in the clean industry or promoting green projects are the first to develop environmentally friendly technologies and products with the financial support of GC and achieve value-added business benefits with the “first-mover advantage” achieved in market competition, thereby generating an “innovation compensation effect.”

We have verified the role of ERs and GC for GTI in manufacturing based on independent perspectives. Next, this paper explores the role of the combination of GC and ERs on GTI from the perspective of coordination and cooperation. Table 2 shows the regression results of ERs and GC. Among them, model 3) contains the regression results of the interaction term with the inclusion of GC and ERs. The main effect and spatial lag coefficients of the interaction term, ERs*GC, are negative, indicating that the combination of GC development and ERs inhibits GTI behavior in manufacturing. These results suggest the negative GC effect as a new type of environmental governance instrument in combination with traditional ERs instruments in terms of driving effects on technological innovation. The reason may be that, after a certain level of ERs intensity, the combination of GC and ERs exerts a greater negative effect on GTI. High-intensity ERs contributed to a rapid increase in production costs for companies in a short time, leading to great social and economic pressure on these companies and hindering the development of GTI. Although GC provides financial support to green enterprises and projects, the interaction term between GC and ERs is negative because GTI is characterized by high risks and long-term lags, and the economic benefits to enterprises are highly uncertain. Therefore, GC does not significantly moderate this negative effect.

The previous section outlines the verification of the applicability of the empirical analysis using time-fixed effects SDM, but we consider possible errors in the spillover effects using point estimation tests, and this section highlights our use of partial differential methods to estimate the direct, indirect, and total effects of ERs and GC on technological innovation (Table 3). First, from the spatial decomposition term, the effect of ERs on local and neighboring GTI is significantly positive, and the technology innovation due to the strengthening of ERs may have had the same promotion effect on other regions (Qu, 2018).

As can be seen from Table 3, there is a positive spillover effect of ERs on GTI in manufacturing in surrounding areas. First, according to the “pollution refuge hypothesis”, if the intensity of ERs in the region increases, polluters will choose to move out of the region because of the rising cost. However, for other regions, the relaxed regulatory policies afford the region a comparative cost advantage, and polluting industries move in. Under the effect of the transfer of polluting industries, the industrial structure of the transferred areas gradually tends toward a lower level. Under the “race to the bottom” effect of local governments, there are ERs policy games and GDP competition in each region. When a certain location implements a more stringent environmental access policy, other regional governments adopt the “race to the bottom” strategy of ERs to achieve GDP growth and attract the inflow of resources by not raising or lowering environmental standards, thereby inhibiting industrial restructuring (Wheeler, 2001). In addition, certain studies have pointed out that in the context of high-quality economic development, the central environmental protection inspectors require the upgrading of ERs policies in each region, and the current competition model between governments is more of a “race to the top” model, where there is a demonstration learning effect when a certain location upgrades their ERs intensity (Holzinger and Sommerer, 2011). However, recently, the central government has abandoned the past “GDP-only” performance appraisal system and fully incorporated green development indicators, which, to an extent, has promoted the GTI process.

Recently, driven by policies such as ecological civilization construction and high-quality urban development, intergovernmental competition in pursuit of mobility elements has gradually weakened, and the concept of green development has increasingly become an important part of the assessment. Technology-intensive industries have become the driving force of regional development and exert a pivotal impact on economic development, government financial growth, the attraction of investment, raising the share of green innovation inputs, and evolving the industrial structure toward cleanliness to an extent. Our model results may provide theoretical implications for designing effective environmental regulatory policies and avoiding inefficiency losses from government intervention.

Second, Table 3 shows that GC exerted significant positive direct effects and negative spillover effects on manufacturing technology innovation. GC development improves the mismatch between the cost and benefit of GTI through the transmission mechanisms of the “financial support effect,” “capital allocation effect,” and “risk diversification effect” and promotes the transformation of the local manufacturing industry into green and clean innovation by achieving technological progress. A significant negative spillover relationship exists between GC and GTI in neighboring regions, mainly because the scarcity property of GC, as a financial resource, leads to an increase in credit funds in the region accompanied by a decrease in credit funds in other regions, which reinforces the status quo of interregional competition for resource elements.

