Air Pollution and Agricultural Economic Resilience in China: The Moderating Role of Environmental Regulation

Abstract

Sustainable agricultural development in China in the face of growing environmental concerns relies critically on how well regulatory policies strengthen agricultural resilience. This study aims to systematically investigate the impact of air pollution on agricultural economic resilience and its mechanisms of action and to explicitly assess the moderating role of environmental regulation. This study develops a thorough index system that evaluates agricultural economic resilience in three areas: risk resistance and recovery, adaptive adjustment capacity, and restructuring innovation. Panel data from 30 Chinese provinces from 2000 to 2023 is used to achieve this. The implications of air pollution and its diverse consequences on agricultural economic resilience are systematically assessed using a two-way fixed-effects and moderating-effects model. The following are the primary conclusions: First, air pollution has a significant negative impact on the economic resilience of agriculture. This conclusion holds after considering the endogeneity problem and a series of robustness tests, such as the exclusion of samples, random sampling, and quantile regression. Second, different dimensions of agricultural economic resilience, intensity levels, and economic growth phases influence how much air pollution reduces agricultural economic resilience. Notably, at various stages of economic growth, air pollution steadily weakens the economic resilience of agriculture. In particular, the impact is more pronounced in the post-financial-crisis phase of domestic demand expansion and the phase of financial clearing and high-quality development. According to a dimensional perspective, air pollution significantly reduces the farm sector’s capacity to endure and recover from dangers while also making adaptive modifications easier, and the impact on transformational innovation is not significant. In terms of intensity, in contrast to places with higher resilience, those with lower resilience are disproportionately more adversely affected by air pollution. Third, environmental control mitigates some of the detrimental effects of air pollution on agricultural economic resilience. Based on these results, this study calls for stricter air pollution control measures, strengthens environmental regulatory support for agricultural resilience, and demonstrates region-specific governance solutions to guarantee the stability and sustainability of the agricultural economic framework.

1. Introduction

In the past four decades, China’s economy has expanded rapidly and continuously. However, environmental contamination caused by China’s reckless development pattern is now the most significant obstacle to the country’s superior economic growth. Of the 339 cities in the People’s Republic of China that were at the prefecture level and above in 2023, 203 (or 59.9% of the total) met ecological air quality standards, based on the Ministry of Ecology and Environment’s 2023 National Ecological Quality Profile. The typical annual PM2.5 concentration of 30 µg/m³ remains significantly higher than the World Health Organization (WHO)’s standard value of less than 10 µg/m³. NASA’s satellite monitoring program shows that since 2000, China’s average PM2.5 concentration has been significantly higher than the World Health Organization’s global air quality standards. One of the main challenges to China’s sustainable economic and social development is air pollution, an international environmental issue.

The agriculture sector is the foundation of the nation’s economy, and its continuous expansion is crucial to increasing the economy’s endogenous momentum. It is closely linked to the continuous growth and progress of the country’s economy. China is a carbon emitter and a major agricultural country. On the one hand, due to a high base and long duration of air pollution concentrations, air pollution reduces agricultural productivity and crop production, which in turn hinders food security [1]. Jeopardizing agricultural systems’ ability to withstand crises and recover from disasters creates instability in agricultural economic systems. On the other hand, agriculture plays a pivotal role in the overall economic performance, and the complexity of the structure of the agricultural system and the uneven regional development have led to significant differences in the adaptive capacity to external shocks. In recent years, changes in the international trade environment, especially in the context of the trade frictions between China and the United States, have resulted in the imposition of punitive tariffs on some Chinese agricultural products, which have directly pushed up the cost of agricultural production and reduced the price advantage of the domestic agricultural products, leading to a contraction of the export market. Export trade is being impacted by higher tariff policies through direct cost transmission and market reconfiguration, which is altering the composition of global supply chains. Moreover, frequently adjusted tariff policies in the context of trade frictions increase market uncertainty, which amplifies long-term risks and inhibits investment confidence, with firms tending to delay entry into new markets or reduce the size of their exports, increasing sunk costs. Tariffs have raised domestic prices, impacted agricultural consumer markets, and decreased the economy’s agricultural economic resilience, even though supply chain diversification, technological innovation, and the efficacy of policy collaborations are factors that affect agricultural economic resilience.

The primary objectives of promoting sustainable agricultural growth are to lessen the effects of uncertainty shocks on the agricultural system, improve the stability of the agricultural economy, and swiftly recover and rebuild in response to external environmental risks and changes. This objective is intimately tied to the concept of “resilience”. China’s agricultural economic development will reach new heights if it can transform from a “significant agricultural nation” to an “agricultural powerhouse” and increase agricultural economic resilience. This will assist the country’s economy in growing steadily and sustainably. It has been pointed out that environmental regulation plays a regulating role in the process of air pollution, inhibiting agricultural product export trade and reducing the negative impact of air pollution on agricultural product export [2]. The government has created a number of environmental initiatives and policies to address the issue. The State Council’s 2013 Action Plan for Preventing and Controlling Air Pollution included ten phases (Atmosphere 10). This is China’s strictest environmental control action plan. However, reducing air pollution requires a systematic and sustained effort. The general situation is still worrisome, even though the air quality has dramatically improved. Despite a 37% drop in the nation’s PM_2.5 concentration, 180 million acres of lightly polluted arable land still requires treatment, according to the National Soil Pollution Survey Bulletin 2014. In this context, a deeper understanding of the impact of air pollution on the resilience of the agricultural economy is crucial from both an academic and a practical perspective. Thus, from an academic and practical perspective, a comprehensive understanding of how air pollution impacts the resilience of the agricultural economy, the effective implementation of environmental regulations, and the promotion of green transformation and agricultural modernization are all extremely valuable.

1.1. Literature Review

Physics first defined resilience as the capacity of something to return to its initial form after having been deformed by external impacts. According to continuing research, resilience is the ability of an ecosystem to tolerate environmental damage and recover [3]. This concept is the foundation for research on economic resilience in the agriculture sector. Four perspectives on economic resilience are presented by Martin (2015): the ability to innovate and reform, the ability to reorganize, the ability to recover to a previous state, and the ability to cope with stress [4]. In examining ecological and social systems, Folke also proposed the concept of resilience, contending that it includes not only the capacity to endure external shocks but also the capacity of systems and their functions to persist and adapt. His theories have found widespread application in the realm of agricultural economic resilience [5]. Agricultural economic resilience frequently captures the growth and stability of local agricultural economies [6]. The overall ability of an agricultural economic system to tolerate, bounce back from, and adjust to external shocks and pressures is known as agricultural economic resilience. To quantify agricultural economic resilience, researchers use the economic resilience method, which primarily uses the indicator and core variable approaches. There has been no agreement on the investigation of indicators to gauge the degree of agricultural economic resilience due to the intricacy of the composition involving various relationships and topics associated with this concept. This article creates a thorough indicator framework for agricultural economic resilience that takes into account three factors: innovation and reconstruction, adaptation to change, and risk resistance. The entropy weight method measures the tiers of agricultural economics resilience. Concerning the factors that influence agricultural economics resilience, scholars have noted that rural industrial integration [7], agricultural technological innovation [8], and the digital economy [9] can contribute to enhancing agricultural economic resilience. Certain scholars have also argued that the effectiveness of fiscal measures to support agriculture exerts a moderating effect on increasing agricultural economic resilience and can successfully mitigate how climate change affects this resilience [10].

