The Transmission Mechanism and Spatial Spillover Effect of Agricultural New Quality Productive Forces on Urban–Rural Integration: Evidence from China

Abstract

Urban–rural integration (URI) plays a crucial role in advancing rural revitalization and the modernization of agriculture. Nevertheless, numerous nations encounter persistent obstacles, including inefficient resource mobility across urban–rural divides and uneven industrial distribution, while striving to foster such integration. Agricultural new quality productive forces (ANPFs) offer an innovation-led production framework fueled by advances in agricultural technology, allowing urban–rural integration (URI) through improved resource mobility between cities and rural regions. Utilizing panel data from 30 Chinese provinces (2013–2022), this study employs a two-way fixed effects model, mediation analysis model, threshold regression model, and the spatial Durbin model to investigate the transmission mechanism and spatial spillover effect of agricultural new quality productive forces (ANPFs) on urban–rural integration (URI). The findings show the following: (1) Agricultural new quality productive forces (ANPFs) significantly influence urban–rural integration (URI). (2) The influence is significantly stronger in western China than in the eastern and central regions. (3) Industrial restructuring and upgrading (IND) function as a mediating influence in this connection. (4) The role of informatization (INF) has a dual-threshold effect. (5) Geographically, while these forces promote local integration, they may impede progress in nearby regions. This study provides new empirical insights into the factors that influence urban–rural integration (URI) and proposes policy solutions to promote sustainable regional development.

1. Introduction

Urban–rural integration (URI) is now known as a key part of the worldwide sustainable growth strategy. It is no longer limited to traditional methods, but instead aims to close the long-standing gap between both rural and urban areas [1]. These gaps are mostly expressed in employment opportunities, access to services, and environmental quality. These issues provide challenges for countries at different levels of development. The goal of URI is to close these gaps, create a more balanced and integrated growth path, and encourage equal growth of urban and rural populations. In recent years, the worldwide discussion about sustainable development highlights the significance of closing the gap between urban and rural areas.

Countries around the world have explored a variety of techniques to connect both rural and urban areas. In developed nations such as the United States and the European Union, governments invest in rural roads, internet connections, and small businesses to boost competitiveness. Recent studies indicate that from 2014 to 2022, the Common Agricultural Policy (CAP) has dedicated €8.6 billion to foster non-farming initiatives in rural regions. This highlights CAP’s dual function in not only assisting agricultural producers but also enhancing diverse enterprises and rural communities [2].

In developing nations, efforts focus on balanced land distribution, modern farming, and basic services such as schools and hospitals. Brazil’s Zero Hunger’ program, for example, supports farmers while also giving food aid to combat poverty [3]. In Africa, organizations such as Comprehensive Africa Agriculture Development Programme (CAADP) aim to increase crop yields and rural income through better roads, tools, and training. As part of this framework, participating nations across Africa have committed to directing a minimum of 10% of their annual governmental expenditures toward enhancing agricultural systems and rural infrastructure, with an accompanying objective of attaining yearly sectoral growth exceeding 6% [4]. Sudan provides a notable illustration of progress toward CAADP’s fiscal objectives, dedicating more than 9% of its government spending to agricultural advancement—approaching the program’s recommended 10% benchmark for national budget allocations [5].

Although efforts by countries around the world to promote URI, some challenges remain. Cities continue to attract major resources and investment, contributing to growing urbanization and possible rural decline. Rural areas usually face challenges with market access, lack of resources, and the lack of skilled labor. For instance, United Nations Development Programme (UNDP) (2023) data indicate that global resource allocation for education and healthcare in rural areas amounted to merely 50% of urban investment levels during the reporting year [6]. In addition, these challenges are getting worse because of global trends such as climate change, which particularly impacts rural populations, and the digital gap, which worsens the existing gaps. URI has accelerated urban growth, resulting in extensive rural land acquisition and intensifying land scarcity. As this integration progresses, urban cultural influences may erode traditional rural practices, potentially diminishing or even erasing them. Furthermore, certain peri-urban rural regions experience compounded environmental pressures, including industrial pollution from relocated urban enterprises and agricultural runoff, degrading their ecological quality.

In this situation, agricultural new quality productive forces (ANPFs) bring hope for resolving these challenges. ANPF can be improved through technological innovation and analysis of data mechanisms [7], encouraging environmentally friendly agriculture and rural economic growth, and contributing to integrated urban and rural development [8]. This strategy makes use of the development of digital technology, ecological agriculture, and automated farming to strengthen the competitiveness of the rural economy. This study uses panel data from 30 main provincial administrative regions in China from 2013 to 2022 to assess the significance of ANPF in empowering URI. The study’s findings provide policy recommendations for encouraging integrated and sustainable development across urban and rural regions.

2. Literature Review

In recent years, scholars have focused more on ANPF, studying their means, evaluation, and growth paths from many different views. Research on URI primarily focuses on three aspects: its connotation, measurement, and paths.

2.1. Relevant Studies on ANPF

Regarding the definition, Jiang [9] defined ANPF as a higher level of productivity that combines digital and intelligent characteristics with traditional agricultural production variables. Gao and Ma [10] argued that ANPF needs to reorganize agricultural production processes with modern technology. This method also included forming new rural relationships. Together, these measures increased value in the agriculture sector and rural areas. Kong and Xie [11] emphasized that ANPF is a modern, flexible, and forward-looking. These forces relied on high-tech tools and materials, resulting in significant changes in farming tools, methods, and objects through technological advancement. Luo and Geng [12] defined ANPF as a new type of productivity that is driven by technical innovation, focuses on high-quality development, and aims to construct a strong agricultural sector. This is accomplished through innovative factor allocation and industrial transformation, which propels agriculture toward digitalization and intelligence while also increasing total factor productivity and ensuring sustainable development.

