Unlocking the Power of Economic Agglomeration: How Digital Finance Enhances Urban Land Use Efficiency Through Innovation Ability and Rationalization of Industrial Structure in China

Abstract

With the rapid progression of urbanization, Chinese cities are encountering an increasingly severe shortage of land resources. To examine the role of digital finance (DF) in promoting economic agglomeration (EA) and subsequently improving urban land use efficiency (ULUE), this paper focuses on 274 prefecture-level cities from 2013 to 2020. By employing an interaction effect model, the study analyzes how DF influences the relationship between EA and the enhancement of ULUE. The findings demonstrate that DF positively moderates the effect of EA on enhancing ULUE. The analysis reveals that DF can accelerate resource flows and more effectively promote the enhancement of ULUE by amplifying the positive influence of EA on regional innovation ability (IA) and the rationalization of industrial structures (RIS). Furthermore, the heterogeneity analysis indicates that this moderating effect is particularly pronounced in China’s central and eastern regions, areas with higher rates of internet penetration, and regions with stronger innovation capacities. Based on these findings, this paper proposes policy recommendations such as promoting regional integration and supporting innovation in DF. These insights contribute to the theoretical discourse on EA in the digital era and provide critical and practical guidance for optimizing the development of agglomeration economies and constructing a high-quality development framework.

1. Introduction

With the rapid advancement of its economy, China has experienced a swift process of urbanization. Given its substantial population, the country is particularly vulnerable to the misallocation of land resources [1] and exhibits low utilization efficiency [2]. As illustrated in Figure 1, the growth rate of output per unit of urban land in China has consistently declined in recent years, even registering negative growth over the past two years. This trend underscores the inefficiency of land use in Chinese cities and highlights the pressing need for a comprehensive reevaluation of existing land use models. This inefficiency is primarily due to the lack of scientific planning, a focus on short-term economic gains [3], and an absence of sustainable development principles [4]. Yanjiao, located in Langfang, Hebei Province, exemplifies a common issue. Its excessive dependence on real estate development has resulted in a significant number of unoccupied residential units, leading to the land idle phenomenon and an inefficient allocation of land resources. Additionally, the problem of underutilized land is compounded by the region’s narrow industrial base and limited employment prospects, which hinder its ability to attract and sustain a stable, long-term population. Inefficient urban land use has led to numerous adverse outcomes [5], directly impacting the city’s economic benefits and environmental sustainability. Regarding the economic implications, the low urban land use efficiency (ULUE) can precipitate an unwarranted enlargement of the urban built-up area [6], thereby escalating the expenses associated with infrastructure and public services. The phenomenon of urbanization characterized by extensive scale yet subpar quality poses a threat to vehicular traffic safety [7]. Consequently, the exorbitant expenditures on public security it engenders represent a substantial economic strain. In the aggregate, these factors can diminish the economic advantages derived from urban development [8]. Moreover, the mismanagement and inefficient utilization of land resources have led to a scarcity in the availability of urban land [9]. This scarcity, coupled with the persistent escalation of land prices [10], intensifies the financial burden on housing and commercial properties. Consequently, this situation adversely impacts the living standards of the populace and the operational expenses for enterprises. Environmentally, the unrefined development model of urban land use often leads to pollution and ecological damage, complicating urban management and environmental governance [11]. Therefore, enhancing the ULUE is a pressing issue in China’s urbanization process. Economic agglomeration (EA), an intensive economic activity [12], can generate various spillover effects and economies of scale [13], thereby increasing land output per unit and improving ULUE. This has been substantiated by multiple studies [14,15]. First, as enterprises and talent concentrate in an area, the competition pressure of innovation intensifies, potentially hindering small and medium-sized enterprises (SMEs) that may struggle to cope with the fierce market competition, limiting their innovation ability (IA) [16]. Second, excessive concentration of similar industries can lead to intensified intra-industry competition, resource wastage, and reduced profit margins [17]. Currently, with the development of digital technology, China has gradually moved into the digital age. Digital finance (DF), emerging as a novel financial model driven by advancements in digital technology, is characterized by its broad accessibility and efficiency [18]. It holds the potential to influence resource allocation and information flow, which can, in turn, enhance and advance the agglomeration economy. Additionally, the inclusive nature of DF fosters enterprise innovation and promotes the improvement of production technologies, improving the efficiency of enterprise production and product quality and ultimately contributing to more efficient land use. In this context, examining the mechanism through which EA improves ULUE and how DF can enhance the EA is of great practical significance. This analysis can provide feasible recommendations for promoting national high-quality development.

EA is a type of economic development that reflects the concentration and spatial distribution of economic activities within a particular area [19]. This phenomenon primarily arises from market forces such as economies of scale, economies of scope, and externalities, including knowledge spillover effects [20]. China has actively pursued EA since its reform and opening up, fostering regional economic prosperity through the creation of various forms of industrial agglomerations [21]. As illustrated in Figure 2, different types of EAs in China have proliferated rapidly, both in terms of the number and diversity of agglomerations. Current studies on the socio-economic aspects of EA mainly focus on its scale and spillover effects [22], which have a substantial impact on regional economic development. Specifically, the concentration of economic activities can effectively enhance knowledge and technology spillover effects among industries [23]. Through frequent interactions and information exchanges among enterprises, firms within agglomeration areas can more readily access external knowledge and advanced technologies, thereby improving their IA and production efficiency. This promotes the transformation and upgrading of economic structures and supports the development of knowledge-intensive and high-tech enterprises, enhancing regional economic resilience [24] and contributing to the innovative and coordinated nature of economic development [25]. However, the limitations of EA can present challenges to economic progress. If enterprises in a region fail to form coordinated relationships, the increased competition resulting from agglomeration may impose severe survival pressures on SMEs. Given that SMEs are disadvantaged relative to large enterprises in terms of resources, technology, and financial capital, EA may prevent these firms from acquiring sufficient resources for development, leading to “excessive competition” and “crowding effects” [26]. The impact of EA on ULUE, which is vital to China’s pursuit of economic development, has also been explored in the literature. EA can generate labor market externalities, technology externalities, and capital market externalities, thereby enhancing ULUE. However, the limitations described above may prevent EA from fully fulfilling its role in improving ULUE [27]. Consequently, relaxing the limitations of EA to empower its more effective functioning is critical for China to promote innovative and coordinated development. DF, as a novel financial service model emerging in the digital era, has the potential to induce far-reaching changes in the distribution of capital resources and facilitate the transmission of information elements (transaction data, market information, credit information, etc.) [28], thereby indirectly promoting the flow of technological, labor, and capital resources [29]. In contrast, the traditional financial model, characterized by inefficient information acquisition and processing, limited access to financial services, and a narrow range of financing products, has hindered the pace of capital flow. As a result, capital cannot be promptly directed towards highly efficient and innovative industries within the region, leading to an irrational industrial structure and restricted innovation capacity. Therefore, DF may offer a viable pathway to maximize the benefits of EA. Most of the existing literature examines the impact of either EA or DF on ULUE from a singular perspective. However, improvements in ULUE typically result from the combined influence of multiple factors. There is limited research that considers DF as a tool to enhance EA or explores how DF contributes to the improvement of EA. This paper aims to address this gap by offering insights and strategies to enhance effective economic agglomeration and promote innovative and coordinated development in China.

