Introduction

Green development is guided by the goal of sustainable development, emphasizing the relationship between economy and environment1, including connotations such as green ecology, green economy, and green society. Economic growth and environmental protection interact and constrain each other, and their positive interaction promotes green development2. In 2015, the United Nations General Assembly proposed 17 Sustainable Development Goals (SDGs), which set requirements for the economy, society, and environment of all humanity3. A series of regional contradictions and imbalances such as economic growth, environmental pollution, and resource scarcity have long existed in China4. The promotion of green development in less developed areas helps to achieve the SDGs.

The old revolutionary base areas (ORBAs) in China were created during wartime, mostly located in less developed areas bordering multiple provinces. The prolonged war has caused serious damage to the environment in the ORBAs, making ecological restoration and protection a tough task. Restricted by geographical location, natural environment, and transportation conditions, most of the ORBAs are lagging behind in social change and relatively backward in economic development5. The process of green development is slow, and the contradiction of uneven and insufficient regional development in ORBAs is still prominent. Due to the dual constraints of internal conditions and external history, less developed ORBAs face enormous challenges in infrastructure, ecological environment, transportation, and characteristic industries. The weak industrial foundation and harsh environment may lead to the dual dilemma of economic growth and environmental protection in the ORBAs4, which poses obstacles to achieving green development. Boosting the joint improvement of economy and environment in ORBAs is an urgent task to make up for the development gaps and achieve green development.

Since the 18th National Congress, China has implemented a series of policies to support the priority development of ORBAs and consolidate the achievements of poverty alleviation. It has successively introduced the National Revitalization Plan for Old Revolutionary Base Areas (NRP for ORBAs), including Shaanxi-Gansu-Ningxia (SGN), Dabie Mountain (DBM), Jiangxi-Fujian- Guangdong (JFG), Left-Right River (LRR), and Sichuan-Shaanxi (SCS) (Fig. 1). To increase the funding subsidies to the ORBAs, from 2001 to 2021, the central financial authorities have cumulatively arranged 121.38 billion yuan of transfer funds for the ORBAs, of which as much as 19.87 billion yuan have been transferred to the ORBAs in 2021 (https://www.mof.gov.cn/zhuantihuigu/cczqzyzfglbf/ybxzyzf_7774/lsbqdqzyzf/index.htm). Up to 2021, except for the LRR areas, all four other key ORBAs have reached the end of the planning period.

Based on the above analyses, our study aims to solve three questions: (1) after the implementation of NRP for ORBAs, can NRP promote both economic growth and environmental quality, and then realize green development in less developed ORBAs? (2) If can, how do these ORBAs achieve green development? (3) And is there heterogeneity in green development in different regions? These are practical issues that urgently need scientific evaluation. Overall, answering these questions is not only conducive to the accumulation of experience and improvement of the implementation for the sustainability of the NRP but also provides useful references for the promotion of green development.

Fig. 1
figure 1

Spatial distribution of the NRP for ORBAs pilot counties.

Full size image

The effectiveness evaluation of NRP has gradually received widespread attention from the academic community. Most existing research focuses on rural revitalization, common prosperity, urban-rural income gap, and other socio-economic development areas. Research has found that the NRP can help to improve the level of rural revitalization, and not weaken due to the pilot counties being in remote areas and far from central cities6. The large urban-rural income gap in the ORBAs hinders the realization of common prosperity7. Nevertheless, the implementation of NRP can narrow the urban-rural income gap in the county and increase the income of rural residents. NRP mainly reduces the urban-rural income gap by strengthening infrastructure construction, increasing local financial inputs, and raising urbanization8. Digital rural governance can also narrow the income gap in ORBAs and promote common prosperity9. In addition, red resources have driven the development of tourism in ORBAs, significantly narrowing the urban-rural income gap, assisting resource flow and wealth redistribution, and promoting the process of common prosperity10–11.

Green development is the internalization of high-level protection of the environment, ecology, and resources into the process of development12. Green development is constrained by the combined influence of internal and external factors13–14. The internal factors mainly include technological innovation15–16, resource endowment foundation, etc. The external conditions mainly include industrial restructuring17, financial development18, urbanization19, and policy shocks20–21, etc. Furthermore, narrowing regional differences is the key to solving the problem of imbalanced green development22. The impact of cultural and social factors such as economic strength and openness is greater than that of natural factors23. It is worth noting that there is still a lack of research on how to realize green development in the ORBAs. Existing research has evaluated the effects of NRP from the perspective of green total factor productivity. The NRP can improve green total factor productivity and promote green development in ORBAs24. Scholars have also explored the green development path of ORBAs from the boundary effect perspective of adjacent regions of provincial boundaries4.

While sorting and summarizing previous research has broadened understanding of the ORBAs, several gaps remain: (1) Lack of studies linking the implementation of the NRP for ORBAs to green development. Previous studies have mainly focused on the economic impacts of ORBAs on agricultural growth, common prosperity, rural revitalization, and urban-rural income gap while lacking the necessary attention to green development. It is worth noting that the ORBAs have unique ecosystems and biodiversity. Paying attention to the green development of ORBAs can help these areas achieve long-term and stable development. (2) There are few studies on how to improve green development in the implementation of the NRP for ORBAs. (3) The Central Government has introduced special NRP in batches for the five ORBAs. Does the implementation of the NRP for each ORBA show heterogeneity? How effective is their green development? The existing literature does not provide an in-depth study on this issue. The win-win situation between the economy and environment is the key to realizing green development25. Therefore, this paper uses county-level balanced panel data from 2003 to 2021 and takes the NRP released since 2012 as policy shocks. The multi-period DID model is used to test the impact of NRP on economic growth and environmental quality, and then examine whether the NRP helps ORBAs realize green development. In addition, we explore the possible transmission mechanisms and heterogeneity involved.

