The best support for this study we found the theories that are related to the Green Economy and Ecological Modernization. Technology innovation and systemic changes in economic practices promote economic growth while protecting the environment. In this way, environmental performance can also be brought into consideration in the context of green finance and is therefore aligned with this notion. By adopting green technologies and focusing on environmental factors, sustainable development can be achieved. This study is also supported by the Green Economy theory which prioritizes ecological stability and resource conversion Daly (1970) which is based on the notion that natural assets continue to meet the resources and environmental services that are essential to our health and wellbeing while promoting economic growth and development. Due to this notion, researchers emphasize the importance of green finance and its role in transforming economies into greener trends. Likewise, the TBL theory by Høgevold et al. (2019) and Sholeh et al. (2020) explains that apart from economic focus, businesses should also prefer to focus on social and environmental factors to ascertain sustainability. The theory of green finance also suggests that green initiatives would be very impactful to achieve environmental sustainability (Wang et al., 2022a). Our paper variables relationship is based on the notion of the above theories.

A green finance activity is a financial activity that seeks to improve the environment through the organized use of financial resources. Although green finance plays a critical role in mitigating the effects of climate change, it remains underinvested despite its importance. A green finance program includes a variety of environmental objectives, such as reducing emissions from industry, reducing CO₂ emissions, preserving biodiversity, and improving water sanitation. As part of this program, green finance can also be used to finance social initiatives (Akram et al., 2020). The importance of Green Finance and Green Technology Innovation in advancing sustainable development can be illustrated by the recent research conducted on the intersection of these two fields. With the increasing use of green finance as a tool for transitioning towards a low-carbon economy, it has gained momentum as a catalyst for promoting investments, loans, and financial instruments that promote environmentally friendly initiatives (Tamasiga et al., 2022). Simultaneously, various studies have consistently demonstrated the close connection between successful Green Finance and groundbreaking Green Technologies. These technologies play a vital role in mitigating the impacts of climate change (Lazaro et al., 2023). The findings of the study emphasize the significant relationship between both of these domains, which exist in a mutually beneficial relationship. The Green Finance sector channels funds toward the exploration, advancement, and application of eco-friendly technologies to foster environmental wellbeing (Wang and Wang, 2021). Green Finance has gained more credibility from investors due to the scalability and profitability of these technologies, making it more appealing to new investors. This dynamic synergy has the potential to generate long-term ecological benefits, foster economic growth, and address climate change on a global scale (Moreno et al., 2023). Early studies in this field examined the effects of green financing policies on GTI in enterprise sectors, along with the various factors that drive green technology innovation within these sectors (Jiakui et al., 2023). The Chinese government has implemented several environmental policies due to its status as one of the world’s largest polluters and the world’s largest developing country. The policy aims to assist businesses in decreasing their energy consumption and emission levels by implementing environmentally-friendly innovations (Hao and Chen, 2023). Multiple executive orders were unable to bring about the “innovation offsets” anticipated by Porter because of this coercion; in fact, rather than promoting enterprise innovation, they stifled it (Gray and Shadbegian, 2003; Zhong and Peng, 2022). Corporate green innovation has been shown to be insufficiently impacted by many market-based policies (Ghisetti and Pontoni, 2015). There are fundamental differences between ordinary innovation projects and green innovation projects, and the latter have higher risks and a lower success rate as compared to the former, so they require a longer R&D process and increased costs, and external financial backing is necessary (Tseng et al., 2013). China has implemented various sustainable financing initiatives, including green bonds and green credits, to support its environmental policies and foster sustainable economic growth. This became more comprehensive and influential due to China’s implementation of the GFPP (Green Finance Pilot Policy) in 2017. As part of the government’s efforts to promote green finance reform, a pilot zone for green finance reform and innovation has been chosen in eight cities across four provinces (Zhang et al., 2023). These cities were motivated to enhance and revolutionize their existing green finance tools by establishing a robust green technological innovation system. According to this theory, GFPP has the potential to impact a company’s external financing and green innovation choices by allocating financial resources to environmentally friendly firms that meet certain environmental criteria (Aizawa and Yang, 2010; Huang et al., 2022). Does the Green Business Partnership Program (GFPP) effectively promote corporate green innovations and have a significant impact on the green economy? There is still much to discover about these subjects, as they remain largely unexplored. Global warming has always been associated with ecological disasters and resource depletion since the 1960s Zheng et al. (2022), land problems Yang et al., 2022 reserve lessening (Abbas and Dogan, 2022). Globalization, the depletion of resources, pollution of the environment, and food crises are all factors that are bringing an increasing number of challenges to the world (Zhang et al., 2022a).