ULUE refers to the optimal allocation and effective use of land resources within a specific urban area to maximize economic, social, and ecological benefits [30,31]. This concept encompasses not only the physical use of land but also the strategic allocation of resources during land use, the full realization of land value, and environmental protection. Enhancing ULUE is crucial at both economic and ecological levels. Economically, improved ULUE implies a more strategic allocation and utilization of land resources. This leads to the optimal distribution of resources, reducing land waste and underutilization, thereby increasing the overall efficiency of economic activities [32]. Efficient land use can also minimize the area of land occupied, thus lowering leasing or purchasing costs for businesses. This allows enterprises to allocate more capital toward production and research and development activities, thereby bolstering their competitiveness [33]. At the socio-ecological level, efficient land use is often associated with more strategic land development planning, which alleviates the excessive consumption of natural resources, alleviates environmental pressures, and promotes sustainable development [34]. Therefore, amidst the challenges of accelerated urbanization, population growth, and land resource scarcity, improving ULUE has become a focal point for governments and scholars. As the key to ULUE lies in land output, enhancing the production efficiency of economic activities per unit of land is crucial to improving overall ULUE. Industry, as the primary actor in regional economic activities, directly influences ULUE through its characteristics. Industrial agglomeration can fully exploit the diverse values of land and increase output per unit of land. Moreover, a well-structured and advanced industrial form can increase value added, thereby further increasing land output per unit [35,36]. In the context of the digital era, industries undergoing digital transformation are more innovative and service-oriented, capable of providing higher added value to products. This transformation is, therefore, an essential strategy for enhancing productivity. DF is an important part of this. Its development can alleviate information asymmetry and the financing constraints of enterprises [37], thereby improving regional innovation capabilities and overall regional productivity. DF can also promote the flow of resources and the rationalization of the industrial structure [38] (RIS), thereby improving production efficiency. By leveraging big data and artificial intelligence technologies, DF platforms can offer more precise financing and investment analyses for urban land projects [39], thereby enhancing the ULUE. Additionally, the application of blockchain technology in the financial sector can streamline processes related to land collateralization, investment, and financing. This reduces intermediary fees and time costs, expediting the initiation of land development projects and helping to prevent land idleness [40]. However, most of the existing research emphasizes the outcomes of EA’s impact on regional economic development in the digital era, with limited attention given to the specific processes through which digital economy factors transform EA and, in turn, influence economic development. We investigate how DF can moderate EA, to enhance regional economic output. This research holds significant theoretical value for informing government policies on EA and industrial transformation. Additionally, it offers important practical implications for the advancement of DF and the adjustment of industrial structures.

This paper uses prefecture-level cities of the People’s Republic of China as a sample from 2013 to 2020 to examine how DF intervenes in EA to enhance its impact on improving ULUE. The study reveals several key findings. First, it confirms that EA indeed has a positive effect on ULUE. Second, this research demonstrates that DF plays a positive role in enhancing EA to improve ULUE. It achieves this by fostering the synergistic development of industries within the region and promoting sustainable competition among them, thereby strengthening the EA and improving ULUE. Third, the study identifies a heterogeneous effect of DF on EA’s ability to enhance ULUE across different regions. This effect is particularly pronounced in economically developed areas and regions with a robust digital foundation, where the regulatory impact of DF is more substantial.

The research contributions of this paper are as follows: First, by introducing DF as a moderating variable, this paper expands the understanding of the factors influencing ULUE, specifically revealing how the interactive effects of DF and EA improve ULUE. This provides a new theoretical framework for understanding the optimal allocation of land resources. This contribution not only enhances the understanding of DF’s role in regional economic development but also offers a new perspective for regional economic policy formulation, highlighting the crucial role of DF in promoting the coordinated development of regional economies.

Second, this paper enriches the research on DF and social benefits by delving into the regulatory effect of DF on EA, a topic that has been underexplored in the current literature. Unlike existing studies that primarily focus on the impact of DF on isolated areas such as enterprise innovation [41] and financing costs [42], this paper takes a holistic approach by investigating how DF can enhance ULUE through the lens of regional EA. It examines how DF promotes industrial synergistic development and moderates competition, thereby broadening the scope of DF-related studies. This research introduces a new paradigm for land resource management in the digital economy era.

Fourth, this paper innovatively uncovers the heterogeneity of DF’s moderating effect across different regions. Existing literature primarily addresses the direct impact of DF on the economic development of various regions [42,43]. This study breaks new ground by considering DF as a regulatory tool and examining the variation in its regulatory effects across different regions. This analysis not only enriches the body of research on DF but also holds significant practical value for regional governments seeking to utilize digital tools to optimize local economic development in the digital era.

The structure of this paper is organized as follows: The second section presents the theoretical analysis and formulation of hypotheses. The third section outlines the research model design, as well as the selection and construction of variables. The fourth section provides the empirical analysis, which includes descriptive statistics, basic regression, and robustness testing. The fifth section offers a further analysis, concentrating on mechanism testing. Finally, the sixth section concludes the study and provides policy recommendations.

2. Theoretical Analysis and Research Hypothesis

Since the reform and opening up, China’s economy has experienced rapid growth in terms of “volume”. Inefficient land use has contributed to a slowdown in economic growth, highlighting that such unsustainable development practices are no longer aligned with China’s current national conditions. Consequently, China’s economy needs to shift to an innovation-driven, moderate-growth economic development model. High-quality economic growth necessitates the efficient utilization of natural resources, particularly for a country like China [44], which is characterized by a large population and limited land resources. Therefore, enhancing the ULUE is essential. As the primary actors in land use, the behaviors of economic agents play a crucial role in influencing ULUE.

2.1. Digital Finance, Economic Agglomeration, and Urban Land Use Efficiency

EA can generate economies of scale and synergistic effects, optimize resource allocation, and thereby promote the efficient use of land resources. Firstly, according to the theory of economies of scale [45], EA results in a significant reduction in production costs through the geographic concentration of enterprises and industries [46]. This concentration allows enterprises within the agglomeration area to share public infrastructure, reducing the need for redundant construction and minimizing resource waste. This externality effect, in turn, enhances the efficiency of intensive land utilization [47]. Secondly, from the perspective of resource allocation theory in neoclassical economics [48], EA facilitates the optimal allocation of factor resources and increases the marginal productivity of land. In areas of agglomeration, the geographic concentration of production factors enables the market mechanism to allocate limited land resources more effectively, promoting the flow of resources, achieving optimal land use, and ultimately improving ULUE. Finally, EA enhances ULUE by promoting the spillover effects of knowledge and technology. This process strengthens regional innovation capacity, advances production technologies, and ultimately boosts regional output.