China is a vast country, and there are significant differences in resource endowments, economic development levels, and historical and cultural factors between coastal and inland regions. To further investigate regional ERs, GC, and their heterogeneous effects on technological innovation, we divide the sample into the two aforementioned regions and conduct an empirical analysis based on the geographical adjacency matrix (Tables 4, and 5).

The study shows that the local and spatial spillover effects of ERs on GTI in coastal areas are significantly negative, which in turn undermines the GTI process in neighboring areas. If there is no reasonable environmental compensation mechanism among local governments, it is difficult to bridge the benefit gap caused by the governance costs and opportunity costs paid by each region for green transformation, which in turn restricts the equity of regional manufacturing innovation development. The impact of ERs in inland areas on GTI is significantly negative, and the impact on GTI in surrounding areas is insignificant, which is mainly constrained by resource endowment, industrial structure, and institutional culture, thereby hindering the stimulation of the dynamics of regional GTI activities.

GC in coastal areas exerts a positive effect on green technologies in the region and a negative effect on GTI in neighboring regions. At the firm level, this can be explained by the fact that the strict credit granting policy of GC, by setting environmental access thresholds, induces heavily polluting firms to focus on investment in production factors at the end of pollution reduction, which in turn inhibits local GTI activities. As GC resources become the focus of competition between regions, the abundance of local GC resources further exacerbates the innovation financing dilemma of neighboring regions, thereby inhibiting the GTI level in neighboring regions (Zhou et al., 2021). The local effect of GC on GTI in inland regions is significantly positive, while the spatial spillover effect does not pass the significance test, indicating that the competition for GC resources in inland regions does not intensify. The reason for this phenomenon is that coastal regions have advantages in terms of economic conditions, market system formations, and technological innovation resource reserves. In particular, the pace of economic green transformation has always been at the forefront for China, with more complete ERs policy measures and mature pollution control experience. Therefore, GC can be combined with traditional ERs instruments to create a gaining effect of regional technological innovation activities. However, the impact of the cross-sectional term of ERs and GC on local technological innovation in inland regions is insignificant, indicating that GC policies and traditional ERs instruments in inland regions have not yet produced incentive-matching effects.

To better verify the mechanism of action between ERs, GC, and GTI, we use the following methods to test the robustness of the empirical results and the results are reported in Table 6.

Considering the inteslrraction of ERs strategies and the spillover effects of GTI levels occurring in neighboring regions, as well as multiple factors, geographical and cognitive proximity also exert an impact. Therefore, here, the economic spatial weight matrix is used instead of the 0–1 geographical neighborhood weight matrix, and the results of the study do not show a significant change in the sign and significance of the regression coefficients of the core variables, confirming that our findings are extremely robust.

By incorporating both environmental regulation and green credit policies into the analytical framework of GTI, this paper empirically tests the joint effect of GC and ERs on GTI. Based on inter-regional interaction strategies and differences, this paper also explores the local effects and spatial spillover effects of environmental regulation and green credit at the regional level, and empirically examines the uneven locational characteristics of the technological innovation effects of ERs and GC. The main findings of this paper include the following: 1) the local impact and spatial spillover effect of ERs on GTI in manufacturing industries at the national level is mainly reflected in the inhibitory effect. The negative effect of “following cost” is greater than the “compensation effect” of technological innovation; that is, environmental regulation policies that raise firms’ production costs and lack economic incentives are ineffective in promoting innovation. GC policies can stimulate the transformational development of GTI in local manufacturing industries, but the scarcity of resources weakens the innovation level of green technology in manufacturing industries in neighboring provinces through spatial spillover effects. 2) The combination of GC development and ERs can produce synergistic and complementary incentive-coordinated technology innovation driving effects. The complementary effects of GC fund supply and product, process, and emission reduction are significant, making the economic performance of ERs for technology innovation prominent. Considering the ecological environment has quasi-public goods characteristics, GTI is a complex systemic project, and green finance, as a new market-based instrument for ecological and environmental governance, can play a complementary role with traditional environmental regulatory systems in a more effective, equitable and sustainable manner. However, it is difficult to promote green innovation transformation by purely relying on the regulation means of GC. The traditional ERs means are fundamental in stimulating technology innovation despite the hard constraint of commercial banks and emission enterprises (Feng and Liang, 2022). 3) The eastern coastal regions with strong innovation vitality and endogenous dynamics. For inland regions with little experience in environmental governance and a weak innovation base, a “circular-coordination” mechanism for GC resources should be constructed. By region, there are regional differences in the technological innovation effects of ERs and GC. The local effect of ERs on GTI in coastal areas is significantly positive; that is, strong innovation vitality and endogenous dynamics of coastal regions enable environmental regulation measures and green financial development produces incentive-matching effects. Although environmental regulation measures for inland areas with little experience in environmental governance and a weak innovation base are significantly negative for local GTI and insignificant for neighboring GTI, GC effectively stimulates the dynamic role of local GTI activities and exerts a significant marginal effect on the financial dilemma of local manufacturing investment in research and development (R&D) green technology.

