Technological Innovation, Industrial Structure Upgrading, and the Coordinated Development of Regional Economies

Abstract

The purpose of this study is to systematically examine the impact of technological innovation on the coordinated development of regional economies and its internal mechanism. It is aimed at revealing whether and how technological innovation promotes the coordinated development of regional economies, and further identifying its heterogeneity characteristics and boundary conditions in the space–time dimension. The research was conducted using panel data for 258 prefecture-level cities in China from 2011 to 2021. This study found that technological innovation significantly promoted the coordinated development of regional economies; this effect was more prominent in China’s eastern region and the Yangtze River Economic Belt. The mechanism test shows that technological innovation can optimize regional resource allocation and narrow the development gap by promoting industrial structure upgrades and rationalization. Further analysis shows that the level of marketization has a nonlinear regulatory effect on the coordination effect of technological innovation, with two threshold levels. A heterogeneity analysis reveals significant differences in the effects of technological innovation in different regions in China, especially in the western region and the northwest side of the Hu Changyong line. The research leads to four key policy recommendations. First, it is important to strengthen the core driving role of technological innovation and implement regionally differentiated innovation support policies. Second, industrial structure upgrades should be encouraged through industrial chain coordination. The third recommendation is to improve the market-oriented institutional environment and minimize barriers to factor flow. Fourth, supporting coordinated policies, such as optimizing human capital and introducing high-quality foreign capital, is necessary to establish a sustainable long-term mechanism for regional coordinated development.

1. Introduction

The coordinated development of regional economies is an important strategy for encouraging China’s high-quality economic growth; the goal is to narrow the regional development gap, optimize resource allocation, and promote common prosperity. However, due to factors such as geographical location, resource endowments, and institutional barriers, there remains a marked imbalance in development in eastern, central, and western China, as well as between urban and rural areas. Given profound adjustments in global economic patterns and China’s economic transformation, an imbalance in regional development has become an significant factor restricting high-quality economic growth. A 2023 World Bank report shows that China’s regional development imbalance index exceeded the international warning line of 0.4. According to China’s National Bureau of Statistics, in 2024, the gross domestic product (GDP) of eastern China was CNY 70.24 trillion, which was 2.45 times and 2.44 times higher than that of the central and western regions, respectively. The outline of China’s 14th Five-Year Plan (2021–2025) clearly emphasizes a strategy of coordinated regional development. However, China’s coordinated regional development faces many problems and challenges, including a significant gap in regional economic development, a prominent digital divide, and slow progress in the equalization of regional public services. These factors restrict the in-depth promotion of a regional coordinated development strategy and the realization of its sustainable development goals.

With the rise of the scientific and technological revolution in the 1950s, the neoclassical growth theory began to incorporate technological progress as an endogenous factor. Solow (1957) first revealed the contribution of technological progress to economic growth at the empirical level by introducing the “Solow residual” [1]. However, his model regarded technology as an exogenous variable and failed to explain the intrinsic mechanism of innovation [1]. In the theory deepening stage, Enos (1962) applied innovation process theory to argue that technological innovation is a multi-dimensional systematic activity that includes technology selection, resource allocation, organizational change, and market development [2]. Freeman (1997) further highlighted the commercialization characteristics of technological innovation; this became an important theoretical basis for subsequent innovation research [3].

Technological innovation has become a key force driving economic and social transformation. This trend can be explained by the “techno-economic paradigm” theory, which posits that every technological revolution reconstructs production relations and spatial patterns. Technological innovation forms the core of the fifth technological revolution, and its innovativeness, permeability, and network externalities effectively break through geographical constraints and reshape regional economic ties. Specifically, technological innovation provides a new path for regional coordinated development by reducing transaction costs [4], optimizing factor flows [5], and promoting knowledge spillover [6]. Its powerful enabling role should be able to break down economic barriers between regions and promote the coordinated development of regional industry, innovation, and talent.

However, in the process of promoting regional economic development, technological innovation faces the dual challenges of the digital divide and the digital dividend. The digital divide can be explained using “technology diffusion theory” [7] and “spatial disequilibrium theory” [8]; the phased characteristics of technology diffusion and the differences in regions’ initial conditions may lead to the “Matthew effect,” whereby successful regions experience more successes over time. Rogers (2014 noted that technology adoption follows an “S-shaped curve” pattern [7]. This means there is a significant time difference between early adopters and late adopters, and this difference in stages enlarges the gap in technology returns due to different regional initial conditions [7]. Myrdal (1957) further proposed the “circular cumulative causality theory,” emphasizing that the scale effect formed by technological innovation in developed regions continues to attract capital and talent, while less developed regions fall into a negative cycle of “system locking” [8]. These two theories jointly explain the internal mechanism by which technological innovation may strengthen rather than exacerbate the imbalance in regional development [8]. The marginal income obtained via technological innovation in developed regions is often higher than that in less developed regions [9].

In China’s economic environment, the digital divide is reflected in the uneven distribution of digital infrastructure and in the differences in digital technology application ability and digital talent reserve. These differences lead to different application effects of technological innovation across different regions; this restricts the economic development of less developed regions and may increase imbalances in regional economic development. The release of digital dividends depends on the inclusive application of digital technology. As a positive economic spillover effect promoted by technological innovation, it is specifically embodied in three dimensions: production efficiency improvements, industrial structure upgrades, and market boundary expansion.

Based on the theoretical framework of endogenous growth, the generation of the digital dividend is first reflected by a jump in total factor productivity. The production function model constructed by Bloom et al. (2012) shows that the penetration of digital technology has significantly improved the allocation efficiency of production factors, resulting in a 1.8% growth effect on total factor productivity in the manufacturing sector [10]. This has occurred by reducing information asymmetry, optimizing production processes, and improving management efficiency. This finding has been verified in the context of China. The widespread adoption of industrial Internet platforms has improved the production efficiency of key industries by 25–30%.

From the perspective of industrial structure evolution, Perez’s (2002) technology economy paradigm theory positions digital technology as a key enabling technology that is reshaping the composition and distribution of the industrial value chain [11]. Melitz (2003)’s heterogeneous enterprise trade theory provides a theoretical basis for understanding the mechanism by which digital technology reduces market entry barriers [12]. The degree of released digital dividends essentially depends on the degree of the coupling of technology diffusion and region-specific factors, such as the institutional environment and human capital. This discovery provides an important theoretical tool for describing the spatial heterogeneity of digital economy development.

The core of regional coordinated development is to reduce the development gap between regions through continuous dynamic regulation. This means recognizing the rationality of a moderate gap and using the kinetic energy effect generated by regional potential energy differences to control the development gap within an economically and socially affordable threshold [13]. The goal is to optimally allocate development factors in the spatial dimension, promote each region’s evolution toward a relatively balanced state through the difference convergence mechanism, and achieve a multi-dimensional dynamic coordination pattern [14]. This process reflects the dialectical unity between the development gradient and synergy effect and reflects the continuous optimization of the regional system from a non-equilibrium state to high-order equilibrium.