H1

Green finance and Green Technology Innovation have a positive relationship.

The hypothesis above describes that investments in green finance positively influence the development and adoption of green technologies, suggesting a synergetic relationship between green technology and green finance. According to this hypothesis, financial mechanisms will play a vital role in enabling technological innovation, which is necessary for sustainable environmental practices to be achieved.

Green finance has been researched for a long time as a means of improving a company’s environmental performance, and prior research indicates that there is a favorable correlation between these two variables. In essence, an organization’s level of financial responsibility is directly tied to its environmental impact. However, Zhang et al., 2022b, documented the concept of green finance, which involves providing financial support to projects that promote environmental sustainability, such as clean energy and technology. Based on existing literature, it has been found that the environmental impact of companies can be improved through the implementation of green finance, leading to better performance (Guang-Wen and Siddik, 2022; Zeeshan et al., 2022a). Moreover, Shi et al., 2022 documented that green finance had a positive impact on the performance of enterprises in regards to green operations. Indriastuti and Chariri (2021) highlighted that green investments lead to better financial performance as well as a sustainable business model. According to the findings of this study conducted by Alola et al., 2021 it was found and concluded that to attain environmental sustainability at a high-income level, there may be a need to compromise on investment risk. Extensive research has shown that green finance has a notable impact on improving environmental outcomes within a firm. Thus, it is anticipated that green finance will have a crucial role in enhancing the sustainability capabilities of banking institutions in the midst of this epidemic due to its positive environmental impact.

Additionally, Hsu et al., 2021 examine the correlation between green finance and environmental performance in China. It is interesting to see the insights that emerge from studying how Chinese firms respond to environmental concerns through Green Finance mechanisms. The study revealed a strong link between the adoption of green finance practices and the enhancement of environmental performance among companies. This finding was based on an extensive dataset that spanned from 2014 to 2018. Based on the results of this study, it is evident that companies involved in Green Finance activities, such as issuing green bonds or accessing green funds, have a significantly higher success rate in achieving their environmental goals compared to companies that do not participate in such activities (Zeeshan et al., 2021b; Baldi and Pandimiglio, 2022). In addition to exploring the relationship between green finance and environmental performance, the study also investigates the impact of institutional factors, such as government regulations and policy support. Given the substantial investment China has made in advancing green finance over the last 20 years, these findings have a noteworthy influence on the governance of its industrial sector, ensuring it remains environmentally sustainable and economically competitive. The growing awareness of global climate change and environmental issues has led governments to adopt a variety of policies aimed at promoting the adoption of green technologies, including green brands and environmentally friendly technology, across different industries. Likewise, Weinhofer and Hoffmann (2010) documented that, there is green finance, such as green bonds, to minimize the impact on the environment (Zeeshan et al., 2022b; Li et al., 2023). Green finance, however, is just a representation of the government’s attitude toward protecting the environment, whereas environmental innovation refers to their attitude towards resolving environmental degradation, as well as innovation in generating renewable energy and improving the efficiency of the utilization of energy (Liang et al., 2021; Wang et al., 2022b). There have been only a few studies that have explored the potential connections between green finance and environmental performance. However, no empirical tests have been conducted to determine if green finance can influence environmental performance. Developing sustainable products and processes based on the current performance of green finance and the development of environmental quality can provide valuable insights for governments to understand the relationship between green finance and environmental performance. By promoting green technology innovation, the government has the potential to enhance economic development and environmental performance. It is crucial for global development to prioritize environmentally friendly, green, and sustainable practices. Therefore, this study aims to understand the impact of green finance and green technology innovation on environmental performance, both in the short and long term.

H2

There is a significant association between Green Finance and the Environmental Performance

In this hypothesis, we examine the impact of green finance on environmental performance from a quantitative perspective. Based on the results, it is suggested that a major factor associated with sustainable policies and outcomes is financial investments in environmental sustainability. The purpose of this hypothesis is to explore how financial initiatives may influence environmental quality by directly impacting the overall quality of the environment.

