Impact of Rural Industrial Integration on Rural Air Quality: Evidence from Prefecture-Level Cities in China

Abstract

In China, the sustainable development of rural areas is of great significance to sustainable development. Especially with the acceleration of rural industrial integration in China, its environmental impact not only affects the resilience of rural areas but also affects the regional environment. According to the multiple and different dimensions of the impact of rural industrial integration on rural ecosystems, this research focuses on prefecture-level cities in China, utilizing regional economic and environmental data from 2014 to 2023. To analyze the impact of rural industrial integration on rural air quality using the entropy method, a rural industrial integration index was constructed, and the fixed effects model was used to analyze the findings. The results showed that a 0.01 increase in the rural industrial integration index will result in a decrease of 1.335% in the AQI index of an entire prefecture-level city, with empirical validation of the moderating effects of greening, rural waste treatment, and the mediating effects of infrastructure development. The study found that rural industrial integration and air quality levels in mainland China’s prefecture-level cities exhibited a steady upward trend.

1. Introduction

In China’s accelerated urbanization process, the imbalance in development between urban and rural areas is becoming increasingly prominent [1]. In order to address structural issues, China has launched the rural revitalization initiative to promote sustainable rural development [2,3]. The most important means is to promote the integration of agriculture with other industries through rural industry integration, thereby increasing residents’ income and enhancing the resilience of rural areas. At the same time, industry integration will also have an impact on the rural environment and the regional environment, which in turn will further affect regional sustainable development.

Regional air quality is determined not only by cities but also by rural areas. Rural areas, which usually occupy vast areas of a country, are an important part of the ecosystem, and the quality of their natural environment has a direct impact on the regional ecological balance and the performance of ecological services. In 2015, the contribution of household (mainly rural households) coal burning to the average monthly PM2.5 concentration in the Beijing, Tianjin, and Hebei region was 46–57% (Zhang et al., 2024) [4]. Thus, when discussing air quality and sustainable development issues in the region as a whole, it is inevitable that the important role of rural areas needs to be discussed.

In recent years, rural areas in China have faced increasingly severe air pollution problems [5]. Agricultural production activities, particularly livestock manure emissions, and rural waste incineration practices exert adverse effects on regional air quality. Concurrently, the advancement of rural industrial convergence offers innovative pathways to address these environmental challenges. This dualistic mechanism manifests through two countervailing effects: while the expansion of industrial scale during this convergence process often correlates with increased fossil fuel consumption, thereby negatively impacting air quality [6], the integration of green industries presents opportunities to enhance ecological conditions and improve atmospheric environments. Consequently, a systematic investigation into the complex interplay between rural industrial convergence and regional air quality holds critical implications for advancing sustainable rural economic development and fostering ecological modernization [7,8].

Air quality within agro-ecological systems constitutes a fundamental prerequisite for achieving ecological livability under China’s rural revitalization strategy. The ongoing integration of agricultural industries continues to reshape modern agricultural frameworks while accelerating the transition toward sustainable development paradigms. This structural transformation, however, engenders a dualistic environmental mechanism: the scaling of industrial convergence may inadvertently intensify resource consumption patterns, whereas the institutionalization of green production practices potentially fosters atmospheric remediation.

The Japanese scholar Naritomi Imamura pioneered the conceptual framework of the “Sixth Industrialization” (a term denoting the integration of primary, secondary, and tertiary industries in rural areas) [9]. Concurrently, China has strategically synthesized international experiences in rural industrial integration, leveraging localized resource advantages to cultivate specialized industries, extend agricultural value chains, and advance rural revitalization. Against this backdrop of rapid rural industrial integration, evaluating its socioeconomic impacts emerges as a pressing academic inquiry. Rural industrial convergence constitutes a systemic transformation driven by value chain extension, functional diversification, factor agglomeration, technology infusion, and institutional innovation. This multidimensional transition fosters symbiotic integration between primary industries and secondary/tertiary sectors, thereby actualizing agrarian modernization and enhancing rural livelihood sustainability through synergistic value creation mechanisms (Ding, 2023) [10]. To date, scholarly discourse predominantly centers on the economic implications of rural industrial integration.

Contemporary academic inquiries into rural industrial integration primarily focus on three theoretical dimensions: conceptual definitions, driving forces, and developmental pathways. While scholars exhibit nuanced interpretations of rural industrial integration, a consensus exists regarding its essence (Luo & Geng, 2023) [11]. A representative conceptualization defines it as “the synergistic convergence, functional permeation, and structural reorganization among primary, secondary, and tertiary rural industries, manifested through agricultural value chain extension, industrial boundary expansion, and functional transformation. These processes engender novel technologies, business formats, and operational models, thereby achieving comprehensive optimization of rural resources, production factors, and technological configurations”(Li et al., 2023; Jiang, 2015) [12,13].

Scholarly investigations can be divided into two principal strands regarding the socioeconomic impacts of rural industrial integration. The first strand predominantly examines its macroeconomic effects. For instance, Freeman and Soete (1997) [14] posit that industrial integration facilitates economies of scope, whereas Cao and Chen (2021) [15] and Amir et al. (2011) [16] identify polarization effects that may disproportionately disadvantage vulnerable sectors. Regarding rural contexts, the academic consensus confirms the efficacy of industrial integration in alleviating poverty. Tu (2020) [17] establishes industrial poverty reduction as a fundamental strategy for sustainable poverty eradication, operating through rural economic growth and urbanization mechanisms (Zhao & Liu, 2022) [18], digital penetration in rural areas (Peng et al., 2024) [19], and enhanced educational investments. Concurrently, empirical evidence demonstrates that industrial integration significantly enhances agricultural productivity resilience (Yao et al., 2024) [20] and elevates farmer incomes (Qi et al., 2021) [21], thereby structurally reinforcing rural revitalization.