The following shortcomings exist for this study: 1) It is appropriate to use prefecture-level city data to explore the spatial effects of ERs, GC, and GTI in manufacturing based on regional scales. Given the availability of GC-related data, provincial panel data, and the large error in measuring GC variables by indirect methods, there are limitations in the generalizability of the study findings. 2) The “Porter hypothesis” and the regulatory role of GC are closely related to the type of ERs, which include formal ERs instruments (command and control and market incentives) and informal ERs instruments (information disclosure, public participation, and voluntary regulation) (Zhou et al., 2022). However, the influence factors considered in this study are relatively single, and different types of ERs tools can be included in the same model in future studies to examine their differential impacts on GTI in manufacturing. 3) Considering that the spatial spillover effect follows the law of distance decay, there is a certain bias in the empirical test based on the entire domain only, and the local spatial matrix of different distance ranges should be set in the future to explore the GTI effect.

Based on the empirical findings of our study, the following policy implications about the integration of ERs and green finance policies are proposed accordingly.

First, based on the spatial dependence of interprovincial manufacturing GTI, local governments should establish a good competitive relationship with each other, strengthen strategic interoperability and positive interaction with neighboring regions, and stimulate local green development with the implementation of green innovation and transformation strategies by regional industries holistically.

Second, the central government should abandon the traditional approach in the design of policies, systems, and processes for environmental and social risk management, and coordinate a regionally differentiated GC system instead. GC policies should be made on a “person-by-person” basis; that is the eastern coastal should leverage the role of commercial banks and other financial sectors in promoting the supply of GC funds to meet the financial needs of high-level technological innovation. It is necessary to improve the green policy system, such as the environmental information disclosure of enterprises, information sharing between environmental protection departments and banks, and strengthen the prescreening and post-supervision mechanism to weaken the influence of information asymmetry. For inland regions, local governments can create an innovative atmosphere through financial support, construction of innovative subjects, and strengthening intellectual property protection, so as to construct a cooperative mechanism and innovation-supportive for interregional credit policies (Liu and Nie, 2022) in the place of the interregional competition for resources with a cooperative mechanism for interregional credit resources to support GTI (Liu and Nie, 2022).

Third, as GTI is a complex systemic project, it is difficult to promote green innovation transformation by purely relying on the regulation means of GC. The traditional ERs means are fundamental in stimulating technology innovation despite the hard constraint of commercial banks and emission enterprises (Feng and Liang, 2022). The ERs system can be planned in a unified manner by establishing a coordinating management institution for GTI, GC, and ERs to coordinate green development data, such as scientific and technological innovation, capital loans, and pollution emissions, and make joint efforts from the market investment and financing and government supervision levels (Chen et al., 2022).

Fourth, considering that GTI activities are characterized by high risks, economies of scale, and innovation spillovers (Wicki and Hansen, 2019), the institutional environment for technology innovation is the key to the “Porter effect.” Local governments can create an innovative atmosphere through financial support, construction of innovative subjects, strengthening intellectual property protection, and other basic systems, which in turn attract the gathering of high-end production factors, such as capital, talents, and knowledge. Concomitantly, the GC policy can also be dovetailed with the science and technology policy, incorporate the green R&D investment and performance of enterprises into the environmental policy assessment, and make joint efforts from the market investment and financing and government supervision levels (Chen et al., 2022).