In terms of measurement, some scholars, drawing on Marxist theory, developed a thorough assessment index system with three dimensions: agricultural laborers, agricultural labor objects, and agricultural labor equipment [13]. Others focused on different definitions of “new quality” in agriculture, using a complex evaluation system that encompasses technological innovation, environmental friendliness, and digitalization levels [14]. Due to the lack of agreement on its meaning, the assessment of ANPF remains exploratory in academia.

With regard to development pathways, Huang et al. [15] emphasized that the creation of ANPF necessitates coordinated efforts in policy and institutional support, technical innovation and application, climate-smart agricultural implementation, and talent development. According to Chen and Rao [16], ecological sustainability must serve as the fundamental position, technological innovation as the primary motivator, and labor quality development as the main guarantee for enhancing ANPF. Tang et al. [17] contend that expanding the scale of local scientific and technological innovation facilitates the diffusion of agricultural technological resources to neighboring regions, thereby promoting balanced regional development of ANPF. Wang and Luo [18] proposed that growing the labor force, optimizing labor tool allocation, and upgrading production processes are critical for significantly increasing the development level of ANPF.

2.2. Related Research on URI

In terms of connotation, researchers primarily agree that URI requires the coordination of many relationships [19] and is typically defined in terms of supporting balanced urban–rural development [20]. For instance, Shi et al. [21] posit that the urban–rural system constitutes an interconnected spatial entity encompassing economic, social, and environmental dimensions, requiring integrated industrial, spatial, and demographic approaches. They demonstrate that elucidating dynamic land-use changes within this system and their ecological and social consequences enables sustainable and coordinated urban–rural development. Xue et al. [22] argue that conceptualizing urban development as an integrated urban–rural composite contributes significantly to sustainable urban planning practices.

Regarding measurement, existing studies evaluate URI both at the national level [23] and in smaller regions [24]. For instance, Sun [25] built a system to measure URI in China. This system uses four key areas: population, economy, society, and environment. Liang et al. [26] took Henan Province as a case study, combining theoretical analysis and empirical research to systematically explore the current characteristics, typologies, driving mechanisms, and obstacles of URI at the county level.

Regarding implementation paths, scholars highlight several key steps. These include coordinating urban and rural development plans, increasing resource sharing between cities and rural areas, developing fresh approaches to enhance integration [27], and focusing on county-level integration [28]. For instance, Tang et al. [29] propose that cultural-tourism integration offers valuable theoretical insights and practical guidance for promoting urban–rural integration and rural revitalization in traditional villages. Wei et al. [30] stated that constructing long-term mechanisms for modern URI necessitates adopting evaluation systems, developing institutional frameworks at the county level, and encouraging innovation through pilot initiatives.

2.3. URI and Sustainable Development

In recent years, URI has emerged as a critical strategy for fostering balanced regional growth and sustainable practices. Empirical evidence demonstrates that URI enhances interregional coordination through more efficient resource distribution, reduces disparities between urban and rural regions, and provides greater equity in access to public services [31]. Concurrently, URI contributes to environmental preservation while boosting economic productivity, thereby reinforcing sustainable development objectives [32]. Notably, the adoption of digital solutions has simultaneously increased agricultural output and accelerated rural digital transformation, enabling improved information exchange and resource mobility across urban and rural sectors [33]. Subsequent studies should prioritize investigating innovative policy frameworks, technology implementations, and participatory social mechanisms to advance shared prosperity and ecological harmony between urban and rural communities [34].

In summary, while the present literature provides a certain theoretical foundation and research tools for studying ANPF and URI, several research gaps exist. Given this, this paper combines ANPF with URI to form a comprehensive analysis framework for furthering scholarly understanding in both research domains. This study makes distinct theoretical contributions beyond prior research by doing the following: (1) elucidating the causal mechanisms through which ANPF affect URI, (2) extending the conceptual framework of ANPF, (3) broadening the analytical scope of URI’s interrelationships, and (4) advancing methodological approaches for investigating ANPF–URI dynamics. This study mainly has the following three highlights: (1) This study uses mediation analysis to identify industrial restructuring and upgrading (IND) as a major transmission channel, clarifying the mechanical pathways by which ANPF influence URI. (2) Using threshold regression techniques and the level of informatization (INF) as a moderating factor, the analysis reveals conditional nonlinear relationships between ANPF and URI, providing a better understanding of these context-dependent dynamics. (3) The study uses the Spatial Durbin Model (SDM) to measure both direct and spillover impacts, showing how ANPF affects URI not only locally but also across nearby regions.

3. Theoretical Analysis

3.1. Impact of ANPF on URI

URI refers to the comprehensive merging of economic, demographic, social, spatial, and ecological aspects between cities and rural regions [35]. This process plays a crucial role in strengthening the linkages between urban and rural areas [36]. In this transformative context, ANPF serves as a primary catalyst, driving the advancement of URI.

First, ANPF facilitates economic convergence between urban and rural areas. By adopting advanced production methods [37], ANPF generates more employment opportunities in rural regions. This narrows the historical economic divide, speeds up urbanization and rural revitalization, and fosters seamless economic integration [38].

Second, ANPF enhances demographic integration. The progress in ANPF boosts the efficiency of farming machinery and technology. Such advancements enable surplus agricultural labor to migrate to urban centers while promoting specialization and industrial equilibrium across regions. Consequently, population movement becomes more fluid and adaptable [39], accelerating the blending of urban and rural populations.

Third, ANPF supports social integration. Establishing accessible digital platforms in rural areas can significantly upgrade services and infrastructure. This approach enhances education, healthcare, social connectivity, and essential amenities in rural communities while reducing disparities in public services [40]. Thus, it strengthens the social safety net, ensuring equitable benefits for all residents regardless of location and bridging the gap between urban and rural societies [41].

