Introduction

Since the Industrial Revolution, the overexploitation of natural resources by human beings in pursuit of rapid economic development has led to a rise in the content of CO2 in the atmosphere1, causing global warming. Global warming triggers extreme weather such as glacier melting, sea level rise, extreme heat and drought2, which seriously threatens the human survival. The effects of global 1warming on the ecological environment have gradually emerged, and CO2 emissions have attracted worldwide attention3. Approximately 200 countries and organizations around the world have signed the Paris Agreement to jointly limit rapid temperature increases and make commitments to reduce emissions. The U.S. has enacted the Clean Air Act4 etc. The EU has formulated carbon pricing policies. China has built a “one + N” policy system of emission reduction goals and has formed a carbon pricing mechanism5. In addition, relevant emission reduction policies, such as pilot low carbon cities, green credit policies and environmental protection tax laws, have been introduced. Among them, NHTDZs also play a vital role in facilitating emission reduction and green low-carbon development.

To accelerate green low-carbon development, NHTDZs play a leading and radiating role in this process. It has become a model of economic transformation and an important engine of green low carbon development. NHTDZs converge many National High-tech Enterprises(HNTEs), forming a strong atmosphere of innovation and possessing fruitful innovation results. In recent years, the innovation capacity and the quantity of patents in NHTDZs have been steadily increasing, indicating that technology is constantly advancing. Technological progress can reduce energy costs and energy service prices6, decreasing the use of energy and reducing CO2 emissions7,8,9. According to the staffing structure in the NHTDZs, the education level of personnel is climbing, which indicates that human capital is accumulating. The accumulation of human capital can inspire inventions10, replace inputs between fossil and non-fossil energy sources11, and reduce energy consumption10 and CO2 emissions. CO2 emission reduction is a continuous process that requires continuous capital investment. The capital investment within the NHTDZs has been increasing annually. Science expenditures promote the advancement of energy-saving technology, which promotes emission reduction12. It also promotes the development of low-carbon transportation systems, improves traffic management and logistics patterns, increases transportation efficiency and reduces carbon emissions. In addition, government financial subsidies for NHTDZs can guide enterprises toward low-carbon development and promote the production of greening products3. Therefore, clarifying the role played by NHTDZs in low-carbon economic transformation provides the necessary empirical support for realizing the win–win goal of ecological protection and economic development.

Studies on CO2 emission reduction have focused on the following aspects. Some scholars argue that the relationship between CO2 emissions and economic growth satisfies the EKC hypothesis. CO2 emissions decrease as economic growth13,14,15,16. Others argue that the relationship between economic growth and CO2 emissions are positive correlation17,18,19. From the macroeconomic perspective, some studies have considered the effects of human capital level20,21,22,23,24, urbanization25,26, foreign inward investment(FID)27,28,29, the digital economy30, digital finance31, population aging32, green innovation33,34, industrial structure optimization35 and energy structure adjustment36 on CO2 emissions. Zhang Tao's study showed that technology transfer in European countries can reduce their own carbon emissions, but increase carbon emissions in Asian countries37. In addition, policies are one of the influencing factors. Cheng J et al. illustrated that low-carbon city pilots reduce carbon emissions through the transformation of technological effects into green technological progress and structural effects38. Allocating resources to green technology innovations can improve environmental sustainability39. Studies by Cheng Z et al. and Xu A et al. have shown that smart cities have a carbon reduction effect40,41. Lingxuan Liu found that industrial symbiosis and renewable energy use can reduce carbon emissions in parks42. Yizheng Lyu studied the impact of four typical industrial parks in China on carbon emission reduction based on the land-industry-carbon integration model43. Xiang Yu et al. studied the effects of 20 national low-carbon pilot zones on CO2 emissions during 2012–201644. Qian L et al. found that there is a suppression effect between NHTDZs and urban carbon emissions. This suppression effect has some lag and spatial spillover effects45. Sun Y and Woldesilassie found that NHTDZs improve innovation capacity to reduce CO2 emissions under government guidance46,47. Li X, and Wang (2023) showed that the NHTDZs policy will reduce per capita CO2 emissions48.