One of the key mechanisms through which EA enhances ULUE is by increasing the mobility of production factors—including information, capital, technology, and land—through a reduction in geographic distances. This will likewise attract rural labor to move to the cities for higher pay. Enhanced mobility of these factors can foster the growth of high-value-added and highly efficient industries, thereby improving regional IA and, consequently, ULUE. In the digital era, however, geographic agglomeration is no longer the only method for facilitating resource mobility. The swift progress of information technology has markedly enhanced both the fluidity and the pace of resource exchange [37]. As a hallmark of the digital era, the development of DF has provided substantial support for the EA. Firstly, DF optimizes the allocation of capital among enterprises through mechanisms such as online payment systems, intelligent credit services, and big data risk management. These innovations reduce financing costs [42] and shorten the capital flow cycle, thereby promoting the movement of capital among various enterprises and industries. In the context of EA, industries and enterprises are diverse, creating favorable conditions for the movement of capital. Thus, the integration of DF into the agglomeration economy can enhance its capacity to improve ULUE. Secondly, digital financial platforms increase transparency and provide real-time information, thereby reducing information asymmetry [49]. This enables production factors such as labor and technology to respond more swiftly to market fluctuations, increasing their mobility within the agglomeration area. Consequently, DF accelerates the reallocation of production factors, further improving the ULUE.

Based on this foundation, this paper proposes the following:

H1a. EA can improve ULUE.

H1b. DF can enhance the effect of EA on ULUE.

2.2. Digital Finance, Innovation Ability, and Rationalization of Industrial Structure

While traditional EAs can enhance ULUE, they exhibit certain limitations. From the perspective of a single industry, EA can foster knowledge spillovers and promote overall innovation within the agglomeration area. However, it may also lead to a Matthew effect [50], where dominant enterprises in the same industry can accumulate more innovation resources, thereby further enhancing their innovative capabilities. In contrast, SMEs in the same industry, despite benefiting from knowledge spillovers, often lack the necessary resources to capitalize on this knowledge or to pursue secondary innovation due to their similarity to larger enterprises. This disparity prevents the full realization of the region’s innovation potential, thereby limiting the improvement of innovation output. From the perspective of multiple industries, EA enriches the regional industrial structure, creating greater opportunities for resource allocation and promoting the rationalization of this structure to some extent. However, the lagging effect of traditional resource allocation [51] hinders the swift and efficient transfer of resources to high-productivity industries, thereby constraining the rationalization process. Given the rapid changes in the economic landscape driven by technological advancements, if the regional industrial structure fails to adapt quickly to market shifts and rationalize its resource allocation, it may result in issues such as industrial overcapacity and excessive market concentration. Consequently, these limitations inhibit the EA’s potential to fully enhance ULUE.

As a crucial development in the financial sector of the digital era, DF can effectively enhance the accessibility and mobility of resources, thereby maximizing the positive impact of EA. First, the realm of DF facilitates an array of avenues for the distribution of financial resources, thereby accelerating the rate at which production factors are directed towards SMEs. Such a mechanism can effectively counteract the Matthew effect prevalent within the industry, subsequently enhancing the aggregate regional innovation capability. Leveraging technologies such as big data, artificial intelligence, and blockchain, DF can more accurately evaluate the creditworthiness of SMEs, reducing the cost of risk assessment for financial institutions and broadening SMEs’ access to financing. With the support of DF, SMEs can secure timely financial backing and improve resource acquisition efficiency through digital platforms, providing them with greater financial security and flexibility in their innovation activities. Moreover, DF facilitates information sharing and the establishment of cooperative networks among enterprises, enabling SMEs to access more innovative resources and market opportunities, thereby advancing their technological research and product innovation. Thus, DF plays a pivotal role in alleviating the over-concentration of resources in large enterprises within the industry, enhancing the knowledge dissemination and innovation spillover effects of EA, and elevating the region’s overall IA. An improved regional IA not only fosters technological progress and boosts production efficiency [52] but also increases product added value [53] and output per unit of land. Second, by enhancing the efficiency of resource flows, DF has substantially accelerated the allocation and transfer of resources across different industries, thereby promoting RIS. In traditional financial systems, factors such as information asymmetry, high transaction costs, and delayed capital flows often restrict the movement of resources between industries, leading to resource over-concentration in certain sectors while others face shortages. Through the application of big data technology, DF is able to collect, integrate, and analyze vast amounts of multidimensional data, enabling a comprehensive, dynamic, and precise evaluation of an enterprise’s creditworthiness. In addition, blockchain technology within DF facilitates the use of smart contracts, which execute automatically when predefined conditions are met, thereby reducing transaction costs and enhancing the efficiency of financial services. These advancements have the potential to expand financing channels for SMEs and increase their access to available funding. These tools allow resources such as capital, technology, and labor to migrate more swiftly from low-efficiency industries to high-efficiency ones, promoting the optimal allocation of inter-industry resources. DF has significantly bolstered the resilience to risks [54] and the total factor productivity [55] within the commercial banking sector. It has facilitated the provision of more extensive and stable financing avenues for emerging industries and SMEs. This enhancement enables these entities to secure substantial resource support during the nascent phases of their development. It also optimizes the external business environment and the survival conditions of these entities, thereby fostering further industrial diversification. Therefore, by enhancing the efficiency of resource flows, DF facilitates the rational allocation of resources across various industries, accelerates the process of RIS, and expedites the reallocation of resources from traditional industries characterized by low value-added, inefficiency, and excessive land use to new industries that are high-value-added and high-efficiency [56], thus improving ULUE. With its inherent digital advantages, DF provides more resources to SMEs within EAs and improves the efficiency of resource flows between industries, thereby amplifying the positive impact of EA on ULUE. Based on this analysis, this paper proposes the following:

H2. DF boosts the effect of EA through IA/RIS on ULUE.

The theoretical mechanism for this article is shown in Figure 3.

3. Research Design

3.1. Data Sources and Data Cleaning

In this study, 274 cities at the prefectural and municipal levels and above in China were selected as the research subjects, covering the period from 2013 to 2020. The data sources included the Digital Finance Index published by the Digital Finance Research Center of Peking University, the China Urban Statistical Yearbook, the China Urban Construction Statistical Yearbook, provincial and city statistical yearbooks, and the Wind database. After the elimination of missing values, a total of 2178 sample data points were obtained. Additionally, this paper employed a 1% winsorization to mitigate the influence of outliers on each continuous variable [56].

