Available online at www.HighTechJournal.org HighTech and Innovation Journal Vol. 6, No. 2, June, 2025 537 ISSN: 2723-9535 Local Economic Autonomy and Enterprises’ Green Total Factor Productivity: A Policy Substitution Perspective Long Wang 1* 1 School of Economics and Management, Northwest University, Xi'an, 710127, China. Received 17 December 2024; Revised 11 May 2025; Accepted 20 May 2025; Published 01 June 2025 Abstract Global climate change and environmental degradation necessitate a transition toward sustainable economic development. The factors influencing green total factor productivity (GTFP), a crucial measure for sustainable economic growth, have garnered significant attention. This study investigates the impact of local economic autonomy on enterprises’ GTFP in China, integrating both fiscal and environmental autonomy. Using panel data from 2008 to 2021, a generalized difference- in-differences (DID) model combined with the non-radial SBM-ML index measures GTFP. Findings indicate that while fiscal autonomy promotes GTFP, environmental autonomy hinders it, resulting in an overall negative effect of economic autonomy. A policy substitution effect emerges; wherein local governments prioritize environmental regulation over support for science and technology. Additionally, industrial structure upgrading plays a role in mitigating the negative impact of autonomy, offering empirical evidence relevant to sustainable development policies in transition economies. Keywords: Local Economic Autonomy; Environmental Decentralization; Fiscal Decentralization; Generalized Difference-In-Differences Method; Green Total Factor Productivity. 1. Introduction Global climate change and environmental degradation owing to economic activity threaten socioeconomic status and human health. Changes in environmental conditions increase the likelihood of severe weather incidents, such as floods and droughts, which cause substantial property losses. Exposure to climate warming, air pollution, and toxins increases individuals’ risks of respiratory issues, cardiovascular problems, and cancer [1]. These concerns underscore the significance of fostering green economic growth, with GTFP being widely recognized as a key metric for assessing the progress of such development. GTFP incorporates the environmental impact of production into productivity calculations, thereby reflecting both the environmental impact of economic units and their economic output. Studies on GTFP have focused on its definition/measurement methods [2-6] and determinants as well as the impact of economic, social, and environmental effects [7-11]. However, among the multifaceted determinants of GTFP, the role of local governments' economic autonomy remains under-explored in extant literature. Related literature, such as studies on political, fiscal, and environmental decentralization, posits that local economic autonomy might exert its influence via multiple mediating mechanisms, specifically (1) governments' accountability incentives, (2) information acquisition costs for policy formulation, (3) corruption control effectiveness, and (4) intergovernmental policy coordination (see details in the literature review section). By integrating measures of fiscal autonomy and * Corresponding author: longwang2234@outlook.com http://dx.doi.org/10.28991/HIJ-2025-06-02-012  This is an open access article under the CC-BY license (https://creativecommons.org/licenses/by/4.0/). © Authors retain all copyrights. https://creativecommons.org/licenses/by/4.0/ https://orcid.org/0009-0007-8388-0216 HighTech and Innovation Journal Vol. 6, No. 2, June, 2025 538 environmental regulatory autonomy within the generalized DID model, this conceptual framework aligns seamlessly with the dual-dimensional nature of GTFP, which inherently encompasses both economic outputs and environmental performance. This theoretical congruence offers valuable insights into the complex interplay between governance structures and sustainable productivity growth. Over the past few decades, China has experienced a significant economic shift from focusing on growth quantity to emphasizing development quality. The persistent trade-offs between economic expansion and environmental protection in this context reveal how government policies both influence and adapt to economic progress. Using data from the Chinese government and listed companies, this study analyzes how local economic autonomy affects GTFP. Results indicate that while fiscal autonomy promotes enterprises’ GTFP, environmental autonomy tends to hinder it, leading to an overall negative effect from increased local governments’ economic autonomy. Additionally, the cross-sectional impact of industrial structure upgrading and its mediating role between local economic autonomy and enterprises’ GTFP is assessed. This influence appears stronger in cities where the tertiary sector surpasses the secondary sector in value contribution. Moreover, local governments balance various intervention measures, including introducing environmental regulations to substitute for their support of science and technology, a phenomenon referred to as the policy substitution effect. The principal contributions of this research are summarized as follows. First, the study broadens existing research on the relationship between governments and green economic development. Although many studies have examined government influence on GTFP, most focus on how specific economic behaviors—such as local debt levels or environmental regulation impacts—affect GTFP. In contrast, this analysis combines two dimensions of government characteristics to measure local economic autonomy. Second, the distinct effects of government actions on GTFP have rarely been explored, and this study advances understanding by incorporating the role of industrial structure upgrading. Third, previous research has largely emphasized conceptual analyses of decentralization. Here, the impact of decentralization on green economic development is examined through both conceptual and empirical approaches, highlighting the mediating and policy substitution effects, wherein environmental regulations are implemented as alternatives to direct support for science and technology. The remainder of this paper is structured as follows: Section 2 reviews the literature; Section 3 develops hypotheses; Section 4 describes the data and methodology; Section 5 presents the results and robustness tests; Section 6 concludes with policy implications. 2. Literature Review 2.1. Influencing Factors of GTFP Following Zhang et al. [6], existing studies on the influencing factors of Green Total Factor Productivity (GTFP) can be categorized into three interconnected streams: technical, economic, and governmental streams. Technical streams collectively highlight that technological advancements, including digital economy [12], fintech [13], green technology innovation (GTI) [7], and internet development [14], play a significant role in enhancing Green Total Factor Productivity (GTFP). While the specific mechanisms vary, these technologies generally improve GTFP through direct efficiency gains and indirect structural transformations. Economic streams also influence GTFP. Upgrading to cleaner industries (e.g., tertiary sectors) optimizes resource allocation and accelerates green technology adoption, fostering GTFP growth [15]. Other contributing factors include green finance [8] and high marketization, which promotes resource mobility and industrial upgrading [16]. Government streams involve factors such as environmental regulation intensity, fiscal decentralization, infrastructure levels, and intellectual property protection [9, 14, 17]. Among these factors, environmental regulation, which comprises pollution abatement costs and environmental emission standards, serves as an important factor influencing economic output and environmental pollution. While environmental regulation is often used as a mediating variable in prior research, consensus on its impact on sustainable development and GTFP is lacking. Environmental regulation has traditionally been considered to directly increase the production costs of enterprises, which may hinder economic growth [18]. However, recent studies mostly agree that environmental regulation may promote economic growth by pressuring enterprises to conduct green technological innovation as well as improve their utilization and production efficiency of resources [19]. Concurrently, some scholars believe a U-shaped correlation between environmental regulation and economic expansion exists [20], whereas others believe the impact of environmental regulation on economic growth is not significant [21]. 