The research on carbon emission reduction is relatively rich. This study establishes a framework for further study. Considering the variations in research contents and perspectives, there is still some research space. This study empirically investigated the effects of establishing NHTDZs on CO2 emissions in China, using a multitemporal asymptotic difference approach. This demonstrates that the establishment of NHTDZs clearly reduces CO2 emissions about 2.53% after controlling influence factors. This indicates that NHTDZs achieve the goal of decreasing CO2 emissions.

The research contributions are as follows. First, the environmentally friendly and sustainable development of NHTDZs has become a key topic of study in recent years. The literature has been conducted in foreign countries, while little literature has been transferred to the Chinese context. This paper focused on the importance of the environmental aspects of establishing NHTDZs and systematically explored the CO2 emission reduction effect of NHTDZs policy, supplementing the assessment of the effect of establishing NHTDZs. Second, this paper adopts apparent carbon emission data from the China Carbon Accounting Database (CADS) to provide more evidence for the CO2 emission reduction effect of NHTDZs. Third, this research reveals the mechanism by which NHTDZs reduce CO2 emissions. Existing studies mainly focus on innovation level, environmental regulations49 and energy48. This paper additionally analyzes this topic from the perspective of human capital level and R&D investment. This approach supplements the transmission mechanism. This study provides certain reference opinions for the government on the development of NHTDZs.

The rest of the paper is organized as follows: part two provides the policy background and theoretical analysis; part three presents the research data and econometric modeling setup; part four reports the empirical results and robustness tests; part five conducts the mechanism test; part six conducts the heterogeneity analysis; and part seven presents the conclusions and policy recommendations.

Policy context and theoretical mechanisms

Policy context of NHTDZs

As an effective way to promote the integration of science technology industries, science technology parks first appeared in the United States and gradually became popular around the world. To grasp the scientific and technological revolution, China gave birth to the idea of establishing NHTDZs. In 1988, the State Council encouraged the establishment of NHTDZs in intellectually intensive areas. NHTDZs rely on domestic science and technology, economic strength, the reform of preferential policies, and the ability to fully absorb advanced foreign scientific technological resources, to achieve local optimization of hard and soft environments. In the same year, the State Council approved the establishment of the first Beijing NHTDZ and carried out the Torch Plan. Under the impetus of the Torch Program, various places have combined local characteristics and conditions to actively create NHTDZs, which opened the prelude to the construction of NHTDZs.

After that, in 1991 and 1992, the State Council approved the construction of 51 NHTDZs in two stages, which formed the preliminary scale of construction of NHTDZs. After 2007, the State Council approved the construction of new NHTDZs at different stages. Especially after 2012, the speed of construction of NHTDZs further accelerated. By the end of 2022, the number of NHTDZs will reach 173. The time of establishment are shown in Fig. 1.

Figure 1
figure 1

Number of NHTDZs number.

Full size image

At the early stage of reform and opening to the outside world, China's industrial base was weak. The high-tech industry was basically blank, taking the development path of "industry first". At this stage, the construction of NHTDZs focused on production factors. The construction path is to create hard conditions for parks to carry out production and attract investment, with the formation of the industrial base and economic scale as the main construction goals. China's economy started to develop. CO2 emissions were low. With the expanding scale of economic construction, the state began to realize the initial purpose of constructing NHTDZs. In 2001, the slogan of second ventures was proposed. It was proposed that the construction of NHTDZs should focus on promoting the "five transformations". NHTDZs around the world choose reasonable leading industries according to their own development characteristics and resource endowments.

For example, Xiamen Torch NHTDZ focuses on the development of strategic emerging industries; Jining NHTDZ focuses on the allocation of human resources and innovation platform resources; and Suzhou NHTDZ focuses on incubating environmentally friendly innovative enterprises.

During this period, NHTDZs achieved significant results in terms of innovation, talent introduction, and the transformation of scientific and technological fruits. Figures2, 3, 4 present the NHTDZs in terms of innovation, talent accumulation and R&D expenditures.