Fourth, ANPF promotes spatial integration. By leveraging digital platforms and advanced technologies, these forces eliminate physical constraints and improve transportation networks between urban and rural zones [42]. This facilitates regional industrial growth, encourages collaboration among businesses, and advances holistic urban–rural development.

Fifth, ANPF aids ecological integration. Environmentally sustainable by nature [43], they minimize resource consumption and ecological harm from the outset while fostering synergy between urban and rural ecosystems [44]. Furthermore, when combined with IoT and modern monitoring systems, these forces enable the creation of a digital environmental protection framework. Through intelligent surveillance and centralized management, this system ensures the steady and healthy progression of urban and rural ecosystems [45].

Based on this analysis, the study proposes the following hypothesis:

H1. 

ANPF positively influences URI.

3.2. The Mediating Effect of Industrial Restructuring and Upgrading (IND)

The strong relationship between ANPF and information technology is causing major changes in the agricultural industrial structure. This shift is first seen in the modernization and improvement of traditional farming techniques, which make agricultural output more efficient and scientific [46]. At the same time, it broadens the agricultural value chain, increasing the diversity of the agricultural industry. For example, developments in agricultural processing technology and the rapid growth of rural e-commerce have not only raised the value of agricultural products but have also opened up fresh sources of income for farmers [47]. IND also fosters the integration of various industries, creates new business models and service forms, generates new productivity, and improves regional linkages [48]. These developments have offered new economic growth opportunities for rural areas and created more balanced growth among urban and rural communities [49].

Based on this, this study proposes the hypothesis:

H2. 

IND is mediating the process of ANPF empowering URI.

3.3. Threshold Effect of the Level of Informatization (INF)

The impact of ANPF on URI shows distinct threshold behavior depending on regional INF. When INF reaches critical mass through key factors such as comprehensive network infrastructure development, sufficient digitally skilled labor pools, and widespread digital technology adoption [50], it greatly boosts agriculture’s transformative capacity to drive URI. Significantly, INF is both an essential element of ANPF and a powerful link between the urban and rural industries. At developed implementation periods, it reduces traditional communication barriers while enabling service standardization across critical areas such as education [51] and healthcare. This dual process generates a breakthrough environment for equitable regional growth.

Based on this, this study proposes the hypothesis:

H3. 

INF acts as a critical threshold in ANPF to drive URI.

3.4. The Spatial Spillover Effect of ANPF on URI

The spatial spillover impact of ANPF on URI can vary. When such innovation fosters integrated development in nearby regions, it is termed a positive spatial spillover; conversely, if it hinders progress, it is considered a negative spatial spillover [52].

For one thing, due to the geographical closeness, a favorable environment for exchanges and cooperation has naturally formed between neighboring provinces [53], and advanced production factors empower the flow between these neighboring provinces. In provinces with advanced development levels of ANPF, production elements like capital [54], technology, and talent are gradually spreading to less developed areas. This makes way for the improvement of ANPF in undeveloped areas through technological diffusion, speeding up URI. Additionally, the demonstration effect encourages provinces with slower progress in agricultural innovation to adopt strategies from neighboring regions with more advanced productivity. By consistently learning from innovative approaches, refining local development frameworks, and optimizing policy systems, these regions can elevate their ANPF, thereby fostering URI within their areas [55].

Based on this, this study proposes the hypothesis:

H4. 

ANPF has a positive spatial spillover effect on URI.

First, provinces that developed earlier may use their advantages to continue to attract resources from surrounding provinces that are developing more slowly. This interaction enhances the progress of ANPF in the early-mover provinces while weakening the ability of surrounding late-mover provinces to catch up, creating a “backwash effect” [56]. Second, to stay ahead in political competition [57], some provinces with more advanced ANPF may adopt restrictive measures to prevent the flow of their technological achievements [58] and key factors to the outside. This practice will hinder the effective and optimal allocation of resources by late-developing provinces, forcing them to invest more energy and resources in developing their own ANPF. This will increase the challenges in advancing ANPF in these provinces, ultimately creating a restraining impact [59].

Based on this, this study proposes the hypothesis:

H5. 

ANPF has a negative spatial spillover effect on URI.

Building on the preceding theoretical discussion, this study establishes an analytical framework to examine how ANPF influences URI, as shown in Figure 1.

4. Research Design

4.1. Model Construction

To assess how ANPF impacts URI, the study constructs the following benchmark model:

$$UR{I}_{it}={\alpha }_0+{\alpha }_{1}ANP{F}_{it}+{\alpha }_2{X}_{it}+{\delta }_i+{\tau }_t+{\epsilon }_{it}$$

(1)

where Uri_it denotes urban–rural integration; Anpf_it represents agricultural new quality productive forces; i and t represent province and year; X_it indicates the control variable; α₀ is the constant term; α₁ and α₂ are the estimated coefficients of ANPF and the control variable; δ_i is the fixed effect of province; τ_t is the fixed effect of time; and ε_it indicates the error term.

To further explore the transmission mechanism through which ANPF influences URI, this study introduces IND as a mediating variable, based on the earlier theoretical framework. This study constructs the following mediating effect model:

$$IN{D}_{it}={\beta }_0+{\beta }_{1}ANP{F}_{it}+{\beta }_2{X}_{it}+{\delta }_i+{\tau }_t+{\epsilon }_{it}$$

(2)

$$UR{I}_{it}={\eta }_0+{\eta }_{1}ANP{F}_{it}+{\eta }_{2}IN{D}_{it}+{\eta }_3{X}_{it}+{\delta }_i+{\tau }_t+{\epsilon }_{it}$$

(3)

The preceding equations contain two fixed parameters (β₀ and η₀) acting as constants, while β₁, β₂, η₁, η₂, and η₃ serve as coefficients for their associated variables.