2.2. Impact of Decentralization on GTFP The nexus between decentralization and GTFP/sustainable economic growth has long been controversial among scholars. Some believe that decentralization can improve GTFP. For example, Song et al. [9] showed that fiscal decentralization directly increases GTFP. As the autonomy of local governments increases, they must take more HighTech and Innovation Journal Vol. 6, No. 2, June, 2025 539 responsibility for local economic development [22]. Furthermore, local governments’ information advantage concerning local economic development (i.e., they can more easily obtain information about local needs and the cost of providing services) promotes the rational allocation of resources [23, 24]. Shah [25] found that a government with more local power can reduce corruption in the long run because decentralization narrows the gap between governments and residents. This influence, effected by the government’s higher autonomy, is generally considered conducive to economic development and GTFP. Conversely, some scholars believe that decentralization is negatively related to GTFP. For example, fiscal decentralization leads to increased competition among local governments for greater fiscal advantages, exacerbating environmental deterioration [4]. A potential ‘race to the bottom’ scenario may arise: when environmental regulation in the surrounding area is less stringent than that within its own jurisdiction, the local government will reduce its environmental regulation or capital tax to attract industry transfer [26]. Hwang [27] found that regional competition can lead to lower capital tax and worse environmental pollution. Some research also indicates that decentralization exacerbates corruption in local governments because there are more opportunities for corruption at the local level [28]. More importantly, local government policies (e.g., pollution abatement measures) rarely affect only local regions and tend to have spillover effects at the national or international level since they are public goods. Local governments may ignore the benefits of basic research investments, pollution information exchanges, and pollution control technologies in other regions [29, 30]. The resulting underestimation of the broader national or global benefits of investment and responsibility in this area reduces enthusiasm for providing these public goods. Improvements in economic autonomy aggravate the degree and possibility of such neglect by local governments, resulting in a decline in both economic and environmental benefits. Prud'Homme [28] pointed out that the problem lies in whether to provide these services with positive externalities and the difficulty of the joint provision of such public services at different levels. 3. Research Gaps and Hypotheses Development 3.1. Research Gaps Although some studies have examined the impact of government autonomy on GTFP, most have focused on a single dimension of autonomy, yielding inconsistent findings. Moreover, the varying relationships between government autonomy and GTFP have not been sufficiently explored. Pollution levels are known to differ by industry, and the presence of tertiary industries may enhance technological innovation, leading to regional differences in industrial structure. Additionally, the influence of decentralization on GTFP remains largely theoretical and lacks robust empirical validation. 3.2. Research Hypotheses While some studies suggest that decentralization promotes GTFP, a significant body of research has identified a negative correlation between them. Therefore, the overall impact of economic autonomy on GTFP is considered negative for four primary reasons: reduced policy coordination, excessive government intervention, increased corruption, and diminished support for science and technology. According to externality theory and the core/periphery model of the regional economy, economic activities in one region impact economic performance in other regions. For example, Lejour & Verbon [31] argued that higher taxes have a negative impact on investment activities in one region, which is then passed onto other regions. When different regions can coordinate with each other and formulate joint policies (e.g., joint tax policies), such collaboration can help improve redistribution and raise economic growth. However, decentralization leads to a lack of coordination among regions as well as between local and central governments, producing inefficient tax rates and impacting economic development negatively [32]. If local economic autonomy increases, the reduced dependence on the central government may increase regional competition and reduce the level of regional policy coordination. Consequently, the allocation of human resources, information, technology, and other resources between regions becomes inefficient, hindering environmental governance and scientific and technological innovation in local areas. Concurrently, the improvement in economic autonomy may also lead to local government intervention in economic affairs and the intensification of corruption by officials, such as rent-seeking behaviors, which increase the operating costs of enterprises and negatively impact innovation [33, 34], ultimately hindering the increase in enterprises’ GTFP. Hypothesis 1: Local economic autonomy hinders the improvement of GTFP. In addition, government support and regulation can affect GTFP by influencing scientific and technological innovation and pollution control capacity. Reduced support for scientific and technological research and fewer environmental regulations may discourage enterprises from establishing innovative production processes and attempting product innovation as well as from implementing effective environmental pollution control measures, which can lead to a decline in GTFP. However, differences in the characteristics of environmental regulation and support for science and HighTech and Innovation Journal Vol. 6, No. 2, June, 2025 540 technology in terms of the risk/return, performance measurement, and level of positive externalities can affect governments’ policy choices after decentralization because they have more say on how their jurisdiction develops according to their policy preferences. First, scientific and technological innovation usually incurs greater costs and risks than pollution control, especially basic scientific research with far-reaching impacts, which is highly uncertain and delivers benefits only in the long term [35]. Most environmental regulations implemented by coercive means have obvious and faster effects [36]. Second, the difficulty of measuring the economic and social value of basic research can reduce the enthusiasm of local governments to invest. Conversely, direct environmental regulations are easier to supervise and evaluate. Third, the positive externalities of science and technology, especially basic scientific research achievements, affect the whole industry and society at large, making the cost of imitation and learning much lower than that of innovation; this leads to free-riding behavior. By contrast, the environmental impact is mainly limited to the local administrative region. Hence, with decentralization in economic regulation, local governments have gained more selectivity in choosing the mode of economic development for their regions and tend to ignore direct support for scientific and technological innovation and are more inclined to take responsibility for environmental pollution. While the former leads to a decrease in enterprises’ GTFP in that region, the latter serves as a catalyst. Hypothesis 2: The increased economic autonomy of local governments negatively impacts governments’ support for science and technology and thus enterprises’ GTFP. Hypothesis 3: The increased economic autonomy of local governments positively affects the level of environmental regulation and thus enterprises’ GTFP. Further, owing to the decline in support for science and technology, the government may increase environmental regulation to promote local economic performance to compensate for the adverse effects that may arise from this decline in regional development. Therefore, there are substitution effects between different policy directions, especially when local governments have some discretion over choosing their mode of economic regulation. This results in the unidirectional policy substitution effect of environmental regulations from support for science and technology (Figure 1). Hypothesis 4: There is a one-way policy substitution effect to focusing on environmental regulation from