The contributions of the paper are the following: (1) This study comprehensively uses socio-economic statistics and ecological remote sensing data to scientifically assess whether NRP can promote ORBAs realize green development with a balance between economy and environment. New perspectives are provided for the study of green development in less developed areas. (2) This study deeply explores the transmission mechanism and heterogeneity of NRP and provides new ideas for the precise implementation of green development strategies in different ORBAs. (3) This study explores how less developed areas realize green development, and provide new references for the green development of special types of areas.

Theoretical analysis and research hypothesis

Green development endogenizes the ecological environment in the economy and society, to achieve sustainable development of the economy, society, and environment. Although the ORBAs have rich political resources26, mineral and tourism resources27, their development is far more hindered than other non-ORBAs. There exists a high spatial overlap between ORBAs and National Key Ecological Function Areas6. They all bear important ecological functions such as soil and water conservation, and biodiversity maintenance, and have high ecological vulnerability. The NRP for ORBAs has incorporated the construction of ecological security barriers, ecological protection compensation mechanisms, and broken through the resource and environmental dilemma28. So, how NRP can realize green development in ORBAs? Research has confirmed that factors such as industrial structure29, technological innovation30, and public service31 can influence green development. In addition, optimizing ecological space and production space by adjusting the spatial layout of the territorial space can also affect green development. Given this, this paper intends to further test the transmission mechanism of NRP for green development through industrial restructuring, technological progress, ecological space governance, and public service provision.

Industrial structure serves as a bridge between economic activities and ecological environment and promotes green development by optimizing industrial structure in the process of economic construction. Diversified industrial support policy is an important part of the NRP. Different ORBAs need to expand and strengthen their characteristic industries according to the resource characteristics and their advantages. For ORBAs with red resources and revolutionary culture, it is important to actively explore the potential of tourism. The ORBAs should avoid focusing on fixed asset investment and natural resource development32. The NRP is to smooth the flow channels of production factors by improving infrastructure and undertaking industrial transfer. Infrastructure drives the adjustment of industrial structure in ORBAs33, promotes the coordinated development of industrial sectors such as agriculture, manufacturing, and service industries, and realizes the optimal combination of various production factors. The rationalization of industrial structure not only improves production efficiency but also promotes the optimal allocation and efficient use of resource factors and reduces production consumption and pollution34. In general, industrial restructuring can reduce the constraints of resources and environment on economic growth, promote both economy and environment in the ORBAs, and then realize green development.

Technological progress is an important driving force to lead green development35. The NRP enhances the innovation force by improving the mechanism of technological innovation, promoting the transformation of scientific and technological achievements, and enhancing the cultivation and introduction of talents. Innovation drive can realize the agglomeration of technology, capital, and other factors, improve the efficiency of the allocation of production factors, and promote the saving and efficient use of resources36. Element agglomeration promotes technological progress and upgrading, creates more employment opportunities, and drives industries toward intelligent development37. In addition, technological progress can assist in energy saving and emission reduction and reduce ecological vulnerability. Technological innovation can help enhance regional green benefits and promote balance between economy and environment. In the long term, technological input is conducive to improving productivity and transforming the development mode. It can reduce the pressure of industrial transformation, promote the transformation of growth momentum, and realize the compatible development of environment and economy38. Therefore, the innovation investment and the resulting technological progress in the ORBAs by NRP can activate the development vitality, which in turn brings positive effects for green development.

The rational allocation of territorial space is also an important means of carrying out ecological governance in the NRP. Territorial space generally includes ecological space, production space, and living space. Infrastructure construction and urbanization development lead to an increase in production space, which may conflict with ecological space to a certain extent39. Especially for ORBAs with high ecological vulnerability, the conflict between production space and ecological space is more prominentt4. To expand the ecological space, more emphasis should be placed on the intensive utilization of resources and efficient environmental governance, to alleviate the conflict between production space and ecological space. To this end, the NRP promotes the layout adjustment of ecological space through the integrated protection of mountains, water, forests, fields, lakes grasses, etc. The optimization of territorial space patterns is conducive to improving its intensive efficiency, promoting the co-benefits of environment and economy, and thus realizing green development.

The NRP has emphasized the importance of public service provision in the development process. Reducing differences in the level of urban-rural public service helps to achieve equalization. The ORBAs have introduced a series of policies to strengthen the foundation of basic public service, for example, raising the level of primary medical care and improving transportation infrastructure. Improving transportation infrastructure can reduce the cost of labor mobility and strengthen economic interaction between less developed areas and other areast40. Enhancing the supply of public service represented by science, education, culture, health, and infrastructure can enhance people’s well-being. It not only lays a solid foundation for economic transformation and development but also promotes the harmony of humans and nature, which in turn guarantees the realization of green development in the ORBAs.

Based on the above analysis, the NRP can realize green development in ORBAs mainly through industrial restructuring, technological progress, ecological space governance, and public service provision (Fig. 2.). Therefore, this paper puts forward the following research hypotheses:

H1

NRP can promote both economic growth and environmental quality, then realize the green development in ORBAs.

H2

NRP realizes green development in ORBAs through industrial restructuring, technological progress, ecological space governance, and public service provision.

Fig. 2
figure 2

Theoretical mechanism of NPRs for realizing green development in the ORBAs.