To examine potential nonlinear relationships between ANPF and URI, this study establishes a threshold effect model, outlined below:

$$UR{I}_{it}={\varphi }_0+{\varphi }_{1}ANP{F}_{it}\times I(IN{F}_{it}\le \theta )+{\varphi }_{2}ANP{F}_{it}\times I(IN{F}_{it}>\theta )+{\varphi }_3{X}_{it}+{\delta }_i+{\tau }_t+{\epsilon }_{it}$$

(4)

where φ₀ indicates the constant term, φ₁, φ₂, and φ₃ indicate the estimated coefficients of the corresponding variables, INF_it indicates the threshold variable, θ is the threshold value, I (·) is the indicator function, and the other variables have the same meanings as described above.

Additionally, to investigate whether ANPF positively influences URI in neighboring regions, this paper uses spatial econometric methods for further analysis. The specific model is:

$$UR{I}_{it}={\gamma }_0+\rho W\times UR{I}_{it}+{\gamma }_{1}ANP{F}_{it}+{\gamma }_{2}W\times ANP{F}_{it}+{\displaystyle \sum {\gamma }_3}{X}_{it}+{\displaystyle \sum {\gamma }_4}W\times X_{it}+{\delta }_i+{\tau }_t+{\epsilon }_{it}$$

(5)

where γ₀ is the constant term, γ₁, γ₂, γ₃, and γ₄ are the regression coefficients of the variables, ρ denotes the strength of spatial autocorrelation, with W defining the spatial weights matrix.

4.2. Variable Description

4.2.1. Explained Variables

The explained variable in this research is URI. Reflecting the theoretical essence of URI, this study establishes a multidimensional evaluation framework. URI manifests through five interconnected dimensions—economic, demographic, social, spatial, and ecological—which collectively shape regional development trajectories. The economic dimension drives efficient resource distribution and boosts aggregate productivity, while demographic integration enables workforce mobility and mitigates population distribution imbalances. Social convergence reduces disparities in access to public services and advances equitable development. Spatially, integration improves settlement patterns and increases land utilization effectiveness. Lastly, ecological harmonization safeguards environmental assets while supporting long-term sustainability. Building on the work of Zhao et al. [60], Xiong and Dai [61], five key dimensions—economic, population, social, spatial, and ecological integration—were identified, further broken down into 18 specific indicators [62,63]. These form the provincial URI evaluation system presented in

Table 1. Evaluation indicator system for URI.

Primary	Secondary	Definition and Description	Properties	Weight
Urban–rural economic integration	Ratio of urban–rural income	Per capita income of urban residents/Per capita income of rural residents	−	0.0172
	Urban–rural household expenditure ratio	Per capita consumption of urban residents/Per capita consumption of rural residents	−	0.0149
	Urban–rural Engel coefficient differential	urban Engel’s coefficient/ rural Engel’s coefficient	+	0.0141
	Ratio of industrial output value	Secondary and tertiary industry output value/ Primary industry output value	+	0.3658
Urban–rural population integration	Ratio of non-farm-to-farm employment	Proportion of people employed in the secondary and tertiary industries/ Proportion of people employed in the primary industry	+	0.2430
	Urbanization rate	Urban population/total population	+	0.0297
	Ratio of urban–rural population density	Urban population density/Rural population density	−	0.0095
Urban–rural social integration	Urban–rural disparity in education and entertainment expenditure	Per capita expenditure on education and entertainment in urban households/ Per capita expenditure on education and entertainment in rural households	−	0.0037
	Level of urban–rural medical security	Hospital bed density per 10,000 people by residency (urban/rural)	+	0.0312
	Level of urban–rural social security	Urban and rural social security and employment expenditure/ General budget expenditure	+	0.0476
	Comparative coefficient of per capita healthcare in urban and rural areas	Per capita healthcare expenditure in urban areas/ Per capita healthcare expenditure in rural areas	−	0.0041
Urban–rural space integration	Road area per capita	Road surface/Population size	+	0.0303
	Per capita park green space	Green space/Population size	+	0.0236
	Transport network density	(Road mileage + Railway operating mileage)/ Total land area	+	0.0519
	Ratio of built-up area	Built-up area/city area	+	0.0402
Urban–rural ecology integration	Sewage treatment rate	Sewage treatment capacity/Total wastewater discharge	+	0.0061
	Environmental protection expenditure	Local financial expenditure on environmental protection	+	0.0641
	Household waste sanitization level	Quantity of household waste treated in an environmentally sound manner/ Household waste generation	+	0.0030

. To enhance the objectivity of weight determination and improve assessment accuracy, we employed the entropy method for quantifying integration levels. This established weighting approach calculates indicator importance according to their statistical variability and degree of dispersion within the dataset. Notably, evaluation metrics demonstrating higher dispersion values exert proportionally stronger influence on composite index outcomes. Widely recognized in empirical research, this technique represents a robust objective weighting methodology.

4.2.2. Core Explanatory Variable

The core explanatory variable in this study is the ANPF. ANPF essentially falls within the category of productivity but with a particular emphasis on the “new” aspect, i.e., novelty and qualitative improvement compared with existing productivity. Its characteristics include three elements: new and high-quality agricultural means of production, new and high-quality agricultural labor, and new and high-quality agricultural objects of labor. Referring to the indicator selection and processing methods of related studies, a comprehensive evaluation system covering three aspects—agricultural laborers, agricultural-related labor objects, and means of agricultural labor—with a total of 17 specific indicators, is finally constructed. The indicator weights were calculated using the entropy method. The specific indicator system and description are shown in

Table 2. ANPF evaluation index system.