There have been two primary areas of study in the past about how air pollution affects the agricultural economy. The first is the effect on the environment where crops are grown. Particulate matter (PM_2.5 and PM₁₀), O₃ [11], NO [12], acid rain [13], and other air pollutants undergo biochemical reactions by changing their physiological forms and soils, etc., which alter crop growing conditions and inhibit crop development, thereby lowering agricultural product yields. Chen discovered that agriculture’s total factor productivity, input use, and yields can all suffer significant losses and decreases due to short-term warming. However, adaptation strategies like labor inputs, fertilizer application, and irrigation are greatly reduced with time [14]. Second is the impact on labor. The work currently accessible on the representation of labor impacts focuses on three main topics: the labor force, population consumption, and population health. First, local workers are harmed by air pollution [15], and the area loses some of its labor capital due to its diminished attractiveness to workers. According to Lewis’s turning point, the agricultural sector has not yet completed the appropriate substitution of capital for labor in the setting of the disparity between urban and rural areas. This has led to a discrepancy between labor force declines and productivity gains, which has reduced regional productivity and negatively impacted agricultural development and output. Air pollution generally reduces the efficiency of output from agriculture. Second, the negative effects of pollution in the air on the labor supply or productivity will directly affect household earnings [16]. According to the Keynesian theory of the consumption function, air pollution will affect consumption costs by lowering income because income is the main factor influencing consumption. Furthermore, Wang and Zheng argue that air pollution reduces the frequency of travel and consumption [17]. These factors could reduce rural residents’ purchasing power by lowering their consumption prices. Because there would be less demand for consumption, the food supply chain might “reverse contract,” and the synergistic effect of the agricultural economy would be diminished. Thirdly, pollution in the air not only impairs public health but also dramatically increases mortality rates and the danger of obesity, cardiovascular disease, respiratory diseases, and other ailments [18]. It might also have a direct effect on the residents’ mental health. Because air pollution had an emotional influence, it affects the residents’ subjective well-being and mental health [19]. Both agricultural productivity and the efficiency of the rural labor force are impacted by the population’s physical and psychological well-being, which is linked to the productivity and quality of the rural labor supply.

The impact of air pollution on the economic growth of agriculture is a major worry for scholars both domestically and abroad. Based on the existing research, we discovered the following limitations: First, as resilience research advances, resilience is becoming more widely acknowledged as a crucial topic in economics. While previous studies have provided helpful information, some areas still need further research. It is essential to keep in mind that, compared to the broader concept of economic resilience, agricultural economic resilience is still in its infancy. The majority of the currently published literature addresses agricultural economic resilience from viewpoints such as agroecology, and the main study results are still centered on other nations. The national framework for studying agricultural economic resilience is insufficiently comprehensive. Very little research fully explores the fundamental elements driving agricultural economic resilience, and the study’s data coverage year is behind schedule. This makes accurately comprehending the dynamics of real-time research challenging. Second, agricultural economic resilience is measured by the academic community using a variety of techniques, but no universal framework or agreement has been reached. Third, it is better known how air pollution influences agricultural economic resilience than how climate change affects agricultural production. Accordingly, this study explores the mechanisms that underlie the connection between air pollution and the economic resilience of agriculture. It provides some contribution to three crucial areas: (1) Based on the existing literature, a system of indicators of agricultural economic resilience is constructed and improved in three dimensions, covering long-term panel data for 30 provinces across the country for the period 2000–2023. And from the perspective of the comprehensive indicator system of air pollution on the agricultural economic resilience, through in-depth analyses of its impact pathways, it expands the research horizons in this field, further improves the theoretical mechanism, and provides a richer and more systematic theoretical foundation for the subsequent research. (2) The level of agricultural economic resilience in each region of China is quantitatively measured, and a ranking of the agricultural economic resilience index for the whole country and different regions is constructed, exploring the stage of the economic development of agricultural economic resilience, different dimensions of agricultural economic resilience, and the heterogeneity of the intensity of agricultural economic resilience, which provides a more comprehensive empirical test. (3) A new perspective on environmental control is provided by clarifying the rational connection between air pollution and agricultural economic resilience. We examine the relationship between agricultural economic resilience and environmental regulation using the moderating-effects model, providing a new policy perspective.

Direct academic inquiry on the association between air pollution and agricultural economic resilience is scarce. This study looks at agricultural economic resilience from three angles: risk-resistant resilience, adaptive adjustment ability, and transformation and innovation ability. This is accomplished in order to develop a thorough indicator methodology. The total value of agricultural economic resilience is then calculated using the entropy weight method. By researching the variances in agricultural economic resilience and their impact on air pollution throughout the country’s provinces, this study aims to respond to the following three queries: (1) What is the impact of air pollution on the economic resilience of agriculture? (2) What distinct effects does air pollution have on agricultural economic resilience in terms of different stages of economic development, different dimensions of agricultural economic resilience, and the intensity of agricultural economic resilience? (3) The agricultural economic resilience is assessed, and the moderating impact of environmental laws is ascertained. Taking into consideration the previous discussion, the remainder of this study is structured as follows: Section 1 presents the literature review, theoretical analysis and research hypotheses. Section 2, Materials and Methods, provides a brief overview of the variables selected and discussed, the econometric model setup, and the data sources. Section 3 of this research presents the analysis of the empirical results. Section 4 is the Discussion and Section 5 summarizes the main conclusions and related policy recommendations.

1.2. Theoretical Analysis and Research Hypotheses

This study traces the relationship between risk resistance, recovery, adaptive adjustment, innovation capacity reconstruction, and agricultural economic resilience. It finds that the following routes are the main ways that air pollution impacts agricultural economic resilience. The allocation of agricultural production components is hampered by air pollution, which is detrimental to reaching the goal of environmental sustainability and raising productivity in agriculture. The factors of agricultural production include four main areas: land, labor, capital, and means of production. Soil acidification caused by the deposition of air pollutants leads to a decline in soil quality; as a result, farmers may need to invest more in soil improvement, which increases capital requirements and reduces the efficiency of capital utilization. Additionally, air pollution impacts labor. It reduces the productivity of labor factors by affecting health conditions [20]. Declining labor productivity affects the allocative and technical efficiency of agricultural productive factors, which may ultimately result in a decline in agricultural production and economic efficiency. The decline in economic efficiency weakens capital accumulation, squeezes resources for resistance and recovery, and reduces adaptive adjustment capacity. On the other hand, the structure of agricultural output may be jeopardized by air pollution. In order to accommodate crop development, farmers may be forced by air pollution to change the kinds of crops they plant as well as their agricultural production methods and techniques [21]. In order to transition to a sustainable pattern of agricultural development, they have also modified management techniques in the agricultural sector. Ecosystems are seriously threatened by air pollution and abnormally high particulate matter concentrations [22]. The agricultural production environment is deteriorating, resource absorption capacity is compromised, restructuring and innovative capacity are impeded, and ecosystems are threatened by air pollutants, including extremely high concentrations of particulate matter, which also increase the risk of disease and crop pests and enhance the probability of extreme weather events. The development of agricultural economic resilience and increased agricultural productivity is not supported by it. In conclusion, environmental deterioration and air pollution will affect the agricultural economy’s output level by impeding the allocation of production elements, raising the price of agricultural production, and preventing the modernization and transformation of production structures. This makes agricultural economic resilience even more vulnerable, impairing its capacity to withstand and bounce back from risk, adjust to change, and rebuild and innovate. The research hypotheses were as follows:

H1: 

Air pollution is detrimental to agricultural economic resilience.