3.2. Model Setup

3.2.1. Existence Test Model

This paper first aims to verify that EA can enhance ULUE and that DF has a positive moderating effect on this relationship. Taking into account the results of the correlation analysis and the multiple variance inflation factor test from the ordinary least squares (OLS) model, and following the outcomes of the Hausman test for random effects and fixed effects, this study constructs the following two-way fixed effect model and interaction effect model to examine the relationship between EA and ULUE, as well as the moderating effect of DF:

$$U{LUE}_{ct}={\alpha }_0+{\alpha }_1{EA}_{ct}+{\gamma }_1\sum {Controls}_{cjt}+{\mu }_c+{\delta }_t+{\epsilon }_{ct}$$

(1)

$$U{LUE}_{ct}={\beta }_0+{\beta }_1{DFEA}_{ct}{+\beta }_2{EA}_{ct}+{{\beta }_{3}DF}_{ct}+{\gamma }_2\sum {Controls}_{cjt}+{\mu }_c+{\delta }_t+{\epsilon }_{ct}$$

(2)

Subscript c is the region, t is the year, and j indicates the type of control variable. The explanatory variable $U{LUE}_{ct}$ denotes urban land use efficiency, the explanatory variable ${EA}_{ct}$ denotes the scale of economic agglomeration, and ${Controls}_{cjt}$ denotes the control variables at the regional level. ${\mu }_c$ is the individual city fixed effect, and ${\delta }_t$ is the year fixed effect. Models that include interaction terms provide an empirical means to assess whether moderating variables influence the relationship between the explanatory and dependent variables. Using these models, one can determine whether there is a moderating effect by observing the interaction term coefficient. According to the previous section, if EA promotes ULUE, it is expected that ${\alpha }_1$ in the model should be positive and significant. And if DF has a positive moderating effect on EA to improve ULUE, ${\beta }_1$ should be positive and significant.

3.2.2. Moderated Effects Model with Mediation

To explore how DF influences the relationship between EA and ULUE, this paper employs a two-step method [57] to test the mediating effect of the moderating variable and construct a channel model. The model is set as follows:

$$M_{ct}={\theta }_0+{\theta }_1{DFEA}_{ct}+{\theta }_2{EA}_{ct}+{{\theta }_{3}DF}_{ct}+{\gamma }_3\sum {Controls}_{cjt}+{\mu }_c+{\delta }_t+{\epsilon }_{ct}$$

(3)

$M_{ct}$ is a mechanism variable.

3.3. Variable Definitions and Descriptions

3.3.1. Explained Variable: Urban Land Use Efficiency

The primary explanatory variable in this paper is ULUE, which primarily assesses the diversity and economic productivity of a city’s land use. Drawing on existing literature [58], this study employs the non-farm economic output per unit of land as a measure of ULUE. Non-farm economic output encompasses the value generated by economic activities such as industry, commerce, and services, which typically have higher added value compared to agricultural production. A higher non-farm economic output per unit of land indicates a more economically efficient use of land. By analyzing non-agricultural economic output per unit of land, this approach allows for the identification of regions with high ULUE and areas where there is potential for improvement. Such analysis facilitates the implementation of targeted measures to optimize land-use structures and enhance the overall efficiency of land use.

3.3.2. Explanatory Variables: Economic Agglomeration

The explanatory variable in this paper is the level of EA, which is measured using the number of urban employment per unit of land [59]. The number of employed individuals serves as a direct indicator of a region’s economic activity and labor market dynamics. Urban employment levels are closely linked to the degree of industrial development and the region’s economic attractiveness, thereby reflecting the concentration of economic activities. EA typically involves the formation of industrial clusters, which have the capacity to attract additional employment, leading to a further concentration of population and resources. Consequently, a higher number of urban jobs per unit of land indicates a more pronounced EA effect within the region.

3.3.3. Moderating Variable: Digital Finance

DF comprehensively reflects the breadth of coverage, depth of use, and degree of digitization of financial services within a region. This paper utilizes 1% of Peking University’s Digital Financial Inclusion Index as a proxy variable for DF [60,61]. The Digital Financial Inclusion Index offers a comprehensive portrayal of the prevalence and utilization of digital financial services, highlighting regional disparities in access to and use of these services.

3.3.4. Mechanistic Variables

The first mechanistic variable is RIS. The index of RIS is a measure of the degree of coupling between the structure of factor inputs and the industrial structure, reflecting the results of comprehensive coordination among industries and the effect of resource utilization. At present, academics generally use the degree of structural deviation to measure the RIS, and its formula is

$$E={\sum }_{i=1}^n\left|{\displaystyle \frac{Y_i/L_i}{Y/L}}-1\right|={\sum }_{i=1}^n\left|{\displaystyle \frac{Y_i/Y}{L_i/L}}-1\right|$$

(4)

$E$ represents the degree of structural deviation, $Y$ denotes output, $L$ denotes employment, $i$ denotes industry, and $n$ denotes the number of industry sectors. According to classical economic theory [62], if the economy is in equilibrium, productivity levels are uniform across all industrial sectors. Here, $Y/L$ denotes productivity. When the economy reaches an equilibrium state, $Y_i/L_{i }$= $Y/L$, resulting in $E$= 0. In this context, $Y_i/Y$ denotes the output structure, while $L_i/L$ reflects the employment structure. The variable $E$ captures the interplay between the output structure and the employment structure. A more balanced economic situation leads to a more rational industrial structure, which in turn results in a lower value of $E$. In a perfectly balanced economy, $E$ can be reduced to zero. However, the structural deviation index fails to account for the varying importance of each industry within the economy. Additionally, the use of absolute values in its calculation introduces challenges in data processing. To address these issues, this paper proposes a new Tyrell Index to measure the RIS.

The Theil Index is a widely used indicator of income inequality. In line with the mainstream literature [63], this paper constructs the Theil Index as follows:

$$RIS=TL={\sum }_{i=1}^n\left({\displaystyle \frac{Y_i}{Y}}\right)\mathrm{l}\mathrm{n}\left({\displaystyle \frac{Y_i}{L_i}}/{\displaystyle \frac{Y}{L}}\right)$$

(5)

The new Theil Index is developed to overcome the limitations associated with absolute value calculations while considering the proportional significance of various industries. It also retains the economic implications of structural deviation. In the analysis of industrial structure, the Tyrell Index can quantify disparities in the weights of different industries. A lower Tyrell Index indicates a more balanced distribution of weights among industries.