supporting science and technology following an increase in local economic autonomy. Figure 1. Policy substitution effect of local economic autonomy The different impacts of decentralization on the economy and environment can also be related to other factors. Khan et al. [37] found that enhancing institutional quality and human capital could amplify the negative influence of fiscal decentralization on carbon emissions. Therefore, differences in regional industrial structures may impact the relationship between regional autonomy and enterprises’ GTFP differently. Measuring the returns and risks of research investment is thus crucial for enterprises and governments to make investment decisions. Improving the tertiary industry’s development locally can accelerate the agglomeration effect of research investment, gradually reducing the risks and costs of research investment failures for companies and regions while promoting technological innovation. Simultaneously, industries based on technology and science are more responsive to research outcomes and more likely to enjoy their benefits. For example, Wang et al. [38] found that agglomeration in the information and communication technology industry reduces carbon emissions through increased technological innovation and economies of scale. The agglomeration of tertiary industries may generate diffusion effects that promote coordination among regional governments and facilitate resource allocation [38]. Therefore, it has a restraining effect on the negative impact of increased local government autonomy. Moreover, upgrading the local industrial structure prompts a shift in economic development patterns from resource- or labor-intensive to technology- and knowledge-intensive [39], which directly Environmental regulations Science and technology support Effect of autonomy on substitution effect HighTech and Innovation Journal Vol. 6, No. 2, June, 2025 541 increases the share of non-polluting enterprises within the economy. This reduces the difficulty for governments to regulate the environment and enhances their proactive attitudes towards environmental regulation. Furthermore, upgrading industrial structures improves the efficiency of coordinating resources, promoting increased GTFP. Hence, this study proposes the following: Hypothesis 5: Industrial structural upgrades weaken the negative relationship between local governments’ economic autonomy and enterprises’ GTFP. Figure 2 illustrates the influence pathway from economic autonomy to GTFP, considering both the direct and indirect effects. Figure 2. Impact mechanism of local economic autonomy on enterprises' GTFP 4. Data and Method 4.1. Core Explanatory Variables Fiscal Autonomy (FA) Many countries have decentralized their fiscal power to promote economic development and reduce poverty [40]. Contrary to the decentralization of environmental regulations, fiscal decentralization improves the government’s overall capacity to utilize fiscal resources. As it is not targeted at a particular field, it allows for a broad enhancement of local governments’ freedom in fiscal expenditure, policy formulation, and execution. In this context, the fiscal decentralization index is introduced as the first criterion for measuring government autonomy. Drawing on [9], the degree of self-financing by local governments—represented by the ratio between fiscal revenue and expenditure—is used to reflect this dimension of autonomy. Environmental Autonomy (EA) Fiscal autonomy reflects only the basic level of autonomy. As environmental issues receive increasing emphasis in economic development, relying solely on fiscal autonomy fails to capture the full scope of governmental decision- making considerations. Consequently, environmental autonomy serves as a more comprehensive measure of government autonomy in economic regulation. Environmental autonomy is primarily determined by environmental decentralization. Traditionally, two methods represent environmental decentralization: one measures the number of staff within the government’s environmental sector, and the other assesses whether local governments possess legislative authority over environmental matters. The delegation of legislative authority to local governments in environmental affairs constitutes a clearer manifestation of environmental decentralization, granting greater discretion in regulating environmental issues and enhancing autonomy in environmental governance. In contrast, the size of environmental management departments is a relatively indirect indicator, influenced by government financing and environmental awareness, and does not directly signify environmental decentralization. Thus, changes in local governments’ legislative authority over environmental matters are used as a proxy for variations in environmental autonomy to measure levels of autonomous control over environmental regulation. In 2015, a new revision of the Legislation Law of the People’s Republic of China (hereafter “Legislation Law”) granted over 270 prefecture-level cities the authority to legislate on environmental issues, strengthening local Policy substitution effect Science and technology support Environmental regulation Corruption Policy coordination Market intervention Between cities Between central and local Industrial structure Heterogeneity Local governments’ economic autonomy Fiscal autonomy Environmental autonomy Green total factor productivity of enterprises HighTech and Innovation Journal Vol. 6, No. 2, June, 2025 542 governments’ capacity to regulate local economic and environmental development. As the implementation of this legal revision took effect after October 2015, and given the typical lag in the impact of legal measures, 2016 is regarded as the year when policy effectiveness became evident. The other 53 cities not affected by this revision in 2015 serve as the control group, while the remaining cities constitute the experimental group. Economic Autonomy (AUTO) Constructing a comprehensive indicator of local government autonomy poses significant challenges, as it requires accounting for multiple factors such as political and economic considerations, as well as the weighting assigned to these factors. Legislative and administrative oversight and organizational elements also play a role [41]. To determine whether government autonomy affects enterprises’ GTFP—which encompasses both economic output and environmental output—economic autonomy is defined by combining fiscal autonomy and environmental autonomy. This autonomy can thus be characterized as the regulatory autonomy of local governments concerning sustainable economic development. However, because fiscal autonomy is a continuous variable and environmental autonomy is a dummy variable, integrating the two while highlighting their combined impact as the core explanatory variable necessitates a specific approach. Accordingly, the general DID model from [42] is employed, multiplying fiscal autonomy and environmental autonomy to achieve this integration. (Details are provided in the description of Equation 1 in the Benchmark regression model section). 4.2. Explained Variable: GTFP To measure enterprises’ production and environmental efficiency comprehensively, this study adopts a non-radial SBM-ML index to calculate enterprises’ GTFP [43]. This calculation method incorporates energy inputs and environmental pollution into the evaluation system, addressing the limitations of traditional total factor productivity, which fails to fully account for enterprise factor productivity. The measurement of input and output indicators for enterprise GTFP is detailed as Table 1. Table 1. The input and output indicators of enterprises’ GTFP Input factors Labour input The number of employees Capital input Net fixed assets Energy input Calculated on industrial electricity consumption in the city where the enterprise is located, considering the proportion of employees in relation to urban employment Expected output Enterprise operating income Non-expected output Emissions of industrial sulphur dioxide, industrial wastewater, and industrial smoke and dust are converted based on the proportion of employees to urban employment to represent non-expected output. 