Full size image

Materials and methods

Model building

Since the implementation of the “Twelfth Five-Year Plan for National Economic and Social Development of the People’s Republic of China” by the CPC in 2012, China has successively formulated the NRP for five ORBAs, including SGN, DBM, JFG, LRR and SCS. Considering that they have different impacts on economic growth and environmental quality whether they are implemented or not, and before and after they are implemented. This paper takes the five ORBAs as a “quasi-natural experiment” and adopts a multi-period DID model to analyze the impacts of NRP on economic growth and environmental quality in ORBAs. The DID model is set as follows:

$${Y_{it}}={\beta _0}+{\beta _1}polic{y_{it}}+\theta {X_{it}}+{\gamma _i}+{\delta _t}+{\varepsilon _{it}}$$
(1)

In Eq. (1), Yit refers to the green development status of county i in year t, including economic growth and environmental quality, policyit is a core explanatory variable that means dummy variable of NRP. If the NRP for ORBAs cover county i in year t, the value is 1, otherwise it is 0. β1 represents the net effect of NRP on the local economic growth or environmental quality. Xit shows a series of control variables that will influence economic growth and environmental quality. γi indicates county fixed effect, δt is year fixed effect, and εit is the random error term.

The premise of policy assessment using the multi-period DID is that the treatment and control groups satisfy the parallel trend assumption. The treatment group is the ORBAs covered by the NRP, while the control group is the counties not covered. The parallel trend assumption is satisfied if there is a consistent trend in economic growth and environmental quality between the treatment and control groups before the implementation of NRP. Following the event study approach (ESA) to test the parallel trend assumption41, the time dummy variable of the implementation of NRP is set as the core explanatory variable for estimation. To facilitate observation in the same period, a total of 13 periods before and after the implementation of NRP are selected for presentation. The year before the first NRP (2011) is chosen as the base period. The formula for the model is as follows:

$${Y_{it}}={\beta _0}+\sum\limits_{{t=2005,t \ne 2011}}^{{2017}} {{\beta _t}polic{y_{it}}} +\theta {X_{it}}+{\gamma _i}+{\delta _t}+{\varepsilon _{it}}$$
(2)

In Eq. (2), policyit represents the dummy variable of NRP for ORBAs. βt shows the core coefficient of the parallel trend test. If βt is not significant in most cases before the policy implementation, then pass the parallel trend test. Since the NRP was first introduced in 2012, and then diverse NRP were introduced in 2014, 2015, and 2016. Therefore, the status of a county being in the treatment or control group is variable.

To investigate the transmission mechanism of NRP on economic growth and environmental quality in ORBAs, we constructed the following mediating effect models42. The first is to test the effect of NRP on the mediating variables. The second is to test the effect of the interaction term between policy and the mediating variables on economic growth and environmental quality43.

$$\left\{ \begin{gathered} {M_{it}}={\beta _0}+{\beta _1}polic{y_{it}}+\theta {X_{it}}+{\gamma _i}+{\delta _t}+{\varepsilon _{it}} \hfill \\ {Y_{it}}={\beta _0}+{\beta _1}{M_{it}} \times polic{y_{it}}+{\beta _2}{M_{it}}++\theta {X_{it}}+{\gamma _i}+{\delta _t}+{\varepsilon _{it}} \hfill \\ \end{gathered} \right.$$
(3)

In Eq. (3), M denotes the mediating variable, includes industrial restructuring (indus), technological progress (tech), ecological space governance (eco), and public service provision (public). The remaining variables are set as before.

Variable selection

Explained variables

Economic growth and environmental quality. Green development needs to balance economic growth and environmental quality. This paper considers both economic growth and environmental quality to reflect green development. We use the natural logarithm of per capita gross domestic product (lnpgdp) to characterize economic growth and the ecological index (EI) to measure environmental quality (eq). EI is a comprehensive index that includes the biological abundance index, vegetation cover index, water network density index, land stress index, etc4. We utilize China’s High-resolution Eco-Environmental Quality (CHEQ) as a proxy variable for the ecological index44. The CHEQ includes indicators of greenness, dryness, heat, humidity abundance index, etc. Besides, CHEQ is highly consistent with EI provided by the Ministry of Ecology-environment of the People’s Republic of China, which can objectively reflect the environmental quality.

Core explanatory variable

NRP for ORBAs. Five documents for the NRP, including the “National Revitalization Plan for the Old Revolutionary Base Areas of Shaanxi-Gansu-Ningxia”, etc., define the scope of counties and the planning period covering ORBAs. The core explanatory variable is revitalization policy (policy). Policy takes the value of 1 when a county i is included in the NRP in year t, and 0 otherwise. Based on the aggregation and collation of the policy texts, 308 counties are categorized as treatment groups and other counties are classified as control groups.

Mediating variables

(1) Industrial restructuring (indus). Industrial restructuring is an important manifestation of the transformation from extensive to green development45. Currently, the ORBAs are in the key stage of industrialization development. From 2010 to 2021, the average annual growth rate of the value-added of the secondary industry in the five ORBAs reached 8.83%, with a remarkable growth trend. Therefore, the proportion of the value-added of the secondary industry in GDP is used to measure the industrial structure.

(2) Technological progress (tech). Technological progress is an important support for regional economic growth. Limited to the difficulty of obtaining data, referring to existing studies46, technology progress is measured in the form of the logarithm of the number of domestic invention patent applications received, to reflect the productivity gains brought to the ORBAs with a weak innovation base.