The second mechanistic variable is IA. First, this paper measures IA from an innovative quantitative perspective. In this paper, the number of inventions produced, including utility model patents and design patents by a region within a given year, is used as a proxy variable for IA. This is done by first adding one to the number of inventions and then taking the logarithm [64]. The number of regional inventions directly indicates the quantity of new technologies, products, or processes developed in a region over a specific period. This serves as a direct measure of the outcomes of innovation activities. A higher number of inventions within a region suggests a greater intensity of regional IA. Second, this paper evaluates IA through the lens of innovation value. It employs the China Urban and Industrial Innovation Index, jointly developed by the Research Institute of Innovation and Digital Economy (RIDE) and the Fudan University Industry Development Research Center (FIND), as a proxy variable for measuring innovation capability. Specifically, the legal status update information of micro-invention patents authorized by the State Intellectual Property Office, along with the annual fee structure for invention patents at different stages, is used to estimate the average value of invention patents across various ages. This estimation follows the Pakes and Schankerman [65] patent renewal model to derive value weighting coefficients (V1, V2, … V20). Then, using the end of each year as the observation point, valid invention patents (those that are authorized and remain in effect) are selected for analysis. The total annual patent value for each dimension—categorized by city, industry, or “city-industry”—is then calculated by applying the relevant weighting factors to these patents. This method ensures that the patent values are appropriately weighted for each respective category, providing a comprehensive measure of innovation capability across different geographic and industrial sectors.

3.3.5. Control Variables (Controls)

Based on relevant mainstream literature [66,67], this paper selects several control variables, including the level of economic development (PGDP), the degree of openness (OL), population size (PDEN), the extent of government intervention (GOV), and the level of human capital (HCL). The variables utilized in this study are summarized in the

Table 1. Main variable definition.

Variable Type	Variable Name	Variable Symbol	Variable Construction
Explained variable	Urban land use efficiency	ULUE	Non-farm economic output per unit of urban construction land.
Explanatory variable	Economic agglomeration	EA	Urban employment per unit of urban construction land.
Mechanism variable	Digital finance	DF	Digital inclusive financial index/100.
	The interaction term between DF and EA	DFEA	The interaction term between DF and EA.
	Rationalization of industrial structure	RIS	Tyrell’s Index. The smaller the Tyrell’s Index, the more even the industry is.
	Innovation ability (number of patent perspectives)	IA1	The logarithm of the number of inventions awarded in the year.
	Innovation ability (innovative value perspective)	IA2	The number of patents is weighted using a value weighting factor.
Control variables	Level of economic development	PGDP	Logarithm of GDP per capital.
	Level of opening up	OL	Foreign capital in use/gross regional product.
	Size of population	PDEN	The logarithm of total population at the end of the year.
	Government expenditure	GOV	General government expenditure/gross regional product.
	Level of human capital	HCL	Number of students enrolled in general undergraduate programs/total population at the end of the year.
	Infrastructure level	INFR	Road area per capita.

.

4. Empirical Results

4.1. Descriptive Statistics

According to the descriptive statistical results in

Table 2. Descriptive statistics.

Variable	N	SD	Mean	Min	p25	Median	p75	Max	Skewness	Kurtosis
ULUE	2178	0.9139	1.7954	0.0906	1.1375	1.6503	2.2361	5.0075	1.0716	4.2968
EA	2178	2.8282	1.7488	0.0704	0.4267	0.8830	1.8239	21.9332	4.5272	28.5562
DF	2178	0.4965	2.4426	0.9554	2.2457	2.5995	2.8096	3.2921	−0.7729	2.416372
IA1	2178	1.6934	5.1075	0.6931	3.9318	4.8675	6.2086	9.3071	0.3585	2.7248
IA2	2178	38.6505	17.3533	0.0276	1.2906	3.3820	12.0642	227.1568	3.7676	18.1516
RIS	2178	0.1977	0.2848	0.0032	0.1230	0.2449	0.4131	0.9084	0.7426	2.9376
PGDP	2178	0.5277	10.7876	9.2193	10.3985	10.7567	11.1512	11.9703	0.1932	2.5650
OL	2178	0.0167	0.0160	0.0000	0.0032	0.0110	0.0230	0.1038	1.7180	6.7691
PDEN	2178	0.6642	5.8822	3.8286	5.4943	5.9417	6.3750	7.1682	−0.5932	3.2652
GOV	2178	0.0909	0.2037	0.0564	0.1379	0.1799	0.2438	0.5815	1.3313	4.8856
HCL	2178	0.0253	0.0198	0.0003	0.0057	0.0104	0.0212	0.1272	2.4688	8.8552
INFR	2178	7.0623	18.3009	6.2300	12.9489	16.8930	22.9600	40.2102	0.7394	3.1727

, the mean value of the explained variable, ULUE, is 1.7954, while the median is 1.6503, indicating a right-skewed distribution of ULUE. The calculated skewness of 1.0716 confirms this right-skewness. This suggests that, under the influence of the rapidly growing digital economy, non-agricultural economic output across various cities is relatively high. A spatial distribution map of ULUE in Figure 4 further reveals that land use efficiency is generally higher in cities along China’s eastern coast, with improvements observed over time. For the explanatory variable, EA, the mean value is 1.7488, with a minimum of 0.0704, a standard deviation of 2.8282, and a kurtosis of 28.5562, indicating the level of EA in the city is concentrated around the mean value, with relatively small differences. Spatial analysis in Figure 5 shows that EA is more pronounced in the central and eastern regions, particularly in the Yangtze River Delta and Pearl River Delta areas. These coastal regions benefit from well-developed waterway transport and proactive external policies, fostering a favorable environment for economic growth and attracting significant industries and talent. Regarding the development of DF, the standard deviation is 0.4965, with a median of 2.4426 and a mean of 2.5995, reflecting minimal variation between the two values. The kurtosis of 2.4163 suggests limited regional differences in DF development overall. However, spatial distribution data in Figure 6 indicate that, while regional disparities in DF development were minimal in 2013, by 2020, a noticeable gap had emerged between eastern and western regions. This divergence is likely due to the eastern coastal cities’ stronger economic foundations and greater capacity for innovation, leading to faster development of digital technologies compared to central and western regions. The descriptive statistics of the control variables are consistent with the results in the mainstream literature, which indicates that the data we used are reliable [68,69].

4.2. Correlation Analysis

Figure 7 presents the results of the correlation analysis between all variables, revealing several key findings. Consistent with the mainstream literature [70], the correlation coefficients of all variables are below 0.7, indicating the absence of multicollinearity. Both DF and EA exhibit a positive correlation with ULUE. Additionally, the cross-multiplication term of DF and EA is also positively correlated with ULUE. Furthermore, both the explanatory and explained variables, as well as the mechanism variables and explained variables, are significant at the 1% level, thereby passing the correlation analysis test.

4.3. Benchmark Regression and Moderating Effects Tests for DF

To examine the relationship between EA and ULUE, this paper performs a regression analysis on the full sample following Equations (1) and (2). The results are presented in columns (1) to (3) of

Table 3. Benchmark regression and moderating effects results for DF.