4.3. Control Variable As enterprises’ GTFP could be affected by a multitude of macro and micro factors, based on the existing literature, this study selects control variables from two dimensions: the city and enterprise levels. At the city level, the per capita GDP (Affl) indicates the wealth level of urban residents [12]. The total number of green patents authorized in a region (Autho) reflects the advancement of green technology in that area, which is crucial for mitigating environmental pollution [19]. The proportion of the sum of imports and exports volume to regional GDP is used to reflect the openness level (Open), which also affects the overall economy [12]. In line with the pollution halo hypothesis, which posits that foreign direct investment (FDI) can enhance environmental conservation, this study quantifies FDI by the ratio of actual foreign capital inflow to the regional GDP. The level of urbanization (City), which is typically associated with economic growth, is gauged by the percentage of the urban population relative to the total population within the locality. Infrastructure construction is represented by the per capita road area coverage (Road) and gas penetration rate (Gas). More details of the measurements are presented in Table 2. At the enterprise level, firm size (Size) is taken into account, as it is generally believed to influence production efficiency and technological innovation [44]. Additionally, the company’s return on equity (ROE) and asset-liability ratio (lev) are included to control for their potential effects. As a company may have been influenced by its productivity in the previous year, a lagged one-period GTFP (L1.GTFP) is controlled. HighTech and Innovation Journal Vol. 6, No. 2, June, 2025 543 4.4. Mediating Variable Science and Technology Support of Governments (Scitechsup) Although enterprises are direct participants in technological R&D, owing to the strong externalities of such activities, it is difficult for companies to achieve effective research investment without external incentives. Therefore, the government’s scientific expenditures are particularly important. From a fiscal expenditure standpoint, this study employs the ratio of local fiscal scientific expenditure to total expenditure as an indicator of the support provided by local governments for technological research and development. This reflects the balancing result of government fiscal spending: a larger proportion indicates stronger government support for science and technology. Environmental Regulation (Enviregu) To prevent the incomplete construction of indicators that might result in biased estimations of the intensity of government environmental regulations, this study measured government environmental regulations based on their direct outcomes. These outcomes include the overall utilization rate of general industrial solid waste, the centralized treatment rate of wastewater by treatment plants, and the harmless treatment rate of domestic waste. Table 2. Descriptive statistics Variables Definition/measurement Mean Std. Dev. Min Max GTFP Green total factor productivity of the enterprise 0.943 0.111 0.747 1.148 FA Fiscal autonomy: Local government’s general public budget revenue/general public budget expenditure 0.733 0.206 0.081 1.107 EA Environmental autonomy: Determined according to the implementation of the Legislation Law 0.116 0.321 0 1 AUTO Autonomy of local governments in economic regulation 0.06 0.178 0 0.945 L1.GTFP One-period-lagged corporate GTFP 0.916 0.111 0.72 1.12 Affl Natural logarithm of regional GDP per capita 11.254 .577 8.549 13.056 Autho Natural logarithm of the total number of green patents granted in the region 6.552 1.867 0 9.811 FDI Ratio of actual utilisation of foreign capital to regional GDP 0.03 0.02 0 0.229 Open Total imports and exports/gross regional product 0.588 0.586 0 3.279 City Town population/total population 0.729 0.167 0.151 1.001 Road Road area per capita 2.524 0.56 0.351 4.096 Gas Gas penetration rate 0.977 0.061 0.11 1.063 Size Natural logarithm of a business’ total assets 22.703 1.323 19.317 26.452 ROE Return on equity 0.068 0.544 -45.737 1.117 Lev Ratio of total liabilities to total assets of a business 0.493 0.191 0.007 0.979 Scitechsup Government technology support: General public budget science expenditure/Total general public budget expenditure 0.037 0.023 0.004 0.12 Enviregu Environmental regulation: General industrial solid waste comprehensive utilisation rate, sewage treatment plant centralised treatment rate, and household garbage harmless treatment rate 0.348 0.029 0.167 0.484 Struc Industrial structural upgrading: Added value of tertiary industry to that of secondary industry 1.479 0.974 0.139 5.349 4.5. Sample Selection and Data Sources Data on China’s listed companies from 2008 to 2021 serve as the basis for analysis. The following samples were excluded prior to further processing: (1) ST and *ST enterprises, (2) research samples lacking major data, and (3) financial companies and those engaged in financial operations. To mitigate potential impacts on the interaction term of environmental autonomy and preserve data quality, samples involving companies that relocated between general prefecture-level cities and larger cities after 2016 were removed. City panel data were matched to the remaining sample companies, with further exclusions applied to samples with substantial missing data; linear interpolation was employed to complete variables with minor gaps. Additionally, listed company samples lacking continuous main data across years were excluded to convert the non-balanced panel data into balanced panel data, ensuring more robust empirical analysis of same-dimensional sample data. Data sources include the China Urban Statistical Yearbook, China Environmental Statistical Yearbook, annual reports of listed companies, social responsibility reports of listed companies, and the websites of listed companies. 4.6. Benchmark Regression Model The revised Legislation Law of the People’s Republic of China in 2015 granted certain local governments the authority to legislate on environmental issues, thereby enhancing autonomy in environmental management. This legislative change enables the application of a DID method to measure the autonomy of local governments’ HighTech and Innovation Journal Vol. 6, No. 2, June, 2025 544 environmental regulations. Given the necessity to integrate fiscal autonomy and environmental autonomy as the core explanatory variable, a general DID method is adopted as the baseline regression model. Moreover, to provide further policy insights, the impact of environmental and fiscal autonomy on firms’ GTFP was examined separately. In particular, to mitigate sample selection bias and improve comparability between the control and experimental groups, this study used the synthetic DID method [45] to test the effect of environmental autonomy on firms’ GTFP. This method integrates the strengths of the traditional DID approach with synthetic control methods, effectively reducing the reliance on parallel trend assumption testing. Furthermore, time and firm fixed effects were introduced into all models to address endogeneity concerns and prevent the omission of important variables. The model specifications are as follows: 𝐺𝑇𝐹𝑃𝑖𝑡 = 𝛼0 + 𝛼1𝐴𝑈𝑇𝑂𝑖𝑡 + ∑𝜂𝑗𝐶𝑜𝑛𝑡𝑟𝑜𝑙𝑗𝑖𝑡 + 𝜇𝑖𝑡 + 𝜆𝑖𝑡 + 𝜀𝑖𝑡 (1) (�̂�𝑠𝑑𝑖𝑑 , �̂�, �̂�, �̂�) = 𝑎𝑟𝑔𝑚𝑖𝑛 𝜏,𝜇,𝛼,𝛽 {∑ ∑ (𝐺𝑇𝐹𝑃𝑖𝑡 − 𝜇 − 𝛼𝑖 − 𝛽𝑡 − 𝐸𝐴𝑖𝑡𝜏) 2�̂�𝑖 𝑠𝑑𝑖𝑑�̂�𝑡 𝑠𝑑𝑖𝑑𝑇 𝑡=1 𝑁 𝑖=1 } (2) 𝐺𝑇𝐹𝑃𝑖𝑡 = 𝛽0 + 𝛽1𝐹𝐴𝑖𝑡 +∑𝜂𝑗𝐶𝑜𝑛𝑡𝑟𝑜𝑙𝑗𝑖𝑡 + 𝜇𝑖𝑡 + 𝜆𝑖𝑡 + 𝜀𝑖𝑡 (3) Equation 1 represents the generalized DID model used in this study. In this equation, GTFP refers to the GTFP of an enterprise. AUTO=Treat×Post×FA is the key explanatory variable, reflecting local economic autonomy. Here, FA represents local fiscal autonomy, Treat×Post indicates environmental autonomy (EA), and Treat is a dummy variable distinguishing between treatment and control group cities; Treat=1 denotes the cities that have acquired environmental legislative power after the revision of legislation law, whereas Treat=0 represents the other cities. The variable Post is a binary indicator for the policy implementation period, with 1 indicating the years after 2016 and 0 representing the years before. Μi and λt denote the firm-specific and time-specific fixed effects, respectively, and εit is the stochastic error term. Equation 2 represents a synthetic difference model, in which individuals in the control group with characteristics similar to those of the experimental group’s GTFP are assigned higher weights, ensuring approximate parallel trends between the two groups in pre-policy samples; this weight is denoted as ωi. Assigning weights to the pre-policy period serves to align the trends of GTFP for control group individuals before and after the policy intervention, represented by λt. Minimizing Equation 2 enables estimation of the average effect of EA on GTFP, where αi denotes firm-fixed effects and βt represents year-fixed effects. Equation 3 corresponds to a normal two-way fixed effects model used to assess the impact of FA on GTFP; the other variables have similar meanings as those in Equation 1. 