(3) Ecological space governance (eco). The first manifestation of green development is the rational development and utilization of territorial space. The territorial space can be divided into three categories: ecological space, production space, and living space. Among them, ecological space is the area of green space, which is used to provide ecological products or services. It is dominated by forest land, grassland, and water, and includes sandy land and saline land47. We use the proportion of ecological space in territorial space to reflect the changes in the space carrying capacity of territorial.

(4) Public service provision (public). Improvements in health care and communications are important foundations for ensuring public service provision. The NRP emphasizes the need to strengthen the construction of primary medical services to ensure basic public service. As communication infrastructure, fixed-line telephones also reflect the level of local public service provision. Due to the difficulty of obtaining data, the number of hospital beds per 10,000 people and fixed-line telephone subscribers per 10,000 people are used to measure the public service provision.

Control variables

Economic growth and environmental quality are also influenced by a variety of factors such as socio-economic conditions and the natural climate. These exogenous factors need to be controlled. Based on the availability of data and concerning existing studies48, control variables are selected comprehensively from both socio-economic and natural conditions. The socio-economic variables include agricultural development base, population density, and level of fiscal expenditure. The agricultural development base (agri) is characterized by the value-added of primary industry to reflect the level of development in the agricultural sector. Population density (den) is represented by the number of people per unit of land area to reflect the sparseness of population distribution and the expansion of population size. The level of fiscal expenditure (fiscal) is expressed by the general budget expenditure of the local finance to reflect the fiscal support of the local government. The natural climate includes temperature (tem), precipitation (pre), and sunshine hours (sun), which reflect the impact of natural climate on the economy and environment.

Data sources

Missing data in this paper were supplemented using interpolation. We obtained balance panel data for 1967 county-level cities (excluding Hong Kong, Macao, and Taiwan) from 2003 to 2021. The sample data of the treatment group of NRP were manually collated based on the relevant documents issued by the National Development and Reform Commission (NDRC). The socio-economic data were obtained from the China County Statistical Yearbook. The data on the number of domestic invention patent applications received comes from the State Intellectual Property Office (SIPO, http://www.cnipa.gov.cn), extracted and matched to counties by Python crawler technology. The EI data were obtained from the CHEQ dataset (2003–2021) released by the National Earth System Science Data Center (NESSD, http://www.geodata.cn), with a spatial resolution of 0.0089°. The natural climate data were obtained from the annual value dataset of China’s surface climate data released by the China Meteorological Data Network (CMDN). The land use data come from the global land cover product data of the Climate Change Initiative (CCI) of the European Space Agency (ESA, http://www.esa-landcover-cci.org), with the temporal-spatial resolution of annual scale and 300 m×300 m, respectively. They were converted to raster data by inverse distance interpolation and extracted using ArcGIS. The variable definition and descriptive statistics are shown in Table 1. The variance inflation factors (VIF) of respective variables are all significantly less than 10, and the mean value of VIF is 1.70, indicating that there is no obvious multi-collinearity problem among them.

Table 1 Variable definition and descriptive statistics.
Full size table

Results and discussion

Benchmark regression

Table 2 reports the regression results of benchmark model. Columns (1) and (2) do not control for county and year fixed effect nor introduce control variables, while columns (3) and (4) control for county and year fixed effect, and columns (5) and (6) further introduce control variables. The results show that the coefficients of policy on economic growth and environmental quality are both significantly positive. The NRP can contribute to the increase of economic growth and the improvement of environmental quality in ORBAs. According to columns (5) and (6), compared to non-ORBAs, NRP can improve economic growth and environmental quality of ORBAs by 17.2% and 1.2%, respectively, in terms of the average treatment effect. Overall, these revitalization plans help China’s less developed ORBAs realize green development with the promotion both of economic growth and environmental quality. The benchmark regression preliminarily verifies hypothesis H1, and the estimated coefficients of policy change only slightly in value, which reflects the robustness of the results, side by side.

In addition, about one-half of the counties covered by the ORBAs are concentrated and contiguous poverty-stricken areas (CCPAs). The ORBAs located in CCPAs have a weaker economic foundation and higher ecological vulnerability, but they have the latecomer advantages. To this end, the role of NRP in influencing the green development of CCPAs is further examined by constructing the interaction term between CCPAs and NRP (CCPAs×policy). The results in columns (7) (8) show that NRP has a stronger role in promoting economic growth and environmental quality in CCPAs than in non-CCPAs bringing about 24.7% and 1.3% improvement, respectively. The CCPAs have more space for improvement in infrastructure construction and urbanization. The policy favoritism and fiscal support for poor areas help them unleash their latecomer advantages and give strong kinetic energy to green development.

Table 2 Benchmark regression results.
Full size table

Parallel trend test

The parallel trend test is put into effect and visualized using the year before the implementation of NRP (2011) as the base period (Fig. 3). The solid dots indicate the effects of NRP in different years, and the vertical solid lines indicate the confidence intervals at the 95% level. Before the implementation of NRP, there is no significant difference in the trend of economic growth and environmental quality between the treatment and control groups. Consequently, the parallel trend assumption is valid.

According to Fig. 3, it also can be found that the implementation of NRP has a persistent dynamic effect. Specifically, the estimated coefficients of policy on economic growth and environmental quality are significantly positive after the implementation of NRP and show positive changes with fluctuating upward trends. It shows that there is a temporal persistence enhancement effect of NRP on economic growth and environmental quality in the covered counties. The NRP combines ecological protection with economic growth and livelihood protection in practice. A multi-level revitalization policy system has been formed through ecological restoration and compensation, cultivation of industries with distinctive advantages, and sound provision of public service, etc. The inclusive and differentiated support mechanism ensures the long-term and sustained growth of green development in the ORBAs.