Regional coordinated development is a systematic project that covers the multi-dimensional coordinated evolution of economy, society, culture, and ecology. Many internal and external variables impact this process. Endogenous variables include resource endowment differences, industrial layout characteristics, and population spatial distribution [15]. Exogenous variables include policy orientation, market mechanisms, and global patterns. Systematically analyzing these impact mechanisms has significant theoretical value for optimizing the regional development path.

From an endogenous perspective, regional resource endowments and industrial structure constitute the material basis for coordinated development, and their heterogeneity determines the diversity of development models. The scale, quality, and spatial flow characteristics of population factors are core dynamic factors that profoundly impact regional synergy efficiency [16]. The degree of development of the public service system is a key indicator measuring people’s well-being and regional balance. For the exogenous environment, institutional arrangements and market efficiency have dual regulatory effects on regional coordination. An effective policy framework and mature market mechanism provide institutional support for factor flows [17].

The path to achieving regional coordinated development can be discussed from multiple perspectives. Public data openness has also been shown to break down information barriers and promote fair resource use [18]. The upgrading of industrial structure promotes regional coordinated development by improving the level of public services and optimizing resource allocations [19]. The construction of urban agglomerations and market-oriented mechanisms encourages the specialized division of labor in terms of technology, forming an inverted “U” shape for assessing coordinated region development. China’s “Belt and Road” initiative has improved transportation infrastructure in the west, reduced transportation costs, expanded the scale of trade, and narrowed the gap between the east and west [20].

Since neoclassical growth theory was proposed, scholars have generally recognized the importance of innovation in economic development. Studies on the coordinated development of technological innovation and regional economies worldwide have mainly focused on the impact of technological innovation on economic growth, the mechanisms involved in technological innovation and economic growth, and the impact of technological innovation on regional economic gaps. Starting with Schumpeter’s concept of technological innovation, the theoretical research explores the core role of technological innovation in economic activities. Studies have noted that new ideas are key to technological innovation; this theory has laid the foundation for subsequent research [21]. Romer (1986) further proposed that knowledge and technological progress are sources of economic growth [22] and that high-quality patents can accelerate national economic growth [23]. This approach counters the assumption that technology is exogenous. Both technological innovation and technological spillover can improve total factor productivity [24], which in turn promotes the efficiency and level of economic growth more significantly in developed regions. The effect for underdeveloped regions is weaker, which may exacerbate the regional development gap [25].

The literature review above shows that technological innovation has a “double-edged sword” effect on the coordinated development of a regional economy. Most studies have verified its positive role in promoting coordinated development through factor allocation optimization, knowledge spillover, and a late development advantage. However, some scholars have noted that the digital divide and differences in infrastructure may exacerbate regional disparities. Most past research has focused on the macro growth effect of the digital economy or the application of a single technology; it has not, however, explored the systematic mechanism by which technological innovation promotes regional coordinated development.

Therefore, this study builds a theoretical framework for the coordinated development of regional economies enabled by technological innovation and reveals the impact of technological innovation on regional economic development. The first research question is as follows: how should China balance efficiency improvements with the fair distribution of technological innovation in the digital era? The permeability and network effect of technological innovation may promote regional synergy by reducing transaction costs, or they may increase the development gap due to the digital divide. Research has not yet fully discussed this impact.

A second research question is as follows: how does technological innovation affect regional coordinated development through changes in industrial structure? The literature suggests that technological innovation can affect the coordinated development of regional economies, but the specific transmission path is unclear. In particular, differences in industrial base, factor endowments, and market maturity across different regions may lead to significant heterogeneity in the optimization effect on the industrial structure of technological innovation. This highlights the need to further clarify the intermediary role of industrial structure upgrades in promoting regional coordinated development through technological innovation.

A third research question is as follows: how does the level of marketization influence the regional coordination effect of technological innovation? Technology diffusion theory and spatial disequilibrium theory explain the disequilibrium of technology spillovers. However, marketization may change this trend by optimizing factor flow and reducing institutional barriers. Past research on the regulatory role of marketization in the relationship between technological innovation and regional coordinated development has not conducted systematic tests, highlighting the need for further analysis. Given this background, this paper conducts in-depth research and analyses addressing these issues.

2. Theory and Hypothesis

2.1. Impact of Technological Innovation on Regional Coordinated Development

As the primary driver of regional coordinated development, technological innovation has reshaped the regional economic pattern through a multi-dimensional dynamic mechanism. From the perspective of the technology diffusion effect, the networked characteristics of the new generation of information technology, such as artificial intelligence and the Internet of Things, have accelerated the cross-regional flow of knowledge, effectively shortened the time lag of technology gradient transfer, and promoted technological convergence between regions. This diffusion has not only overcome the spatial barriers of traditional industries but also reconstructed industrial ecology through digital and intelligent transformation, forming a new path of technology industry collaborative evolution. From the perspective of synergy, technological innovation has spawned systematic changes throughout the whole industrial chain. For instance, intelligent manufacturing restructures production modes; digital logistics optimizes circulation systems; and new retail innovates consumption patterns. This whole-chain digital transformation not only strengthens the cluster advantages of developed regions but also creates high-value-added embedding opportunities for less developed regions through the decomposition of the value chain [26]. More importantly, technological innovation has reconstructed the regional division of labor mode through new business forms, such as the platform economy and the sharing economy. Moreover, blockchain, big data, and other technologies have enabled regional governance, reduced institutional transaction costs, and provided technical support for the construction of a unified big market.

The impact mechanism of technological innovation on the coordinated development of a regional economy indicates significant stage characteristics and environmental dependence. In the initial stage of development, technology diffusion is dominated by one-way spillover. Developed regions drive industrial upgrading in less developed regions through technology transfer. However, due to differences in technology absorption capacity, this may aggravate imbalances in regional development. After entering the mature stage, the networked technology ecology has promoted the formation of multi-directional interaction among regions, gradually improved the multisynergy mechanism of the technology industry space, and converged regional development. Regarding environmental conditions, regions with perfect infrastructure are better equipped to realize the scale effect of technological innovation, and regions with a small digital divide can form a coordinated development pattern faster. Regions with superior institutional environments can accelerate the release of technological dividends and promote sustainable economic development through policy adaptation. This discussion leads to Hypothesis 1.

Hypothesis 1 (H1):

Increased technological innovation increases the coordinated development of regional economies, and this influence is non-linear.

2.2. Technological Innovation Affects the Upgrading of Industrial Structure and Enables Regional Coordinated Development

Technological innovation is a core force driving economic growth, impacting the coordinated development of regional economies mainly through the path of industrial structure upgrading. Two key dimensions are the upgrading and rationalization of the industrial structure. Technological innovation affects the coordination of regional economies by directly or indirectly acting on these two dimensions. Specifically, technological innovation promotes the deepening of the inter-regional industrial division of labor and improves resource allocation efficiency by enhancing inter-industry labor productivity. This optimizes the industrial proportion, corrects structural deviations, and ultimately achieves the goal of coordinated economic development.