Indicator Category	Definition and Description	Properties	Weights
Agricultural laborers	Average years of education of rural labor force	+	0.0094
	Full-time equivalent R&D staff	+	0.0981
	Labor productivity in primary industry	+	0.0354
	Rural disposable income per capita	+	0.0351
Agriculture-related labor objects	Intensity of chemical fertilizer use	−	0.0155
	Carbon emissions from pesticides	−	0.0137
	Annual income from leisure agriculture and rural tourism	+	0.0573
	Ratio of green agricultural cooperatives to primary sector workforce size	+	0.1352
	Agricultural product processing industry operating income	+	0.0854
	Output value of agriculture, forestry, animal husbandry, and fishery	+	0.0770
Means of agricultural labor	Ratio of rural road mileage to rural population	+	0.0553
	Length of optical cable routes per unit area	+	0.1137
	Rural broadband access volume	+	0.0568
	Rural Digital Inclusive Finance Index	+	0.0186
	Average mobile phone ownership per 100 rural households	+	0.0243
	Area of machine-transplanted land per capita	+	0.0931
	Agricultural R&D investment	+	0.0761

.

4.2.3. Mediating Variable

Combining the theoretical analysis above, this paper selects IND as the mediating variable. Following the approach of Cheng et al. [64], the industrial structure hierarchy coefficient is used to quantify IND, calculated based on the output value shares of the primary, secondary, and tertiary sectors. The formula is as follows:

$$IN{D}_{it}={\displaystyle \sum _{m=1}^{3}m\times Y_{imt}}$$

(6)

where m indexes economic sectors: 1 (primary industries), 2 (secondary industries), 3 (tertiary industries). Y_imt denotes industry m’s output share in region i during period t.

4.2.4. Threshold Variables

Drawing on the preceding theoretical analysis, this study chooses INF as the threshold variable. Specifically, the total amount of post and telecommunications business as a proportion of gross domestic product (GDP) is used to measure INF [65,66].

4.2.5. Control Variables

To more accurately depict the link between ANPF and URI, this study introduces four control variables: (a) Consumer demand (CONS): represented by the ratio of total retail sales of consumer goods to GDP. (b) Economic development (ECO): measured by per capita GDP in each province. (c) Opening up to the outside world (OPE): measured by the annual ratio of total import and export trade to regional GDP. (d) Government intervention (GOV): measured by the ratio of total social fixed asset investment to GDP to represent government actions [27,61,67].

4.3. Data Sources and Statistical Characteristics

The study considers data continuity over ten years from 2013 to 2022 throughout China’s 30 provincial-level administrative regions. Four regions—Tibet, Hong Kong, Macao, and Taiwan—were removed due to insufficient data availability. The analysis includes 35 distinct indicators measuring ANPF and URI. The indicator system for URI incorporates three ecological metrics derived from the China Environmental Statistics Yearbook, with all other variables obtained from the China Statistical Yearbook. Regarding ANPF measures, the “Full-time equivalent R&D staff” data originates from the China Science and Technology Statistics Yearbook, whereas the remaining indicators are extracted from the China Rural Statistics Yearbook. To maintain dataset completeness, minor missing values were corrected using linear interpolation before empirical analysis. Detailed data are presented in

Table 3. Descriptive statistics for variables.

Variable Category	Variables	Symbol	N	Mean	SD	Min	Max
Explained variable	Urban–rural integration	URI	300	0.204	0.101	0.087	0.788
Core explanatory variable	Agricultural new quality productive forces	ANPF	300	0.189	0.085	0.053	0.492
Mediating variable	Industrial restructuring and upgrading	IND	300	2.401	0.123	2.194	2.836
Threshold variable	Level of informatization	INF	300	0.065	0.056	0.015	0.290
Control variables	Consumer demand	CONS	300	1.165	0.949	0.055	4.488
	Economic development	ECO	300	0.627	0.311	0.221	1.903
	Opening up to the outside world	OPE	300	0.254	0.262	0.008	1.362
	Government intervention	GOV	300	0.250	0.101	0.107	0.643

.

5. Analysis of Empirical Results

5.1. Benchmark Regression

To assess the influence of ANPF on URI across provinces, a Hausman test was conducted, rejecting the null hypothesis that random effects are preferable to fixed effects. Consequently, a fixed effects model was adopted, with regression results detailed in

Table 4. Results of the benchmark regression.

Variables	URI
	(1)	(2)	(3)	(4)	(5)
ANPF	0.666 *** (21.49)	0.894 *** (15.35)	0.180 *** (2.67)	0.286 *** (4.55)	0.268 *** (4.17)
CONS		−0.041 ** (−4.56)	−0.035 *** (−4.99)	−0.042 *** (−6.58)	−0.041 *** (−6.41)
ECO			0.247 *** (14.01)	0.208 ** (12.34)	0.214 *** (12.25)
OPE				−0.139 *** (−7.58)	−0.137 *** (−7.38)
GOV					0.052 ( 1.28 )
Fixed effects	Yes	Yes	Yes	Yes	Yes
N	300	300	300	300	300
R2	0.632	0.658	0.803	0.838	0.839

Note: ** and *** denote passing 5% and 1% significance levels.

. Column (1) presents the baseline relationship between ANPF and URI, with a statistically significant coefficient of 0.666 (p < 0.01). Columns (2) to (5) show the outcomes after adding control factors such as consumer demand (CONS), economic development (ECO), opening up to the outside world (OPE), and government intervention (GOV). ANPF continues to have a strong positive impact on URI, with all passing the 1% significance test. At the same time, the coefficient of determination (R²) improved from 0.632 to 0.839, demonstrating that ANPF had a strong positive effect on URI, confirming H1.