The second strand investigates organizational transformations induced by rural industrial integration. Agricultural cooperatives exemplify such institutional innovations, enabling deep convergence of primary, secondary, and tertiary industries through collective management models (Li et al., 2016) [22]. Cross-national studies substantiate cooperatives’ welfare-enhancing effects, including improved farmer livelihoods (Wollni & Zeller, 2007; Chagwiza et al., 2017; Mojo et al., 2017; Lee & Van Cayseele, 2024) [23,24,25,26]. Spatiotemporal analyses further reveal that industrial integration catalyzes geographic clustering phenomena. Firms gravitate toward optimal development loci to capitalize on agglomeration economies, thereby forming industrial clusters that reshape regional economic landscapes (Heo & Lee, 2019) [27]. This spatial reorganization manifests as concentrated innovation hubs and value chain networks, fundamentally altering traditional rural production geographies.

With the transformation and upgrading of the manufacturing sector, the environmental effects triggered by rural industrial integration have extended beyond mere economic impacts to become a growing academic concern. Virkanen (1998), through empirical research on air and water environments in southern Finland’s industrial zones, identified industrial agglomeration as a direct contributor to atmospheric and water pollution [28]. Wu et al. (2020) further demonstrated that rural industrial clustering constitutes a significant factor influencing agricultural carbon emissions [29]. The relationship between agricultural economic growth and the rural environment has been explored through the seminal Environmental Kuznets Curve proposed by Grossman et al. (1991), which posits an inverted U-shaped correlation between economic development and environmental quality—initial economic expansion exacerbates environmental degradation until reaching an inflection point, after which environmental conditions progressively improve [30].

The current literature has primarily examined the economic impacts arising from rural industrial integration, whereas insufficient attention has been paid to its environmental implications. Moreover, few studies have specifically investigated the effects on rural air quality from the perspective of rural industrial integration. Furthermore, the research remains largely qualitative in nature, lacking in-depth theoretical or empirical analysis of its mechanisms of action.

This study adopted the panel data of prefecture-level cities in China from 2014 to 2023, incorporating regional economic and environmental indicators, to decipher the sustainability trade-offs inherent in rural industrial integration, probing their implications for air quality governance under the rural revitalization strategy. The research not only establishes measurement methodologies for rural industrial integration levels but also quantitatively examines their causal relationship with air quality dynamics. Furthermore, through mechanism analysis focusing on green coverage effects and waste management effects, the study specifically investigates the mediating pathways through which rural industrial integration influences regional air pollution patterns.

The marginal contributions of this study are threefold: First, while the existing literature predominantly focuses on urban economic development when analyzing regional air quality determinants, this research systematically incorporates rural industrial integration as a critical analytical dimension, thereby expanding the existing analytical framework to encompass rural–urban environmental interactions. Second, building upon conventional economic determinants like technological advancement and industrial agglomeration, this investigation pioneers the identification of rural-specific mechanisms—particularly greening infrastructure enhancement and waste management efficacy improvement—through which industrial integration exerts environmental remediation effects, offering empirical validation of understudied ecological transmission channels. Third, the study innovatively operationalizes county-level green coverage ratios and solid-waste utilization rates as environmental performance metrics, establishing quantifiable linkages between rural industrial restructuring and atmospheric quality optimization.

2. Materials and Methods

2.1. Research Hypotheses

Rural industrial integration significantly amplifies greening effects through systematic intervention measures. These include the adoption of ecological agricultural technologies, expansion of forest coverage, and development of agricultural tourism landscapes, collectively increasing vegetation coverage. The enhanced greening function acts as a biological filtration system, directly improving regional air quality by adsorbing particulate matter (PM_2.5 and PM₁₀), absorbing gaseous pollutants (SO₂ and NO_x), and producing oxygen.

Additionally, rural industrial integration has improved rural waste management efficiency by providing better technical and financial support for waste treatment, such as constructing waste collection stations, material recycling facilities, and centralized treatment systems. This has reduced reliance on open burning and unregulated landfilling, thereby suppressing emissions of methane (CH₄), dioxins, and airborne particulate matter. This reduces anthropogenic pollution sources and complements air purification driven by greening initiatives.

Furthermore, well-developed infrastructure is a prerequisite for rural industrial integration, driving ecological and environmental improvements. For example, improvements in transportation facilities, the enhancement of water conservancy facilities, and the construction of environmental protection facilities create better conditions for rural industrial integration, thereby reinforcing its positive role in improving air quality.

Therefore, Figure 1 illustrates the impact of rural industrial integration, including greening effect, rural waste treatment effect, and infrastructure construction, on regional air quality.

As industrial integration progresses, rural economies transition toward diversification and modernization, exemplified by emerging sectors such as agricultural product processing and rural tourism. Compared to traditional agriculture, these industries generate lower environmental pollution and may even facilitate ecological restoration and protection through practices like eco-tourism. Furthermore, rural industrial convergence serves as a dynamic conduit for capital mobilization and technology diffusion, catalyzing structural transitions in rural energy matrices through resource recombination mechanisms. This shift reduces reliance on high-pollution energy sources like coal, thereby curbing airborne pollutant emissions. Collectively, the advancement of rural industrial integration fosters air quality improvement and promotes sustainable rural ecological development. Therefore, this paper proposes the following hypothesis:

Hypothesis 1.

Enhancing rural industrial integration improves air quality. Rural industrial integration improves air quality by increasing greening rates.

Industrial integration may amplify rural economic activity and waste generation. However, it also introduces advanced technological and financial resources for waste management. For instance, circular economy industries can establish comprehensive waste recycling and treatment systems, elevating waste treatment rates. Effective waste management reduces air pollution caused by open burning or indiscriminate disposal. Traditional practices like uncontrolled incineration emit hazardous gases and particulates, severely degrading air quality. By improving waste treatment efficiency, such pollution sources are mitigated, directly benefiting air quality. Consequently, rural industrial integration indirectly enhances air quality through improved waste management, elevating rural ecological standards. Therefore, this paper proposes the following hypothesis:

Hypothesis 2.

Rural industrial integration improves air quality by enhancing waste treatment efficiency.

Rural industrial convergence—where farming, processing, and green-energy enterprises cluster within the countryside—creates economies of scale that turn scattered village waste into a shared resource. Centralized bio-digesters, compost hubs, and recycling cooperatives collect crop residues, household organics, and agri-plastics, converting them into biogas for cooking, organic fertilizer for fields, and recycled feedstock for local manufacturers. This closed-loop system eliminates open burning and indiscriminate dumping, cutting airborne emissions while restoring soil health and rural incomes. In this way, integrated rural industries transform waste management from a cost into a driver of cleaner air, carbon savings, and long-term sustainability for the entire countryside. Therefore, this paper proposes the following hypothesis:

Hypothesis 3.