5. Results 5.1. Benchmark Regression Table 3 displays the impact of different factors on firms’ GTFP across the three models. The first and second columns present the outcomes of Equation 1, revealing the study’s primary focus: the influence of local government economic autonomy on GTFP, with and without control variables. After controlling for other variables, the coefficient of AUTO on GTFP is significant at a 5% level, with an estimated effect of -0.0026. This indicates that an increase in local governments’ economic autonomy is detrimental to firm performance and significantly reduces enterprises’ GTFP, confirming Hypothesis 1. The remaining columns present the experimental results for Models 2 and 3, with the additional control variables of fiscal and environmental autonomy introduced in Equations 2 and 3, respectively. After controlling for other variables, environmental autonomy significantly inhibits the improvement of firms’ GTFP at a 5% significance level, whereas fiscal autonomy has a promoting effect on GTFP at a 5% significance level, with impact effects of -0.0017 and 0.0071, respectively. Thus, although fiscal autonomy promotes GTFP, environmental autonomy leads to a GTFP decline in firms. The findings indicate that local economic autonomy negatively affects enterprises’ GTFP (β = −0.0026, p < 0.05), aligning with theoretical predictions concerning policy coordination challenges and substitution effects. This result contrasts with Song et al. [9], who found that fiscal decentralization positively impacts GTFP in China. The discrepancy arises because this study incorporates environmental autonomy (a binary policy shock) alongside fiscal autonomy, whereas Song et al. [9] focused solely on fiscal decentralization. These results align with the arguments of [46], suggesting that environmental decentralization may trigger a “race to the bottom” in regulatory stringency, ultimately undermining productivity. Notably, fiscal autonomy positively correlates with GTFP (β = 0.0071, p < 0.05), consistent with [22], who highlight fiscal autonomy’s role in improving resource allocation. Environmental autonomy, however, hinders GTFP (β = −0.0017, p < 0.05), corroborating [26]’s observation that decentralized environmental regulation may reduce innovation incentives. HighTech and Innovation Journal Vol. 6, No. 2, June, 2025 545 Table 3. Effects of different variables on the firm’s GTFP under the three models Models Variables (1) (2) (3) (4) (5) (6) General DID Synthetic DID Fixed effects model GTFP GTFP GTFP Auto -0.0023* (0.0012) -0.0026** (0.0013) EA -0.0015* (0.0008) -0.0017** (0.0008) FA 0.0029 (0.0029) 0.0071** (0.0031) Control variables No Yes No Yes No Yes Year fixed effect Yes Yes Yes Yes Yes Yes Firm fixed effect Yes Yes Yes Yes Yes Yes Sample size 7588 7588 7588 7588 7588 7588 Note. Standard errors in parentheses; *** p<0.01, ** p<0.05, * p<0.1 5.2. Robustness Tests 5.2.1. Parallel Trend Test A parallel trend test in the control and experimental groups was a prerequisite for using the DID approach. To assess the impact of the local government’s overall economic regulation autonomy using the DID method, it was necessary to ensure that the GTFP of enterprises in the control and experimental groups were aligned prior to the enactment of the new legislation. By constructing time dummy variables for pre- and post-implementation years and experimental group dummy variables, the model was estimated, yielding the results presented in Table 4. None of the interaction term coefficients prior to legislation implementation passed significance tests—an approximately similar trend was observed in changes in the GTFP between the control and experimental groups, which satisfied the assumption of parallel trends, validating the results (see Figure 3). Figure 3. Parallel trend test results 5.2.2. Placebo Test To eliminate the influence of omitted variables and random factors on the results, fictitious and randomized treatment groups were established, and the baseline model was subsequently re-run 1,000 times to ensure robustness. The scatterplot in Figure 4 displays the p-values of the AUTO coefficient across the randomized treatment groups, with the vertical dashed line representing the true AUTO coefficient of the actual treatment group and the horizontal dashed line indicating a p-value threshold of 0.1. Most estimated coefficients from the fictitious treatment groups cluster near zero and lie above the 10% significance level line, far from the true estimated coefficient, with only a few values approaching the true values. Consequently, the estimated coefficients from the fabricated treatment groups are not statistically significant, indicating that the results are not driven by random factors and are robust. HighTech and Innovation Journal Vol. 6, No. 2, June, 2025 546 Figure 4. Placebo test results 5.2.3. Propensity Score Matching–Difference-in-Differences (PSM⁃ DID) As the majority of the control group in this study comprised provincial capitals, municipalities directly governed by the central government, heterogeneity exists compared with ordinary prefecture-level cities. Therefore, to address endogeneity issues and improve the robustness of the experimental results, this study employs the propensity score matching (PSM) method to match the treated sample with cities that have similar characteristics in all aspects except for AUTO. Subsequently, the matched group was subjected to regression analysis. Table 4 displays the re-estimated results—even after controlling for the control group, AUTO’s estimated coefficient on GTFP remains significantly negative. Thereby reinforcing the robustness of the findings. Changing the sample As major events may directly or indirectly affect enterprise productivity, as research indicates that the 2008 economic crisis reduced firms’ innovation drive [47, 48], governments may relax their consideration of environmental issues and adopt more lenient intervention measures to stimulate economic recovery. During the COVID-19 pandemic, local governments imposed varying degrees of restrictions on businesses to protect the health of residents, which also impacted productivity. Therefore, this study excludes the years affected by major events and re-estimates the regression coefficients of local governments’ economic regulatory autonomy. Table 4 encapsulates the summarized findings. Column (3) in Table 4 presents the re-estimated results after excluding the data from 2008 to 2009, the years impacted by the global financial crisis. Column (4) presents the re-estimated results, with data from 2020 and 2021 excluded to eliminate the impact of the COVID-19 pandemic. Even after removing the years affected by major events, the estimated coefficient for local governments’ economic regulatory autonomy remains significantly negative—this study’s findings are not altered by the occurrence of major economic events. Table 4. PSM⁃DID, results, and changing samples results Models Variables (1) (2) (3) (4) PSM⁃ DID Excluding years affected by the economic crisis (2008, 2009) Excluding years affected by COVID-19 (2020, 2021) GTFP GTFP GTFP GTFP AUTO -0.0028* (0.0014) -0.0030** (0.0014) -0.0031** (0.0014) -0.0034** (0.0014) Constant 0.7657*** (0.0008) 0.9023*** (0.0237) 0.9215*** (0.0224) 0.8544*** (0.0202) Control variables No Yes Yes Yes Year fixed effect Yes Yes Yes Yes Firm fixed effect Yes Yes Yes Yes R2 0.9831 0.9834 0.9809 0.9827 Observations 5366 5356 6504 6504 Note. Standard errors in parentheses; *** p<0.01, ** p<0.05, * p<0.1 HighTech and Innovation Journal Vol. 6, No. 2, June, 2025 547 5.3. Mechanism Test 5.3.1. Mediating Effect and Masking Effect To examine whether autonomy affects firms’ GTFP through science and technology support and environmental regulations, this study employs the following three-step mediation model: 𝑀𝑒𝑑𝑖𝑡 = 𝛽0 + 𝛽1𝐴𝑈𝑇𝑂𝑖𝑡 + ∑𝜂𝑗𝐶𝑜𝑛𝑡𝑟𝑜𝑙𝑗𝑖𝑡 + 𝜇𝑖𝑡 + 𝜆𝑖𝑡 + 𝜀𝑖𝑡 (4) 𝐺𝑇𝐹𝑃𝑖𝑡 = 𝛾0 + 𝛾1𝐴𝑈𝑇𝑂𝑖𝑡 + 𝛾2𝑀𝑒𝑑𝑖𝑡 +∑𝜂𝑗𝐶𝑜𝑛𝑡𝑟𝑜𝑙𝑗𝑖𝑡 + 𝜇𝑖𝑡 + 𝜆𝑖𝑡 + 𝜀𝑖𝑡 (5) In this model, Med represents the mediating variable. In this study, science and technology support (Scitechsup) and environmental regulation (Enviregu) are used as mediating variables. Table 5 details the mediation effect outcomes. Column (2) reports the results of Equation 4—after controlling for other variables, local economic autonomy has a significant negative estimate coefficient at the 1% level for Scitechsup (-0.0050). Column (3) displays the test results for Equation 5, showing that the two-period lagged Scitechsup has a significantly positive estimate coefficient at the 10% level for GTFP (0.0333). This