The diversity in agricultural economic resilience is influenced by the number of economic development phases, and there are also variances in the distinctive features and extent of agricultural economic resilience. As a result, while researching how air pollution causes agricultural economic resilience, heterogeneity cannot be disregarded, indicating that the effects of air pollution on agricultural economic resilience vary. First, we consider the economic development stage. With China’s accession to the WTO in 2001, the most prominent feature of the macroeconomic operation of the national economy is the closer connection with the world economy. At the same time, trade liberalization will also increase China’s carbon emission intensity through the scale effect and structural effect [23]. During this stage, air pollution may be somewhat impacted. China’s economy entered an era of dependence on domestic demand after the financial crisis in 2008 [24]. Increased air pollution issues were a result of the growth of heavy industry and building, especially following the crisis. The Chinese economy entered a phase of superior expansion in 2017, and the government tightened its environmental laws. Environmental contamination has decreased to a certain degree [25]. As a result, the level of air pollution in China changes according to different stages of economic development, and it is crucial to investigate how air pollution affects agricultural economic resilience at different points in time. The second consideration is the diversity of the aspects and levels of agricultural economic resilience. Air pollution affects agriculture differently in different locations. Communities with a greater concentration of industry are likely to be more seriously affected by air pollution. On the other hand, locations that are primarily agricultural or have more excellent natural surroundings are probably less impacted. Region-specific adaptive adjustment capacities and agricultural recovery resistance to risk may therefore differ. However, air pollution and carbon emissions encourage creativity and propel the growth of green technologies in urban areas [26]. There are typically more resources available for revolutionary agricultural innovation in areas with stronger economic development. Through technical measures that impact agriculture’s capacity to adapt and rebuild, they can lessen the consequences of air pollution. In light of this, this paper introduces the following theory:

H2: 

Heterogeneity exists in the impact of air pollution on agricultural economic resilience.

The preservation of agricultural economic resilience and sustainable agricultural development depends heavily on environmental regulation, a mechanism created by governments to safeguard the environment, lessen pollution, and encourage the prudent use of resources. Two opposing theories about the moderating function of environmental regulation can be found in the literature. Effective environmental regulations, according to the Porter hypothesis, can incentivize businesses to innovate, thereby offsetting some or all of the costs associated with regulatory compliance. This may end up in a situation that benefits the economy and the environment [27]. As a means to address environmental degradation and reduce hazards to the climate and secure the long-term prosperity of the nation, China, a signatory to the Paris Agreement, has enacted substantial policy reforms in the areas of industry, transportation, urban planning, agriculture, and energy production. By 2060, China wants to achieve carbon neutrality. By lowering greenhouse gas emissions from agriculture, optimizing energy consumption, and encouraging resource efficiency, these programs’ main goal is to lessen the burden on the environment. By reducing air pollution emissions, enhancing the environment for agricultural output, and boosting the agrarian economy’s ability to respond, environmental regulation improves agricultural economic resilience. However, environmental legislation requires organizations engaged in pollution-related activities to look for technical advances that can minimize pollution [26]. Stricter environmental laws, however, would raise the cost of compliance and impede the expansion of green production in agriculture, according to some studies [28]. Instead of increasing agricultural efficiency, this scenario will lead to an additional fee for inputs in agricultural development, which could worsen agrarian pollution in nearby and local places [29]. As a result, less money would be available to invest in environmentally friendly technical advancements [30]. Air pollution is generally less affected by the environmental regulation of carbon emissions than green technological progress [31]. It is still unclear how environmental laws pertaining to air pollution affect agriculture’s ability to survive economically. Therefore, the following theories are put forth:

H3a: 

Environmental regulatory policies are beneficial in mitigating air pollution affecting the economic resilience of agriculture.

H3b: 

Environmental regulatory policies do not help to reduce air pollution affecting the economic resilience of agriculture.

2. Materials and Methods

2.1. Model Construction

To analyze how air pollution affects the economic resilience of agriculture, this paper sets up the regression model as follows:

$${AGRES}_{it}=\alpha +\beta {PM}_{it}+\gamma X_{it}+{\theta }_i+{\delta }_t+{\epsilon }_{it}$$

(1)

${AGRES}_{it}$ denotes the index of agricultural economic resilience in the region; ${PM}_{it}$ denotes the degree of air pollution in the area; the coefficient to be estimated $\beta$ represents the influence of air pollution on agricultural economic resilience; $X_{it}$ denotes a set of control variables including the GDP per capita, infrastructure level, degree of foreign trade, rate of environmentally sound treatment of domestic wastes, financial commitment to the entire society’s fixed assets, and the level of green innovation, etc.; ${\theta }_i$ and ${\delta }_t$ denote the region and the year fixed effects, respectively; and ${\epsilon }_{it}$ is the error term.

Testing the mitigating role of environmental regulations in the agricultural economic resilience of air pollution, this paper further constructs a moderating-effects model based on the baseline regression model as follows:

$${AGRES}_{it}=\alpha +\beta {PM}_{it}+{\phi }_1{\mathrm{e}\mathrm{r}}_{it}+{\phi }_2{PM}_{it}\times {\mathrm{e}\mathrm{r}}_{it}+\gamma X_{it}+{\theta }_i+{\delta }_t+{\epsilon }_{it}$$

(2)

${AGRES}_{it}$ represents the index of regional agricultural economic resilience, ${PM}_{it}$ represents the extent of air pollution in the area, ${er}_{it}$ represents the level of environmental regulation in the province and year; ${PM}_{it}\times {\mathrm{e}\mathrm{r}}_{it}$ represents the interplay between environmental regulation and air pollution, and ${\phi }_2$ reflects the moderating role of environmental regulation in how air pollution affects agricultural economic resilience. If ${\phi }_2$ has the same sign as $\beta$, it means that environmental regulation positively moderates (exacerbates) the impact of air pollution on agricultural economic resilience; conversely, it means that environmental regulation negatively moderates (mitigates) the impact of air pollution on agricultural economic resilience.

2.2. Variable Definitions

2.2.1. Dependent Variable: Agricultural Economic Resilience (AGRES)

This study makes reference to the work of Hao and Tan [32], who systematically built a comprehensive evaluation system for agricultural economic resilience and measured it using the coefficient of entropy weight method. They utilize the capability to resist and recover from risk, adaptive adjustment, reconstruction, and innovation as benchmarks for agricultural economic resilience. On the basis of this basic literature, we improve and refine the comprehensive indicator system of agricultural economic resilience and enrich and innovate the construction of the indicator system. In addition, we construct a serial ranking of agricultural economic resilience indices for the whole country and different regions, which enriches the empirical expansion of the existing resilience studies at the time-series and spatial levels. As shown in

Table 1. System of indicators of agricultural economic resilience.

Goal Level	Indicator Layer	Indicator Interpretation	Attributes	Weight
Resistance and recovery to risk	Rural consumption capacity	Consumer price index for rural residents	Positive	0.006
	Rural household Engel’s coefficient	Rural household food consumption index	Opposite	0.006
	Effective irrigation rate	Effective irrigated area in cultivated area	Positive	0.075
	Total power of agricultural machinery per acre	Total power of agricultural machinery/cultivated land area	Positive	0.237
	Total area of crops planted	Total sown area of crops (thousand hectares)	Positive	0.053
	Rural education level	Rural per capita years of education	Positive	0.076
	Agricultural fertilizer application	Discounted amount of agricultural fertilizer applied	Positive	0.089
Adaptive adjustment	Value-added index of primary industry	GDP growth rate of the primary industry	Positive	0.004
	Rural delivery routes	Rural delivery routes	Positive	0.049
	Consumption expenditure of rural residents	Rural per capita consumption expenditure	Positive	0.080
Reconstruction and innovation	Strength of financial support for agriculture	Expenditures of local finance on agriculture, forestry, and water affairs	Positive	0.099
	Stock of human capital	Number of undergraduate and specialized students in higher education institutions	Positive	0.063
	Rural electricity consumption	Electricity consumption of rural residents	Positive	0.163

, the ability to resist and recover from risk means the capacity of an agricultural economic system to uphold stability despite external pressures and perturbations, as expressed in its ability to resist, both in terms of productive base and ecological conditions. Adaptive adjustment capacity refers to the ability of the agricultural economic system to rapidly adapt and quickly recover to the pre-disturbance status after encountering external perturbations, which manifests as the regulating ability of the agricultural economy; reconstruction and innovation capacity refer to the ability of the agricultural economic system to achieve self-improvement and upgrade after encountering external disruption or even reach leapfrog development, which shows the upgrading capacity of the agricultural economic system.