Infrastructure development strengthens the air quality benefits of rural industrial integration.

2.2. Empirical Model

Drawing on the methodological framework of Jie Zhou et al. (2023) [31], we construct the following econometric model:

$$\mathrm{l}\mathrm{n}{\mathrm{A}\mathrm{Q}\mathrm{I}}_{\mathrm{i}\mathrm{t}}={\mathsf{\alpha }}_0+\mathrm{R}\mathrm{I}\mathrm{I}\mathrm{i}\mathrm{t}+\mathsf{\beta }\sum {\mathrm{X}}_{\mathrm{i}\mathrm{t}}+{\mathsf{\mu }}_{\mathrm{i}}+{\mathsf{\lambda }}_{\mathrm{i}}+{\mathsf{\epsilon }}_{\mathrm{i}\mathrm{t}}$$

(1)

lnAQIit denotes the natural logarithm of the Air Quality Index for city i in year t. Higher AQI values indicate more severe air pollution and greater health risks.

RIIit indicates rural industrial integration index for city i in year t. Higher RII values denote stronger rural industrial integration and more advanced rural economies.

Xit denotes a set of control variables, including Industrial Structure (IS), which denotes the ratio of tertiary sector GDP to total GDP; Financial Development Level (FDL); Government Intervention (GOV); Consumer Market Size (CON); and environmental policy (Policy). ${\mathsf{\mu }}_{\mathrm{i}}$ indicates individual characteristics of city i, ${\mathsf{\lambda }}_{\mathrm{i}}$ denotes time fixed effects for city i, and ${\mathsf{\epsilon }}_{\mathrm{i}\mathrm{t}}$ denotes the error in the data for city i in year t.

2.3. Description of Variables

The explanatory variable of this paper is the rural Air Quality Index and the core explanatory variable is the rural industrial integration development index. In this paper, control variables such as Industrial Structure (IS), Financial Development Level (FDL), Government Intervention (GOV), and Consumer Market Size (CON) are added to the econometric model.

2.3.1. Explained Variables

Air Quality Index (AQI) is a non-linear dimensionless index that quantitatively describes the air quality situation, whose value depends on the air quality sub-index of the corresponding area and the corresponding concentration index table of pollution items. The Air Quality Index (AQI) serves as a non-linear dimensionless composite index for quantitatively characterizing atmospheric conditions. Its numerical value is derived through the aggregation of constituent air quality sub-indices, each calibrated to specific pollutant concentration thresholds. AQI operates on a scale where elevated values correlate with intensified atmospheric contamination and heightened public health risks. The computation of AQI integrates multipollutant monitoring data, encompassing fine particulate matter (PM_2.5), inhalable particles (PM₁₀), sulfur dioxide (SO₂), nitrogen dioxide (NO₂), ground-level ozone (O₃), and carbon monoxide (CO). These pollutants are classified into different levels and categories, with each level corresponding to a different color, to facilitate public understanding and appropriate health protection measures (Zhou et al., 2023) [31].

The range of AQI values is usually from 0 to 500, with different value intervals corresponding to different air quality levels and categories, e.g., 0–50 means good air quality, 51–100 means moderate, 101–150 means unhealthy for selected groups, 151–200 means unhealthy, 201–300 means very unhealthy, 301–400 means hazardous, and 401–500 means hazardous (Horn & Dasgupta, 2024) [32]. Long-term exposure to PM_2.5, PM₁₀, and NO₂ was associated with increased blood pressure and hypertension in rural Chinese adults (Li et al., 2020) [33].

2.3.2. Explanatory Variables

The core explanatory variable of this paper is the level of rural industrial integration development (RII). Considering the current status quo of rural industrial integration development covering multiple dimensions and levels, Rural Industrial Chain Extension, Agricultural Technology Penetration, Urban–Rural Income Gap, and Rural Digital Development, Rural Digital Development and Urban–Rural Income Gap are the core explanatory variables of this paper. Urban–Rural Income Gap, Rural Digital Development, and four other aspects of the indicator system construction were applied (

Table 1. Indicators of rural industrial integration variables.

Level 1 Indicators	Level 2 Indicators	Unit	Attribute Value
Rural Industry Chain Extension	Value-added of Primary Industry/Gross Regional Product (RICE1)	%	Positive (+)
	Primary Business Revenue of Agricultural Product Processing Enterprises Above Designated Size (RICE2)	Billions	Positive (+)
Agricultural Technology Penetration	Per Capita Total Agricultural Machinery Power (ATP1)	Kilowatt	Positive (+)
	Agricultural Labor Productivity (ATP2)	CNY/person	Positive (+)
Urban–Rural Income Gap	Ratio of Rural to Urban Disposable Income Per Capita (URIG)	%	Negative (−)
Rural Digital Development	Proportion of Administrative Villages with Broadband Internet Access (RDD)	%	Positive (+)

), and based on this, the entropy method was also applied for measurement. Descriptive statistics are shown in

Table 2. Descriptive statistics of rural industrial integration.

Variable	Obs.	Mean	Std. Dev.	Min	Max	Unit
RICE1	2722	0.122	0.079	0	0.487	%
RICE2	2722	3041.747	1008.883	606.079	7254.422	Billions
ATP1	2722	14,246.96	5117.421	2292.710	40,658.630	Kilowatt
ATP2	2722	3044.233	1008.961	625.389	6943.141	CNY/person
URIG	2722	16.120	5.424	2.981	39.845	%
RDD	2722	0.528	0.096	0.286	0.979	%

Value-added of Primary Industry/Gross Regional Product (RICE1), Primary Business Revenue of Agricultural Product Processing Enterprises Above Designated Size (RICE2), Per Capita Total Agricultural Machinery Power (ATP1), Agricultural Labor Productivity (ATP2), Ratio of Rural to Urban Disposable Income Per Capita (URIG), Proportion of Administrative Villages with Broadband Internet Access (RDD).

.