Figure 2
figure 2

Innovation.

Full size image
Figure 3
figure 3

NHTDZs staff.

Full size image
Figure 4
figure 4

Expenditures.

Full size image

As a "testing ground" for national progressive reform, the NHTDZs are driven by the national mission23,48. This process realizes the efficient recycling of resources with the specialized agglomeration of HNTEs. It adheres to circular economy and intensive development as a means of promoting coordinated environmental and ecological development. In recent years, NHTDZs have vigorously developed in term of atmospheric control, greening and energy conservation. For air pollution management, the Wuhan East Lake NHTDZ is at the leading level in air management in China. Nanning NHTDZ has greatly reduced CO2 emissions, sulfur dioxide and other gases by replacing coal with biomass as boiler fuel. On the greening side, Suzhou NHTDZ has created green projects to realize 2.5 million square meters of new urban green space. On the energy conservation front, the Zhaoqing NHTDZ vigorously promoted photovoltaic power generation; and the Guiyang NHTDZ introduced energy-saving science technology enterprises, realizing an annual emission reduction of 1.1 million tons of CO2. By strengthening ecological construction and environmental protection, NHTDZs are actively promoting low-carbon economic development.

In view of this, the authors used the establishment of NHTDZs as a natural experiment and selected samples collected between 2003 and 2019 to explore carbon emission reduction effects. After 2019, due to the new coronavirus epidemic, restrictions on economic activity and production resulted in reduced CO2 emissions50,51,52, which need to be excluded from this shock.

Theoretical mechanisms

NHTDZs, as entrepreneurial highlands, talent highlands and science technology bases, are important for China's strategic development. Its core competitiveness is innovation capacity. NHTDZs can improve the environment by combining human capital with innovation53, reducing carbon emissions54. Human capital has a positive moderating effect on the environmental impact of green innovation53.

Human capital stimulates society's willingness to use energy saving and environmental technologies. It increases individual productivity to achieve emission reductions55. The improvement in China's human capital is largely attributable to increased educational attainment56. The education level of practitioners in China's NHTDZs is international first-class and team structure is constantly optimized (see Fig. 4). NHTDZs drive regional transformation and upgrading through the improvement of the human capital structure and industrial structure from labor-intensive and low-value-added industries to capital-intensive and high-value-added industries, realizing green development.

NHTDZs have attracted ample HNTEs and innovative talent through preferential policies. Through financial incentives, loan interest subsidies, project grants, the transfer of rights and interests and risk compensation, NHTDZs are growing quickly. For example, the Changsha NHTDZ arranges 500 million yuan annually as a science technology innovation and industrial development fund. In addition, government subsidies are considered labels that increase companies’ recognition of capital markets, and enterprises will go further into green development planning57. The following hypothesis is proposed.

H1: The establishment of NHTDZs can reduce CO2 emissions.

HNTEs are the main innovation force. Technological innovation can transform traditional factor-driven approaches into innovation-driven development approaches58. According to Schumpeter, technological change enters the production process in the form of invention. It is possible to improve environmental quality when production processes are optimized59. Endogenous growth theory suggests that the function of technological change is crucial in the economic growth process. In the past, nonrenewable energy sources were used more frequently. Economic growth may increase in carbon emissions, leading to environmental degradation60. As an effective means to combat climate change, innovation can improve the efficiency of energy utilization61, promote industrial upgrading62 and reduce carbon emissions by replacing fossil fuels with cleaner energy sources63.

From the perspective of technological innovation, through the implementation of green innovation strategy, enterprises carry out comprehensive greening. They reduced carbon emissions through end-of-pipe technologies, cleaner production technologies, carbon capture and other "negative emission" technologies64,65. The application of clean energy decreases public health risks66,67 and the consumption of solid fuel and improves air and water quality66,67, thereby reducing air pollution68. The distributed photovoltaic power generation project promoted by the Zhaoqing NHTDZ optimized the energy structure. The annual power generation will reach 21.33 million kWh, which will reduce CO2 emissions by 17,512 tons of annually. The following hypothesis is proposed.