When considering industrial structure upgrading, technological innovation promotes the dynamic transfer of industrial dominance by changing the technology intensity and production efficiency of different industries. Technological innovation significantly improves the production efficiency of primary industries through modern means, such as agricultural mechanization and biotechnology. This enables many individuals in the labor force to transfer to secondary and tertiary industries, changing the proportion of employment and output value among the three industry types. Furthermore, technological innovation in industrial fields and technological penetration in the service industry further strengthen the competitive advantages of secondary and tertiary industries. This accelerates upgrades to the industrial structure.

This process is reflected in the coordinated development of the regional economy. The leading technology regions form high-value-added industrial agglomerations through industrial upgrading, while the regions that are catching up optimize their industrial structure by undertaking industrial transfers. This narrows the development gap between regions through dynamic adjustment. Technological innovation promotes the evolution of the industrial structure toward higher efficiency levels and higher added value by improving labor productivity in various industries and optimizing collaborative relationships between industries [27]. The upgrading of industrial structure focuses on changes in industrial proportions and emphasizes the balanced improvement of inter-industry technological linkages and labor productivity.

Technological innovation affects this process in two ways. First, it directly improves the labor productivity of a single industry. For example, using industrial robots significantly improves the output efficiency of the manufacturing industry, and cloud computing technology optimizes the operating costs of the service industry. Second, technological innovation promotes collaborative innovation among industries through technology spillover effects. An example includes digital technology enabling traditional agriculture to evolve into smart agriculture. This promotes the integration of primary, secondary, and tertiary industries. These improvements have shifted the industrial division of labor between regions from low-end homogeneous competition to high-end differentiated cooperation. Technology-leading regions focus on R&D and high-end manufacturing, while other regions rely on their own resource endowments to develop support industries. This forms a complementary development pattern [28]. In addition, technological innovation also reduces factor flow restrictions; reduces information asymmetry; optimizes cross-regional talent, capital, and technology allocations; and further strengthens the positive effect of the upgrading of industrial structure on regional coordinated development.

The core index of this study must measure the efficiency of resource allocation among industries. Optimizing industrial structure rationalization is a complex interaction between dynamically adjusting the industrial structure and rebalancing economic systems driven by technological innovation [29]. Theoretically, rationalizing the industrial structure requires recognizing two dimensions of synergy. At the micro level, the proportion of output value and employment in various industries is important; this conforms to the comparative advantage of factor endowment [27]. At the macro level, the economic weight of different industries needs to be adapted to the regional development stage [30] (Chenery et al., 1986).

Technological innovation promotes this process through two paths. First, skill-biased technological progress promotes labor transfers from low-productivity sectors to high-value-added sectors by changing the elasticity of capital–labor substitution. This significantly improves structural deviations between traditional agriculture and manufacturing [31]. Second, digital technology penetration has led to new business forms, such as the platform economy, providing an “employment buffer” for the unemployed in traditional industries. Empirical studies show that the digital economy has increased employment elasticity in the manufacturing industry by 23–35% during the process of servitization [32].

When considering industrial linkages, the externality of the innovation network reshapes the industrial position through forward and backward linkage effects. For example, breakthroughs in new energy technology drive the research and development of upstream materials and promote the expansion of downstream applications. This better balances the GDP contribution of relevant industrial chains. Research on regional heterogeneity shows that differences in initial technological innovation capability may lead to a “Matthew Effect” in the rationalization process [33]; however, the upgrading speed of the industrial structure in underdeveloped regions can be increased by more than 40% by building a cross-regional technology transfer alliance [34]. This approach provides empirical support for China’s regional coordination policies, such as “counting East and counting West”.

Recent research has shown that general-purpose technologies such as AI can improve the convergence rate of the industrial structure rationalization index by 1.8–2.5 times, through the multiplier effect of the industrial permeability coefficient [35]. This suggests that the systematic importance of technological innovation continues to increase. The impact of green technology innovation on the rationalization of industrial structure follows a U-shaped curve [36]. The synergy between industrial policy and market mechanisms can significantly improve the rationalization adjustment efficiency [37]. This discussion leads to Hypothesis 2.

Hypothesis 2 (H2):

Increased technological innovation promotes the coordinated development of regional economies by increasing the upgrading and rationalization of industrial structure.

2.3. Adjustment Effect of Technological Innovation on Regional Coordinated Development

The mechanism of technological innovation is affected by the level of marketization. This highlights the need to clarify the internal logic of technological innovation and regional coordinated development and analyze the nonlinear regulatory role of marketization. The ultimate goal is to provide theoretical support for constructing a balanced regional development model.

Marketization is a key dimension of the institutional environment, and it has a dual regulatory effect on the impact of technological innovation in promoting coordinated regional development. This has been widely verified in academic circles [38,39] (North, 1990; Acemoglu et al., 2004). A higher degree of marketization increases the efficient flow of technological innovation resources among regions through effective price signals and competition mechanisms. This aligns with the theoretical expectation of neoclassical economics with respect to the importance of free-flowing factors to improve allocation efficiency [21]. When the factor market is mature, the technology spillover effect breaks through administrative boundaries and forms a cross-regional innovation network based on comparative advantage. One example of this spatial spillover effect has been seen in the practice of EU Innovation Integration [6,40].

Research shows that in regions with a high marketization index, the marginal output elasticity of enterprise R&D investment to surrounding regions can be increased by 30–40% [41,42]. This knowledge diffusion effect has significantly reduced the technological generation gap between regions, especially in the Yangtze River Delta and the Pearl River Delta in China, which shows a clear convergence trend. However, excessive marketization may lead to the excessive agglomeration of innovation factors in high-return regions; this could exacerbate regional differentiation. This is referred to as the “core periphery” dilemma by economic geographers [43].

When an institutional environment inadequately protects intellectual property rights, the negative externalities of technology imitation may inhibit the independent innovation power of underdeveloped regions, forming an “innovation depression.” This is common during technology catch-up in developing countries. This market-oriented paradox requires policymakers to establish an interregional innovation income redistribution mechanism [44]. It should accompany promoting market-oriented reform and balancing the regulatory effect of marketization on the spatial distribution of technological innovation through institutional designs, such as tax incentives for cross-regional technology transfer and patent pool sharing [45]. This policy concept was successfully verified through the eastern scientific and technological innovation revitalization plan after the reunification of Germany [46,47]. This discussion leads to Hypothesis 3.

Hypothesis 3 (H3):

The level of marketization regulates the role of technological innovation in promoting regional coordinated development. A higher degree of marketization is associated with technological innovation playing a more significant role in increasing regional coordinated development.

3. Study Design

3.1. Model Setting

The research hypotheses are first tested using a regression model to assess the impact of technological innovation on regional coordinated development:

$${RCD}_{i,t}={\alpha }_0+{\alpha }_1{DT}_{i,t}+{\alpha }_2{Z}_{i,t}+{\mu }_i+{\delta }_t+{\epsilon }_{i,t}$$

(1)

where ${RCD}_{i,t}$ is the level of regional coordinated development; ${DT}_{i,t}$ is the level of technological innovation; $Z_{i,t}$ represents the control variable; ${\mu }_i$ represents the individual fixed effect; ${\delta }_t$ represents the time fixed effect; and ${\epsilon }_{i,t}$ represents the random disturbance term.