	(1)	(2)	(3)
Variable	ULUE	ULUE	ULUE
DFEA			0.020 ***
			(4.505)
DF			0.029
			(0.778)
EA	0.035 ***	0.039 ***	−0.015
	(2.801)	(3.554)	(−0.925)
PGDP		0.870 ***	0.889 ***
		(15.418)	(15.701)
OL		0.097	0.342
		(0.124)	(0.440)
PDEN		0.350 **	0.047
		(2.239)	(0.279)
GOV		−1.444 ***	−1.294 ***
		(−5.710)	(−5.113)
HCL		−1.268	−1.415
		(−0.746)	(−0.838)
INFR		−0.005 **	−0.004 *
		(−2.247)	(−1.904)
Year FE	YES	YES	YES
City FE	YES	YES	YES
Constant	1.735 ***	−9.300 ***	−7.835 ***
	(76.296)	(−7.707)	(−6.337)
Observations	2178	2178	2178
R2	0.890	0.913	0.915

Note: t-values in parentheses; ***, **, and * represent 1%, 5%, and 10% significance levels, respectively.

. The regression coefficient for EA is 0.038, which is positive and significant at the 1% level, even after controlling for the fixed effects of year and region. Economically, this suggests that for every unit increase in urban EA, the local ULUE increases by 0.038. So, hypothesis H1a is not rejected. Column (3) of Table 3 incorporates the moderating effect of DF, revealing that the coefficient of the DFEA is 0.020, which is also positive and significant at the 1% level. This indicates that for every unit increase in EA driven by DF, the local ULUE rises by 0.020. Overall, the results in columns (1) and (2) indicate that a higher degree of EA is associated with an increase in ULUE, thereby supporting hypothesis H1a of this study. Additionally, when considering the results in columns (2) and (3), it is evident that DF significantly amplifies the positive impact of EA on ULUE, thus not rejecting hypothesis H1b. This conclusion partially agrees with the existing literature. The similarity lies in the fact that previous studies have also highlighted that DF can indeed enhance the ULUE [71]. However, the key distinction is that the existing literature has not explored the moderating effect of DF, a dimension that this paper specifically addresses.

4.4. Robustness Testing

4.4.1. Lagged Period Test

Due to the potential lag in the effects of EA and the development of DF on ULUE, the impact of these factors may not be immediately observable. This delayed response necessitates a more long-term perspective when assessing their influence, as the benefits of EA and DF may take time to manifest in measurable improvements in ULUE. To address this issue, this paper follows the approach used in mainstream literature [72] by lagging the explanatory variables by one period for re-empirical analysis (L.DEFA, L.EA, L.DF). The regression results, displayed in column (1) of

Table 4. Robustness testing.

	(1)	(2)	(3)	(4)
Variable	ULUE	ULUE (SBM)	ULUE	ULUE
L.DFEA	0.011 **
	(2.267)
L.DF	0.102 ***
	(2.796)
L.EA	−0.019
	(−1.138)
DFEA		0.006 ***	0.013 ***	0.017 ***
		(8.323)	(2.746)	(3.700)
DF	0.081 **	−0.027 ***	0.060	0.048
	(2.137)	(−4.452)	(1.514)	(1.153)
EA	0.014	−0.023 ***	−0.022	−0.005
	(0.770)	(−8.834)	(−0.954)	(−0.334)
Controls	YES	YES	YES	YES
Year FE	YES	YES	YES	YES
City FE	YES	YES	YES	YES
Observations	1902	1872	1907	1954
R2	0.927	0.872	0.927	0.917

Note: t-values in parentheses; *** and ** represent 1% and 5% significance levels, respectively.

, show that DFEA remains positive and significant at the 1% level, consistent with the findings of the benchmark regression.

4.4.2. Replacement of Explanatory Variables

To account for potential biases in land use efficiencies under different construction models, this paper employs the SBM (Slack-Based Measure) model to re-evaluate land use efficiencies. The regression results, presented in column (2) of Table 4, indicate that DFEA remains positive and significant at the 1% level. These findings are consistent with the results of the benchmark regression, further supporting the robustness of the original analysis.

4.4.3. Reducing the Sample Period

The year 2013 is widely recognized as the beginning of China’s Internet finance (important types of DF) era [68], marked by the rapid emergence of Internet finance models such as P2P lending and third-party payment systems. During this period, the scale of financial technology applications also expanded significantly. However, despite the swift growth of Internet finance in 2013, the accompanying regulatory framework lagged behind, and the market remained immature, leading to heightened volatility in the interactions between this emerging sector and the traditional financial market. The regulatory delay may have allowed the rapid expansion of financial technology to excessively stimulate economic agglomeration in certain regions in the short term, which was neither effectively managed by market mechanisms nor adequately guided by policy interventions. As a result, this short-term effect tended to normalize over time. The combination of market instability and institutional deficiencies may introduce uncertainty and disruption to the role of financial technology in moderating the relationship between economic agglomeration and land use efficiency. Therefore, this article refines the sample by narrowing the time interval and excluding data from the year 2013. The regression results, shown in column (3) of Table 4, indicate that DFEA is positive and significant at the 1% level, consistent with the benchmark regression results.

4.4.4. Excluding the Sample of Xinjiang

The positioning and regional planning of cities in autonomous minority regions such as Xinjiang and Inner Mongolia are unique in the Chinese landscape and may, therefore, bias regression results. To address this, this paper excludes the city sample of the ethnic minorities and conducts the regression analysis again. The results, as shown in column (4) of Table 4, indicate that DFEA remains positive and significant at the 1% level, aligning with the baseline regression results and confirming the consistency of the findings.

5. Further Analysis

5.1. Mechanism Analysis

5.1.1. Improve IA

The preceding theoretical analysis suggests that digital finance can expedite capital flows and ease the financing constraints faced by SMEs. This, in turn, mitigates the regional Matthew effect within the same industry and amplifies the spillover effects of knowledge and technology associated with economic agglomeration. This improves the level of regional innovation output, thereby promoting the role of EA in enhancing ULUE. The regression results of the mechanism test are presented in

Table 5. Mechanism testing.

	(1)	(2)	(2)
Variable	IA1	IA2	RIS
DFEA	0.013 **	3.982 ***	−0.004 ***
	(2.488)	(21.193)	(−3.358)
DF	0.073 *	7.346 ***	−0.021 **
	(1.714)	(4.710)	(−1.978)
EA	−0.011	−7.359 ***	0.006
	(−0.588)	(−10.929)	(1.281)
Controls	YES	YES	YES
Year FE	YES	YES	YES
City FE	YES	YES	YES
Observations	2178	2178	2178
R2	0.967	0.917	0.857

Note: t-values in parentheses; ***, **, and * represent 1%, 5%, and 10% significance levels, respectively.

. The results in column (1) indicate that DFEA has a positive effect on innovation output, and column (2) shows that DFEA has a positive effect on innovation capability from the perspective of innovation value, confirming a positive moderating effect of DF on the innovation spillover effect of EA. Thus, hypothesis H2 of this paper is verified.