H2: The establishment of NHTDZs facilitates CO2 reduction by enhancing innovation levels.

Human capital, as a country's soft power, can also reduce CO2 emissions69,70. Human capital theory suggests that the evolution of factor endowment structure changes regional production patterns and development trends. Endogenous growth theory suggests that knowledge spillovers can generate innovations that promote economic development71.

The most intuitive manifestation of human capital is investment in higher education10. Formal education is the main way of acquiring knowledge, skills and abilities. It also affects people’s attitudes and behaviors toward environmental ecology72. An increased education level produces income effects and changes cognitive abilities. The Energy consumption structure changes, which reduces the use of nonrenewable energy sources73,74. The inputs of energy can decrease with increasing human resources, while the total output remains constant75.

Human resources improve industrial structure, shifting from primary industry senior human capital and industrial primary technology to secondary and tertiary senior human capital and new cutting-edge technology, optimizing the allocation of production factors and promoting energy conservation77. The knowledge spillover effect and teamwork effect of human capital enhance the knowledge stock of enterprises and promote the absorption of foreign pollution control and clean technology. According to Nelson and Phelps, a country's ability to introduce and use new technology comes from its domestic human capital stock. A high level of human resources can drive the realization of emission reduction targets78. The following hypothesis is proposed.

H3: The establishment of NHTDZs contributes to carbon emission reduction by increasing the level of human capital.

A green low-carbon transition is a continuous process that requires substantial financial support, active policy support, adequate subsidized funding and diversified financing channels. Government spending on R&D reduces the CO2 emissions of countries79 and decreases the CO2 emissions of other countries80.

First, R&D is costly, long and risky. Most enterprises do not have enough capital to invest and have excessive concerns81. The government not only provides public services, but also provides policies that promote enterprises to expand strategic investment82. For example, Hefei city has taken the lead in taking the green low-carbon industry as the new growth momentum of the NHTDZ, skillfully handling the relationship between carbon emission reduction and economic development.

Second, the government fosters the development of new energy and low-carbon industries through industrial policies, in which enterprises establish positive linkages with energy and technology industries and promote green consumption3. This further provides reliable sources of funding for clean, energy-efficient and low-carbon technologies83 which enhances environmental quality and combats climate change.

Once again, financial technology spending improves the regional digital economy, facilitating regional greening84. Monitoring environmental change by reducing energy usage and carbon pollution85. Improving the efficiency of energy use in other sectors and reducing the burden of natural resource use86.

Finally, there is a relationship between external investment and CO254. According to opportunity cost theory, R&D investment is generally countercyclical87. Because of financing constraints, R&D investment increases with the prosperity of business operations88. External investment can transform from theory into profitable projects that generate profits and are supported by sustaining capital. The environmental quality will improve when commerce operations prosper 89,90. The following hypothesis is proposed.

H4: The establishment of NHTDZs can promote carbon emission reduction by increasing investment in scientific research.

Materials and methods

Econometric methodology

This paper adopts an asymptotic double difference model to measure the emission reduction function of NHTDZs and the model is set as follows:

$${lnco2}_{it}={\alpha }_{0}+{\alpha }_{1}{park}_{it}+\theta {control}_{it}+{\sigma }_{i}+{\pi }_{t}+{\varepsilon }_{it} $$
(1)

In the above equation, the dependent variable \({lnco2}_{it}\) denotes the CO2 emissions of city i in year t plus 1 to take the logarithm, from the CADS.

\({park}_{it}\) indicates that city i established an NHTDZ in year t. If city i established an NHTDZ for the first time in year t, then \({park}_{it}\)=1, and conversely \({park}_{it}\)=0. The coefficient \({\alpha }_{1}\) is attention, which indicates the policy effect. \({\sigma }_{i}\) is the control city fixed effect. \({\pi }_{t}\) is the year fixed effect. \({\varepsilon }_{it}\) is the random error and is clustered at the city level. The list of NHTDZs from the Ministry of Science and Technology Torch High Technology Industry Development Center.