2.2.2. Core Independent Variable: Air Pollution (PM)

The explanatory variable chosen to describe the level of air pollution was the yearly average concentration of PM_2.5, or tiny particulate matter. Air pollutants come in a variety of forms, such as CO₂, SO₂, NO_x, haze, and soot. Haze is the main contributor to air pollution in China. Haze, of which PM_2.5 is the primary component, has long had a major influence on China’s social and economic growth. PM_2.5 is thought to be the most indicative of total air pollution levels and is the carrier and collector of a number of dangerous compounds. Based on the State Environmental Protection Department’s official Ambient Air Quality Index (AQI) Regulations (Trial) technique, this document computes the air pollution index values using the yearly average concentration of PM_2.5, the main pollutant that causes smog. We used Chen’s suggested research methodology to guarantee data continuity for the data resources because the PM_2.5 concentration data for China’s main cities only started to be released in 2013, and some data from 2000 to 2012 are significantly missing for the PM_2.5 concentration reference. The data provided are based on global PM_2.5 raster data published by the Atmospheric Composition Analysis Group at Dalhousie University in Canada and annual averages of global PM_2.5 concentrations measured by satellites published by the Center for Socio-Economic Data and Applications at Columbia University in the United States. These data have been extensively utilized in research on air pollution in China, among other areas. The PM_2.5 data for every Chinese province were analyzed using ArcGIS software (ArcGIS 10.8), and linear interpolation was used to add to the data.

2.2.3. Control Variables

Citing previous research according to Dong, D., and Wang, J. [33], Hering, L. [34], and others, several provincial-level control factors that may affect air pollution levels and the resilience of the agricultural industry are selected in this study: The integrity of the supply chain for agriculture is maintained, and the gross domestic product per person (pergdp), which is determined by the GDP per capita, represents the area of the economy. The degree of infrastructure (infra), which is determined by dividing the administrative division’s area by the road length, indicates how effectively resources are allocated and regulates the adaptability of the agricultural system. After adjusting for its nonlinear impact on the agricultural economic resilience, the degree of open foreign trade, as indicated by the proportion of regional GDP to overall exports and imports, indicates how dependent and diversified the agricultural economy is on external markets. The rate of the sound treatment of household garbage (trash), which is calculated by dividing the total quantity of domestic waste created by the total amount of soundly treated domestic waste, affects the carrying capacity threshold and steady-state resilience of the agroecosystem. An indicator of the level of capital investment in agriculture that accounts for its path-dependent impacts on agricultural resilience is societal investment in fixed assets (INVEST), which is calculated as the total amount of societal investment in fixed assets. The role of technological progress in the agricultural economy is indicated by the extent of green innovation (inno), as evidenced by the total green patents granted.

2.3. Data Sources and Descriptive Statistics

The National Bureau of Statistics’ China Statistical Yearbook, each province’s statistical yearbook, and the Ministry of Ecology and Environment of China’s China Environmental Statistical Yearbook—all regarded as reliable sources—provided the provincial data used in this work. Prior to conducting empirical analyses, this study performs the following data cleaning and screening steps to guarantee the accuracy of the data. First, it deals with high values and outliers and eliminates samples that lack essential factors. Second, control variables are processed logarithmically. Lastly, linear interpolation is used to add the missing data.

Table 2. Variable meanings and descriptive statistics.

	Variable	Description of Indicators	Obs	Mean	Std. Dev.	Min	Max
Dependent variable	agres	Gross agricultural economic resilience	720	0.183	0.115	0.017	0.534
Explanatory variable	PM	Air pollution levels	720	37.752	15.235	4.392	85.653
Control variables	pergdp	GDP per capita	720	10.276	0.890	7.923	12.168
	infra	Level of infrastructure	720	−0.525	0.850	−3.481	0.830
	open	Level of foreign trade	720	−1.345	1.307	−4.605	2.551
	trash	Harmless treatment rate of domestic waste	720	4.277	0.448	2.911	4.605
	invest	Social fixed-asset investment	720	8.748	1.354	5.018	11.225
	inno	Level of green innovation	720	6.265	2.034	0.010	10.722
Moderating variable	er	Environmental regulation intensity	720	33.027	32.488	1.035	156.718

presents the descriptive statistics for the primary variables in this study.

2.4. Analysis of Trends Regarding the Resilience of the Agricultural Economy Across All Regions of the Country

This study examines national and regional trends and variations in the field of agricultural economic resilience. It makes use of panel data from 2000 to 2023 from 30 Chinese provinces. Each province’s agricultural economic resilience index was calculated using the entropy weight method and then averaged to obtain each year’s national agricultural economic resilience index. Due to space constraints,

Table 3. Regional rankings of agricultural economic resilience at different stages.

Year		2000–2004			2009–2014			2019–2023
Region	Province	Index Score	National Ranking	Regional Rank	Index Score	National Ranking	Regional Rank	Index Score	National Ranking	Regional Rank
Eastern Region	Shandong	0.2161	1	1	0.3803	2	2	0.4734	2	1
	Guangdong	0.1725	4	3	0.3338	4	3	0.4280	4	3
	Jiangsu	0.2112	2	2	0.4408	1	1	0.4515	3	2
	Zhejiang	0.1176	10	5	0.2504	8	5	0.3149	11	5
	Fujian	0.0774	16	6	0.1595	17	6	0.2262	17	6
	Hebei	0.1676	5	4	0.3023	5	4	0.3787	7	4
	Shanghai	0.0470	25	7	0.1133	25	7	0.1527	26	7
	Beijing	0.0464	26	8	0.0866	27	8	0.1332	27	8
	Hainan	0.0214	29	10	0.0558	29	10	0.1072	28	9
	Tianjin	0.0342	27	9	0.0712	28	9	0.0936	29	10
Central Region	Henan	0.1916	3	1	0.3578	3	1	0.4848	1	1
	Hubei	0.1323	7	2	0.2513	7	2	0.3611	8	2
	Anhui	0.1272	9	4	0.2327	10	4	0.3502	10	4
	Hunan	0.1317	8	3	0.2402	9	3	0.3577	9	3
	Jiangxi	0.0861	13	5	0.1613	16	5	0.2620	16	5
	Shanxi	0.0683	19	6	0.1355	22	6	0.1920	23	6
Western Region	Inner Mongolia	0.0840	14	4	0.1702	13	4	0.2633	15	5
	Sichuan	0.1357	6	1	0.2646	6	1	0.4152	5	1
	Guangxi	0.0819	15	5	0.1628	15	5	0.2675	14	4
	Xinjiang	0.0924	12	3	0.1741	12	3	0.3003	12	2
	Yunnan	0.0966	11	2	0.1824	11	2	0.2991	13	3
	Shaanxi	0.0652	21	7	0.1429	20	7	0.2192	18	6
	Chongqing	0.0675	20	6	0.1457	19	6	0.1819	24	9
	Gansu	0.0566	23	8	0.1199	23	8	0.1948	20	7
	Guizhou	0.0307	28	10	0.0943	26	10	0.1944	21	8
	Ningxia	0.0192	30	11	0.0509	30	11	0.0862	30	11
	Qinghai	0.0499	24	9	0.1154	24	9	0.1590	25	10
Northeast Region	Heilongjiang	0.0614	22	3	0.1648	14	1	0.3811	6	1
	Liaoning	0.0683	18	2	0.1539	18	2	0.2154	19	2
	Jilin	0.0722	17	1	0.1408	21	3	0.1935	22	3

displays the economic resilience scores and rankings of agricultural provinces in the eastern, central, western, and northeastern parts of the country across three stages of economic development: 2000–2004, 2009–2014, and 2019–2023. A line graph showing the resilience tendency of the regional and national agricultural economies throughout time is shown in Figure 1.