5.2. Robustness Test

To assess the robustness of our findings, we employ three validation methods. (1) Instrumental Variable Method. To avoid the problems of reverse causality and ignoring key control variables that lead to endogeneity, this paper adopts the two-stage least squares method, referring to Vega’s method, and uses the first-stage lag value of the core variable as the instrumental variable to test for endogeneity. Model (1) in

Table 5. Results of the robustness test.

Variables	Model (1)		Model (2)		Model (3)
	Coef.	S.E.	Coef.	S.E.	Coef.	S.E.
ANPF	0.395 **	0.197	0.216 ***	0.052	0.231 **	0.109
Control variable	Yes		Yes		Yes
N	300		300		150
95%CI	[0.009, 0.781]		[0.114, 0.318]		[0.015, 0.446]
R2	0.831		0.874		0.591

Note: ** and *** denote passing 5% and 1% significance levels.

presents the endogeneity test results. The significant influence of ANPF on URI persists, confirmed at the 5% significance level. Additionally, the results pass both the weak instrument identification and the instrumental variable underidentification test, validating the appropriateness of the instrumental variables selected. (2) Winsorization. To mitigate the influence of outliers, the data on ANPF and URI underwent 1% bilateral truncation, followed by re-running the regression analysis. Model (2) in Table 5 demonstrates that, after this adjustment, the significance and direction of the impact of ANPF remain consistent with the original findings, confirming the robustness of the study’s results. (3) Stagewise Regression. The robustness of the conclusion is tested by performing a regression analysis using data from 2018 to 2022. As shown in Model (3) of Table 5, even after reducing the sample size, the result remains significantly positive, further affirming the robustness of the findings.

5.3. Heterogeneity Analysis

Due to notable disparities in development stages and resource allocation across regions, the impact of ANPF on URI varies. In accordance with the official classification standards established by China’s National Bureau of Statistics, the sample provinces were categorized into three major regions: eastern, central, and western China. The eastern region comprises 11 provincial-level administrative units, including Beijing, Tianjin, Hebei, Shanghai, Liaoning, Jiangsu, Zhejiang, Fujian, Shandong, Guangdong, and Hainan. The central region consists of 8 provinces such as Shanxi, Henan, Hubei, Jilin, Heilongjiang, Hunan, Anhui, and Jiangxi. The western region encompasses 11 administrative areas, notably Inner Mongolia, Guangxi, Chongqing, Sichuan, Guizhou, Yunnan, Shaanxi, Gansu, Qinghai, Ningxia, and Xinjiang.

Table 6. Results of the heterogeneity analysis.

Variables	Eastern	Central	Western
ANPF	0.294 ** (2.33)	0.196 *** (3.49)	0.544 *** (8.22)
Control variable	Yes	Yes	Yes
N	110	80	110
95%CI	[0.043, 0.546]	[0.084, 0.308]	[0.413, 0.676]
R2	0.854	0.951	0.938

Note: ** and *** denote passing 5% and 1% significance levels.

outlines the findings of the regional heterogeneity analysis. ANPF significantly enhances URI, with the western region experiencing a notably stronger impact compared to the eastern and central regions.

Specifically, the central and western regions show significance at the 1% level, while the eastern region is significant at the 5% level. The influence coefficients for the eastern, central, and western regions are 0.294, 0.196, and 0.544, respectively, reflecting a pattern of “western > eastern > central” in the impact of ANPF on URI.

The variation in impact across the eastern, central, and western regions may stem from differences in economic development levels and industrial structures. The eastern and central regions, being more economically advanced with diversified industries, have a smaller agricultural share in their economies, resulting in a relatively lower marginal effect of ANPF on URI. In contrast, the less developed western region relies heavily on agriculture for household income, making ANPF a more influential factor in driving URI.

5.4. Mediation Analysis

The mediating effect analysis of IND is presented in

Table 7. Results of the mediation analysis.

Variables	(1)	(2)
	IND	URI
ANPF	0.759 *** (5.74)	0.223 *** (3.28)
IND		0.060 ** (2.03)
Control variable	Yes	Yes
Fixed effects	Yes	Yes
N	300	300
R2	0.568	0.842

Note: ** and *** denote passing 5% and 1% significance levels.

. Initial findings in column (1) reveal a statistically significant positive relationship (p < 0.01) between ANPF and IND, indicating that ANPF is conducive to the development of IND. Column (2) reveals statistically significant coefficients for both IND and ANPF, with the latter showing positive significance at a slightly reduced magnitude compared to baseline regression outcomes. These results confirm that IND partially mediates the relationship between ANPF and URI, thereby supporting Hypothesis 2.

The observed phenomenon may be attributed to the transformative impact of ANPF on IND. By facilitating the convergence of agricultural production and information technologies, ANPF has catalyzed the expansion of agricultural value chains and enhanced their sophistication. Such structural transformations enable traditional sectors to achieve technological leapfrogging through the development of advanced, high-value products. Digital innovation serves as a key driver in this process, accelerating the modernization of conventional industries through digital transformation and intelligent upgrading [68]. This technological integration facilitates the convergence of emerging infrastructure with traditional production systems, effectively breaking down sectoral barriers between urban and rural industries while strengthening their vertical and horizontal linkages. The resultant optimization of resource allocation mechanisms enhances the efficiency of factor mobility across urban–rural divides, fostering more balanced regional economic development [69].

5.5. Threshold Effect Analysis

This study investigates the nonlinear relationship between ANPF and URI by employing INF as a threshold variable. The likelihood ratio function plot for the two thresholds is shown in Figure 2. The empirical findings, presented in

Table 8. Threshold value.

Threshold Variables	Threshold Type	Threshold Value	F Value	p Value
INF	Single threshold	0.049 **	38.330	0.033
	Double threshold	0.054 ***	50.620	0.000

Note: ** and *** denote passing 5% and 1% significance levels.