Fig. 3
figure 3

Parallel trend test.

Full size image

Robustness test

The benchmark regression preliminarily verifies that NRP is conducive to green development of ORBAs. To ensure the stability and reliability of the estimation results, a series of robustness tests are conducted.

Placebo test

To verify whether the benchmark regression results are affected by omitted variables or random factors, we conduct a placebo test49. The specific method is as follows: among all observation samples, counties are randomly selected as the sham treatment group, and the rest of the samples are used as the sham control group. After re-estimation, if the estimated coefficients of random samples fluctuate around 0 and the true estimated coefficients are non-zero significantly, the estimation results are shown to be robust. To avoid the contingency of quasi-natural experiments, the above process is repeated 500 times randomly to obtain the corresponding regression coefficients, p-values, and density distributions (Fig. 4). The coefficients obtained based on random samples fluctuate around 0, obey normal distribution, and are significantly different from those obtained from the benchmark regression. It shows that the effect of NRP is not affected by other omitted variables and random factors, and the results of the benchmark regression are robust.

Fig. 4
figure 4

Placebo test. The x-axis in the figure represents the estimated coefficient value of NRP, the dots are the kernel densities corresponding to the estimated coefficient value, and the vertical dashed line is the true estimate of the baseline regression results.

Full size image

Propensity score matching-difference-in-difference (PSM-DID)

The selection and establishment of ORBAs are selective, which is not completely random. That is, there may be estimation error problems arising from omitted variables and sample selection bias in the treatment and control groups. The multi-period DID does not eliminate this problem. In view of this, PSM-DID is further developed. The control variables in the benchmark regression model are first used to predict the probability that NRP covers each county. Then the k-neighborhood 1:2 matching method within caliper is used to match the control group of counties with similar characteristics as theirs for each treatment group. So that they can satisfy the stochastic premise of ORBAs (Columns (1) (2) of Table 3). It can be found that the estimated coefficients of policy are all significantly positive, and NRP has the role of significant contribution to economic growth and environmental quality in the ORBAs. Thus, the conclusion that ORBAs realize green development is robust.

Tailing treatment (5%)

The presence of extreme outliers in the data is likely to have a large impact on the estimation results, which can be mitigated by tailing treatment. In this paper, the explained variables lnpgdp and eq are re-estimated by tailing treatment at the 5% and 95% quantiles. Columns (3) (4) of Table 3 show that the estimated coefficients of policy on economic growth and environmental quality in the covered counties remain significantly positive after tailing treatment, which is basically consistent with the results of benchmark regression. It shows that the promotion effect of NRP on green development has robustness.

Table 3 A series of tests of robustness.
Full size table

Replacement of explained variables

There is a significant positive correlation between nighttime light and economic growth. Nighttime light brightness can characterize the degree of regional economic development50. We used the logarithmic value of nighttime light brightness (lnlight) to replace lnpgdp. Normalized differential vegetation index (NDVI) can effectively reflect the regional vegetation coverage and ecological environment changes51, and NDVI was used as a proxy variable for eq. The results of re-estimation after replacing explained variables are shown in columns (5) (6) of Table 3. The results show that the estimated coefficients of policy are still all significantly positive, consistent with the benchmark estimation results, and the green development effect of NRP is robust.

Exclusion of other policy interventions

Economic growth and environmental equality in the ORBAs may be affected by other relevant policies during the same period, such as the National Integrated Pilot Policy on New Urbanization (NIPPNU) and the National-level Poverty-stricken County Support Policy (NPCSP), which may interfere with the benchmark estimation results. The NDRC has published a list of NIPPNU pilots for a cumulative total of 513 counties across the country since Feb 2015. The NIPPNU works on a series of activities such as urbanization transformation, green and smart development, and integrated urban-rural development. The NPCSP was implemented at the end of 2012, and a total of 832 National-level Poverty-Stricken Counties have been established so far. Both policies are related to rural development, and there are a lot of overlaps between the covered counties and the ORBAs. In order to exclude the interference of related policies, two dummy variables, NIPPNU and NPCSP, are added to the benchmark regression. When a county belongs to the list of NIPPNU pilots in a certain year, NIPPNU takes the value of 1, otherwise, it takes the value of 0. The same applies to NPCSP. The regression results are shown in columns (7) (8) of Table 3, the estimated coefficients of policy are still all significantly positive, implying that the implementation of other pilots in the same period did not interfere with the estimation results, and the benchmark regression results are still robust.

Mechanism analysis

Benchmark regression and a series of robustness tests have amply demonstrated that NRP is able to strike a balance between economy and environment and promote local green development. However, how ORBAs realize green development needs to be further explored. Based on the actual implementation of NRP, combined with theoretical analysis, we explored the transmission mechanism of ORBAs to realize green development from four aspects: industrial restructuring, technological progress, ecological space governance, and public service provision (Tables 4 and 5).

Industrial restructuring

The impact of NRP on industrial restructuring is significantly positive. After the implementation of NRP, the proportion of value-added of the secondary industry in the covered counties increased by 3.3%. This indicates that the NRP has effectively promoted the industrial restructuring of the covered counties. Columns (2) (3) of Table 4 show that the estimated coefficients of the impact of industrial restructuring on economic growth are significantly positive, and environmental quality is negative but not significant. Industrial restructuring can help promote economic growth in the covered counties, but it is not yet sufficient to improve environmental quality. The main reason may be that the industrial base of the ORBAs is relatively weak, and the effect of structural restructuring is more prominent in economic aspects. However, the estimated coefficients of interaction terms between indus and policy are all positive. In other words, the implementation of NRP strengthens the enhancement effect of industrial restructuring on economic growth and weakens the inhibiting effect of industrial restructuring on environmental improvement, thereby contributing to the green development of the ORBAs. Industrial restructuring is one of the effective paths for NRP to realize green development in the ORBAs.