To further explore the possible mechanism by which technological innovation impacts regional coordinated development, this study tests the intermediary role of industrial structure upgrading. This involves building a regression model that assesses the impact of technological innovation on industrial structure upgrading. The specific form of the model is as follows:

$${INS}_{i,t}={\beta }_0+{\beta }_2{DT}_{i,t}+{\beta }_2{Z}_{i,t}+{\mu }_i+{\delta }_t+{\epsilon }_{i,t}$$

(2)

where ${INS}_{i,t}$ is the upgrading or rationalization of the industrial structure; ${DT}_{i,t}$ is the level of technological innovation; $Z_{i,t}$ represents the control variable; ${\mu }_i$ represents the individual fixed effect; ${\delta }_t$ represents the time fixed effect; and ${\epsilon }_{i,t}$ represents the random disturbance term.

3.2. Index Selection

3.2.1. Explained Variable: Regional Coordinated Development Level (RCD)

In the context of the new development pattern, building a multi-level and multi-dimensional evaluation system to assess regional coordinated development has become a key issue in academic research and policymaking. Drawing on the theoretical framework of Zhang et al. (2021) [48], this study focuses on key areas such as narrowing the regional development gap, promoting the equalization of basic public services, improving the balanced accessibility of infrastructure, and enhancing people’s livelihood security to comprehensively measure the level of coordinated regional development. Coordinated regional development must coordinate the carrying capacity of resources and the environment and achieve green development while optimizing the allocation of resources. Based on this concept, this study selects four dimensions—infrastructure, living standards, resources and environment, and economic development—to construct an evaluation system comprising 18 indicators that cover four core dimensions.

This study aims to comprehensively reflect the coordinated development level of a regional economy, society, environment, and infrastructure. (1) From the perspective of infrastructure construction, it evaluates the investment and coverage level of the region in terms of public services, transportation, communications, and other hardware facilities and reflects the region’s ability to support residents’ lives and economic activities. (2) The dimension of people’s living standards reflects residents’ income, consumption, and overall welfare level, as well as the inclusive nature of regional economic development. (3) The dimension of resource and environmental carrying capacity evaluates the coordination between regional economic growth and the ecological environment and reflects the sustainable development capacity. (4) Regarding economic development potential and coordination, Porter’s (1990) competitive advantage theory emphasizes that regional development needs to rely on innovation, investment, and external market linkage [49]. Economic development potential and coordination assess the power, openness, and long-term sustainability of regional economic growth. This comprehensive evaluation framework responds to China’s strategic needs to build a modern economic system and provides a scientific policy tool for realizing the people’s desire for a better life. Indicator construction is shown in

Table 1. Construction of a regional coordinated development index.

	Secondary Indicator	Measurement Indicator	Unit	Nature
Coordinated development of the regional economy	Infrastructure Development	Per Capita Public Library Collection	Thousand volumes	+
		Number of Hospital Beds per Capita	Item	+
		Basic Medical Insurance Coverage Rate	%	+
		Basic Pension Insurance Coverage Rate	%	+
		Per Capita Telecommunications Revenue	CNY	+
		Mobile Phone Subscribers per 100 People	Ten thousand units	+
		Internet Users per 100 People	Persons	+
		Per Capita Road Area	Square meters per person	+
	Living Standards	Average Wage of Employees	CNY	+
		Per Capita Household Consumption Expenditure	CNY	+
		Per Capita GDP	CNY	+
	Resource and Environmental Carrying Capacity	Urban Green Coverage Rate	%	+
		Industrial Wastewater Discharge per Unit GDP	Tons per CNY
		CO2 Emissions per Square Kilometer	Tons per square kilometer
		Industrial Smoke and Dust Emissions per Unit GDP	Tons per CNY
	Economic Development Potential and Coordination	Ratio of imports and exports of goods to GDP	%	+
		Ratio of Fixed Asset Investment to GDP	%	+
		Regional GDP Growth Rate	%	+

.

3.2.2. Explanatory Variable: Technological Innovation (DT)

Given the particularities of evaluating regional innovation, this study’s core indicators are all comprehensive. This has multiple advantages. First, it comprehensively reflects the multi-dimensional characteristics of innovation activities. Comprehensive indicators avoid the limitations of a single indicator by integrating multiple dimensions, such as innovation input and innovation output. Second, this improves the scientific nature and objectivity of the evaluation. The entropy method applied in this study automatically determines the weight based on the discrete degrees of the data. This reduces deviations from human intervention. Third, the study approach enhances horizontal and vertical comparability. A comprehensive index normalizes the indicators of different dimensions to facilitate comparisons across regions, enterprises, or time periods. Fourth, this approach provides precise support for formulating policy and optimizing resources. Decomposing secondary indicators enables the identification of the short board in the technological innovation chain. This can help decision-makers optimize resource allocations, such as shifting funds to regions or enterprises with high innovation efficiency, to improve overall levels.

Table 2. Technological innovation indicator system.

	Secondary Indicator	Tertiary Indicator	Nature
Technological Innovation	Innovation Input	Number of R&D Personnel	+
		R&D Expenditure	+
		Proportion of Computer Software Personnel	+
	Innovation Output	Patent Applications	+
		Patents Granted	+

lists the indicators used to assess the level of substantive technological innovation.

3.2.3. Mechanism Variable: Industrial Structure Upgrading

This study focuses on the upgrading of the industrial structure. Based on Yuan H. and Zhu C. (2018) [50], the associated index system includes the two dimensions of industrial structure: upgrading and rationalization. These capture changes in industrial structure. They are calculated as follows:

➀

For industrial structure upgrading (z), the calculation formula is as follows:

$$Z_{i,t}={\sum }_{m=1}^3{y}_{i,m,t}\times {lp}_{i,m,t}, m=1,2,3$$

(3)

In Equation (3), $Y_{i,m,t}$ has the same meaning as in Equation (1). The term ${lp}_{i,m,t}$ represents the labor productivity of $m$ industry in region $i$ during period $t$. The calculation formula is as follows:

$${lp}_{i,m,t}=Y_{i,m,t}/L_{i,m,t}$$

(4)

where $Y_{i,m,t}$ represents the added value of $m$ industry in region $i$ during period $t$, and $L_{i,m,t}$ represents the employment of $m$ industry in region $i$ during period $t$. In Equation (3), the proportion of the output value, $y_{i,m,t}$, has no dimension, whereas the labor productivity, ${lp}_{i,m,t}$, has a dimension. Therefore, this study adopts the averaging method to eliminate the dimensionality; therefore, the assessed quality of industrial structure upgrading is dimensionless.