These indicators comprehensively and scientifically capture the current status and developmental trends of rural industrial integration across multiple dimensions. Their robust rationality provides strong theoretical and empirical support for research and practice in rural industrial integration. The descriptive statistics of the indicators of each dimension are shown in Table 2.

In the operation of the entropy method, it is first necessary to eliminate the influence of variable units and perform normalization processing. For positive indicators:

Firstly, standardized processing is performed.

Positive indicator:

$$\begin{array}{c}{X^{\prime }}_{ij}={\displaystyle \frac{X_{ij} - min\left\{X_j\right\}}{max\left\{X_j\right\}-min\left\{X_j\right\}}}\end{array}$$

(2)

Negative indicator:

$$\begin{array}{c}{X\mathrm{’}}_{ij}={\displaystyle \frac{{max\left\{X_j\right\}-X}_{ij}}{max\left\{X_j\right\}-min\left\{X_j\right\}}}\end{array}$$

(3)

where, ${X^{\prime }}_{ij}$ represents the standardized index value, $X_{ij}$ represents the original value of the j index in sample i, $max\left\{X_j\right\}$ characterizes the maximum value of the j index, and $min\left\{X_j\right\}$ is the minimum value of the j index.

Secondly, calculate the proportion of the j index in sample i. The calculation formula is as follows:

$$\begin{array}{c}{P}_{ij}={\displaystyle \frac{{X^{\prime }}_{ij}}{{\sum }_{i=1}^m{X^{\prime }}_{ij}}}\end{array}$$

(4)

where m represents the sample size and the meanings of the remaining variables are the same as those in Equation (2).

Third, calculate the information entropy of the indicator. The calculation formula is as follows:

$$\begin{array}{c}{e}_i=-k{\displaystyle \sum _{i=1}^m}\left(P_{ij}ln{P}_{ij}\right), k={\displaystyle \frac{1}{ln m}}, 0{\le e}_j\le 1\end{array}$$

(5)

Fourth, calculate the information entropy redundancy. The formula is as follows:

$$\begin{array}{c}{d}_j=1-e_j\end{array}$$

(6)

Fifth, calculate the weights of each indicator. The formula is as follows:

$$\begin{array}{c}{W}_j={\displaystyle \frac{d_j}{{\sum }_{j=1}^n{d}_j}}\end{array}$$

(7)

where n represents the number of indicators in the evaluation index system.

Sixth, calculate the score of the j indicator of sample i. The formula is as follows:

$$\begin{array}{c}{S}_{ij}=w_j{X^{\prime }}_{ij}\end{array}$$

(8)

The level of rural industrial integration was obtained after the measurement, and for the convenience of data analysis, the normalization process was carried out to obtain the level of rural industrial integration in each city from 2014 to 2023, as shown in the following Figure 2:

In this scatter plot, the horizontal axis represents the year and the vertical axis represents the level of rural industrial structure integration, illustrating the distribution across different years.

2.3.3. Control Variables

The following control variables were selected: Industrial Structure (IS), Financial Development Level (FDL), Government Intervention (GOV), and Consumer Market Size (CON). Among them, the change in industrial output value adopts the value added of the secondary industry and takes the logarithm, the rural financial development adopts the ratio of the balance of loans of financial institutions to the GDP at the end of the year, the Government Intervention adopts the ratio of the general budget expenditure of local finance to the GDP, and the industrial structure adopts the ratio of the value added of the tertiary industry to the value added of the secondary industry. And,

Table 3. Descriptive statistics.

Variable	Obs.	Mean	Std. Dev.	Min	Max
lnAQI	2709	4.152	0.293	3.184	5.026
RII	2709	0.415	0.149	0	1
IS	2709	0.458	0.087	0.198	0.805
FDL	2709	1.595	0.722	0	20.096
GOV	2709	0.21	0.106	0.043	1.014
CON	2709	0.398	0.121	0.027	2.990
POLICY	2709	0.182	0.386	0	1

Rural industrial integration (RII), Industrial Structure (IS), Financial Development Level (FDL), Government Intervention (GOV), and Consumer Market Size (CON).

presents the descriptive statistics of the control variables.

2.4. Data Source

Based on data availability, panel data at the prefecture level of 273 cities in mainland China from 2014 to 2023 are used as the sample. The missing data for some years and districts were filled in using linear interpolation. The data were mainly obtained from the China Agricultural Yearbook, the China Statistical Yearbook, and the China Rural Statistical Yearbook, as well as the National Bureau of Statistics (NBS) and local statistical yearbooks.

3. Results

3.1. Benchmark Regression

Based on the previous theoretical analysis and research design, the results of the benchmark regression are shown in

Table 4. Benchmark regression results.

	(1)	(2)	(3)	(4)
VARIABLES	lnAQI	lnAQI	lnAQI	lnAQI
RII	−2.231 ***	−2.198 ***	−1.422 ***	−1.335 ***
	(0.0572)	(0.0567)	(0.0535)	(0.0538)
POLICY		−0.0700 ***	−0.0268 ***	−0.0294 ***
		(0.00941)	(0.00798)	(0.00787)
IS			−1.452 ***	−1.500 ***
			(0.0480)	(0.0477)
FDL			−0.0299 ***	−0.0423 ***
			(0.00464)	(0.00504)
GOV				0.194 ***
				(0.0624)
CON				0.199 ***
				(0.0260)
Constant	6.535 ***	6.577 ***	6.728 ***	6.594 ***
	(0.0634)	(0.0629)	(0.0529)	(0.0543)
City	YES	YES	YES	YES
Year	YES	YES	YES	YES
Observations	2709	2709	2709	2709
R-squared	0.829	0.833	0.884	0.887

Rural industrial integration (RII), Industrial Structure (IS), Financial Development Level (FDL), Government Intervention (GOV), and Consumer Market Size (CON). The values in parentheses are robust standard errors. *** indicates significance at the 1% confidence levels.

. The first and second columns do not consider city or time fixed effects. The third and fourth columns consider urban and time fixed effects. The regression results show that an increase of 0.01 in the level of rural industrial integration at the 1% significance level will result in a decrease of 1.335% in the AQI of the entire prefecture-level city, i.e., the air quality will be improved.