suggests that government autonomy reduces the intensity of science and technology support and negatively affects enterprises’ GTFP, validating Hypothesis 2. Columns (4) and (5) show the results of Equations 4 and 5, using Enviregu as the mediating variable. The estimated coefficient of AUTO on Enviregu is 0.0110, significant at the 1% level. In Column (5), the estimated coefficient of Enviregu’s impact on GTFP is 0.0224 at the 5% significance level, but the estimated coefficient of AUTO is significantly negative at the 1% level—local government autonomy reduces its negative impact on GTFP by promoting the improvement of the Enviregu level. Therefore, the final total effect of AUTO on GTFP (− 0.0026) is weaker than its direct effect (− 0.0029); expressly, there is a masking effect, which verifies Hypothesis 3. The negative effect of autonomy on science and technology support (Scitechsup, β = −0.0050, p < 0.01) and positive effect on environmental regulation (Enviregu, β = 0.0110, p < 0.01) reveal a policy substitution effect. This aligns with [6] framework, where government streams (e.g., environmental regulation) interact with technical streams (e.g., innovation). Specifically, the substitution effect reflects a trade-off between short-term regulatory outcomes (government stream) and long-term innovation investments (technical stream), as identified by Addy & Jiří [36]. The masking effect (β = −0.0029 vs. −0.0026) suggests environmental regulation partially offsets autonomy’s negative impact, consistent with [20]’s finding that moderate environmental regulation can enhance efficiency. However, the net negative effect underscores the dominance of innovation suppression over regulatory gains, echoing [21] critique of compliance costs. Table 5. Results of mediating and masking effects Models Variables (1) (2) (3) (4) (5) GTFP Mediating effect of research support Inhibitory effect of environmental regulation Scitechsup GTFP Enviregu GTFP AUTO -0.0026** (0.0013) -0.0050*** (0.0010) -0.0031** (0.0014) 0.0110*** (0.0016) -0.0029*** (0.0013) Scitechsup 0.0333* (0.0189) Enviregu 0.0224** (0.0094) Constant 0.8564*** (0.0183) 0.0304*** (0.0005) 0.9194*** (0.0224) 0.2296*** (0.0232) 0.8513*** (0.0184) Control variables Yes Yes Yes Yes Yes Year fixed effect Yes Yes Yes Yes Yes Firm fixed effect Yes Yes Yes Yes Yes R2 0.9864 0.1966 0.9809 0.3011 0.9864 Sample size 7588 Note. Standard errors in parentheses; *** p<0.01, ** p<0.05, * p<0.1. In conducting the mechanism test, the objective is to demonstrate that autonomy influences GTFP through its effect on support for science and technology. Given that the impact of science and technology on GTFP may exhibit a lagged effect rather than manifesting in the same year, the two-period lagged Scitechsup variable was employed to verify its mediating effect. 5.3.2. Substitution Effect of Policies To verify whether an increase in local economic regulation autonomy results in unidirectional policy substitution effects, Scitechsup serves as the mediating variable, and the influence of autonomy on environmental regulation is examined in three steps, following an approach similar to a mediation effect model. HighTech and Innovation Journal Vol. 6, No. 2, June, 2025 548 𝐸𝑛𝑣𝑖𝑟𝑒𝑔𝑢𝑖𝑡 = 𝛼0 + 𝛼1𝐴𝑈𝑇𝑂𝑖𝑡 + ∑𝜂𝑗𝐶𝑜𝑛𝑡𝑟𝑜𝑙𝑗𝑖𝑡 + 𝜇𝑖𝑡 + 𝜆𝑖𝑡 + 𝜀𝑖𝑡 (6) 𝑆𝑐𝑖𝑡𝑒𝑐ℎ𝑠𝑢𝑝𝑖𝑡 = 𝛽0 + 𝛽1𝐴𝑈𝑇𝑂𝑖𝑡 + ∑𝜂𝑗𝐶𝑜𝑛𝑡𝑟𝑜𝑙𝑗𝑖𝑡 + 𝜇𝑖𝑡 + 𝜆𝑖𝑡 + 𝜀𝑖𝑡 (7) 𝐸𝑛𝑣𝑖𝑟𝑒𝑔𝑢𝑖𝑡 = 𝛾0 + 𝛾1𝐴𝑈𝑇𝑂𝑖𝑡 + 𝛾2𝑆𝑐𝑖𝑡𝑒𝑐ℎ𝑠𝑢𝑝𝑖𝑡 +∑𝜂𝑗𝐶𝑜𝑛𝑡𝑟𝑜𝑙𝑗𝑖𝑡 + 𝜇𝑖𝑡 + 𝜆𝑖𝑡 + 𝜀𝑖𝑡 (8) Among them, α1 represents the overall effect of AUTO on environmental regulation; β1 in Equation 7 denotes the impact of AUTO on science and technology support (policy to be substituted); γ1 in Equation 8 represents the direct impact of AUTO on environmental regulation after controlling for the influence of science and technology support and other control variables, whereas γ2 reflects the extent to which environmental regulation responds to science and technology support after controlling for the influence of AUTO and other control variables, indicating a potential substitution effect. Table 6 presents the experimental results. As examined within the context of mediating and masking effects, the findings indicate that greater local economic autonomy leads governments to reduce support for science and technology while increasing environmental regulation. Additionally, Table 6 shows that the estimated coefficient of Scitechsup on Enviregu is significantly negative, approximately -0.2535. Given that both estimated coefficients β1 and γ2 are significantly negative at the 1% level, it can be inferred that local governments employ environmental regulation as a substitute for science and technology support, thereby validating Hypothesis 4. Table 6. Results of substitution effect Variables (1) (2) (3) Enviregu Scitechsup Enviregu AUTO 0.0110*** (0.0017) -0.0050*** (0.0010) 0.0097*** (0.0016) Scitechsup -0.2535*** (0.0202) Constant 0.2296*** (0.0232) 0.0304*** (0.0005) 0.2018*** (0.0230) Control variables Yes Yes Yes Year fixed effect Yes Yes Yes Firm fixed effect Yes Yes Yes R2 0.3011 0.1966 0.3164 Sample size 7588 Note. Standard errors in parentheses; *** p<0.01, ** p<0.05, * p<0.1 5.3.3. Cross-Section Effect Test To explore how regional heterogeneity affects the correlation between government behavior and corporate performance, this study measures industrial structural upgrading (Stru) by employing the tertiary-to-secondary industry value-added ratio. The sample is stratified into two groups at the median value: cities with industrial structures below the median and those above it. After conducting a group regression analysis on the baseline model, as shown in Table 7, a significant negative correlation between AUTO and GTFP was found in cities with lower levels of industrial structure, with an estimated regression coefficient of -0.0030 at a significance level of 10%. However, no similar significant negative impact was observed for economic regulation autonomy on corporate GTFP in cities with more industrial structures. This difference verifies Hypothesis 5, suggesting that industrial structural upgrading helps mitigate the negative effects of local government regulatory autonomy on corporate GTFP. The mitigating role of industrial structure upgrading (Struc ≥ 1.17) aligns with [15], who emphasize that tertiary industries reduce pollution intensity and foster innovation. This result extends [16] work on marketization, demonstrating that structural upgrading enhances policy coordination and reduces regulatory burdens. In contrast, regions with lower industrial structure (Struc < 1.17) exhibit amplified negative effects, corroborating [38] analysis of innovation agglomeration in technology-intensive sectors. HighTech and Innovation Journal Vol. 6, No. 2, June, 2025 549 Table 7. Results of heterogeneity test Groups Variables (1) (2) Struc < 1.17 Struc ≥ 1.17 GTFP GTFP AUTO -0.0030* (0.0017) 0.0005 (0.0106) Constant 0.9285*** (0.0276) 0.8814*** (0.0323) Control variables Yes Yes Year fixed effect Yes Yes Firm fixed effect Yes Yes R2 0.9834 0.9806 Sample size 3795 3793 Note. Standard errors in parentheses; *** p<0.01, ** p<0.05, * p<0.1 6. Conclusions 6.1. Main Findings Exploring the relationship between government behavior and corporate economic performance holds significant importance for advancing sustainable economic development. Using sample data from Chinese listed companies from 2007 to 2021, this study calculates enterprise GTFP with a non-radial SBM-ML index and thoroughly examines the impact of local governments’ economic regulation autonomy on enterprises’ GTFP across two dimensions: environmental and fiscal decentralization. Furthermore, the analysis investigates the roles of different types of government policies and regional differences in industrial structure in shaping government-enterprise relations. Based on the findings of this study, three principal conclusions are identified. First, based on a difference-in-differences model, fiscal and environmental decentralization were used to measure the autonomy of local governments’ economic regulation. The results show that an increase in local government autonomy has a hindering effect on enterprises GTFP. This is because increased autonomy reduces policy coordination between regions and reduces dependence on the central government; intensified competition between regions leads to increased corruption and excessive intervention in economic activities. Separately, fiscal autonomy promotes enterprises’ green total factor, whereas environmental autonomy hinders its improvement. Second, owing to the existence of policy substitution effects, local governments weighed their options between environmental regulations and support for science and technology. Higher levels of science and technology support imply lower levels of environmental regulation. The increased autonomy of local governments effects greater policy discretion, making them more willing to enhance environmental regulations than science and technology support. Improvements in environmental regulations are also partly aimed at compensating for the reduction in science and technology support. Therefore, autonomy leads to an increase in the level of environmental regulation and reduces the negative impact on GTFP. However, autonomy decreases science and technology support and suppresses corporate GTFP. Finally, considering the differentiated impact of regional industrial structure levels, evidence shows that in regions with lower levels of industrial structure, government autonomy continues to significantly inhibit enterprises’ GTFP. However, this negative effect is absent in regions with higher levels of industrial structure, suggesting that upgrading the regional industrial structure influences local governments’ economic regulatory behavior. 