To complete the dimensionless processing, the average value is taken as the unit, and all data are removed as the average value. This is expressed as follows:

$$y_{ij}={\displaystyle \frac{x_{ij}}{x_j}}, \overline{x_j}={\displaystyle \frac{1}{m}}{\sum }_{i=1}^m{x}_{ij}$$

(5)

Averaging is often used in comprehensive evaluations, such as in gray relational analysis. This method assumes that all data are greater than 0; otherwise, it is not a suitable approach.

➁

Rationalization of the industrial structure (H)

This study uses the Theil index to measure the degree of rationalization of industrial structure in various cities. The index effectively considers the structural deviation in measuring different industrial output values and employment, as well as the different economic status of various industries. The specific calculation formula is as follows:

$$H_{i,t}={\sum }_{m=1}^3{y}_{i,m,t}\mathit{ln}\left({\displaystyle \raisebox{1ex}{$y_{i,m,t}$}\!\left/ \!\raisebox{-1ex}{$l_{i,m,t}$}\right.}\right),m=1,2,3$$

(6)

where $y_{i,m,t}$ represents the proportion of $m$ industry in $i$ region in the regional GDP during $t$ period, and $l_{i,m,t}$ represents the proportion of employees in $m$ industry in $i$ region as a part of the total employment during $t$ period. The rationalization of industrial structure reflects the output value structure and employment structure of China’s three major industries. A value of 0 indicates that the industrial structure is at an equilibrium level. A value other than 0 indicates that the industrial structure deviates from the equilibrium state, and the industrial structure is ineffective.

3.2.4. Adjusting Variables

The marketization index (MI) is calculated using Fan Gang’s marketization index.

3.2.5. Control Variables

The level of industrialization (SEC) is measured using the ratio of the secondary output value of region I to the regional GDP in t. Regional economic growth (RE) is measured using the per capita gross national product (GNP); this variable assesses the overall level of a country or region’s economy, making GNP an effective alternative variable of regional economic growth. The ln (GDP) is used to represent the logarithm of the per capita GDP. The level of foreign investment (FI) is measured using the proportion of FDI as part of the GDP. The level of human capital (RL) is measured using the proportion of ordinary college students and above per million permanent residents in the region. The logarithm of population density (PD) is measured using the amount of regional land area above the ratio of the total population of the region. The urbanization rate (UR) measures a region’s urbanization level, representing the percentage of the urban population as a part of the total population.

The descriptive statistics of each variable are shown in

Table 3. Descriptive statistics.

Variables	Obs	Mean	Variance	Min	Max
RCD	2558	0.1540	0.0454	0.0293	0.3460
DT	2559	0.0306	0.0471	0.0013	0.5460
Z	2559	0.5110	0.2670	0.0517	1.9800
H	2559	0.2900	0.2090	−0.0156	1.7210
MI	2569	11.9000	2.3060	4.9600	19.6900
SEC	2559	0.3680	0.1560	0.0000	1.0000
RE	2559	10.7200	0.5660	8.8420	13.0600
FI	2416	0.0173	0.0178	1.77 × 10−6	0.2100
RL	2537	0.0200	0.0259	4.08 × 10−6	0.1330
PD	2559	5.7600	0.9230	0.6830	7.8820
UR	2559	0.5550	0.1490	0.1810	1.0000

.

3.3. Data Sources

This study mainly uses panel data for 258 prefecture-level cities in China from 2011 to 2021. City-level data are mainly from the statistical yearbooks of China’s land and resources and the statistical yearbooks of China’s cities. Missing values are supplemented by interpolation.

4. Empirical Analysis

4.1. Benchmark Regression Analysis

Table 4. Benchmark regression of technological innovation’s influence on coordinated regional economic development.

	(1)	(2)
Variables	RCD	RCD
DT	0.0650 ***	0.0570 ***
	(0.0117)	(0.0117)
SEC		0.0044 *
		(0.0023)
RE		0.0180 ***
		(0.0015)
FI		0.0021
		(0.0196)
RL		−0.1430 ***
		(0.0446)
PD		0.0023
		(0.0018)
UR		0.0003
		(0.0064)
Constant	0.1520 ***	−0.0542 ***
	(0.000414)	(0.0192)
Year fixed	YES	YES
Urban fixed	YES	YES
Observations	2557	2400
R-squared	0.952	0.956

Notes: *** p < 0.01, and * p < 0.10; the values in parentheses are standard errors.

shows the benchmark regression analysis assessing the impact of technological innovation on the coordinated development of regional economies, using a two-way fixed effects model. The results show that in Models (1) and (2), the coefficients for technological innovation are 0.0650 and 0.0570, respectively; both results are significant at the 1% level, indicating that technological innovation has a significant role in promoting the coordinated development of regions. For each unit of technological innovation, the level of regional coordination increases by about 0.065 and 0.057 units, respectively. This result is consistent with the new economic growth theory and technology diffusion hypothesis. In other words, technological innovation promotes the coordinated development of the regional economy by improving production efficiency, optimizing resource allocation, and promoting industrial upgrading. After controlling for other variables, the coefficient for technological innovation decreased slightly but still remained highly significant, indicating that its impact was robust. These results support Hypothesis 1.

When examining the individual control variables, the regression result of the industrialization level is positive (0.0044 **) at a 5% significance level. This indicates that industrialization supports the coordinated development of the regional economy. Regions with a higher degree of industrialization usually have more perfect industrial systems and infrastructure, which drives the coordinated development of surrounding regions. The logarithmic regression result of per capita GDP is positive (0.0180 **) at a 5% significance level. This indicates that a higher level of economic development is associated with a higher degree of regional coordination. This is consistent with the expectation of the “Kuznets Curve,” which posits that after economic growth reaches a certain stage, the regional gap gradually narrows.

The regression result of human capital level is negative (−0.143 **) at a 5% significance level. This may reflect the “siphon effect” of human capital; this occurs when highly skilled talent concentrates in developed regions, thereby increasing the imbalance in regional development. This finding is consistent with Autor (2019) [51]. The coefficients of foreign investment level, logarithm of population density, and urbanization rate are not significant; as such, these factors have a nonsignificant direct impact on regional coordination.

4.2. Endogenous Test

Table 5. Instrumental variable: urban light data.

Variables	RCD
DT	0.5650 ***
	(0.0953)
Controls	YES
Observations	2400
R-squared	−0.733

Notes: *** p < 0.01; the values in parentheses are standard errors.

presents the endogenous test results. Using the instrumental variable (IV) method and the urban lighting data as the IV of technological innovation, this study analyzes the impact of technological innovation on regional coordinated development. Technological innovation is often closely related to the density of economic activities and the process of urbanization, while nighttime light intensity can effectively reflect regional economic activity, the level of infrastructure development, and the state of industrial agglomeration. Therefore, nighttime light intensity has a natural correlation with technological innovation, which meets the basic requirements of the correlation conditions of tool variables. From an exogenous perspective, lighting data indirectly affect the coordinated development of a regional economy, mainly by influencing technological innovation activities, rather than directly interfering with the economic balance between regions. Light-intensive areas usually attract a large number of high-tech enterprises, R&D institutions, and innovative talents, forming the spatial carrier of technological innovation. However, lighting brightness itself does not directly determine the degree of coordinated development of a regional economy, thus ensuring the exogeneity of the IVs.