Technological progress

There is a significant positive impact of NRP on technological progress. After the implementation of NRP, the patent level of ORBAs increased by 27.5%. This indicates that NRP has effectively promoted technological progress and innovation in the covered counties. Columns (5) (6) of Table 4 show that there is a significant positive impact of technological progress on environmental quality, while the negative impact on economic growth is not significant. Technological progress, as characterized by patents, promotes environmental improvement in the covered counties, but is not yet sufficient to drive economic growth. The estimated coefficients of interaction terms between tech and policy are all significantly positive. The NRP weakens the insignificant negative effect of technological progress on economic growth and strengthens the effect of technological progress on environmental quality, which in turn promotes the green development of the ORBAs. Technological progress is one of the effective paths for NRP to realize green development in the ORBAs.

Ecological space governance

There is a significant positive impact of NRP on ecological space governance. After the implementation of NRP, the proportion of ecological space covering ORBAs increased by 0.2%. This indicates that NRP has effectively expanded the ecological space of covering counties and promoted the intensive use of territorial space. Columns (8) and (9) of Table 4 show that the impacts of strengthening ecological space governance on economic growth and environmental quality are both significantly positive. The expansion of ecological space not only promotes economic growth of the covered counties but also contributes to the improvement of local ecological environment, then promotes both. Reasonable expansion of ecological space implies that the exploitable territorial space is optimized, and production efficiency is enhanced through moderate environmental regulation, thus driving economic growth52. The interaction term has a significant positive impact on both economic growth and environmental quality. The NRP strengthens the role of ecological space governance in enhancing economic growth and environmental quality. Ecological space governance is one of the effective paths for ORBAs to realize green development.

Table 4 Test of transmission mechanism (I).
Full size table

Public service provision

The NRP has a significant positive impact on health care and communication (Table 5). After the implementation of NRP, the health care and communication in ORBAs improved by 10.0% and 17.6%, respectively. This indicates that the NRP has effectively improved the local health care and communication and promoted the effective supply of public service provision. The coefficients of health care and communication are both significantly positive, and increasing the supply of public service provision in the covered counties can enhance economic growth and improve environmental quality, then promote both. The estimated coefficients of interaction terms of health care and communication with policy are also significantly positive, so NRP enhances the role of public service provision in economic growth and environmental quality. Public service provision is one of the effective paths for ORBAs to realize green development.

In summary, NRP can promote economic growth and environmental quality through the paths of industrial restructuring, technological progress, ecological space governance, and public service provision, then realize green development in the ORBAs. The H2 is verified.

Table 5 Test of transmission mechanism (II).
Full size table

Heterogeneity analysis

Since ORBAs with unbalanced development and differences in endowment bases, the heterogeneity of NRP in playing the role of promoting green development is systematically examined. Heterogeneity analysis is carried out from the following perspectives: different ORBAs; different terrain features; and different quantiles of economy and environment.

Different ORBAs

The National Government Council successively introduced key NRP for ORBAs, including SGN, DBM, JFG, LRR and SCS. The implementation period, infrastructure construction, and resource endowment of these five ORBAs are different, so the promotion effect of NRP on green development will be different among ORBAs. Table 6 reports the regression results of introducing five different dummy variables from the ORBAs. Overall, there is heterogeneity in whether the NRP has realized green development in different ORBAs. The JFG, LRR and SCS areas all have a significant positive impact on economic growth and environmental quality. The launch of NRP helps to promote economy and environment in these ORBAs and realize green development. In comparison, SCS areas have the strongest promoting effect on economic growth, while LRR areas have the strongest enhancing effect on environmental quality. The impact of SGN areas on environmental quality is significantly negative, while the positive impact of DBM areas on economic growth is not significant, these two ORBAs have not yet realized green development.

For SGN areas, the starting point for development is relatively low, and a perfect environmental management system may not have been formed at the initial stage of economic growth. Neglecting the carrying capacity of environment and exchanging “green mountains for silver mountains” is not conducive to the improvement of environmental quality. Although the current environmental conditions in the SGN areas have improved, upgrading the environmental quality is an improvement process that needs to be put into lasting action. More time and resources need to be invested, and the task of ecological management is arduous.

For DBM areas, they are both soil and water conservation ecological functions and important water supply areas for the middle reaches of Huai River and the lower reaches of the Yangtze River. Influenced by the complex terrain of low mountains and hills, DBM areas have prominent contradictions between resource development and environmental management, and they are areas of key concern for environmental protection. There is a large space for environmental improvement, and the intervention of external policies can effectively enhance the local environmental quality. However, the economic structure of DBM areas is relatively homogeneous, market competitiveness is weak, and green development is subject to multiple constraints. This may lead to a weakening role of NRP in enhancing economic growth.