The footprints regarding the relationship of these variables exist in the body of knowledge. Singh et al. (2020), and data was collected from 669 manufacturing firms in the United Arab Emirates. Their findings indicate that a firm’s green innovation processes a crucial role in determining its environmental performance. Another study on 244 Malaysian large manufacturing companies was conducted and found that there is a strong connection between green innovation and both human resource management and green intellectual capital to improve the environment (Kraus et al., 2020). Moreover, Rehman et al., 2021a highlighted the results of 244 large manufacturing companies and found that green innovation was related to environmental performance. However, Yan and Zhang (2021) highlighted an analysis of green practices and environmental management in energy-intensive units showed a positive impact on the environment through a stochastic Frontier model that was developed between 2011 and 2017. Furthermore, Zameer et al. (2021) and Xie et al. (2022) documented an efficient and effective way to monitor, develop, and implement green practices that can contribute to the improvement of the environment. Their research suggests that by closely monitoring, developing, and implementing green practices, significant improvements can be made to the overall environmental conditions. However, (Luo, 2020), found that the use of green technology innovation to improve environmental performance is significantly correlated with the use of green technology innovation. However, this study analyzes the positive impact of technology innovation on environmental performance in the energy sector, using data from 136 companies over 10 years (2009–2019). The connection between green innovation, government subsidies, and environmental performance has become more evident in recent years, although there are still some controversial findings. Several studies have explored the correlation between different types of government subsidies and environmental performance s (Zeeshan et al., 2021a; Xie et al., 2022). However, previous studies have not been able to distinguish between different categories of government subsidies, including those aimed at environmental protection. The potential impact of green technology innovation within both of them has not yet been fully demonstrated in the current situation (Sharif et al., 2022). Environmental protection subsidies aim to guarantee the preservation of the environment and the regulation of pollution. To assess the impact of China’s environmental subsidy policy, it is crucial to directly analyze the influence of these subsidies on the performance of heavy-polluting industrial sectors, specifically those involved in the production of heavy pollutants. This analysis will help determine the effectiveness of these subsidies in safeguarding the environment (Pei and Pei, 2022).

H3

Green Technology Innovation is Postulated to Improve Environmental Performance

The hypothesis above suggests that advancements in green technologies directly contribute to the improvement of environmental performance, which illustrates that innovative solutions are essential to dealing with the challenges associated with the environment. Using green technologies for research and development will lead to more efficient resource use and reduced environmental degradation. Table 1 denote the variable definition and measurement of the study the same table pattern used by many researcher like (Zeeshan, Rehman, Ullah, Hussain, and Afridi, 2022)

After thoroughly examining the available literature and analyzing the data presented in this paper, we were able to carry out a comprehensive study on the effects of green finance, green technology innovation, and environmental performance in 280 Chinese cities having 16 types of heavy-polluting industries by using the data source which is available on: Ministry of Ecology and Environment, China City Statistical Yearbook and from CNRDS (http://www.cnrds.com) database which has been reported in this paper for the period 2003–2022. We selected those firms for which the green finance and GTI data are available in the Chinese database. The most relevant econometric techniques of this study are as follows Eq. 1: E P i t = α 0 + β 1 G F i t + β 2 G T I i t + e i t (1)

Where EP depicts environmental performance Shao et al. (2023), α is the coefficient, GF denotes green finance Lee and Lee (2022) GTI represents green technology innovation Yang et al. (2023) and e is the error term. It was important to conduct this analysis to assess the normality of the data and evaluate the strengths and weaknesses of each variable. In addition, the CSD test was utilized to analyze cross-sectional dependency, which was considered suitable given the cross-sectional nature of the data. The cross-sectional dependency test is given in bellow equation Eq. 2: C S D I T = I T T − 1 2 1 2 ρ ¯ ^ T (2) ρ ¯ ^ T denotes T stands for time, and I stand for cross-section units, this is what we see as a pair-wise correlation coefficient.