From the empirical results, all the statistics of the IV test performed well. Anderson canonical correlation LM statistics significantly rejected the original assumption that the IV was not related to the endogenous variable at the 1% level. The Cragg Donald Wald F-statistic was far beyond the critical value of a weak IV, indicating that the light data have a strong explanatory power on technological innovation. Sargan test results further confirmed the exogenous nature of the IVs, that is, lighting data only affect the coordinated development of a regional economy through technological innovation channels. The final regression analysis results indicate that after controlling for the endogenous problem, the promotional effect of technological innovation on regional coordinated development is still significantly positive at the 1% level. This not only verifies the key role of technological innovation but also indicates that as a tool variable, urban lighting data can effectively address the endogenous bias of the model, providing more reliable causal evidence for the research conclusions.

4.3. Robustness Test

To assess the robustness of the research results, this study uses two robustness test methods: a replacement model analysis (OLS) and replacement of the core explanatory variables. The regression results are shown in

Table 6. Robustness test.

	(1)	(2)	(3)
	OLS	OLS	Replace the Explanatory Variable
Variables	RCD	RCD	RCD
DT	0.1720 ***	0.1190 ***	0.0005 *
	(0.0276)	(0.0248)	(0.0003)
Constant	0.1480 ***	−0.1540 ***	−0.0614 ***
	(0.00112)	(0.0181)	(0.0193)
Controls	NO	YES	YES
Observations	2558	2401	2400
R-squared	0.032	0.190	0.955

Notes: *** p < 0.01 and * p < 0.10; the values in parentheses are standard errors.

. The OLS regression method is used for Models (1) and (2). The coefficient of technological innovation is 0.172 in Model (1) and 0.119 in Model (2); both are positive at 1% significance level. This indicates that technological innovation plays a significant role in promoting regional coordinated development. Model (3), which replaces the core explanatory variable, “technological innovation,” with the “logarithm of the number of R&D personnel” results in a positive coefficient (0.000541), but with a reduced significance level of 10%. This is far lower than the significance level of the original technological innovation variable. This shows that the impact of R&D personnel’s input on regional coordinated development is weak. This may be because the variable only reflects the human capital input of innovation, rather than the direct technological output. However, the positive sign still indicates that technological innovation has a small promotional effect on regional coordination.

4.4. Heterogeneity Test

As presented in

Table 7. Heterogeneity analysis of technological innovation on the coordinated development of regional economies.

	(1)	(2)	(3)	(4)	(5)	(6)
	Eastern Region	Central Region	Western Region	Changjiang Economic Belt	Pearl River Delta Region	Yellow River Economic Belt
Variables	RCD	RCD	RCD	RCD	RCD	RCD
DT	0.0533 ***	0.0385	0.0733	0.1310 ***	−0.0143	0.0254
	(0.0123)	(0.0486)	(0.0681)	(0.0243)	(0.0182)	(0.0501)
Constant	−0.2020 ***	−0.0596	−0.0142	−0.0692	−0.1880 ***	−0.0475
	(0.0594)	(0.0386)	(0.0372)	(0.0591)	(0.0557)	(0.0346)
Controls	YES	YES	YES	YES	YES	YES
Year fixed	YES	YES	YES	YES	YES	YES
Urban fixed	YES	YES	YES	YES	YES	YES
Observations	908	848	644	967	867	794
R-squared	0.960	0.946	0.961	0.958	0.965	0.947

Notes: *** p < 0.01; the values in parentheses are standard errors.

, the impact of technological innovation on the coordinated development of regional economies indicates obvious differences among the eastern, central, and western regions. The technological innovation coefficient of the eastern region is 0.0533, which is significant at the 1% level, indicating that technological innovation has significantly promoted the coordinated development of the region. This result may be related to the strong economic foundation, perfect innovation system, and efficient technology transformation ability of the eastern region. In contrast, the coefficient of technological innovation in the central and western regions did not pass the significance test, reflecting the structural spear in regional development. Although the central region has a certain industrial base, the uneven distribution of innovation resources and the lag of industrial transformation may weaken the marginal effect of technological innovation. Moreover, although the central region has undergone industrial transfer from the east, the proportion of traditional industries is high, and the agglomeration of innovation factors is insufficient. This leads to a weak marginal effect of technological innovation, making it difficult to transform technological innovation into the driving force of coordinated development. The absolute value of the coefficient in the western region is large but not significant, implying that its innovation potential is limited by either an insufficient sample size or the short-term nature of innovation investment. First, the small sample size in the western region may lead to insufficient statistical efficacy, making it difficult to capture the real effect of technological innovation. Second, the innovation investment in the western region may not have formed a scale effect; the innovation resources are scattered, the transformation efficiency is low, and the infrastructure and talent reserves are relatively weak, thereby restricting the actual contribution of technological innovation. Finally, the internal development of the western region is uneven, and some provinces, such as Chengdu and Chongqing, may have demonstrated the positive role of technological innovation, while other underdeveloped regions are still in the stage of innovation accumulation, so the overall effect is diluted.

According to the grouping regression results of the Yangtze River Economic Belt, the Pan Pearl River Delta, and the Yellow River Economic Belt, there are significant differences in the impact of technological innovation. The technological innovation coefficient of the Yangtze River Economic Belt is as high as 0.131 and significantly positive at the 1% level, which is much higher than that of other economic belts. This result is closely related to its regional strategic positioning and highly integrated innovation network. The technological innovation coefficient of the Pan Pearl River Delta is −0.0143, which may be attributed to the imbalance in the distribution of innovation dividends in the region. For example, the technology spillover of the Guangdong Hong Kong Macao Greater Bay Area failed to effectively benefit the surrounding underdeveloped areas and even exacerbated the “siphon effect.” The excessive concentration of high-end innovation resources in core cities may lead to the further marginalization of peripheral areas in technology competition, thereby widening the gap within the region. In addition, the industrial structure of the Pan Pearl River Delta is highly diversified; however, some traditional industries may face the impact of technology substitution, which will exert pressure on employment and economic growth in the short term, offsetting the positive impact of technological innovation. This result also verifies Hypothesis 1, which states that the impact of technological innovation on the coordinated development of a regional economy is nonlinear. The coefficient of the Yellow River Economic Belt is 0.0254, which fails the significance test. This may be due to the constraints of ecological protection policies on industrial innovation. The region pays more attention to green transformation, and the economic pulling effect of traditional technological innovation is limited, implying that ecological constraints may limit the promotion of technological innovation on coordinated development. In addition, the level of foreign investment in the Pan Pearl River Delta is significantly positive, while it is not significant in other economic zones. This indicates that the regulatory effect of opening-up on regional coordination exhibits obvious regional heterogeneity.