Table 3 illustrates the notable regional variations in China’s agricultural economic resilience. Particularly noteworthy is the agricultural and economic resilience of coastal and more economically developed areas like Jiangsu, Guangdong, and Shandong. It is intimately tied to these provinces’ technological application and degree of economic development. Increasing the resilience of agricultural economies is more difficult in some areas, including Ningxia and Hainan, which are comparatively resource-poor or economically underdeveloped. In order to comprehend the variations in resilience among different regions, this study analyzes and contrasts the strength of the agricultural economy in each province. It also identifies the major elements that influence the strength of the agricultural economy. Future policy and practice must strengthen the agricultural economy’s resilience to risks in underdeveloped areas, particularly in light of unequal development. Low-resilience regions can learn a lot from the successful experiences of high-resilience provinces. However, more scientific studies and policy support are needed to effectively solve the issues in the low-resilience sectors and help strengthen the agricultural industry’s resilience in the country.

Figure 1. National trend in agricultural economic resilience from 2000 to 2023.

Figure 1. National trend in agricultural economic resilience from 2000 to 2023.

The analysis of Figure 1 shows that, overall, the national agricultural economy’s resilience index has shown a gradual increase. Specifically, the resilience index grew gradually from 2000 to 2018. After entering the WTO, the Chinese economy was driven by external demand and rapidly transitioned towards industrialization and urbanization [35]. The contribution of agricultural foreign trade in China to the growth of the agricultural economy was remarkable [36]. Apart from that, the government has introduced a policy of expanding domestic demand and has vastly multiplied its agricultural science and technology investment [37]. Technological innovation in agriculture has become the primary driver, propelling agricultural economic growth and enabling steady progress in agricultural economic resilience. In 2019–2023, the trend slowed down due to COVID-19, which had a variety of effects on the agricultural economy, rural development, and farmer livelihoods [38]. These effects included labor shortages and supply chain disruptions, which made the agricultural economy more vulnerable to development, and a brief drop in the resilience index of the agricultural economy. After 2021, the agricultural economy’s resilience steadily recovered due to environmental laws, such as the introduction and application of “carbon-neutral” programs.

3. Results

3.1. Benchmark Regression

This research uses a two-way fixed-effects model for the regression analysis based on the abovementioned theoretical analyses. The estimated effects of air pollution on the agricultural economy’s resilience are shown in

Table 4. Benchmark regression.

Variables	(1)	(2)	(3)	(4)
PM	−0.444 ***	−0.141 **	−0.159 ***	−0.078 ***
	(0.051)	(0.060)	(0.027)	(0.030)
pergdp		0.004	0.033 ***	−0.028 *
		(0.021)	(0.011)	(0.015)
infra		0.012	0.015 **	0.047 ***
		(0.014)	(0.007)	(0.010)
open		−0.005	−0.005 **	−0.002
		(0.005)	(0.002)	(0.002)
trash		0.000	−0.001	−0.004
		(0.014)	(0.005)	(0.005)
invest		0.023 **	0.012 ***	0.018 ***
		(0.011)	(0.005)	(0.005)
inno		0.016 *	0.009 **	−0.007
		(0.010)	(0.004)	(0.005)
_cons	0.351 ***	−0.110	−0.248 ***	0.431 ***
	(0.028)	(0.135)	(0.079)	(0.130)
N	720	720	720	720
R2	0.244	0.826	0.931	0.939
Prov_FE	Yes	No	Yes	Yes
Year_FE	Yes	No	No	Yes
Control_Variables	No	Yes	Yes	Yes

Note: ***, **, and * denote 1%, 5%, and 10% significance levels, standard errors are in parentheses.

. Considering solely the explanatory factors and incorporating the two-way fixed effects for province and time, the regression results are displayed in column (1); after adding only the control variables, they are displayed in column (2); after adding the control variables and the province fixed-effects model, they are displayed in column (3); and after adding the control variables and the two-way fixed-effects model of province and time, they are displayed in column (4). Further, a test for multicollinearity was performed using the variance inflation factor on the variables of interest, and the mean value of the VIF was 4.31, which is less than 10. It shows that there is no multicollinearity problem in the regression equation. Air pollution (PM), the primary explanatory variable, has regression coefficients that are all significantly negative at high levels, according to the findings of models (1) through (4). Air pollution has a detrimental effect on agricultural economic resilience, and the more air pollution there is, the worse the situation is for agricultural economic resilience, according to this theory. This validates hypothesis 1 (H1).

Air pollution (PM) has a statistically significant regression coefficient in model (1) of Table 4, meaning that even when considering additional factors, the influence of PM on agricultural economic resilience is substantial, indicating that air pollution adversely affects agricultural economic resilience. Model (3) adds province fixed effects after controlling for the control variables, and the air pollution (PM) coefficient remains highly adverse at the 1% level. After taking into consideration potential systemic variations within provinces, it is implied that air pollution still limits the growth of agricultural economic resilience. The negative coefficient of air pollution (PM) in model (4) is further diminished when province and time effects in the baseline model are taken into account. At the 1% significance threshold, it is still negative and significant. Adjusting for a wide range of observable and unobservable characteristics at the provincial and temporal levels appears to eliminate certain differences in how air pollution impacts agricultural economic resilience. This represents a more accurate magnitude of the effects.

3.2. Robustness Tests

3.2.1. Other Robustness Tests

In this research, a robustness testing approach has been used to guarantee the dependability of the results. First, all variables were regressed after being cut at the 5% quartile. Model (1) in

Table 5. Robustness regression results.

Variables	(1)	(2)	(3)	(4)
PM	−0.078 ***	−0.078 ***	−0.100 ***	−0.078 ***
	(0.030)	(0.030)	(0.032)	(0.026)
_cons	0.431 ***	0.351 ***	0.283 **	0.267 **
	(0.130)	(0.125)	(0.130)	(0.117)
sigma_u				0.000
				(0.001)
sigma_e				0.028 ***
				(0.001)
N	720	720	624	720
R2	0.939	0.850	0.946
Prov_FE	Yes	Yes	Yes	Yes
Year_FE	Yes	Yes	Yes	Yes
Control_Variables	Yes	Yes	Yes	Yes

Note: ** and *** indicate significance at the level of 5% and 1%, respectively; robust standard errors are given in parentheses, as below.

displays the findings. Second, 500 samples were chosen at random. The negative impact of air pollution on agricultural economic resilience is further supported by model (2) in Table 5, demonstrating that the coefficient of air pollution (PM) is significant at the 1% level. Third, to prevent the regression results from being unduly impacted by the realization of extraordinary growth in particular years due to policy dividends, resource endowments, or globalization dividends, the sample with higher economic growth from 2000 to 2023 was removed. Model (3) in Table 5 displays the findings. The findings’ robustness is further confirmed by the fact that the negative coefficient of air pollution (PM) is still considerable. In order to account for both time and province fixed effects, the regression model is swapped out for a panel fixed-effects Tobit model in the fourth phase.

3.2.2. Quantile Regression

Quantile regression was applied in this study to demonstrate the variations in the impacts of various quartiles and further validate the validity of the conclusions of this paper.