6.2. Policy Implications This article provides several policy recommendations. First, in contrast to the central government, local governments have a clear information advantage in terms of local economic development and are closer to local businesses and residents. It is crucial for them to fully leverage this information and proximity advantage by adopting reasonable policies and guaranteeing the efficacious execution of central policies. As local government autonomy increases, they should enhance cooperation among regions and improve coordination between local policies as well as between local and central policies. They should also consider positive externalities when formulating policies, actively implement those with higher positive externalities, and promote the efficient allocation of resources. For example, increasing fiscal support for scientific and technological research can reduce the cost of innovation for businesses, enhancing their GTFP. HighTech and Innovation Journal Vol. 6, No. 2, June, 2025 550 Second, considering the differentiated impacts of fiscal and environmental decentralization, as well as their combined hindrance to the GTFP of enterprises, it is not appropriate to simply consider the positive influence of fiscal decentralization on corporate economic performance. When designing decentralization reform plans, the central government should comprehensively consider the differentiated effects of administrative system reform and economic development integration under different decentralization conditions on microeconomic entities’ input-output efficiency and resource allocation optimization at the local level. For example, by combining carbon emissions accounting with carbon footprint measurements, efforts should be made to strengthen the coordination between government governance and enterprise green development. Third, as the supervisory bodies of the regional economy, local governments do not participate directly in the production and operational activities of enterprises. They lack sufficient understanding of enterprise production and operations. Therefore, local governments should fully coordinate the relationship between government and enterprises, reduce excessive intervention in enterprises, strengthen integrity construction, and lower rent-seeking costs for businesses to provide a favorable political environment. Concurrently, the government must balance at least two objectives: environmental protection and economic growth. Local governments should focus on economic growth and take responsibility for environmental pollution issues, as its deterioration will hinder economic development. It is unacceptable to sacrifice the environment by allowing unrestricted pollution emissions from companies or by excessively restricting business activities owing to environmental protection concerns. Fourth, enhancing the industrial structure not only amplifies the technological innovation agglomeration effect but also eases the difficulty of environmental regulations. This further reduces the risk associated with government science and technology support. Local governments should take appropriate measures to encourage the transformation of the industrial structure from resource- and labor-intensive to knowledge- and talent-intensive. Short-term policies include providing better financial subsidies and preferential tax policies for tertiary industries, whereas long-term policies can consider improving mechanisms for attracting talent and promoting higher levels of human capital, providing a talent foundation for upgrading the industrial structure. 6.3. Limitations and Prospects The impact of local government autonomy on the economy has several aspects. The findings of this study are derived solely from Chinese corporate sample data, without considering differences in political systems and economic structures among countries, thus limiting their generalizability. Baskaran et al. [33] analyzed 31 studies on the impact of fiscal decentralization on the economy and find that such an impact is influenced by regional heterogeneity and the methodologies adopted by researchers. Therefore, subsequent research ought to concentrate on exploring how the factors related to political systems and economic structures contribute to governments’ influence on firms’ GTFP. Furthermore, this study measured the autonomy of local governments in economic regulation using two indicators closely related to sustainable development: fiscal autonomy and environmental autonomy. More indicators could be incorporated to enhance the measurement of local governments’ economic autonomy. Finally, due to the diversity in government policy directions, local governments not only make trade-offs between support for scientific and technological innovation and environmental regulations but may also have substitution relationships in other areas. Further investigation into policy substitution effects would contribute to a better understanding of governmental behavior and provide guidance for innovative governance models. 7. Declarations 7.1. Data Availability Statement The data presented in this study are available in the article. 7.2. Funding The author received no financial support for the research, authorship, and/or publication of this article. 7.3. Institutional Review Board Statement Not applicable. 7.4. Informed Consent Statement Not applicable. 7.5. Declaration of Competing Interest The author declares that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. HighTech and Innovation Journal Vol. 6, No. 2, June, 2025 551 8. References [1] Ostfeld, R. S., & Brunner, J. L. (2015). Climate change and Ixodes tick-borne diseases of humans. Philosophical Transactions of the Royal Society B: Biological Sciences, 370(1665), 1–11. doi:10.1098/rstb.2014.0051. [2] Guo, B., Yu, H., & Jin, G. (2024). Urban green total factor productivity in China: A generalized Luenberger productivity indicator and its parametric decomposition. Sustainable Cities and Society, 106, 105365. doi:10.1016/j.scs.2024.105365. [3] Huang, X., Feng, C., Qin, J., Wang, X., & Zhang, T. (2022). Measuring China’s agricultural green total factor productivity and its drivers during 1998–2019. Science of the Total Environment, 829, 154477. doi:10.1016/j.scitotenv.2022.154477. [4] Li, Y., & Chen, Y. (2021). Development of an SBM-ML model for the measurement of green total factor productivity: The case of pearl river delta urban agglomeration. Renewable and Sustainable Energy Reviews, 145. doi:10.1016/j.rser.2021.111131. [5] Shi, X., & Li, L. (2019). Green total factor productivity and its decomposition of Chinese manufacturing based on the MML index:2003–2015. Journal of Cleaner Production, 222, 998–1008. doi:10.1016/j.jclepro.2019.03.080. [6] Zhang, J., Lu, G., Skitmore, M., & Ballesteros-Pérez, P. (2021). A critical review of the current research mainstreams and the influencing factors of green total factor productivity. Environmental Science and Pollution Research, 28(27), 35392–35405. doi:10.1007/s11356-021-14467-4. [7] Jiakui, C., Abbas, J., Najam, H., Liu, J., & Abbas, J. (2023). Green technological innovation, green finance, and financial development and their role in green total factor productivity: Empirical insights from China. Journal of Cleaner Production, 382, 135131. doi:10.1016/j.jclepro.2022.135131. [8] Lee, C. C., & Lee, C. C. (2022). How does green finance affect green total factor productivity? Evidence from China. Energy Economics, 107, 