After 2019, due to the new coronavirus epidemic, restrictions on economic activity and production resulted in reduced CO2 emissions50,51,52, which need to be excluded from this shock. Furthermore, the explanatory variable data comes from the CADS database, which is updated to 2019. Therefore, the data from 2003 to 2019 are selected as the sample.

Referring to the literature, this paper incorporates a series of city-level control variables from the Statistical Yearbook of China's Cities. See Supplementary table 1.1 of part one of appendix 1 for details.

Empirical results

Baseline regression

Table 1 shows the regression results, with each column controlling for city and year fixed effects. The negative effect gradually increases after the gradual addition of control variables and is significant at the 1% level, indicating that the establishment of NHTDZS can reduce carbon emissions at the city level. The coefficient of the core explanatory variable is − 0.0253, indicating that after the establishment of the NHTDZS, the total carbon emissions of the regions decreased by approximately 3.49%. H1 is verified.

Table 1 Baseline results.
Full size table

Parallel trend testing

Is the discrepancy in total carbon emissions resulting from the establishment of NHTDZs itself? The impacts of other factors are difficult to observe. The following model is constructed for testing:

$${lnco2}_{it}={\alpha }_{0}+{\alpha }_{1}{park}_{it}+\theta {control}_{it}+{\sigma }_{i}+{\pi }_{t}+{\varepsilon }_{it}$$
(2)

The park still indicates whether the region established an NHTDZ, so ji indicates the year in which region i obtained the first NHTDZ. Park(− 7) = 1 when t-j <  = − 7, otherwise it is 0; park(k) = 1; when t-j <  = k, k = − 6, k = -5, k = -4, k = -3, k = − 2, k = − 1, k = 0, k = 1, k = 2, k = 3, k = 4, k = 5, k = 5, k = 6; park(7) = 1, when t-j >  = − 7; otherwise it is 0. Drawing on the classic literature, the regression equation is based on the establishment of the previous year as the benchmark group, and the rest are consistent with Model 1.

The results of the parallel trend test are shown in Fig. 5, which demonstrates the treatment trend in seven periods before and after the event, with the horizontal axis showing the years, the vertical axis showing coefficients and dashed lines indicating the confidence intervals at the 90% level. Figure 5 shows that the model satisfies the assumption of parallel trends.

Figure 5
figure 5

Parallel trend testing.

Full size image

Robustness test

PSM test

Using propensity score matching methods mitigates conclusion bias due to sample selection. In this research, 1:1 and 1:4 proximity matching and kernel matching are utilized for sample matching. The regression results are shown in Supplementary table 1.2 of part one of appendix 1. It suggests that the conclusion is robust.

Placebo test

This paper performed the following 4 aspects of the placebo test to exclude random interference. First, a time placebo test is conducted. The establishment of NHTDZs was preceded by phases 1 to 10 as a pseudo-processing group to examine the significance of the placebo effect. Second, conducting spatial placebo, a number of individuals are randomly selected from the full sample and the pseudo-processing time is set to t1. The remaining sample pseudo-processing time was set to t2, which included 1000 samples. Finally, placebo tests were conducted with and without constraints, with randomized pseudo-treatment times for each individual in the sample within a specified range, without maintaining the community structure (without constraints) and with maintaining the community structure (with constraints). The results of the tests are shown in Supplementary Fig. 1.1–1.4 in of part one of appendix 1.

Excluding contemporaneous policies

To exclude the interference of related policies, this paper excludes low-carbon city pilot areas, key air pollution control areas, emissions trading areas, and carbon emissions trading areas. The regression results are shown in Table 2. After excluding relevant interference policies, the conclusion is still robust.

Table 2 Contemporaneous policies.
Full size table

Incorporation of predetermined variables

When establishing NHTDZs, areas with better initial environmental conditions may be chosen. Referring to Hua yue et al.49, the cross-multiplication terms of municipal wastewater emissions, the comprehensive utilization rate of general industrial solid wastes, SO2 emissions, nitrogen oxide emissions and the initial time are added to model(1). The results are shown in column (1) of Table 3. The core coefficients are negative. The conclusion is still robust.