Table 6 Heterogeneity analysis: distinguish among five ORBAs.
Full size table

Different terrain features

Most of the ORBAs are in mountainous areas with complex and varied terrain. Different terrain features have important impacts on economic growth and environmental quality, especially on agricultural production, engineering construction, and urban development. Areas with undulating terrain may lag behind other areas in terms of transportation construction and industrial development. The density of transportation network and the level of industrial industry in these areas tend to be low, which forms a bottleneck constraint on the green development of ORBAs. In view of this, we refer to the National Water and Rainfall Information Network (NWRIN, http://xxfb.mwr.cn/slbk/slgckc/dm/202004/t20200409_1466880.html) to classify the terrain. Elevations less than 200 m are plains, 200–500 m are hills, 500–3500 m are mountains, and above 3500 m are plateaus. Different counties are divided into plains, hills, mountains, and plateaus as terrain features, and then explore the heterogeneity of NRP under different terrain features. Since the five ORBAs do not involve plateau terrain, Table 7 reports the regression results under plains, mountains, and hills.

The regression results show that NRP has a significant positive contribution to both economic growth and environmental quality in hilly and mountainous terrain, thus achieving green development in ORBAs. Nevertheless, it fails to effectively improve the environmental quality of ORBAs in plains. The coefficients reveal that NRP has a more prominent green development promotion effect on mountain counties. The development of featured service industries based on red and ecological tourism resources can effectively promote local economic growth, thus producing a higher economic effect. For example, the Tiantai Mountain Scenic Area, a national forest park known as the “First Peak of Huainan”, is rich in red tourism resources and cultural values, and has had a wide impact throughout the country.

In addition, the hilly counties are mostly located in Fujian, Jiangxi, and Anhui, with abundant precipitation and humid climate, preserving large areas of primary forests. These areas have several nature reserves and ecological civilization-advanced demonstration zones. For example, Wuyi Mountain Nature Reserve located at the border of Jiangxi and Fujian provinces has 2,527 plant species, with rich forest, and biotic resources. It has a high demonstration role for the construction of ecological civilization and gains economic radiation and overflow from the surrounding neighboring areas. Its endowment conditions can promote industrial layout and upgrading and produce more prominent economic and environmental effects. On the contrary, the plain counties are mostly dominated by agriculture, facing prominent environmental problems such as agricultural surface pollution, and improving environmental quality requires further precise coordination.

Table 7 Heterogeneity analysis: distinguish among terrain features.
Full size table

Different quantiles of economy and environment

Considering that for regions with different levels of economic growth and environmental quality, there may be heterogeneity in ORBAs to the realization of green development. Therefore, a panel quantile regression model is further developed to examine the heterogeneity impact of NRP on economic growth and environmental quality (Fig. 5).

The impact of NRP on economic growth at different quantiles is significantly positive. As the quantile increases, the promotion effect of NRP on economic growth shows a “U”-shaped pattern of first decline, followed by a rise and an overall decline. It has the weakest positive promoting effect around the 0.8 quantile. The promotion effect of economic growth is stronger in counties with relatively low levels of economic growth and weaker in counties with high levels of economic growth. When a county’s economic growth level is low, although the development pattern is crude, it has greater development potential and space. The intervention of external policies can improve production efficiency, thus greatly promoting local economic growth. When economic growth continues to rise, the existence of a marginal diminishing effect makes the promotion effect of NRP gradually weaken. When economic growth reaches a higher level, the strong industrial base can break through the constraints of the law by promoting technological progress and optimizing resource allocation. The diminishing marginal effect tends to weaken so that the policy promotion effect has rebounded.

The impact of NRP on environmental quality at different quantiles has experienced a shift from negative to positive. As the quantiles increase, the impact of NRP on environmental quality gradually changes from a negative inhibitory effect to a positive enhancement effect, and generally shows a stable upward trend of change. Near the 0.6 quantile is the turning point from positive to negative and reaches the strongest positive effect at the 0.9 quantile. The NRP has a positive effect on environmental quality in counties with high ecological endowments. However, it is not enough to improve the environmental quality of counties with medium or lower ecological endowment conditions. When the ecological endowment of a county is low, the environmental foundation is weak, and ecological vulnerability is high, and improving environmental quality is a process that requires sustained efforts. External policy interventions have not been effective in improving the local environmental quality.

In general, the role of NRP in enhancing economic growth and environmental quality is significantly different at various quantiles. As the quantile rises, the role of NRP on economic growth shows a decreasing and then increasing trend, with an overall falling trend. The role of environmental quality shows a trend from negative to positive and a stable rise. On the whole, the ORBAs with relatively high levels of economic growth and environmental quality have greater potential and prominent policy effects for realizing green development.

Fig. 5
figure 5

Quantile estimation curve of economic growth and environmental quality.

Full size image

Discussion

This paper uses the multi-period DID model to assess the impact and transmission mechanism of the NRP for ORBAs on economic growth and environmental quality, and to explore whether the NRP can promote the ORBAs to achieve green development. The benchmark regression model in this study reveals that the implementation of NRP boosts the ORBAs to achieve green development. This finding passes a series of robustness tests. Our finding is in accord with recent study indicating that the NRP for ORBAs not only effectively promotes economic development but also improves the ecological condition index4. In addition, we have confirmed that the NRP for ORBAs has a stronger role in promoting green development in CCPAs than in non-CCPAs.

The mechanism analysis indicates that the NRP realizes green development through industrial restructuring. There are similarities among the current study. In the process of industrial restructuring, through the transfer of factors of production, the NRP can help to improve technological efficiency and thus promote regional green total factor productivity24. We innovatively found that NRP can enhance green development through technological progress, ecological space governance, and public service provision. With the increase of national investment in technological progress, the ORBAs improve green development through learning and sharing effects. Ecospatial construction focuses on protecting and restoring natural assets, enhancing ecosystem service functions and providing an ecological foundation for green development. Strengthening public service provision such as health care and communications can help promote interregional connectivity and greening of the economy.