The likely mechanism is that rural industrial integration encourages cleaner production methods, reduces the concentration of high-emission industries in urban cores, and enhances resource utilization efficiency across regions. Furthermore, the spatial spillover effects embedded in such integration—such as labor mobility, supply chain restructuring, and ecological restoration—may also play a critical role in reducing ambient pollution. These findings underscore the importance of cross-regional coordination and integrated development strategies in achieving both economic upgrading and environmental improvement.

3.2. Dynamic Effects Analysis

This study incorporates the one-period lagged air quality into the regression analysis, recognizing that current air quality may be influenced by preceding period conditions. This methodological treatment enhances the accuracy of variable relationship estimation while mitigating omitted variable bias. Specifically, persistent pollutants from prior periods with poorer air quality could systematically affect current measurements. By including lagged air quality as an explanatory variable, the model effectively captures the dynamic temporal dependencies inherent in air quality evolution, thereby improving the ecological validity of causal inference. Results of dynamic effects analysis are shown in

Table 5. Results of dynamic effects analysis.

	(1)	(2)
VARIABLES	lnAQI	lnAQI
L.lnAQI	0.742 ***	0.584 ***
	(0.0133)	(0.0286)
RII	−0.294 ***	−0.334 ***
	(0.0633)	(0.0673)
POLICY		−0.281 ***
		(0.0604)
IS		−0.328 ***
		(0.100)
FDL		0.000503
		(0.00330)
GOV		−0.0555
		(0.0936)
CON		0.0773
		(0.0508)
Constant	1.396 ***	2.600 ***
	(0.111)	(0.196)
City	YES	YES
Year	YES	YES
Observations	2709	2709
Number of id	274	274

Rural industrial integration (RII), Industrial Structure (IS), Financial Development Level (FDL), Government Intervention (GOV), and Consumer Market Size (CON). The values in parentheses are robust standard errors. *** indicates significance at the 1% confidence levels.

.

The inclusion of this lagged variable not only reflects the realistic evolution of air quality over time but also improves the model’s explanatory power by accounting for unobserved dynamic factors. Moreover, after controlling for this temporal autocorrelation, the coefficient of rural industrial integration remains statistically significant and directionally consistent, indicating that its beneficial effect on air quality is robust and not driven by omitted dynamic processes. This strengthens the validity of the causal interpretation and highlights the importance of addressing temporal dependencies in environmental impact evaluations.

3.3. Non-Linear Effects

The impact of rural industrial integration on air quality may exhibit a potential non-linearity; that is, the effect of integration on air quality varies across different stages. Therefore, this paper employs a threshold effects model, and the results are presented in

Table 6. Results of the test of non-linear effects.

VARIABLES	lnAQI
RII ≥ 0.1187	−0.980 ***
	(0.113)
0.1187 < RII ≤ 0.4602	−1.098 ***
	(0.106)
RII ≥ 0.4602	−1.157 ***
	(0.0970)
POLICY	−0.0312 **
	(0.0121)
IS	−1.522 ***
	(0.116)
FDL	−0.0426
	(0.0289)
GOV	0.231
	(0.181)
CON	0.193 *
	(0.105)
Constant	5.256 ***
	(0.0635)
City	YES
Year	YES
Observations	2700
R-squared	0.603

***, **, and * indicate significance at the 1%, 5%, and 10% confidence levels, respectively.

.

The regression results indicate that a higher rural industrial integration index is associated with greater improvements in regional air quality. When the index exceeds 0.4602, the positive effect on air quality reaches its maximum.

3.4. Heterogeneity Test

In this paper, we consider the heterogeneity of the impact of rural industrial integration on rural air quality from two perspectives: regional differences and temporal differences.

3.4.1. Heterogeneity Between Eastern and Western Regions

China’s eastern, central, and western regions manifest pronounced disparities across multiple dimensions, spanning economic development trajectories, industrial composition, infrastructure endowment, and ecological carrying capacities, that collectively define distinctive regional development paradigms. In this paper, the eastern, central, and western regions are tested separately, and the following regression results can be obtained and shown in

Table 7. Results of the test of regional variability.

	Eastern	Central	Western
VARIABLES	lnAQI	lnAQI	lnAQI
RII	−1.375 ***	−1.229 ***	−1.051 ***
	(0.0838)	(0.100)	(0.0970)
POLICY	−0.0247 **	−0.0467 ***	−0.0557 **
	(0.0103)	(0.0135)	(0.0247)
IS	−1.673 ***	−1.579 ***	−1.335 ***
	(0.119)	(0.0744)	(0.0822)
FDL	−0.0655 ***	−0.0185 ***	−0.0773 ***
	(0.0125)	(0.00605)	(0.0236)
GOV	0.512 ***	−0.227 **	0.715 ***
	(0.169)	(0.0970)	(0.104)
CON	0.357 ***	0.0793 **	0.271 ***
	(0.0455)	(0.0388)	(0.0639)
Constant	6.653 ***	5.857 ***	5.200 ***
	(0.0772)	(0.0534)	(0.0633)
City	YES	YES	YES
Year	YES	YES	YES
Observations	967	924	796
R-squared	0.926	0.854	0.858

Rural industrial integration (RII), Industrial Structure (IS), Financial Development Level (FDL), Government Intervention (GOV), and Consumer Market Size (CON). The values in parentheses are robust standard errors. *** and ** indicate significance at the 1% and 5% confidence levels, respectively.

.

The regression outcomes indicate that enhanced rural industrial integration levels in the eastern, central, and western regions exert positive effects on localized air quality amelioration across all three macro-regional divisions. However, in the eastern region, rural industrial integration has the most significant effect on air quality improvement, with the largest regression coefficient, indicating that industrial integration has a more prominent effect on air quality improvement in this region. This spatial disparity may be attributed to the eastern region’s advanced economic development and industrial structural sophistication, which collectively enable the systematic integration of ecological imperatives into rural industrial transitions, thereby accelerating the adoption of green production modalities. At the same time, the eastern region’s well-developed infrastructure and stronger environmental awareness and investment can provide good support conditions for industrial integration, thus reducing pollution emissions and improving air quality more effectively. In contrast, although the central and western regions also show some air quality improvement effects, their improvement effects are relatively weak due to differences in economic development levels, infrastructure, and environmental protection capabilities. Therefore, the eastern region has a stronger advantage in terms of air quality improvement by rural industrial integration.