Table 3 Adding predetermined variables and replacing explanatory variables.
Full size table

Replacing the explanatory variables

Considering possible measure error, the explanatory variables are reconstructed and included in the regression. Specifically, the ratio of total CO2 emissions to household population, the ratio of total CO2 emissions to urban area and the ratio of GDP to CO2 emissions were regressed separately into the baseline equation. The results are shown in columns (2)-(4) of Table 3. The park coefficients are significantly negative.

Bacon decomposition

In order to estimate the carbon emission reduction effect of the pilot policy of the NHTDZs by using the two-way fixed-effect model, it is necessary to ensure that the treatment effect of the treatment group does not change with time in addition to satisfying the parallel trend。If the treatment effect is likely to vary over time, two-way fixed-effect model should not be used to summarize the estimated effect. Bacon decomposition helps provide a way to judge whether two-way fixed-effect model can provide meaningful causal estimates. Following the approach of Goodman-Bacon (2021)91, the estimators obtained from the two-way effect estimation are decomposed into a weighted average of all the classical 2*2DID estimators. If the estimated effects obtained by multiplying the coefficients of all type groups with the weights and summing them are essentially the same as the treatment effects obtained in the baseline regression model (1), this indicates that the two-way fixed-effects model has meaningful causality. The decomposition results are shown in Table 4. Samples are divided into 12 timing group (Fig. 6), including an always-treated group and a never-treated group. There are two types of treatment in the timing groups, the early treatment group (experimental group) vs. the late treatment group (control group), and the late treatment group (experimental group) vs. the early treatment group (control group). The sum of weighted treatment effect is − 0.0253(P = 0.062, consistent with the results estimated from two-way fixed effects). Two-way fixed model provides meaningful causal estimates.

Table 4 Bacon decomposition.
Full size table
Figure 6
figure 6

Bacon decomposition.

Full size image

Reselection of samples

First, the research retains only the treatment group that establishes a NHTDZ for the regression. Multiple NHTDZs may have a stacking effect. Column 1 of Table 5 shows the results. Second, municipalities were deleted. Municipalities are at the same administrative level as the provincial level. They differ in terms of their built-up areas, population, resources, economic vitality and so on. After, excluding the municipality samples for regression, See column 2 of Table 5 for the results. Third, the sample size was adjusted. The NHTDZS establishment time was concentrated between 1988–1997 and 2007-present. Therefore, this paper excludes the NHTDZS established in 1988–1997. The regression results are shown in column 3 of Table 5. The above tests are significantly negative, demonstrating that the findings are credible.

Table 5 Changing sample.
Full size table

Transmission mechanism analysis

Technology upgrading effect

The innovation data come from the statistical yearbook of each city. The green innovation data come from the China Research Data Service Platform (CNRDS). patents granted, utility patents, appearance patents and green utility patents indicate a city's innovation level respectively. The results are shown in columns (1) -(4) of Table 6. In addition, the invention score of the Peking University Enterprise Big Data Research Center(PUEBDRC) indicates a city’s innovation level. The results are shown in column (5) of Table 6. All of the above results show that the creation of NHTDZs significantly improves the city's innovation level. H2 is valid.

Table 6 Innovation mechanisms.
Full size table

In summary, measuring the level of innovation in different dimensions suggests that the establishment of NHTDZs improves a city’s innovation capacity, further reducing carbon emissions. The impact of the setting of NHTDs on carbon dioxide emissions has a technological improvement effect. It refers to that the establishment of NHTDs increase technological progress. Technological progress reduces the total amount of carbon dioxide emissions or the production of more alternative products to reduce carbon dioxide emissions. The greater the ability to innovate, the lower the CO2 emissions, which may be due to: (1) Technological innovation is NHTDs’ core competitiveness. Low-carbon technology and carbon-free technology reduce the total carbon dioxide emissions; (2) Technological innovation can replace energy-consuming products by producing new environmentally friendly products, so as to achieve the effect of energy conservation and emission reduction. Shao Shuai et al. (2022) showed that the improvement of China's technological innovation capability has a restraining effect on carbon dioxide emissions76. Compared to provincial development zones or other special economic zones, NHTDs have a special status and establishment goals. They can play a better role in reducing CO2 emissions through innovative channels.