Heterogeneity analysis reveals that there are differences in different ORBAs, terrain features, and quantiles of economy and environment. The ORBAs of JFG, LRR, and SCS are capable of realizing green development. The SGN and DBM areas have not yet realized green development. SGN areas are mainly located in the Loess Plateau with long ravines and gullies, with perennial drought and prominent soil erosion problems. Green development of SGN is limited by natural and geographical factors4. Counties situated in mountainous terrain can achieve green development. The reason may be that the ORBAs of mountainous terrain have unique location advantages and rich natural landscapes and cultural heritage. The ORBAs with higher quantiles of economy and environment play a more prominent role in promoting green development. The ORBAs with high economic levels can increase financial support for green development. When the ecological endowment is high, it is possible to strengthen environmental governance more effectively and obviously reduce pollution emissions, thus greatly contributing to the improvement of local environmental quality.

Despite the insights gained in the impact of the NRP for ORBAs on green development, this study still has limitations. The regional differences and source decomposition of green development in the five ORBAs are worth exploring. Moreover, the spatial spillover effects of the ORBAs need to be further analyzed. Identifying the radiation effects of the policy from a spatial perspective can help to support regional synergies in achieving green development.

Conclusions and policy implication

The revitalization plans are important plans to narrow regional disparities and promote high-quality development in less developed areas. This paper takes the NRP that has been successively introduced as a quasi-natural experiment, uses the multi-period DID model to evaluate the impact of NRP on economic growth and environmental quality in ORBAs, and investigates whether green development can be realized in less developed areas. The results are as follows: First, NRP can promote both economic growth and environmental quality, thereby realizing green development in less developed ORBAs. Second, NRP can help the ORBAs realize green development through four ways: industrial restructuring, technological progress, ecological space governance, and public service provision. Third, their differences in different ORBAs, terrain features, and quantiles of economy and environment. JFG, LRR, and SCS areas can realize green development, while SGN and DBM areas can not realize green development. The ORBAs with mountainous terrain and higher quantiles of economy and environment have a more prominent role in realizing green development.

At present, China’s support policy for ORBAs has entered a new stage from the “1258” system to the “1 + N + X” system, which means 1 guideline, N implementation plan, and X special policies. The improvement of policy system not only provides more accurate support to the ORBAs but also expands the scope of support. The NRP provides better policy support for the less developed areas to realize green and high-quality development. In addition, based on the research conclusions, this paper proposes the following policy implications:

Firstly, address both symptoms and root causes, enhance endogenous motivation, and expand external vitality. Ecological environment protection and economic development are common challenges facing the world. China is a staunch supporter and active practitioner of sustainable development, adhering to a new development pattern with domestic circulation as the main body and domestic and international dual circulation promoting each other. External support and internal efforts jointly promote green and high-quality development. The policy support provided by the NRP mainly reflects the intervention of external forces. Nevertheless, external support always “addresses the symptoms but not the root cause”. Once the support stops, it may lead to unsustainable development, and the key lies in “addressing both the root cause and root cause” and “It is better to teach fish than to teach it to fish”. In that way, the external support forces in less developed areas must be coordinated and matched with the endogenous power. On the basis of balancing economic development and environmental carrying capacity, developing the ecological endowment and the advantages of red tourism resources not only helps to promote industrial transformation but also stimulates the natural circulation and circulation among various elements. In addition, it is necessary for multiple entities to participate together, introduce social capital, and leverage the driving role of government support and the regulatory role of market participation, exerting efforts from both external and internal sides helps achieve dynamic balance and mutual promotion between internal and external, and continuously improves the sustainability and vitality of green development in ORBAs.

Secondly, multiple measures should be taken simultaneously to smooth the channels of action. It is necessary to accelerate the adjustment of industrial structure and improve industrial policies that are in line with the actual development situation. Paying attention to the quality of undertaking industrial transfer, helps enhance the development momentum of industries and manufacturing in ORBAs. It should take measures to enhance the competitiveness of service industry systems such as ecotourism and red tourism and accelerate the transformation and upgrading of industrial structure. In terms of technology, we need to increase investment in science and education, and enhance the ability of less developed areas to introduce and absorb advanced technology. It is necessary to improve development quality and production efficiency through technological innovation and attach importance to the accumulation of human capital. In terms of ecology, we ought to increase environmental supervision of polluting and energy-consuming enterprises in ORBAs. Increasing the proportion of green GDP and other ecological environmental indicators in government assessments helps optimize the spatial pattern of territorial space and allocate ecological spaces reasonably. Based on the ecological resource advantages of ORBAs, improving ecological restoration and compensation mechanism, and exploring feasible paths for realizing the value of ecological products, so as to promote the development of green economy. In terms of public service, we are supposed to enhance the supply of public service, and support investment in infrastructure construction, science, education, culture, and health. Consummating transportation facilities and networks in ORBAs and accelerating construction of public service systems lays a solid foundation for green development.

Thirdly, we need to adapt measures to local conditions and promote targeted measures. Different countries and regions have diverse levels of green development, and precise support policies need to be formulated. Accelerating the pace of green development in SGN and DBM ORBAs is urgent. For ORBAs in plain terrain, it is necessary to focus on solving prominent environmental problems and moderately strengthen environmental regulations. For ORBAs with backward endowment conditions, introduce more funding tilt to improve the balance and coordination of the economy and ecology. In addition, various measures need to be taken to strengthen the support for concentrated and contiguous areas in the ORBAs and guide the effective connection and coordinated promotion of green development and rural revitalization.