3.4.2. Hu Huanyong Line

The Hu Line (Hu Huanyong Line), a pivotal geographical demarcation in China stretching from Heihe in Heilongjiang Province to Tengchong in Yunnan Province, reveals stark contrasts in population density and rural production modes between its southeastern and northwestern regions. According to the 2020 Seventh National Population Census, the southeastern side of the Hu Line occupies 43.18% of China’s land area but sustains 93.5% of the national population, while the northwestern side covers 56.82% of the territory with only 6.5% of the population. This divergence is rooted in distinct agricultural systems: the southeastern region, characterized by plains, river networks, and hills, has historically relied on intensive farming as its economic foundation, whereas the northwestern region, dominated by grasslands, deserts, and alpine plateaus, has maintained nomadic pastoralism as its traditional livelihood. To account for these systemic differences, we conducted stratified regression analyses on subsamples from both sides of the Hu Line, yielding the following empirical findings, as shown in

Table 8. Regression results of the Hu Line.

	Southeast Region	Northwest Region
VARIABLES	lnAQI	lnAQI
RII	−1.384 ***	−0.550 ***
	(0.0569)	(0.154)
POLICY	−0.0330 ***	−0.160 ***
	(0.00795)	(0.0385)
IS	−1.547 ***	−0.840 ***
	(0.0503)	(0.154)
FDL	−0.0380 ***	−0.128 ***
	(0.00505)	(0.0315)
GOV	0.157 **	0.295 *
	(0.0662)	(0.174)
CON	0.168 ***	0.631 ***
	(0.0263)	(0.110)
Constant	6.673 ***	4.850 ***
	(0.0550)	(0.102)
City	YES	YES
Year	YES	YES
Observations	2428	259
R-squared	0.896	0.743

Rural industrial integration (RII), Industrial Structure (IS), Financial Development Level (FDL), Government Intervention (GOV), and Consumer Market Size (CON). The values in parentheses are robust standard errors. ***, **, and * indicate significance at the 1%, 5%, and 10% confidence levels, respectively.

.

The empirical analysis reveals that enhanced agricultural industrial integration significantly amplifies air quality improvements in the southeastern region of the Hu Line. As the southeastern region is predominantly engaged in intensive farming, agricultural integration here typically encompasses diversified models such as agro-processing, eco-agriculture, and agritourism. These integrated practices not only elevate agricultural productivity but also reduce pollution emissions from traditional farming by promoting green industries. Furthermore, the region’s stronger environmental awareness and higher investment in ecological governance provide institutional safeguards for industrial integration, synergistically reinforcing its air quality benefits. Consequently, the southeastern Hu Line region demonstrates more pronounced air quality gains from agricultural integration, underscoring the viability of harmonizing economic growth with ecological preservation.

3.4.3. Heterogeneity in Rural Revitalization

In October 2017, the Chinese government proposed the rural revitalization strategy, which has made great changes in the countryside. So, this paper also tested the heterogeneity before and after the implementation of the rural revitalization strategy. The heterogeneous results of rural revitalization are shown in

Table 9. Regression results of the implementation of the rural revitalization strategy.

	Before 2018	After 2018
VARIABLES	lnAQI	lnAQI
RII	−0.296 ***	−0.799 ***
	(0.0919)	(0.0614)
POLICY	1.156 ***	−0.0254
	(0.0921)	(0.0192)
IS	−0.700 ***	−0.488 ***
	(0.122)	(0.0799)
FDL	−0.0256	−0.0311 ***
	(0.0169)	(0.00457)
GOV	−0.389 ***	0.417 ***
	(0.122)	(0.0741)
CON	−0.293 **	0.114 ***
	(0.114)	(0.0275)
Constant	4.476 ***	5.484 ***
	(0.0526)	(0.0824)
City	YES	YES
Year	YES	YES
Observations	1077	1610
R-squared	0.936	0.921

Rural industrial integration (RII), Industrial Structure (IS), Financial Development Level (FDL), Government Intervention (GOV), and Consumer Market Size (CON). The values in parentheses are robust standard errors. *** and ** indicate significance at the 1% and 5% confidence levels, respectively.

.

The results show that before and after the implementation of the rural revitalization strategy, the increased level of rural industrial integration contributes to the improvement of regional air quality, but the effect of this improvement is greater after the implementation of the strategy. The implementation of the rural revitalization strategy provides institutional enablers for industrial convergence, propelling structural optimization of rural production systems. This transition fosters the proliferation of agro-ecological systems, circular agriculture models, and green value chains, thereby advancing the efficacy of rural emission mitigation and atmospheric remediation processes. The strengthening of the rural revitalization strategy in terms of policy support, financial investment, and infrastructure construction has further reinforced the positive effect of rural industrial integration on air quality improvement, reflecting the positive impact of the combination of policy guidance and industrial development on the ecological environment.

3.5. Mechanism Testing

3.5.1. Greening Effect

This paper uses the greening rate of rural areas as a mechanism variable to analyze the role of greening effects in rural areas, the regression result is shown in

Table 10. Regression result of greening effect.

	(1)	(2)
VARIABLES	RGR	lnAQI
RGR		−0.0115 ***
		(0.00182)
RII	17.961 ***	−1.138 ***
	(0.518)	(0.0618)
POLICY		−0.0292 ***
		(0.00780)
IS		−1.484 ***
		(0.0474)
FDL		−0.0410 ***
		(0.00500)
GOV		0.172 ***
		(0.0620)
CON		0.199 ***
		(0.0258)
Constant	2.183 ***	6.619 ***
	(0.574)	(0.0540)
City	YES	YES
Year	YES	YES
Observations	2722	2687
R-squared	0.878	0.889

Regression result of green effect (RGR), rural industrial integration (RII), Industrial Structure (IS), Financial Development Level (FDL), Government Intervention (GOV), and Consumer Market Size (CON). The values in parentheses are robust standard errors. *** indicates significance at the 1% confidence levels.

.