Table 6. Quantile regression results.

	(1)	(2)	(3)
	fe30	fe60	fe90
PM	−0.102 **	−0.170 ***	−0.256 ***
	(0.048)	(0.049)	(0.098)
N	720	720	720
Prov_FE	Yes	Yes	Yes
Year_FE	Yes	Yes	Yes
Control_Variables	Yes	Yes	Yes

Note: *** and ** denote 1% and 5% significance levels; standard errors are in parentheses.

indicates that the 30% quartile of weak agricultural economic resilience, the more moderate 60% quartile, and the 90% high agricultural economic resilience are all significantly negatively impacted by air pollution. This shows that regardless of whether air pollution levels are low, medium, or high, they consistently harm agricultural economic resilience. In support of hypothesis one (H1), this further demonstrates the harmful impact of air pollution on agricultural economic resilience.

3.3. Endogeneity Test

In brief, reverse causality, omitted variable errors, and other issues typically lead to biased and inconsistent parameter estimates. Agricultural economic growth has a direct impact on agri-environmental pollution [39]. Agricultural environmental pollution has a total effect on pollution, influencing the level of air pollution. Additionally, it has been shown that economic growth helps mitigate air pollution [40]. The agricultural economic resilience indicator system’s theoretical framework views the primary industry’s GDP growth rate and local financial spending on forestry, agriculture, and water issues as the system’s fundamental constituent dimensions. Consequently, the growth of the agricultural economy and the enhancement of its resilience are compatible. Thus, agricultural economic resilience has a correlative effect on air pollution. If not adequately handled, this can lead to unreliable empirical test results. To alleviate the endogeneity problem of bidirectional causality and dynamic panel bias, this study employs a regression analysis method with the explanatory variable, lagged first-order air pollution, as an instrumental variable, satisfying two core assumptions. One is the correlation assumption that air pollution has time-series inertia, a physical property of inter-period diffusion that makes lagged first-order air pollution highly correlated with current pollution levels. The other is the homogeneity hypothesis. Lagged first-order air pollution does not act directly on agricultural economic resilience. The selection of lagged first-order pollution as an instrumental variable to explore the influence on agricultural economic resilience is compelling.

Since

Table 7. Endogeneity tests.

	(1)	(2)
	PM	y
L.PM	0.828 ***
	(0.038)
PM		−0.175 ***
		(0.034)
_cons	0.112
	(0.125)
N	690	690
R2	0.968	0.827
Prov_FE	Yes	Yes
Year_FE	Yes	Yes
Control_Variables	Yes	Yes

Note: *** represent the significance levels of 1%; standard errors are in parentheses.

presents the findings of regressions of air pollution lag to the first order, a discussion of potential endogeneity problems in this work is necessary. According to model (1), the correlation assumption is met, and there is a considerable impact of the air pollution level in the lagged period on the air pollution level in the current period. Model (2) presents the regression results for lagged first-order air pollution, illustrating this variable’s negative and significant effect on agricultural economic resilience. These results suggest that air pollution remains detrimental to agricultural economic resilience even after accounting for endogeneity. This outcome aligns with the benchmark regression results and adds to this study’s legitimacy.

3.4. Moderation Analysis of Environmental Regulation

Based on the arguments mentioned above, established research has not reached an agreement regarding whether environmental regulation might improve agricultural economic resilience by reducing the adverse externalities caused by air pollution. By incorporating it into a benchmark model, this study investigates the moderating role of environmental legislation on the influence of air pollution on agricultural economic resilience.

Table 8. Moderating effects of environmental regulation.

	(1)	(2)
PM	−0.078 ***	−0.111 ***
	(0.030)	(0.029)
er		0.000
		(0.000)
c_pm1er		0.001 ***
		(0.000)
_cons	0.431 ***	0.438 ***
	(0.130)	(0.131)
N	720	720
R2	0.939	0.941
Prov_FE	Yes	Yes
Year_FE	Yes	Yes
Control_Variables	Yes	Yes

Note: *** represent the significance levels of 1%; standard errors are in parentheses.

displays the findings. Compared to the baseline regression model (1), the regression coefficient of the moderating effect of model (2) has the opposite sign and is significantly positive at the 1% level. This implies that the impact of air pollution is negatively moderated by environmental control as a moderating variable, which may mitigate the negative effects of air pollution on the economic resilience of agriculture. The fundamental reason for this is that moderate environmental regulation can strengthen ecological responsibility. It compels the subject to take the required steps to lower pollution production and emissions while adhering to stringent environmental rules. It fulfills the cleaner production goals demanded by the government, amplifies the incentive effect on agricultural economic resilience, and partly relieves the negative impact of air pollution on agricultural economic resilience. Thus, hypothesis 3a (H3a) is tested.

3.5. Heterogeneity Analysis

Heterogeneity can arise from different stages of economic development and elements of agricultural economic resilience and variations in the degree of agricultural economic resilience. Thus, using regression analyses categorized into three groups, this section will investigate how air pollution affects the variability in agricultural economic resilience.

3.5.1. Effects of Air Pollution on Agricultural Economic Resilience Through Different Phases of Economic Development

Economic development and air pollution levels, as shown by PM2.5 concentrations, are closely associated. This section separates China’s 21st-century economic growth into three stages: the WTO entrance and external demand expansion period (2000–2008), the phase of domestic demand expansion after the financial crisis (2008–2016), and the phase of financial clearing and high-quality development (2017–present).

Table 9. Heterogeneity of stages of economic development.

	(1)	(2)	(3)
PM	−0.009	−0.069 ***	−0.147 ***
	(0.024)	(0.025)	(0.051)
_cons	−0.362 **	−0.531 ***	−2.040 ***
	(0.147)	(0.127)	(0.642)
N	270	240	210
R2	0.980	0.993	0.975
Prov_FE	Yes	Yes	Yes
Year_FE	Yes	Yes	Yes
Control_Variables	Yes	Yes	Yes

Note: *** and ** denote 1% and 5% significance levels; standard errors are in parentheses.

displays the estimation findings. In Table 9, the model’s estimated coefficient for air pollution (PM) is negative and negligible, suggesting that air pollution has no discernible effect on agricultural economic resilience through the WTO accession and growth in external demand stages. Specifically, among the three stages of economic development, air pollution impacts agricultural economic resilience in the post-financial-crisis phase of domestic demand expansion and the phase of financial clearing and superior development. The effect on the WTO accession and foreign demand expansion stage samples is not significant. The possible explanations are the stage of WTO accession and the expansion of external demand.

Under the industrial priority development model [35], agriculture relies on a labor-intensive production model, and traditional factor inputs and domestic demand still dominate the improvement in agricultural economic resilience. At this time, the increase in pollutant emissions has affected agricultural production to a certain extent, but the sensitivity of agriculture to pollution is fully understood, so the impact of air pollution on agricultural economic resilience is not significant. Second, the marginal damage caused by air pollution to agricultural economic resilience is rising notably because of the overlap between the post-financial-crisis expansion of domestic demand and the modernization of agriculture. The Chinese economic development model, characterized by high investment, pollution, and consumption, has led to economic expansion and fostered intensive and large-scale agriculture. In the meantime, the sensitivity of large-scale cultivation to environmental pollution has increased. Ecological thresholds have been breached, and resources have been over-consumed, limiting the space for agricultural development and affecting root development and the safety of agricultural products. This impairs agricultural ecosystems’ ability to self-repair and threatens the agricultural economic system’s sustainability, which impacts the agricultural economic system’s stability and increases its susceptibility. Thirdly, our economy has come into a phase of high-quality growth as a result of supply-side reforms that have encouraged the clearing of production capacity. We have paid more attention to the transformation of agriculture towards greening and digitalization, and we have strengthened environmental protection inspections. However, the cumulative effects of previous pollution are visible. The increased pressure on agricultural ecological remediation and the difficulty of environmental management have led to rising costs of pollution management and of resources required for agricultural production. There are long-term challenges to agricultural economic resilience. This further confirms hypothesis 2 (H2).