, reveal a statistically significant double-threshold effect (p < 0.01). The analysis identifies two critical threshold values at 0.049 and 0.054, respectively, indicating distinct stages of impact. These results provide robust empirical support for the validation of Hypothesis 3.

Table 9. Results of the threshold effect analysis.

Variables		INF
Threshold value	θ1	0.049
	θ2	0.054
ANPF × I (INFθ1)		0.238 *** (4.21)
ANPF × I (θ1 < INF < θ2)		0.452 *** (7.62)
ANPF × I (INF < θ2)		0.280 *** (4.82)
Control variable		Yes
N		300
R2		0.879

Note: *** denotes passing1% significance level.

presents the results of the threshold regression analysis, revealing significant differences in APNF across different INF levels. In the low INF level regime (≤0.049), the ANPF coefficient of 0.238 demonstrates significant positive effects (p < 0.01). Within the intermediate INF level range (0.049–0.054), the impact intensifies, with the ANPF coefficient rising to 0.452 while maintaining 1% significance. However, in the high INF level regime (>0.054), the ANPF coefficient moderates to 0.280, though remaining statistically significant at the 1% level. Collectively, these findings indicate a nonlinear yet consistently positive relationship between ANPF and URI across varying informatization thresholds.

This phenomenon can be attributed to the differential impacts of informatization across developmental stages. In the initial phase, enhanced information flow efficiency and reduced transaction costs facilitate resource optimization, thereby amplifying the transformative potential of ANPF on URI. As informatization progresses, advanced technologies generate more profound societal impacts, particularly through their comprehensive empowerment of production factors. This technological sophistication significantly augments the overall effectiveness of ANPF in driving URI. However, beyond a certain threshold of informatization maturity, while technological capabilities continue to enhance productivity, their relative contribution diminishes as they cease to be the primary growth driver. Consequently, the transformative impact of ANPF on URI exhibits a moderated effect compared to earlier stages.

5.6. Further Analysis: Spatial Effect Analysis

This research employs an inverse geographical distance matrix for spatial effect analysis. The Moran’s I index, presented in

Table 10. Moran’s Index result.

Year	ANPF		URI
	Moran’s I	Z Value	Moran’s I	Z Value
2013	0.076 ***	3.105	0.100 ***	4.121
2014	0.083 ***	3.307	0.089 ***	3.857
2015	0.082 ***	3.297	0.092 ***	3.920
2016	0.082 ***	3.303	0.073 ***	3.432
2017	0.080 ***	3.218	0.071 ***	3.366
2018	0.083 ***	3.296	0.061 ***	3.096
2019	0.088 ***	3.403	0.061 ***	3.056
2020	0.097 ***	3.652	0.044 ***	2.588
2021	0.095 ***	3.600	0.027 **	2.099
2022	0.096 ***	3.626	0.023 **	1.999

Note: ** and *** denote passing 5% and 1% significance levels.

, reveals statistically significant positive values throughout the 2013–2022 study period. These findings demonstrate a consistent positive spatial autocorrelation, validating the appropriateness of spatial econometric modeling for examining the geographical interdependencies between ANPF and URI.

Table 11. Results of the LM, LR, and Wald tests.

Type of Test	Test Statistical Values	p Value
LM-Error	2.918 *	0.088
Robust LM-Error	0.016	0.899
LM-Lag	10.530 ***	0.001
Robust LM-Lag	7.628 ***	0.006
Hausman	44.150 ***	0.000
LR-ind	36.680 ***	0.000
LR-time	570.750 ***	0.000
Wald-sem	29.040 ***	0.000
Wald-sar	38.330 ***	0.000

Note: * and *** denote passing 10% and 1% significance levels.

presents the model selection diagnostics. The SEM model satisfies one LM test criterion, while the SAR model meets both LM test requirements. Following Elhorst’s methodological framework, the presence of spatial effects warrants consideration of the SDM specification. Both Hausman and LR tests yield statistically significant results, supporting the adoption of a double fixed-effects approach. Furthermore, the Wald test results confirm that the SDM model maintains its distinct specification without degenerating into either SEM or SAR models. Based on these diagnostic outcomes, the study employs a double fixed-effects SDM framework for spatial analysis.

The spatial regression analysis reveals significant findings, as presented in

Table 12. SDM model regression results.

Variables	Main (1)	Wx (2)	LR-Direct (3)	LR-Indirect (4)	LR-Total (5)
ANPF	0.386 *** (4.900)	−0.851 * (−1.690)	0.524 *** (6.400)	−0.692 *** (−3.380)	−0.168 (−0.790)
Control variable	Yes
Fixed province	Yes
Fixed time	Yes
N	300
R2	0.841

Note: * and *** denote passing 10% and 1% significance levels.

. Initial results in columns (1)–(2) demonstrate a positive local impact of ANPF (coefficient = 0.386, p < 0.05) on URI, confirming its spatial significance. However, the negative spatial lag term (−0.851, p < 0.05) suggests a competitive effect, where local ANPF advancement may hinder neighboring regions’ development.

Further decomposition of spatial effects through partial differential analysis (columns 3–5) provides additional insights. The significant direct effect (0.524, p < 0.05) confirms the positive intra-regional impact, while the negative indirect effect (−0.692, p < 0.05) substantiates the inter-regional competitive dynamics. These findings collectively reject Hypothesis 4 while validating Hypothesis 5, indicating that while ANPF fosters local regional integration, it simultaneously creates competitive pressures that impede neighboring areas’ development.

6. Conclusions and Recommendations

6.1. Conclusions and Discussions

This study investigates the influence of agricultural new quality productive forces (ANPFs) on urban–rural Integration (URI) using panel data covering 30 Chinese provinces from 2013 to 2022. After establishing provincial-level evaluation systems for both ANPF and URI, multiple econometric approaches were applied, including bidirectional fixed-effects models, mediation analysis, threshold regression, and spatial Durbin models, to explore the transmission mechanism and spatial spillover effect systematically. The empirical results demonstrate:

(1)

ANPF exhibits a statistically significant positive correlation with URI advancement. Each unit increase in ANPF corresponds to a 0.268-unit rise in URI, with this relationship remaining robust across various specification tests.