The findings demonstrate that enhanced rural industrial integration significantly improves air quality by increasing greening rates. As rural industrial integration advances, the development of green industries such as eco-agriculture and agritourism drives expanded vegetation coverage through initiatives like planting economic forests, cultivating ornamental flora, and constructing ecological landscapes. These measures markedly elevate rural greening rates. The heightened greening levels not only enhance rural aesthetics but also improve air quality via biological mechanisms: vegetation absorbs CO₂ and releases oxygen through photosynthesis while filtering airborne pollutants such as particulate matter (PM2.5/PM10) and degrading harmful gases like sulfur dioxide (SO₂) and nitrogen oxides (NO_x. Consequently, rural industrial integration fosters an eco-friendly development paradigm where greening serves as a critical mediator in translating economic activities into environmental benefits.

3.5.2. Rural Waste Treatment

This paper uses the number of administrative villages with rural waste treatment as a mechanism variable to analyze the role of rural waste treatment effects. Therefore, the regression result is shown in

Table 11. Regression results for rural waste treatment effects.

	(1)	(2)
VARIABLES	DWD	lnAQI
DWD		−0.0119 ***
		(0.00173)
RII	32.261 ***	−0.965 ***
	(0.547)	(0.0758)
POLICY		−0.0289 ***
		(0.00779)
IS		−1.462 ***
		(0.0476)
FDL		−0.0416 ***
		(0.00499)
GOV		0.144 **
		(0.0622)
CON		0.207 ***
		(0.0257)
Constant	5.014 ***	6.647 ***
	(0.606)	(0.0544)
City	YES	YES
Year	YES	YES
Observations	2722	2687
R-squared	0.955	0.889

Rural industrial integration (RII), Industrial Structure (IS), Financial Development Level (FDL), Government Intervention (GOV), and Consumer Market Size (CON). The values in parentheses are robust standard errors. *** and ** indicate significance at the 1% and 5% confidence levels, respectively.

.

The analysis reveals that enhanced rural industrial integration significantly improves air quality by elevating rural waste treatment rates. With the advancement of industrial integration, waste management infrastructure and systems in rural areas have been progressively upgraded, leading to marked improvements in waste treatment efficiency. Effective disposal of domestic waste reduces practices such as indiscriminate dumping, open-air accumulation, and uncontrolled incineration, thereby mitigating emissions of hazardous pollutants, including dioxins, sulfur dioxide (SO₂), and nitrogen oxides (NO_x), as well as particulate matter. These reductions directly contribute to air quality enhancement. Consequently, rural industrial integration fosters a synergistic mechanism where waste treatment serves as a critical mediator, translating economic activities into environmental benefits. This underscores the viability of aligning economic growth with ecological conservation in rural development.

Under normal circumstances, the combined collection and transport cost of household waste in rural China is about CNY 80 per tonne, and conventional afforestation ranges from CNY 800 to 3000 per mu. Yet the resulting improvements in air quality and in residents’ health are likely to far exceed these outlays. In 2019 alone, urban PM₂.₅ exposure in China was responsible for 711,000 premature deaths and an estimated economic loss of CNY 2.75 trillion (Hu et al., 2022) [34]. Consequently, the economic costs of rural industrial integration—and of the environmental measures it finances—are dwarfed by the benefits they generate, quite apart from the additional gains of narrowing urban–rural disparities and reducing income inequality.

3.5.3. Infrastructure Moderating Effects

To analyze the moderating effects of infrastructure, this study employs interaction terms between rural industrial integration levels and two variables: per capita road area and village road pavement rate. From a rural perspective, roads serve as critical links connecting villages to external regions. Increased per capita road area enhances transportation conditions, reduces agricultural product logistics costs, promotes diversified rural industrial development, and thereby drives industrial integration. Simultaneously, improved road networks enhance waste collection efficiency, mitigating issues such as garbage accumulation and open burning caused by poor transportation accessibility, which contributes to air quality improvement. Furthermore, higher village road pavement rates effectively reduce dust emissions from unpaved roads, decreasing air pollution. These two variables thus robustly capture infrastructure’s moderating role in rural industrial integration and environmental quality enhancement, serving as key indicators for studying the dynamic interplay between rural development and ecological sustainability. The regression results of infrastructure moderating effects are shown in

Table 12. Regression results of infrastructure moderating effects.

		(1)
VARIABLES		lnAQI
RII	−0.226 *	−0.731 ***
	(0.128)	(0.0997)
RII*A	−0.00184 ***
	(0.000193)
RII*B	−0.0342 ***	−0.0192 ***
	(0.00774)	(0.00268)
POLICY	−1.460 ***	−0.0347 ***
	(0.0470)	(0.00782)
IS	−0.0388 ***	−1.472 ***
	(0.00496)	(0.0473)
FDL	0.132 **	−0.0393 ***
	(0.0616)	(0.00500)
GOV	0.185 ***	0.157 **
	(0.0255)	(0.0619)
CON	−0.0342 ***	0.191 ***
	(0.00774)	(0.0257)
Constant	6.607 ***	6.616 ***
	(0.0534)	(0.0538)
City	YES	YES
Year	YES	YES
Observations	2687	2687
R-squared	0.891	0.890

Rural industrial integration (RII), per capita road area in interaction with rural industrial integration (RII*A), village road pavement rate in interaction with rural industrial integration (RII*B), Industrial Structure (IS), Financial Development Level (FDL), Government Intervention (GOV), and Consumer Market Size (CON). The values in parentheses are robust standard errors. ***, **, and * indicate significance at the 1%, 5%, and 10% confidence levels, respectively.

.

The interaction term coefficients share the same negative sign as the regression coefficient of RII, indicating that rural infrastructure development, particularly road construction, strengthens the air quality improvement effects generated by rural industrial integration.

3.6. Endogeneity Concerns

This study may face potential endogeneity concerns. A bidirectional causality could exist between rural industrial integration levels (RIIs) and air quality: while industrial integration may influence air quality, improved air quality might also reciprocally promote further integration. Additionally, omitted variable bias could arise, as factors such as regional economic development levels and policy support may jointly affect both industrial integration and air quality.

To address these issues, we employed China’s Digital Inclusive Finance Pilot Policy as an instrumental variable. This policy is exogenous, as its implementation was determined by central government planning rather than being directly influenced by pre-existing rural industrial integration levels or air quality conditions. Concurrently, the policy indirectly affects rural industrial integration by enhancing financial accessibility and optimizing rural credit environments, thereby establishing a valid exclusion restriction.