Furthermore, a cross-sectionally augmented IPS test (CIPS) is performed to assess the stationarity of the variables. Stationarity is a vital aspect to consider before applying a suitable model to a study. As a result, CIPS was also used. Eq. 3 CIPS are as follows: Δ W i , t = ϕ i + ϕ i Z i , t − 1 + ϕ i Z ¯ t − 1 + ∑ 1 = 0 Ρ ϕ i l ∆ W ¯ t − 1 + ∑ 1 = 0 Ρ ϕ i l ∆ W i , t − 1 + μ i t (3) W ¯ depict the mean “cross-sectional” and it is given in Eq. 4 equations: W i , t = ϕ 2 E P i , t + ϕ 2 G F i , t + ϕ 3 L n G T I i , t (4)

Therefore, the CIPS establishes equation (5), wherever, CADF depicts the cross-sectional ADF test.

In previous studies, the Westerlund and Edgerton (2008) cointegration test was used to assess cointegration and determine whether it applies to the relevant model. This step is crucial for employing the appropriate model. In addition, when taking into account the distinctive characteristics of the durable CSD approach and making assumptions about structural breaks, it emerges as a highly efficient method for examining cointegration. Additionally, we explore how the structure varies during periods of regime shifts and non-regime shifts. Below is the equation for the test Eq. 6: 1 Log L = α 0 − 1 2 ∑ i = 1 N Tlog σ i . t 2 − 1 σ i , t 2 ∑ t = 1 T eit 2 (6)

It is also important to note that the previous literature applied the CS-ARDL technique to check the correlations between the constructs under study. There are several assumptions associated with this model, such as endogeneity, CSD, and slope heterogeneity. Furthermore, it can be noted that ARDL is an approach broadly used for panel data analysis, but it does not address the problem of CSD errors. It is therefore appropriate to deal with CSD issues in the best possible way using CS-ARDL. This is a novel approach developed by Chudik and Pesaran (2015) that also has some strict assumptions, which are related to these methods. An example of how the approach equation can be represented is shown below Eq. 7: Δ Y i t = φ i + ∑ I = 1 Ρ φ i t Δ Y i , t − 1 + ∑ I = 0 Ρ φ il ′ E X V s , i , t + ∑ l = 0 1 φ il ′ CSA ¯ i , t − 1 + ε i t (7)

Thus previous research also proposed the cross-sectional ARDL employing the understudy concept and stated as follows equation no Eq. 8: Δ E P i t = φ i + ∑ l = 1 Ρ φ i t ′ Δ E P i , t − 1 + ∑ l = 0 Ρ φ i t ′ GF s , i , t + ∑ l = 1 Ρ φ i t ′ lnGTI s , i , t + ∑ l = 0 1 φ i t ′ CSA i , t − 1 + ε i t (8)

The study additionally used panel quantile regression, a method introduced by Koenker (2004) to understand the different heterogeneity of the dataset. Quantile regression has become a crucial instrument in environmental research in recent years. There are four strong reasons why quantile panels should be modeled based on a regression approach. By employing traditional regression analysis, researchers calculate regression coefficients based on average effects of covariates, rather than estimating coefficients using average effects. For more accurate regression coefficients, it is advisable to utilize this method as it accounts for the significant variations between predicted and observed variables. It has been noticed that the conditional mean does not consistently produce the same outcomes. PQR, on the other hand, does not rely on any assumptions about the distribution, including the absence of a normal distribution (Ouyang et al., 2022). In terms of the distinct heterogeneity of the panel data, there is no distinction between CM and PQR methods, which also incorporate the distributional heterogeneity of the panel data (Akram et al., 2020). When studying PQR, it is important to measure the independent variables alongside the dependent variables to gain a thorough understanding of them. This approach sets itself apart from CM methods by not relying on distributional assumptions. In addition, it avoids making any normality assumptions, unlike the CM method. In comparison to CM, PQR also addresses outliers and produces reliable results. Due to the utilization of various quantiles in PQR, there exists a distinct relationship between the predicted and observed variables. Each cross-section is also explored for unobserved heterogeneity by measuring various parameters within the quantiles (Amin et al., 2020). Considering the coefficient at the extreme is also useful for policy purposes. Besides its econometric advantage, the PQR provides a multidimensional analysis of how GF and GTI are estimated at different stages of EP, offering a comprehensive analysis of these variables.