A heterogeneity analysis comparing cities on both sides of the Hu Huanyong line further shows the differential impact of technological innovation, the results are shown in

Table 8. Heterogeneity of technological innovation on the coordinated development of regional economies—Hu Changyong line.

	(1)	(2)
	Southeast	Northwest
Variables	RCD	RCD
DT	0.0565 ***	0.7300 ***
	(0.0118)	(0.2670)
Constant	−0.1070 ***	−0.0249
	(0.0276)	(0.0696)
Controls	YES	YES
Year fixed	YES	YES
Urban fixed	YES	YES
Observations	2245	155
R-squared	0.956	0.959

Notes: *** p < 0.01; the values in parentheses are standard errors.

. The technological innovation coefficient of the southeast side is 0.0565, which is significantly positive. However, the absolute value is relatively small. This indicates that the traditional development path may lower the marginal effect of technological innovation. In contrast, the coefficients for the northwest and southwest reach a level of 0.730, which is significantly higher compared to the southeast at a 1% significance level. This may be due to the amplification effect of policy preferences (such as the Western Development policies) or greater elasticity due to the low economic base. Overall, the significant differences between the two sides of the Hu Huanyong line indicate that policymakers should consider the heterogeneity of geographical and population distribution when pursuing technology innovation. The northwest side can strengthen innovation investment but should stay alert to the risk of resource mismatch. In contrast, the southeast side should focus on technology diffusion and institutional optimization.

4.5. Mechanism Evaluation

The impact of technological innovation on regional coordinated development is mainly realized through the upgrading and rationalization of industrial structures.

Table 9. Mediating effect test of industrial structure.

	(1)	(2)
Variables	Z	H
DT	0.4760 ***	0.2780 ***
	(0.1120)	(0.0985)
Constant	0.4590 ***	0.2200 **
	(0.1070)	(0.0946)
Controls	YES	YES
Year fixed	YES	YES
Urban fixed	YES	YES
Observations	2401	2401
R-squared	0.881	0.844

Notes: *** p < 0.01, ** p < 0.05; the values in parentheses are standard errors.

presents the regression analysis where the industrial structure acts as the mediator. First, the upgrading of industrial structures indicates that the industry is upgrading in the direction of adding value and high technology; technological innovation is the core driving force of this process. The empirical results show that technological innovation has a significant positive impact on upgrading industrial structures (with a coefficient of 0.476). This indicates that technological innovation directly promotes regional industrial upgrading, narrowing the development gap. In addition, the level of human capital and the urbanization rate also play important roles. This indicates that accumulating high-quality labor and accelerating urbanization further strengthen the role of technological innovation in promoting advanced urbanization. In contrast, the impact of foreign investment and the level of industrialization are not significant. This may be because upgrading depends more on local innovation ability and talent reserve than on external capital or traditional industrial expansion.

Second, the rationalization of industrial structure reflects improvements in resource allocation efficiency among industries; technological innovation plays a key role in this process (the coefficient is 0.278). In contrast to the advancement of industrial resources, rationalization emphasizes coordinated and balanced development among industries. The level of industrialization (0.0541) has a significant positive impact on rationalization. This indicates that optimizing and adjusting the traditional industry can improve the efficiency of resource allocation. However, foreign investment has a significant inhibitory effect. This may be due to the excessive concentration of foreign investment in certain advantageous industries, leading to an imbalanced regional industrial structure. The impacts of human capital and urbanization rate are not significant, indicating that rationalization depends more on adjusting industrial policies and resource flows between regions than on the quality of labor or the size of cities.

Overall, technological innovation promotes the coordinated development of regions through two mechanisms. On the one hand, the advanced path relies on technological innovation and human capital to regionally transform high-tech industries and reduce the development gap. On the other hand, the rationalization path alleviates the structural differentiation within the region by optimizing resource allocations. Policymakers should focus on the core role of technological innovation and take differentiated measures for different paths. This could include strengthening the cultivation of local talent to promote upgrading and optimizing the layout of foreign capital to support rationalization. The desired outcome would be to achieve a more balanced regional development. These results support Hypothesis 2.

4.6. Additional Analysis of the Regulatory Effect

Technological innovation has a significant positive impact on regional coordinated development. The regression results in

Table 10. Moderating effect of the marketization level.

Variables	RCD
DT	0.0558 ***
	(0.0117)
DT × MI	0.0193 **
	(0.00843)
Constant	−0.0536 ***
	(0.0191)
Controls	YES
Year fixed	YES
Urban fixed	YES
Observations	2400
R-squared	0.956

Notes: *** p < 0.01, ** p < 0.05; the values in parentheses are standard errors.

show that the coefficient is 0.0558 and is highly significant at the 1% level of significance. This effect may be achieved by improving production efficiency, promoting industrial upgrading, and optimizing resource allocation. At the same time, the adjustment effect of the marketization level is also significant. Its coefficient is 0.0193, indicating that improving the marketization level further strengthens the positive impact of technological innovation on regional coordination. Marketization provides a favorable environment for diffusing and applying technological innovation by improving the competition mechanism and reducing barriers to factor flow. This amplifies the effect when promoting regional coordination.

The impact of the marketization level is nonlinear. There is a double threshold effect (Figure 1), with specific thresholds of 9.9643 and 12.5361. This means that different intervals in the marketization level have different effects on the relationship between technological innovation and the coordinated development of the regional economy. When the marketization stage is low (marketization index ≤ 9.9643), the imperfection of the market mechanism restricts the diffusion and application of technological innovation. This results in a weak role in promoting coordinated economic development. In the medium marketization stage (9.9643 < marketization index ≤ 12.5361), as the degree of marketization improves, the positive effect of technological innovation increases, and the optimization of the institutional environment provides more efficient support for technological transformation. In the high marketization stage (marketization index > 12.5361), the perfect market mechanism helps technological innovation maximize its potential, significantly increasing the coordinated development of the regional economy.

The likelihood ratio (LR) statistic in the threshold graph further verifies the significance of the double threshold; this indicates that the regulatory effect of the marketization index is nonlinear. This study reveals a key regulatory role for marketization in the relationship between technological innovation and the coordinated development of regional economies. This provides an important basis for policymaking. When promoting coordinated regional development, it is important to combine the local marketization level with the adoption of differentiated technological innovation support strategies to achieve more efficient regional coordinated development. These results support Hypothesis 3.

5. Conclusions and Policy Recommendations

5.1. Research Conclusion

In recent years, global research on technological innovation and regional coordinated development has generally emphasized the core role of technological innovation; however, its effect exhibits significant spatial heterogeneity and conditional dependence. This empirical study reveals that the positive impact of technological innovation on regional coordination is consistent with the research conclusion of European and American developed economies [52]. The effect of the EU’s technological innovation on narrowing the regional gap through industrial chain integration, as observed by Baldwin (2019), is slightly higher than that of this study, which may be related to the EU’s effective factor flow policy [53]. However, in developing countries, the coordination effect of technological innovation is often subject to the institutional environment [53]. A comparison between China and India revealed that the technological innovation coefficient of Indian states is significantly lower than that of eastern China [54].