The two-stage least-squares (2SLS) regression results are as follows in

Table 13. Regression results of 2SLS.

	(1)	(2)
VARIABLES	RII	lnAQI
DID	0.076 ***
	(0.022)
RII		−3.402 ***
		(1.052)
POLICY		−0.193 ***
		(0.0246)
IS		−0.660 ***
		(0.209)
FDL		0.00657
		(0.0159)
GOV		−0.00544
		(0.109)
CON		0.541 ***
		(0.112)
Constant	0.402 ***	5.564 ***
	(0.00270)	(0.316)
City	YES	YES
Year	YES	YES
Observations	2582	2570
R-squared	0.005	0.837

China’s Digital Inclusive Finance Pilot Policy (DID), rural industrial integration (RII), Industrial Structure (IS), Financial Development Level (FDL), Government Intervention (GOV), and Consumer Market Size (CON). The values in parentheses are robust standard errors. *** indicates significance at the 1% confidence levels.

:

The first-stage F-statistic is 11.87, which passes the weak-instrument test. Because the number of instruments equals the number of endogenous variables, no overidentification test was required. After endogeneity is taken into account, its regression coefficient becomes larger in terms of absolute value, showing a strong improvement in air quality by increasing the level of integration of rural industries.

3.7. Robustness Test

To assess the robustness of the results and to determine which air pollutant is most effectively mitigated by the integration of rural industrial structures, this paper examines the impacts on PM2.5, SO₂, and O₃ separately. The findings indicate that integration exerts the strongest effect on PM2.5: a 0.01-unit increase in the integration index is associated with a 1.942-unit reduction in annual average PM2.5 concentration. The primary mechanisms are the expansion of green areas and the reduction of waste incineration, both of which significantly lower the concentration of particulate matter. Results of robustness test are shown in

Table 14. Robustness test results.

	(1)	(2)	(3)
VARIABLES	PM2.5	${\mathbf{S}\mathbf{O}}_2$	${\mathbf{O}}_3$
RII	−194.2 ***	−3.708 ***	−21.45 *
	(26.98)	(0.409)	(11.74)
POLICY	−6.122	−0.455 ***	−9.891 ***
	(3.894)	(0.0591)	(1.695)
IS	−79.02 ***	−1.286 ***	−53.74 ***
	(24.55)	(0.372)	(10.68)
FDL	−2.198	−0.124 ***	2.582 ***
	(2.232)	(0.0339)	(0.971)
GOV	6.959	0.335	−40.59 ***
	(31.31)	(0.475)	(13.63)
CON	62.43 ***	1.306 ***	−0.488
	(11.93)	(0.181)	(5.194)
Constant	312.2 ***	6.397 ***	26.94 **
	(26.22)	(0.398)	(11.41)
City	YES	YES	YES
Year	YES	YES	YES
Observations	1941	1941	1941
R-squared	0.323	0.449	0.464

The above table shows the results for the test of the impact of the level of rural industrial integration on different air pollutants. ***, **, and * indicate significance at the 1%, 5%, and 10% confidence levels, respectively.

.

4. Discussion

Empirical evidence shows that the improvement of rural industrial integration has a certain promotional effect on air quality. Under the promotion of industrial integration, the rural economy is undergoing a transformation toward diversification and modernization, leading to improvements in the rural energy structure and thereby a reduction in the emission of air pollutants. Additionally, industrial integration promotes an increase in rural greening rates, with the development of green industries, such as ecological agriculture and leisure agriculture, further improving air quality. Meanwhile, the rate of rural household waste treatment has also improved alongside the enhancement of rural industrial integration, contributing to better air quality. Effective waste treatment helps prevent the emission of harmful gases and particulate matter. Finally, the development of rural infrastructure has facilitated industrial integration. Well-developed infrastructure helps address issues like waste accumulation, promotes ecological development, and reduces air pollution through regulatory mechanisms.

Compared to previous studies, this research has made the following improvements: Firstly, it has verified the positive effect of rural industrial integration on air quality improvement, which aligns with previous research by Li et al. (2024) and Guo (2024) [35,36]. Secondly, it has innovatively proposed the moderating effects of greening and rural waste treatment efficiency on the impact of rural industrial integration on air quality.

5. Conclusions and Policy Recommendations

5.1. Conclusions

This study synthesizes existing research findings and clarifies the mechanisms of rural industrial integration, examining its impact on air quality in prefecture-level cities across China. A quantitative analysis was conducted using the entropy method to assess the relationship between rural industrial integration and air quality. The study draws the following conclusions: First, the level of rural industrial integration in China’s prefecture-level cities has steadily increased, and rural air quality has been continuously improved. Second, the heterogeneity analysis found that the improvement in rural air quality was more pronounced in the eastern–central region of Chinese prefectures, on the southeast side of the Hu Huanyong Line, and after the implementation of the rural revitalization strategy. Third, the greening effect exerts a positive moderating effect on the relationship between rural industrial convergence and air quality improvement, while rural waste management efficiency demonstrates a comparable regulatory function. Simultaneously, infrastructure development serves as a critical mediating channel through which industrial integration initiatives translate into measurable atmospheric remediation outcomes.

This study also has certain limitations. Due to data availability constraints, the analysis was conducted at the city level. Future research could explore smaller spatial units or conduct more detailed policy evaluations. In addition, although this study addresses some aspects of heterogeneity, there remains room for further improvement in analyzing specific cities in greater depth.

5.2. Policy Recommendations

Based on the findings, this study recommends promoting differentiated rural industrial integration, with eastern provinces focusing on green, high-value-added sectors (e.g., eco-tourism and digital agriculture), led by provincial DRCs and agricultural departments, by 2026. The Ministry of Ecology and Environment should establish a cross-regional ecological compensation mechanism by 2026 to support green transformation in central and western regions. Simultaneously, provincial forestry bureaus should promote “forest economy” models by 2027, while local governments should upgrade waste management systems under MOHURD’ s guidelines. County-level transportation departments should integrate green corridors and drainage planning by 2026, and rural sewage treatment should be improved with support from water and environmental agencies by 2027.