3.5.2. Effects of Air Pollution on Various Aspects of Agricultural Economic Resilience

The three elements of agricultural economic resilience—the capacity to resist and recover from risk, the ability to adapt and adjust, and the ability to reconstruct and innovate—were examined for variation in models (2) to (3) in

Table 10. Heterogeneity in the three dimensions of agricultural economic resilience.

	(1)	(2)	(3)	(4)
	y	y1	y2	y3
PM	−0.078 ***	−0.133 ***	0.156 ***	−0.082
	(0.030)	(0.028)	(0.049)	(0.061)
_cons	0.431 ***	0.336 **	1.057 ***	0.336
	(0.130)	(0.154)	(0.147)	(0.232)
N	720	720	720	720
R2	0.939	0.925	0.955	0.878
Prov_FE	Yes	Yes	Yes	Yes
Year_FE	Yes	Yes	Yes	Yes
Control_Variables	Yes	Yes	Yes	Yes

Note: *** and ** denote 1% and 5% significance levels; standard errors are in parentheses.

. The estimated results are shown in Table 10. The regression coefficients of air pollution (PM) for models (2) and (3) are both statistically significant at the 1% level, and model (4) is not statistically significant. The magnitude of the regression coefficient for air pollution (PM) is specified. Initially, air pollution hampers the capacity to resist and recover from threats. Air pollution directly damages agricultural infrastructure and reduces the efficiency and effectiveness of the basic factors of agricultural production. Agricultural producers may need to make additional investments in land improvement, irrigation facility upgrades, pest and disease control, and other measures as a result of damage to agricultural production factors, which worsens the agricultural production environment and hinders the agricultural economy’s ability to recover. Secondly, on the contrary, it improves adaptive adjustment capacity. The government may simultaneously implement a number of beneficial measures, such as providing subsidies, technical support, and tax incentives. This would help agricultural economic agents quickly adapt to the impact of air pollution and resume production within a short period. Thirdly, given the long-term, time-lagged, cumulative, and policy effects of reconstructing innovations, as well as agricultural scientific innovations and structural upgrading, they tend to require long planning cycles. Because of the short- or medium-term effects of air pollution, agricultural agents might not need to modify their development methods. As a result, there is little impact on reconstructing inventions. Behind this also lies the fact that the country is in a developmental stage of adaptive adjustment and does not yet possess complete capacity for reconstructing innovations. This demonstrates that the effects of air pollution on agricultural economic resilience vary, and hypothesis 2 (H2) is put to the test.

3.5.3. Effects of Air Pollution on Various Agricultural Economic Resilience Levels

This study determines the overall worth of agricultural economic resilience from 2000 to 2023 across all provinces in the nation, determines the overall average, and ranks the provinces accordingly. Models (1) to (4) in

Table 11. Heterogeneity of resilience intensity across agricultural economies.

	Top 10 Provinces in Terms of Agricultural Economic Resilience				The Rest of the Provinces
	(1)	(2)	(3)	(4)	(5)	(6)	(7)	(8)
	y	y1	y2	y3	y	y1	y2	y3
PM	0.061	0.000	0.379 ***	0.032	−0.162 ***	−0.216 ***	0.063	−0.162 ***
	(0.046)	(0.029)	(0.073)	(0.125)	(0.029)	(0.044)	(0.058)	(0.043)
_cons	0.213	−0.363 ***	1.138 ***	0.797 *	0.286 **	0.517 **	0.797 ***	−0.305 *
	(0.164)	(0.106)	(0.194)	(0.464)	(0.135)	(0.205)	(0.157)	(0.176)
N	240	240	240	240	480	480	480	480
R2	0.966	0.982	0.971	0.908	0.915	0.846	0.944	0.843
Prov_FE	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Year_FE	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Control_Variables	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes

Note: ***, **, and * denote 1%, 5%, and 10% significance levels, standard errors are in parentheses.

show the regression analyses of the top ten provinces regarding agricultural economic resilience. Regression studies of the top ten provinces in terms of agricultural economic resilience across the three characteristics are shown in (2) through (4); (5) to (8) are regression analyses of the remaining provinces and municipalities that have obtained agricultural economic resilience; and (6) to (8) are regression analyses of the remaining provinces and municipalities that have achieved agricultural economic resilience across the three dimensions of agricultural economic resilience.

The 10 leading provinces were evaluated regarding their agricultural economic resilience. First, the top ten provinces’ agricultural economic resilience is not significantly impacted by air pollution over time. Generally speaking, high-resilience regions have more complete infrastructures, higher levels of agricultural scientific technology, and better policy support, which gives them an advantage in agricultural technology innovation, the digital economy, and smart agriculture. They can swiftly modify their patterns of agricultural production to minimize the adverse effects of air pollution; thus, air pollution has a favorable effect on the top 10 provinces in terms of agricultural economic resilience. Second, it appears that air pollution encourages adaptive control because the model (3) regression coefficients for PM in Table 11 have significance at the 1% level. They could transform pollution regulation into an opportunity for green agricultural development, where this type of development offsets some of the negative impacts of air pollution. Thirdly, models (2) and (4) are both statistically insignificant, and model (3) is significant. Highly resilient regions have progressed beyond the stage of building resistance and recovery capabilities to risk. They are now entering the phase of adaptive adjustment but have not yet crossed the tipping point of the innovation-driven reconstruction of the agricultural economy.

Then, the remaining provinces and cities were analyzed. First, air pollution significantly suppresses overall agricultural economic resilience, and areas with weaker agricultural economic resilience are more vulnerable to adverse shocks caused by air pollution due to weak infrastructure, a lack of diversified economic support, and a lack of innovative resources. Secondly, models (6) and (8) show significant adverse effects on resistance and recovery from risks, as well as the reconstruction of innovation capacities; in contrast, model (7) lacks substance. The rigid agricultural factors of production and delayed policy implementation are typically problems in areas with low agricultural economic resilience, which have failed to activate the “push effect” that leads to innovation and upgrading. Low-resilience areas lack effective ways to cope with the adverse impacts of pollution through adjustments in industrial structure or technological innovations. As a result, the capacity for adaptive adjustment fails to emerge.

In summary, the effects of air pollution are not significant and may even promote adaptive adjustment capacity in areas of high agricultural economic resilience. Air pollution dramatically lowers total agricultural economic resilience in areas with low agricultural resilience, especially when it comes to the capacity to reconstruct and innovate as well as to resist and recover from hazards. The real reasons for this difference are primarily related to regions with more agricultural economic resilience having a high monetary base and high levels of social services [41], industrial agglomeration [42], agricultural technology innovation [8], digital economy [9], agricultural infrastructure development [43], industrial structure, and so on.

4. Discussion

Agricultural economic resilience is critical for the green transformation of agricultural systems towards sustainable agricultural development objectives. In particular, as a large agricultural country and the world’s largest food producer, enhancing China’s agricultural economic resilience is critical to ensuring global food security. However, changes in precipitation patterns and composition, an increase in extreme weather events, and changes in temperature conditions can affect the resilience of agricultural economies and threaten food security and development [44,45]. Air pollution, as a global environmental problem, is acknowledged by the Chinese Government, which regards it as one of the key challenges to sustainable economic and social development. However, few studies have directly examined the relationship between air pollution and agricultural economic resilience. For this reason, this paper empirically investigates the impact of air pollution on agricultural economic resilience and its mechanisms using a two-way fixed-effects model and a mediated-effects model, using provincial macro-panel data, to elucidate potential theoretical mechanisms and to formulate effective policy measures.