5.1.2. Promote the RIS

From the previous theoretical analysis, it can be seen that digital finance can promote the flow of resources in the region to high-value-added industries and promote the RIS. The RIS can foster the development of high-efficiency and high-value-added industries within a region, thereby enhancing regional ULUE. DF, as an innovative financial resource in the digital era, plays a significant role in promoting the RIS. The regression results of the mechanism test are presented in the following table. Column (2) shows that the regression coefficient of the interaction term between DF and EA on the index of RIS is negative. This finding suggests that DF can further enhance EA’s capacity to reduce the disparity of industrial structure, thereby strengthening the impact of EA on ULUE. Thus, hypothesis H2 of this paper is verified.

5.2. Heterogeneity Analysis

5.2.1. Heterogeneity in Geographic Locations

The positive effect of EA on ULUE, driven by DF, varies across different geographic locations. In the eastern and central regions, higher levels of economic development and a more diversified industrial base provide DF, as a critical tool for resource allocation, with greater flexibility in selecting industries for investment. Entities in these regions can allocate resources to industries with higher productivity, and the more developed market mechanisms facilitate smoother resource allocation through DF. This environment enables DF to effectively promote the flow of factors, leading to a more pronounced improvement in ULUE through EA. In contrast, the western and northeastern regions face challenges such as lagging economic development, a less diverse industrial structure, and an underdeveloped market mechanism. These factors limit the positive impact of DF-driven EA on ULUE. This paper conducts regressions based on the geographical regions of the provinces [27] in the sample. Columns (1) to (4) of

Table 6. Heterogeneity analysis 1.

	(1) Eastern	(2) Central	(3) Western	(4) Northeastern
Variable	ULUE	ULUE	ULUE	ULUE
DFEA	0.027 ***	0.037 **	0.009	−0.018
	(2.785)	(2.448)	(0.596)	(−0.490)
DF	0.138	−0.021	−0.061	0.047
	(1.049)	(−0.260)	(−1.194)	(0.752)
EA	−0.023	−0.068	−0.023	0.167
	(−0.839)	(−1.366)	(−0.397)	(1.301)
Controls	YES	YES	YES	YES
Year FE	YES	YES	YES	YES
City FE	YES	YES	YES	YES
Observations	649	632	622	267
R2	0.885	0.908	0.920	0.932

Note: t-values in parentheses; *** and ** represent 1% and 5% significance levels, respectively.

respectively present the regression results for the eastern, central, western, and northeastern regions. The regression coefficients for all groups are positive. However, the coefficients for DFEA in columns (1) and (2) are significant at the 1% level, while those in columns (3) and (4) are not significant. This indicates that the effect of EA on ULUE, driven by DF, is significantly more pronounced in the eastern and central regions than in the western and northeastern regions.

5.2.2. Heterogeneity in Internet Penetration

The positive impact of DF-driven EA on ULUE varies across regions with different levels of internet penetration. In regions with higher internet penetration, the environment is more conducive to the development of DF, thereby enhancing its promotion effect. Additionally, in these regions, smoother information flows, reduced innovation costs, and more significant collaboration and technology diffusion further amplify the impact of DF. Conversely, in regions with lower internet penetration, the accessibility and efficiency of financial services are limited, and the innovation environment is less developed, leading to a less pronounced positive effect of DF on ULUE. This study incorporates these considerations by constructing a heterogeneity variable for internet penetration. Columns (1) and (2) in

Table 7. Heterogeneity analysis 2.

	(1)	(2)	(3)	(4)
	Regions with High Internet Penetration	Regions with Low Internet Penetration	Regions with High S&T Expenditures	Regions with Low S&T Expenditures
Variable	ULUE	ULUE	ULUE	ULUE
DFEA	0.013 **	0.016	0.010 *	0.016
	(2.141)	(1.156)	(1.834)	(0.720)
DF	0.102	0.022	0.091	−0.057
	(1.451)	(0.506)	(1.274)	(−1.314)
EA	−0.012	−0.008	0.002	−0.015
	(−0.627)	(−0.207)	(0.101)	(−0.216)
Controls	YES	YES	YES	YES
Year FE	YES	YES	YES	YES
City FE	YES	YES	YES	YES
Observations	1099	1040	1066	1072
R2	0.936	0.934	0.901	0.945

Note: t-values in parentheses; ** and * represent 5% and 10% significance levels, respectively.

present the regression results for regions with higher and lower internet penetration rates, respectively. The regression coefficients for both groups are positive. However, the coefficient for the DFEA in column (1) is significant at the 1% level, while in column (2), it is not significant. This indicates that the positive impact of DF-driven EA on ULUE is considerably more significant in regions where internet penetration is higher.

5.2.3. Heterogeneity in Innovation Emphasis

The moderating effect of DF is heterogeneous under regions with different innovation emphases. From the previous analysis, it is clear that IA and RIS are important mediators of the regulatory effect of DF. From the perspective of regional IA, as the government focuses more on innovation, which means that the government invests more in creating an open innovation environment, which brings a better innovation environment and more favorable innovation policies, etc., the positive moderating effect of DF will be more significant in this type of region. From the perspective of RIS, the RIS makes resources flow to high-productivity industries, and innovation is the key to improving the productivity of each industry. The more the government pays attention to innovation, the better the innovation environment of each industry, and the more it can promote vibrant competition among industries, so that the industrial structure is less likely to be concentrated, and DF can better exert its deployment effect. The percentage of local government S&T expenditure can be a good measure of the importance of local government to innovation, and a high percentage of regional S&T expenditure indicates that the government pays more attention to innovation. In this paper, the sample is divided into regions with high innovation importance and regions with low innovation importance according to the median government science and technology expenditure ratio for regression. As shown in Table 7, where the regression coefficients are positive for all groups, the regression coefficient of the DFEA in column (3) remains significant at the 10% level, while the regression coefficient of the DFEA in column (4) is not significant, suggesting that the moderating effect of DF is more pronounced in regions where the government places more emphasis on innovation.

6. Conclusions and Policy Implications

6.1. Conclusions

Based on the premise that DF-driven EA contributes to an increase in ULUE, this paper delves into the potential role and pathways through which urban IA and RIS can enhance this positive effect. The study is structured around two primary perspectives: first, it examines the facilitative impact of EA on ULUE and investigates whether DF can serve as a moderating factor. The findings indicate that the facilitative role of EA should be sustained, and digital financial tools should be consistently utilized to accumulate greater financial resources and provide more advanced financial services, thereby supporting the efficient use and development of land. Second, the study explores the mechanisms by which DF coordinates EA to enhance ULUE. The study found that DF can optimize EA, promote RIS and regional IA by facilitating the flow of production factors, improve production efficiency and product value-added, and then enhance the ULUE. In the long term, the inclusiveness of digital finance may also enable financial services to be more widely available, helping to promote social equity. The development of digital finance may also drive innovation in enterprises, thereby reducing their energy consumption and promoting sustainable social development. The limitations of this paper are the limited timeliness and coverage of the data, the possible omission of other external factors affecting ULUE, and the fact that the interaction effects model may not fully capture the complex multidimensional effects.