Human capital

College graduates are an important force for national scientific technological innovation. Scientific technological talent is an important part of scientific research. This approach is crucial for improving the construction of a scientific research governance system and capacity. The NHTDs is an important platform for high-quality entrepreneurship and employment, and has become a base for continuously attracting high-level innovative talents, bringing together two-thirds of the entrepreneurial talents in the national entrepreneurship plan. The number of fresh graduates from colleges and universities increased from 472,000 in 2012 to 800,000 in 2021. The improvement in Chinese's human capital is mainly due to the increase in educational attainment56. The most visible manifestation of investment in human capital is investment in higher education10. By improving the structure of human capital and industrial structure, we will drive regional transformation and upgrading, shift from labor-intensive industries to capital-intensive industries, and evolve from low-value-added industries to high-value-added industries, so as to achieve green development. To test H3, education expenditures in government finance from the China Urban Statistical Yearbook are selected as proxy variables. The results in column (1) of Table 7 show that the creation of NHTDZs improved the government's financing of education emphasis, promoting human capital accumulation. If human capital is high, technological progress is very significant78, and carbon reduction targets can be effectively promoted. H3 is verified.

Table 7 Human capital and research expenditure mechanisms.
Full size table

Research expenditure mechanisms

Technological progress is conducive to CO2 reduction. R&D expenditure is the key element in promoting technological progress. When R&D investment increases, the more advanced energy-saving and emission reduction technologies and equipment will be acquired, reducing CO2 emissions. Technological transformation is an ongoing process. The innovation of enterprises in the NHTDZs has a stable source of funding, a competitive external environment and policy support, etc. Under the condition of government cultivation and subsidies, venture capital has broken through the counter-cyclical nature and accelerated the turnover speed, which eased the constraints on R&D funds and accelerated the investment process of enterprises, thereby promoting carbon emission reduction. Whether the NHTDZs can promote carbon emission reduction by increasing investment in scientific research, the following steps are taken to verify hypothesis 4. The ratio of government expenditure to science technology expenditure indicates the degree of R&D support. The FDI score and ranking from PUEBDRC measure the degree of foreign investment attraction. The results are shown in columns (2)-(4) of Table 7, indicating that NHTDZs increase their investment in science technology R&D by obtaining government support which promotes carbon emission reduction. H4 is tested.

Heterogeneity analysis

Regional heterogeneity

First, due to differences in administrative level, urban vitality and informatization level, cities are divided into central and peripheral cities. The results are shown in columns (1) and (2) of Table 8. The NHTDZs in the central city have more R&D capital and personnel, which produces a rainbow effect, reducing emissions. Second, the cities were divided into large and small cities according to the median population. It has been shown that larger populations in cities have a greater demand for resource consumption, which generates more CO2. As shown in columns (3)-(4) of Table 8. NHTDZs in large cities have certain advantages in term of their technological level, human capital, openness, and basic transportation, and they can better reduce emissions. Finally, the cities are divided into coastal and inland areas, as shown in columns (5)-(6) of Table 8. The carbon reduction in coastal cities is better. Because coastal cities are developed and are engaged in light industry and the industrial structure is more reasonable, coastal city carbon reduction is better. All the results were subjected to 1000 bootstrap intergroup coefficient difference tests (Subsequent tests for heterogeneity were also examined) , and the results were considered significant.

Table 8 Regional heterogeneity.
Full size table

Resource Heterogeneity

According to the classification criteria of the National Sustainable Development Plan for Resource-Based Cities (2013–2020), cities are divided into resource-based and non-resource-based cities. Columns (1)-(2) of Table 9 show the regression results. The effect of nonresource cities is greater92. It may be that resource-based cities depend on resources in the initial stages. This leads to a development model dominated by resource-based industries, which produce more CO2.