(2)

Regional heterogeneity exists in ANPF’s effects, with western China experiencing substantially stronger impacts than eastern and central regions.

(3)

IND serves as a crucial transmission channel. Through agricultural technological innovation, ANPF drives industrial transformation and upgrading, while IND further optimizes the allocation of production factors between urban and rural areas, ultimately supporting URI progress.

(4)

The influence of ANPF on URI displays nonlinear characteristics concerning INF. Initially, INF strengthens ANPF’s positive effects. While continued INF improvement enhances the overall enabling effect, the marginal benefit diminishes after reaching certain development thresholds.

(5)

Spatial effect analysis reveals that ANPF generates negative spillover effects on neighboring regions. Although ANPF significantly boosts local URI development, it concurrently inhibits URI advancement in adjacent areas.

URI significantly contributes to sustainable development through multiple pathways: enhancing resource allocation efficiency, fostering balanced economic growth, ensuring equitable public service distribution, and conserving ecological systems. As demonstrated in preceding analyses, ANPF development actively promotes URI, consequently supporting global sustainability objectives. Nevertheless, regional disparities in ANPF advancement necessitate tailored approaches, analogous to China′s distinct development strategies for its eastern, central, and western regions. IND has been identified as a crucial mediator in the ANPF-URI relationship. Consequently, particular emphasis should be placed on fostering IND when promoting both ANPF advancement and URI initiatives. Moreover, the effects of ANPF on URI exhibit spatial heterogeneity between local and adjacent areas, demanding differentiated treatment and coordinated regional planning. These evidence-based insights, when translated into context-specific policies, offer vital foundations for achieving sustainable development goals.

6.2. Recommendations

Significant urban–rural disparities persist across China, evidenced by income ratios of 2.2:1 (eastern region), 2.6:1 (central region), and 2.8:1 (western region), highlighting pronounced regional variations. Regional planning adaptability is more advanced in the east compared to central and western areas, where improvements are still needed. In 2023, rural technical training participation reached merely 30% nationally. Additionally, while China had 100 interregional agricultural collaboration platforms, the majority were located in eastern provinces, with limited coverage in central and western regions. Moving forward, policy interventions should prioritize enhanced financial and technical support, improved resource distribution, expanded skills training programs, and broader access to cross-regional agricultural networks to foster equitable urban–rural development and long-term sustainability.

Based on the conclusions, three key policy directions warrant government attention.

First, develop ANPF in line with local conditions and promote URI development. Regional development plans should be adapted to local conditions, with eastern regions focusing on innovation-driven growth, while central and western areas prioritize infrastructure development and market mechanism improvement. Illustrative cases demonstrate distinct regional approaches to rural development. Hangzhou (Zhejiang Province, eastern China) has strategically utilized its abundant ecological assets and cultural heritage to establish thriving agritourism initiatives, yielding dual benefits of stimulating local economic growth and increasing tourist attractiveness. Meanwhile, Chengdu (Sichuan Province, western China) has achieved significant agricultural modernization through technological innovation and digital farming solutions, resulting in enhanced production efficiency and upgraded rural infrastructure.

Second, promote urban–rural IND. The coordinated development of urban and rural industrial sectors enhances resource utilization efficiency, reduces regional disparities, and accelerates ecological transition, thus contributing to sustainable growth. Digital solutions, including e-commerce networks and blockchain-based tracking systems, can bridge supply chain gaps between urban and rural areas, enabling agricultural producers to reach metropolitan and global markets directly. Such integration fosters interconnected industries like agro-processing and eco-tourism, generating local job opportunities and mitigating rural-to-urban migration.

Third, promote the balanced regional development of ANPF. Advancing the equitable growth of ANPF contributes to synchronized agricultural advancement across regions, serving as a critical indicator of sustainable farming practices. To tackle the problem of ANPF generating negative spatial spillover effects that hinder the development of neighboring regions, provinces with ANPF should facilitate knowledge transfer through inter-provincial cooperation, while less developed regions should actively adopt best practices and cultivate local agricultural technology enterprises. As an illustrative case, Jiangsu Province has fostered knowledge diffusion and strengthened ANPF via interregional collaboration. Drawing upon Jiangsu′s pioneering approaches, Gansu Province has implemented a trans-provincial agricultural technology initiative that convenes specialists and sector stakeholders to conduct joint knowledge-transfer and capacity-building programs.

6.3. Limitations and Future Prospects

While this study provides meaningful contributions to the literature, several constraints should be acknowledged. First, the analysis relies exclusively on provincial-level panel datasets, which may produce findings with limited granularity when applied to smaller administrative divisions like county-level jurisdictions. Subsequent investigations would benefit from employing more localized data to enhance the precision of urban–rural integration (URI) assessments.

Second, from a regional perspective, although geographical differentiation serves as a conventional basis for heterogeneity analysis, both agricultural new productivity factors (ANPFs) and URI outcomes are substantially affected by jurisdiction-specific policy frameworks and governance structures. This creates significant variation in policy implementation effectiveness across different localities. An important avenue for future inquiry would be examining ANPF-URI relationships while accounting for these policy-driven variations.

Third, regarding analytical comprehensiveness, while our model incorporates key control variables (CONS, ECO, OPE, GOV), the ANPF-URI interaction mechanism encompasses a broader range of influential elements. Future modeling efforts should therefore integrate supplementary indicators, particularly those measuring rural financial market development and technological adoption in agricultural production systems.