The promotion effect of China’s technological innovation on regional coordinated development indicates obvious spatial heterogeneity. The eastern region drives the development of surrounding areas through innovation diffusion, a process similar to the EU’s “smart specialization strategy” [55]. However, the effect of the western region is not significant, which contrasts with the marginal benefit improvement achieved through policy preference in less developed regions of developing countries, such as India and Brazil, highlighting the dual challenges of geographical constraints and institutional barriers in western China. This finding provides important insights for sustainable regional development, highlighting that simply increasing innovation investment makes it difficult to break through the rigid constraints of geography and system, and it is necessary to establish an innovation governance system that matches regional characteristics.

The double threshold effect of the marketization level, as found in this study, further reveals the key regulatory role of the institutional environment. When the marketization index is lower than the first threshold of 9.96, the coordination effect of technological innovation is almost offset by institutional barriers, which is consistent with Rodrik’s (2018) [56] theory of “institution technology fit” and Acemoglu and Robinson’s (2019) theory of the “inclusive institution” threshold [57]. China’s higher institutional dependence reflects the typical characteristics of transition economies. From the perspective of sustainable development, this implies that coordinated regional development needs to simultaneously promote market-oriented reform and technological innovation, while avoiding the inefficient allocation of innovation resources caused by institutional lag.

In terms of mechanism, industrial upgrading reflects the particularity of developing countries. The leading role of industrial rationalization is more prominent in central and western China, which has policy value for sustainable industrial transformation. This finding has policy value for Sustainable Industrial Transformation. Unlike the advanced path of EU “Industry 5.0”, developing countries need to consider both technological frontier catch-up and the efficiency improvement of existing industries [58], which is supported by the study by Xie and Zhang (2015) based on the data for China’s manufacturing industry [59]. The high-elasticity coefficient on the northwest side of the Hu Huanyong line is consistent with the conclusion of Li and Yu (2023) regarding the quantitative evaluation of the effect of China’s regional policy intervention, that is, short-term policy dividends may mask long-term structural risks [60].

A comprehensive comparison reveals that the regional coordination effect of technological innovation is significantly conditional. The success of the eastern region confirms the theory of “innovation ecosystem” [61], while the similarity of the “innovation island” phenomenon between the western region and Latin America [62] indicates that the spatial mismatch of innovation elements is a common challenge. These findings provide a multi-dimensional basis for the policy design of regional sustainable development. Therefore, we should not only pay attention to the adaptability of innovation investment and the institutional environment but also choose differentiated industrial upgrading paths according to the development stage and be vigilant about the long-term locking effect of policy intervention. Future research needs to explore how to build a sustainable innovation governance system in geographically disadvantaged regions to balance short-term benefits with long-term development resilience.

5.2. Policy Recommendations

Based on the research conclusions outlined above, this study puts forward the following policy recommendations to optimize the role of technological innovation in promoting regional coordinated development:

First, in the eastern technology leading region, efforts should be made to build an open innovation network and strengthen the technology diffusion channel and knowledge spillover effect. Some suggestions include breaking through the barriers of administrative divisions; building a cross-regional innovation consortium based on existing industrial clusters; and reducing the technology absorption cost in the central and western regions through sharing R&D platforms to unify technical standards and other measures. Others involve learning from the “innovation value chain matching” mechanism in the EU smart specialization strategy and establishing a precise docking system between the eastern technology exporter and the midwest acceptor. For the western region, whose marketization level is lower than the first threshold (9.96), the policy focus should be on optimizing the institutional environment and cultivating a market mechanism. This includes simplifying the administrative approval process, strengthening the judicial practice of intellectual property protection, and improving the pricing mechanism of factor market and other basic institutional reforms, rather than simply increasing innovation investment. Cities with appropriate conditions in the west should be selected to carry out the “system technology” collaborative reform pilot, focusing on verifying the coupling relationship between the marketization process and the effect of innovation policies.

At the industrial policy level, it is necessary to implement a dual-path promotion strategy of differentiation. In the central and western regions, efforts should be prioritized to support the rationalization and upgrading of industries, as well as to improve the total factor productivity of traditional industries through special funds for technological transformation. They should also implement subsidies for production process optimization, while combining the characteristics of local factor endowment to avoid blindly copying the high-end path of the east. In view of the special situation on the northwest side of the Hu Changyong line, the policy design needs to balance short-term benefits and long-term sustainability. Referring to Australia’s experience in alleviating the north–south divide, investment in infrastructure connectivity, such as transportation and communications, should be increased, while supporting the implementation of the technological innovation capacity-building plan to prevent the formation of a “policy-dependent” development model.

Finally, it is recommended that a dynamic policy adjustment mechanism be established. Through the construction of a regional development monitoring system that includes core indicators such as marketization index, industrial coordination degree, and technology absorption capacity, the effect of technological innovation policies is evaluated in real time. When the level of marketization in a region exceeds the threshold, the combination of policy tools should be adjusted in time to realize the gradual transformation from “system cultivation leading” to “innovation incentive leading.” This refined policy management approach is not only consistent with Rodrik’s principle of system technology fit but can also effectively avoid the common policy fossilization problems in developing countries. Moreover, all regional coordination policies should maintain a moderate elastic space, allowing local governments to adaptively adjust according to their local technology digestion capacity and industrial base, thereby coping with inherent uncertainty and regional heterogeneity in the process of technological innovation.

6. Research Limitations and Future Prospects

This study has some limitations, which are mainly reflected in the data sources and research methods. First, although the panel data for 258 prefecture-level cities in China from 2011 to 2021 are used, the data mainly rely on the official statistical data of China, and there may be problems associated with inconsistent statistical caliber or lagging data updates. In addition, some missing values are supplemented using the interpolation method, which ensures the integrity of the data but may introduce measurement errors and affect the accuracy of the results. Second, in terms of research methods, although the fixed effects model and robustness test can effectively control for some endogenous problems, the relationship between technological innovation and regional coordinated development may be affected by other unobserved variables, resulting in a certain deviation in causal inference. Furthermore, the research does not cover more microscopic data at the enterprise or individual levels, which may limit the in-depth analysis of the mechanism of technological innovation.

Future research can improve the reliability of the conclusion by conducting cross-validation of multi-source data and undertaking more detailed microscopic analysis. Step-by-step instructions for conducting further research are as follows: First, expand the data sources, combine the enterprise microdata or satellite remote sensing data, as well as other new data, and enhance the comprehensiveness and timeliness of the indicators. Second, enrich the mechanism analysis, including environmental regulation, digital infrastructure, and other variables, and reveal the multi-channel mechanism of technological innovation. Third, strengthen cross-border or cross-regional comparative research, explore the differential impact of technological innovation on regional coordinated development under different institutional background conditions, and provide policy implications for global sustainable development. In addition, the successful experiences and bottlenecks in typical regions can be analyzed in-depth, in combination with case studies or field surveys, to provide empirical support for the theoretical findings.