Several areas of research in this paper warrant further exploration. First, comparative studies across different countries or regions could be undertaken to assess the external validity of DF’s impact on various cities and land use patterns, thereby identifying its global applicability and potential regional differences. Second, the long-term effects of DF on urban land use may extend beyond efficiency improvements, potentially influencing the ecological environment and social equity. These aspects offer valuable opportunities for future research.

6.2. Policy Suggestions

Promote regional integration. First, the government must conduct a comprehensive analysis of each region’s resource endowments, industrial base, and development potential. This should involve formulating a scientifically grounded regional development strategy and establishing a unified regional development plan that clearly defines the functional roles and development trajectories of each region. Such an approach can prevent redundant construction and resource wastage. Through strategic planning, the government can direct resources toward regions with comparative advantages, thereby facilitating EA. To this end, regional coordination offices could be established to specialize in resource rationalization and policy implementation. Second, the government should enhance information and communication networks to ensure the efficient flow of people, goods, and information. Strengthening infrastructure, including transportation, energy, and information networks, is crucial for interconnectivity within the region. This interconnection can reduce logistics costs, improve intra-regional economic linkages, and enhance openness to external markets, creating favorable conditions for EA. The government can utilize the big data platform for the effective allocation of resources and talents to ensure that each region receives development support based on its strengths. During policy implementation, the uneven levels of economic development across regions may pose challenges to effective execution. Moreover, underdeveloped regions with deficiencies in infrastructure, technology, and human resources may experience limited policy effectiveness. Therefore, the government must increase financial subsidies and prioritize policies for regions with weaker infrastructure. By utilizing financial transfers and targeted support, less developed regions can accelerate their development and reduce the regional development gap. This approach would provide a broader market for resource allocation, fostering economic agglomeration and promoting internationalization and agglomeration effects within the regional economy.

Support digital financial innovation. DF, representing an advanced stage in the integration of finance and technology, is pivotal in fostering economic growth, enhancing the efficiency of financial services, promoting inclusivity, and strengthening risk management capabilities. On one front, the government should incentivize financial institutions and technology enterprises to increase their investment in DF research and development through tax incentives, financial support, and policy formulation. This approach would encourage the adoption of cutting-edge technologies, such as big data, artificial intelligence, and blockchain, in the provision of financial services. Simultaneously, the government should focus on improving the payment system, establishing a credit information-sharing platform, and bolstering cybersecurity measures to ensure that digital financial services are both efficient and secure. Furthermore, the promotion of cross-sector and cross-industry collaboration, alongside strengthened international cooperation, is essential. By dismantling information silos and encouraging open data sharing, the government can provide a wealth of data resources to fuel digital financial innovation. This strategy can alleviate information asymmetry and reduce transaction costs, thereby promoting economic growth and enhancing China’s competitiveness in the world market. The advancement of DF presents several challenges. Significant regional disparities exist in China’s digital infrastructure, with the digital divide impeding the collection of information in remote areas, ultimately restricting access to convenient digital financial services. Moreover, cybersecurity risks have intensified alongside technological progress, potentially undermining the interests of DF users and threatening financial stability. Therefore, it is imperative for the government to enhance the development of digital infrastructure in remote regions to bridge the digital divide. Simultaneously, increased investment in cybersecurity is essential, along with the establishment of a specialized regulatory body to prevent information breaches and fraud in digital financial services and promote sound international financial cooperation.

In coordinating industrial relationships and promoting the RIS, first, the government should focus on advancing the sophistication and modernization of various industries. By continuously enhancing the core competitiveness of each sector, the government can drive quality improvement, foster brand development, and elevate industries to the middle and high-end segments of the market. Second, intelligent manufacturing should be at the forefront of industrial-technological transformation and optimization. Accelerating the adoption and application of new intelligent manufacturing technologies is essential for transforming manufacturing models, supporting technological innovation, and facilitating industrial upgrading. This strategy will guide the reallocation of resources toward high-value-added and high-productivity sectors, thereby promoting overall economic growth. Finally, the government needs to solidify the leading position of competitive industries, expedite the transformation and modernization of traditional sectors, and increase the proportion of advanced production capacity. This includes expanding the supply of high-quality products and addressing gaps in areas vital to secure development. Simultaneously, the government should progressively phase out excess capacity and eliminate outdated production methods. In doing so, it can foster and expand strategic emerging industries, creating a new engine for industrial development. By guiding resource allocation toward high-value-added and high-productivity industries, the government can ensure sustainable and balanced industrial growth. Simultaneously, the government should intensify the protection of intellectual property rights by refining legal frameworks, improving the efficiency of law enforcement and judicial processes, utilizing advanced technologies to monitor violations, and increasing efforts in public awareness and education. Through fostering international cooperation and providing support for SMEs, it can more effectively address cross-border infringements and foster innovative growth.

The government should actively develop regional innovation channels and comprehensively enhance regional IA. On one hand, increased investment in public education serves as the foundation for enhancing regional IA. The government can lower the barriers to family investment in education by increasing public expenditure, particularly by subsidizing teachers’ salaries and reducing or eliminating tuition and miscellaneous fees. This approach enables more low-income families to pursue higher education, thereby enhancing human capital accumulation and increasing the workforce in the education sector. Additionally, the government can promote the transformation of scientific and technological achievements by supporting the establishment of joint research and development institutions between enterprises, universities, and research institutes. This collaboration boosts enterprises’ IA and operational efficiency, reduces reliance on high-cost labor, lowers innovation hiring costs for SMEs, and ultimately enhances regional IA. China is currently facing a significant imbalance in educational resources, which poses a major obstacle to enhancing regional innovation capacity. This disparity can result in the migration of talent, further widening the innovation gap between regions. To address this issue, efforts to support regions with limited educational resources must be intensified, and strategies to attract and retain talent should be implemented to bridge the educational resource divide. On the other hand, scientific and technological cooperation and regional collaborative innovation are crucial for advancing regional IA. Horizontally, the government can facilitate the balanced distribution and optimal allocation of innovation resources by strengthening scientific and technological cooperation between the eastern and western regions. Vertically, the government can address the diverse needs of enterprises at various developmental stages through targeted policy guidance and support. This includes steering SMEs toward a path of “specialization, refinement, and innovation”, establishing a complete cultivation system that supports enterprises throughout their entire life cycle, from incubation to growth [62]. Encouraging large enterprises to share resources and establish a benefit- and risk-sharing mechanism with SMEs further enhances the regional innovation ecosystem.