105863. doi:10.1016/j.eneco.2022.105863. [9] Song, M., Du, J., & Tan, K. H. (2018). Impact of fiscal decentralization on green total factor productivity. International Journal of Production Economics, 205, 359–367. doi:10.1016/j.ijpe.2018.09.019. [10] Yu, D., Liu, L., Gao, S., Yuan, S., Shen, Q., & Chen, H. (2022). Impact of carbon trading on agricultural green total factor productivity in China. Journal of Cleaner Production, 367, 132789. doi:10.1016/j.jclepro.2022.132789. [11] Zhang, C., Zhu, H., & Li, X. (2024). Which productivity can promote clean energy transition —total factor productivity or green total factor productivity? Journal of Environmental Management, 366, 121899. doi:10.1016/j.jenvman.2024.121899. [12] Hao, X., Wang, X., Wu, H., & Hao, Y. (2023). Path to sustainable development: Does digital economy matter in manufacturing green total factor productivity? Sustainable Development, 31(1), 360–378. doi:10.1002/sd.2397. [13] Hu, W., & Li, X. (2023). Financial Technology Development and Green Total Factor Productivity. Sustainability (Switzerland), 15(13), 10309. doi:10.3390/su151310309. [14] Wu, H., Hao, Y., Ren, S., Yang, X., & Xie, G. (2021). Does internet development improve green total factor energy efficiency? Evidence from China. Energy Policy, 153, 112247. doi:10.1016/j.enpol.2021.112247. [15] Feng, Y., Zhong, S., Li, Q., Zhao, X., & Dong, X. (2019). Ecological well-being performance growth in China (1994–2014): From perspectives of industrial structure green adjustment and green total factor productivity. Journal of Cleaner Production, 236. doi:10.1016/j.jclepro.2019.07.031. [16] Lu, X. hai, Jiang, X., & Gong, M. qi. (2020). How land transfer marketization influence on green total factor productivity from the approach of industrial structure? Evidence from China. Land Use Policy, 95. doi:10.1016/j.landusepol.2020.104610. [17] Ren, S., Du, M., Bu, W., & Lin, T. (2023). Assessing the impact of economic growth target constraints on environmental pollution: Does environmental decentralization matter? Journal of Environmental Management, 336, 117618. doi:10.1016/j.jenvman.2023.117618. [18] Stavins, R. N. (2025). Environmental Regulation and the Competitiveness of U.S. Manufacturing: What Does the Evidence Tell Us? Environmental Economics and Public Policy, 33(1), 45–76. doi:10.4337/9781035363155.00011. [19] Song, M., Peng, L., Shang, Y., & Zhao, X. (2022). Green technology progress and total factor productivity of resource-based enterprises: A perspective of technical compensation of environmental regulation. Technological Forecasting and Social Change, 174, 121276. doi:10.1016/j.techfore.2021.121276. [20] Cao, Y., Wan, N., Zhang, H., Zhang, X., & Zhou, Q. (2020). Linking environmental regulation and economic growth through technological innovation and resource consumption: Analysis of spatial interaction patterns of urban agglomerations. Ecological Indicators, 112(5), 106062. doi:10.1016/j.ecolind.2019.106062. [21] Becker, R. A. (2011). Local environmental regulation and plant-level productivity. Ecological Economics, 70(12), 2516–2522. doi:10.1016/j.ecolecon.2011.08.019. HighTech and Innovation Journal Vol. 6, No. 2, June, 2025 552 [22] Martinez-Vazquez, J., Lago-Peñas, S., & Sacchi, A. (2017). the Impact of Fiscal Decentralization: a Survey. Journal of Economic Surveys, 31(4), 1095–1129. doi:10.1111/joes.12182. [23] Chen, X., & Chang, C. P. (2020). Fiscal decentralization, environmental regulation, and pollution: a spatial investigation. Environmental Science and Pollution Research, 27(25), 31946–31968. doi:10.1007/s11356-020-09522-5. [24] Mu, R. (2018). Bounded rationality in the developmental trajectory of environmental target policy in China, 1972-2016. Sustainability (Switzerland), 10(1), 199. doi:10.3390/su10010199. [25] Shah, A. (2006). Corruption and decentralized public governance. In Handbook of Fiscal Federalism. World Bank. doi:10.4337/9781847201515.00029. [26] Konisky, D. M. (2009). Assessing U.S. state susceptibility to environmental regulatory competition. State Politics and Policy Quarterly, 9(4), 404–428. doi:10.1177/153244000900900402. [27] Hwang, S. (2022). Is There an Environmental Race to the Bottom in an Endogenous Growth Model of Interjurisdictional Competition? Hitotsubashi Journal of Economics, 63(1), 24–50. doi:10.15057/hje.2022002. [28] Prud’homme, R. (1995). The dangers of decentralization. World Bank Research Observer, 10(2), 201–220. doi:10.1093/wbro/10.2.201. [29] Oates, W. E. (2013). A Reconsideration of Environmental Federalism. Recent Advances in Environmental Economics, Edward Elgar Publishing, 1-32. doi:10.4337/9781843767305.00006. [30] Oates, W. E. (2008). On the evolution of fiscal federalism: Theory and institutions. National Tax Journal, 61(2), 313–334. doi:10.17310/ntj.2008.2.08. [31] Lejour, A. M., & Verbon, H. A. A. (1997). Tax Competition and Redistribution in a Two-Country Endogenous-Growth Model. International Tax and Public Finance, 4(4), 485–497. doi:10.1023/A:1008613015303. [32] Devereux, M. B., & Mansoorian, A. (1992). International Fiscal Policy Coordination and Economic Growth. International Economic Review, 33(2), 249. doi:10.2307/2526893. [33] Baskaran, T., Feld, L. P., & Schnellenbach, J. (2016). Fiscal federalism, decentralization, and economic growth: a meta‐ analysis. Economic Inquiry, 54(3), 1445-1463. doi:10.1111/ecin.12331. [34] Li, Y., & Tina Zhang, X. (2023). Rent-seeking in bank credit and firm R&D innovation: The role of industrial agglomeration. Journal of Business Research, 159, 113454. doi:10.1016/j.jbusres.2022.113454. [35] Wouters, O. J., McKee, M., & Luyten, J. (2020). Estimated Research and Development Investment Needed to Bring a New Medicine to Market, 2009-2018. JAMA - Journal of the American Medical Association, 323(9), 844–853. doi:10.1001/jama.2020.1166. [36] Addy, T., & Jiří, S. (2001). The political economy of transition. Ecumenical Review, 53(4), 501–508. doi:10.1111/j.1758- 6623.2001.tb00135.x. [37] Khan, Z., Ali, S., Dong, K., & Li, R. Y. M. (2021). How does fiscal decentralization affect CO2 emissions? The roles of institutions and human capital. Energy Economics, 94, 105060. doi:10.1016/j.eneco.2020.105060. [38] Wang, J., Dong, X., & Dong, K. (2022). How does ICT agglomeration affect carbon emissions? The case of Yangtze River Delta urban agglomeration in China. Energy Economics, 111. doi:10.1016/j.eneco.2022.106107. [39] Zhao, J., Jiang, Q., Dong, X., Dong, K., & Jiang, H. (2022). How does industrial structure adjustment reduce CO2 emissions? Spatial and mediation effects analysis for China. Energy Economics, 105, 105704. doi:10.1016/j.eneco.2021.105704. [40] Sanogo, T. (2019). Does fiscal decentralization enhance citizens’ access to public services and reduce poverty? Evidence from Côte d’Ivoire municipalities in a conflict setting. World Development, 113, 204–221. doi:10.1016/j.worlddev.2018.09.008. [41] Ladner, A., & Keuffer, N. (2021). Creating an index of local autonomy–theoretical, conceptual, and empirical issues. Regional and Federal Studies, 31(2), 209–234. doi:10.1080/13597566.2018.1464443. [42] Nunn, N., & Qian, N. (2011). The potato’s contribution to population and urbanization: Evidence from a historical experiment. Quarterly Journal of Economics, 126(2), 593–650. doi:10.1093/qje/qjr009. [43] Chung, Y. H., Färe, R., & Grosskopf, S. (1997). Productivity and undesirable outputs: A directional distance function approach. Journal of Environmental Management, 51(3), 229–240. doi:10.1006/jema.1997.0146. [44] Shao, S., Hu, Z., Cao, J., Yang, L., & Guan, D. (2020). Environmental Regulation and Enterprise Innovation: A Review. Business Strategy and the Environment, 29(3), 1465–1478. doi:10.1002/bse.2446. HighTech and Innovation Journal Vol. 6, No. 2, June, 2025 553 [45] Arkhangelsky, D., Athey, S., Hirshberg, D. A., Imbens, G. W., & Wager, S. (2021). Synthetic difference-in-differences. American Economic Review, 111(12), 4088-4118. doi:10.1257/aer.20190159. [46] Li, X., Younas, M. Z., Andlib, Z., Ullah, S., Sohail, S., & Hafeez, M. (2021). Examining the asymmetric effects of Pakistan’s fiscal decentralization on economic growth and environmental quality. Environmental Science and Pollution Research, 28(5), 5666–5681. doi:10.1007/s11356-020-10876-z. [47] Archibugi, D., Filippetti, A., & Frenz, M. (2013). Economic crisis and innovation: Is destruction prevailing over accumulation? Research Policy, 42(2), 303–314. doi:10.1016/j.respol.2012.07.002. [48] Hausman, A., & Johnston, W. J. (2014). The role of innovation in driving the economy: Lessons from the global financial crisis. Journal of Business Research, 67(1), 2720–2726. doi:10.1016/j.jbusres.2013.03.021.