Table 9 Heterogeneity of resources and green finance.
Full size table

Green finance heterogeneity

Green finance can formulate green standards and principles. It provides credit support for low-carbon projects, which helps enterprises in NHTDZs obtain more funds, reducing carbon emissions. Green credit, green investment, green bonds and green support indicate green finance. The results are shown in columns (3) and (4) of Table 9 and Table 10. The emission reduction effect is more obvious in regions with a higher level of green finance.

Table 10 Heterogeneity in green finance.
Full size table

Discussion, conclusion and insights

Discussion

NHTDZs can achieve the goal of protecting the environment and ecology on the basis of economic growth93. Unlike previous studies that focused on economic impacts, NHTDZs policy is included in the analytical framework for influencing CO2 emissions. Hua Y et al. (2023) combined provincial high-tech zones and NHTDZs to study their environmental effects94. In this paper, NHTDZs have the advantages of being high grade and enjoying strong policies, only these zones are taken as the research objects for examining emission reduction intensity. Li X (2023) showed that NHTDZs can reduce CO2 emissions through scale effects and technological innovations48. This paper further enriches the knowledge on the transmission mechanism of human capital stock, research investment and attractiveness of venture capital, and the Bacon decomposition of emission reductions in NHTDZs.

The research in this paper has great practical significance, but there are some limitations. In terms of data use, this study uses the CADS, which only has carbon emission data up to 2019. the data for subsequent years is missing and cannot be estimated, resulting in a sample date that can only be used to 2019. In addition, although the reliability of the results has been verified by various robustness tests, there may be factors that cannot be completely ruled out and have a potential impact on the results.

In future research, the following aspects can be strengthened: in terms of data, the data source of the explanatory variables is only a single database. for example, the reliability of the conclusions can be tested by using the measured values of other databases. In terms of the depth of research on the NHTDZs; As an important base for high-tech industries, the NHTDZs need to study its actual role in the process of China's green development from more angles, so as to provide a theoretical basis for reality.

Conclusion and inspiration

The establishment of NHTDZs is an important initiative to implement the new development concept to realize sustainable development. It has great potential compared with other traditional policies. This paper verifies the impact of the establishment of NHTDZs on carbon emissions based on city-level data in China from 2003 to 2019. This study revealed that the construction of NHTDZs is effective at reducing carbon emissions. The conclusion remains robust after a series of tests such as the exclusion of contemporaneous policies and Bacon decomposition. The mechanism results show that NHTDZs facilitate carbon emission reduction by improving innovation levels, accumulating human capital, and promoting R&D expenditures. The heterogeneity analysis revealed that the carbon emission reduction effect was greater in the central city, which has a high level of green financial development, inland areas and nonresource cities.

The construction of NHTDZs can effectively reduce regional carbon emissions, and has a positive effect on improving the level of regional innovation, research capabilities and strict conditions. Therefore, the regional government should provide certain support to the provincial development zones and economic development zones and add national inspection targets to the assessment system, to meet the conditions for application in the NHTDZs. The importance of NHTDZs for emission reduction is highlighted.

Different cities should be supported to jointly apply to the NHTDZs. Research shows that different geographic locations of NHTDZs have different emission reduction effects. The approval conditions for NHTDZs are generally more stringent. If a joint application can be made across cities, the probability of success increases. This can not only lead to sustainable development, but also competition, increasing synergistic development.

The establishment of NHTDZs has occurred for more than 30 years. Earlier-established NHTDZs may outperform later-established NHTDZs. In addition, the NHTDZs’ targets are also various. In the subsequent research, NHTDZs can be divided into specific divisions with different development goals, highlighting the national policy of "one park, one policy".

Finding

This study appreciates support for a fund: Researches on the Optimization Path of Forest and Grassland Carbon Sink Policy in Inner Mongolia Autonomous Region under the Goal of “Double Carbon” (Project Approval No. GXKY22203).