According to Eq. 9, Akram et al. (2021) the Panel Quantile Regression model can be summarized as follows: Q Y i t τ | X i t = γ τ ′ X i t + α i   i = 1 , … … , N , t = 1 … … , T (9)

Here Y i t is the DV EP, Q Y i t τ / X i t denotes to the τ t h quantile of EP, X i t signifies the vector of IV GF , GTI in year t for i country. While, γ τ denote to unidentified coefficients,   α i shows the effects of a specific country are unknown. While i shows the Chinese cities and t represents the year. This study presents a successive model Eq. 10 Iuliano et al. (2018) that will be used to analyze the data. Q Y i t τ | X i t = γ 1 τ G F i t + γ 2 τ G T I i t + α i (10)

In the case of Panel Quantile Regression, a traditional linear regression model cannot be used to estimate the model. Likewise, Koenker (2004) documented that with the elimination of an unknown part of the estimation process, it is possible to minimize the estimation process by offering a penalty term. Compared to other approaches, this method has two distinct advantages over the others. Additionally, it minimizes variance caused by distinct coefficient estimation by decreasing the estimated parameters efficiently (Akram et al., 2020). It has been estimated that the Eq. 11 in this way: argmin β ∑ m = 1 M ∑ i = 1 N ∑ t = 1 T W M ρ τ m Y i t − γ 1 τ GF i t − γ 2 τ GTI i t − α i + μ ∑ i = 1 N α i (11) ρ t γ = γ τ − 1 y < 0 check function, 1 A It indicates that set A has a function of an indicator. Whereas, Y i t specifies the EP in the country i in time t . In this case, M stands for the quantile index, first, we check whether data have the attributes of normality, stationarity, and heterogeneity along with the econometric method mentioned above. Using a panel quantile regression model, this study confirms non-normality, stationarity, and slope heterogeneity. Data are first normalized, stationary, and heterogeneous in slope before applying the above method. Using a panel quantile regression model, this study confirms non-normality, stationarity, and slope heterogeneity.

To know the impact of Green Finance and Green Technology Innovation on Environmental Performance we conducted a CS-ARDL estimator as a main frame econometric technique and the results in this regard are displayed in Table 8 which provides an overview of these values. This technique shows both short-term and long-term relationships between EP, GF, and GTI, the research done using CS-ARDL analysis suggests that there is a statistically strong association between (GF& CO₂ and GTI and CO₂) these variables in the long run and the short term. GF and GTI are associated with increases in EP, as both show an inverse relationship with CO₂ a proxy for Environmental Performance. In addition, a significant error correction term appears that indicates that any deviation from the long-run equilibrium tends to be corrected by the system when it performs a short-run analysis.

Table 9 PQR results explain how GF and GTI, affect the EP. For a comprehensive analysis of the impact of GF on EP, the study used numerous quintiles for a comprehensive analysis. Moreover, since PQR models an entire conditional distribution, it allows for an exploration of the effects of I.V. on the D.V. simply because the quintiles show a diverse set of effects in an individual conditional distribution. A PQR analysis also examines hidden heterogeneity across all cross-sections and assesses the slope coefficients for different quantiles corresponding to the various cross-sections (Amin et al., 2020). Additionally, decision-makers should keep in mind the value of the coefficient at the level of the distribution.

Table 10 has been placed as a second choice or robustness check estimator. The results in the quintiles show the effect of green finance and green technology innovation on CO₂ emission (EP). The quintiles for green finance are negative and statistically significant which determines that with green finance initiatives in cities in China, the amount of CO₂ emission is reduced which contributes to the environmental quality of China. All quintiles regarding GF are statistically significant. Likewise, the results in numerous quintiles for green technology innovation on CO₂ emission (EP). The quintiles for green finance are negative and statistically significant and also show statistically significant coefficients with negative signs which confirm the negative impact of GTI on CO₂ emission, implying that an increase in green technology innovation on CO₂ emission (EP). The quintiles for green finance are negative and statistically significant leading to an improvement in Environmental Performance.

Tabel 11 the heterogeneity was assessed based on dividing the firms in the Eastern region and Central and Western regions firms. Our results displayed that terms of both green finance and green technology innovation are some how same in both panels, i.e., eastern region, Central and western regions or cities as both green finance and green technology innovation reduce CO₂ emission which indicates that both regions in regards to its cities contribute to the regional environmental quality. Hence it is assumed and obtained that there are favoring results for both green finance and green technology innovation in both regional penal documents that heterogeneity does not exist in our data. The results are some how same in both panels for the Eastern region and Central and Western regions, which documented that there is no such bigger issue of heterogeneity in the data of these firms. Hence, our results would have no spurious coefficients.