The findings and unique contribution of this paper lie in the in-depth analysis and measurement of the impact of air pollution on the system of indicators of agricultural economic resilience from multiple dimensions. This dimension of analysis is highly consistent with the multidimensional resilience measurement framework for agricultural systems proposed by Meuwissen and is an empirical result in the Chinese context, expanding the research horizons in this field [6]. This finding enriches our understanding of agricultural economic resilience. First, this study points to the adverse effects of air pollution on agricultural economic resilience and makes a valuable contribution to existing research. This finding is consistent with existing research on ozone and climate change as threats to global food security and supports the theory of “negative externalities” in environmental economics, which states that environmental degradation imposes structural constraints on the productivity and resilience of agricultural systems and reduces the ability of agricultural systems to adapt and reconfigure themselves in response to shocks. [1]. At the same time, the introduction of environmental regulation is a moderating variable. Research has found that environmental regulation mitigates the negative impact of air pollution on agricultural economic resilience to some extent. It supports Porter’s hypothesis that rational environmental regulation can stimulate green innovation and enhance systemic resilience. This further enhances our understanding of the impact of air pollution on agricultural economic resilience [27]. It provides policymakers and practitioners with important perspectives that are critical for the design and implementation of strategies aimed at enhancing agricultural economic resilience, especially in achieving balanced regional development and improving the efficiency of agricultural resource allocation. The empirical results of the three dimensions of heterogeneity further corroborate Martin’s suggestion that regional resilience evolves as an economy develops [4].

Although the findings of this paper are corroborated by several existing studies, we have introduced the perspectives of “air pollution” and “environmental regulation” into the research framework of agricultural economic resilience and obtained more systematic conclusions at the empirical level, which provide theoretical support and an empirical basis for the current sustainable development of agriculture, environmental governance, and regional policy formulation; however, some limitations must be recognized. Firstly, constrained by data availability and sample generalization, this study uses provincial panel data, and some air pollution data are estimated by interpolation methods. This may hide the characteristics of agricultural heterogeneity at the county or smaller scale level, and it is not clear whether the conclusions apply to other countries or regions with different agricultural economic structures. In addition, this paper has not empirically examined the impact of air pollution on agricultural economic resilience regarding the effect on resource allocation due to difficulty in obtaining data on land quality and health, among others. Secondly, the research perspective focuses on the direct impacts of air pollution. Although this study addressed the air pollution problem, due to the missing data on precipitation, the use of metrological methods to fill in the gaps for this may cause problems with the authenticity of the data, as could the lack of access to data sources on extreme weather events and changes in temperature conditions. We have not yet systematically considered the effects of the synergistic mechanisms of precipitation pattern variability, changes in the frequency and intensity of extreme weather events, and the spatial reconfiguration of temperature gradients, among other factors, on agricultural economic resilience.

In response to the above limitations, in future research, at the data method level, we could consider the introduction of ground-based monitoring data, high-resolution remote sensing data, and field surveys to improve accuracy and enhance the precision of air pollution data. Methods such as machine learning or principal component analysis are applied to the screening of the agricultural economic resilience indicator system to improve the objectivity and generalization of the indicator system. At the level of research horizons, we could establish a multi-factor coupling analysis framework for climate change, create an agricultural economic resilience assessment model that includes climate–pollution interaction effects, and carry out empirical research on the heterogeneity of different countries or regions in agricultural economic resilience, to eventually form a spatially adaptable theoretical system of agricultural economic resilience. By addressing these areas, China can build a more international perspective of the agricultural economic resilience theory system and assessment framework and provide more targeted scientific support for the sustainable development of global agriculture.

5. Conclusions

5.1. Research Conclusions

Considering panel data from 30 Chinese provinces between 2000 and 2023, this paper expands the data, deepens the mechanism identification, and examines how air pollution affects agricultural economic resilience. Agricultural economic resilience is defined as the agricultural economic system’s capacity to resist and recover from risks, make adaptive adjustments, and implement restructuring innovations. It uses models with mutual fixed consequences and moderators to investigate empirically how air pollution affects agricultural economic resilience in a number of ways, providing a multi-perspective empirical expansion of agricultural economic resilience research. Using two-way fixed-effects and moderation-effects models, we conduct an empirical analysis of how air pollution affects agricultural economic resilience through various factors, leading to the following research findings and conclusions: First, after controlling for various factors that affect agricultural economic resilience, air pollution remains a significant and detrimental factor in agricultural economic resilience. After a number of robustness tests and endogeneity corrections, this conclusion remains true. Secondly, there is also variation in how air pollution affects agricultural economic resilience. First of all, agricultural economic resilience is negatively impacted by air pollution at various phases of economic growth. In the stage of WTO accession and external demand expansion, air pollution has no significant effect on agricultural economic resilience. Both during the phase of domestic demand expansion following the financial crisis and the phase of financial clearance and high-quality development, air pollution can significantly exacerbate the vulnerability of agricultural economic resilience. Next, with regard to sub-dimensions, this study found that air pollution notably reduces risk-resistant recovery and reconstructs the innovation aspects of agricultural economic resilience while having a positive and remarkable effect on the adaptive adjustment capacity. Finally, in terms of intensity, air pollution’s effects on areas with less resilient agricultural economies are significantly greater than on areas with higher agricultural economic resilience. Specifically, regarding the adaptive adjustment to air pollution across regions, the greater the agricultural economic resilience, the more robust the adaptive response. The lower the agricultural economic resilience, the more air pollution reduces the adaptive adjustment capacity of agriculture. Thirdly, environmental control is a moderator in lessening the detrimental impacts of air pollution on agricultural economic resilience; in other words, the detrimental impacts of air pollution on the economic resilience of agriculture are mitigated by ecological management.

5.2. Policy Implications

The first step is to control and mitigate the air pollution situation gradually, to eliminate the effect of the deterioration in air quality on the agricultural economy at the source, and to scientifically assess the impact of air pollution on agricultural economic resilience development. It is necessary to consider not only the effects of air pollution on the improvement in the agricultural economy but also its impact on the green transformation of agriculture and sustainable development. To address the challenges of air pollution, it is necessary to develop effective response strategies that reduce the reliance of agrarian production on climatic resources and mitigate agroclimatic risks. There is also a need to encourage and support agricultural organizations to take climate adaptation and air pollution alleviation measures to stabilize the development of the agricultural economy.

The second step is to reinforce support for environmental regulation in climate adaptation and air pollution monitoring measures and to enhance the efficiency of environmental regulatory policies. Environmental regulations are compelling enough to decrease the damaging effects of climate change on agricultural economic resilience; therefore, it is vital to strengthen the environmental regulatory system and establish a climate-resilient agricultural policy framework. As a key element of improved environmental protection laws and regulations, we should construct a diversified system of environmental regulatory tools, strengthen the synergistic effect of policies, and develop a comprehensive environmental regulatory system encompassing source protection and control, process monitoring, and end-of-pipe treatment. Firstly, the complementary mechanism between market-incentive-type and order-control-type tools must be optimized. A composite regulatory system with market-incentive-type tools as the primary component and order-control-type tools as the safeguard should be established, forming a “mandatory constraints + economic incentives” dual-action drive. A dedicated fund for agricultural pollution control and a fiscal incentive mechanism should be established. Through coordinated efforts between central and local governments, an “Agricultural Clean Production Incentive Fund” should be set up to reward agricultural enterprises or entities that meet pollution reduction targets, thereby encouraging continuous improvement in their environmental performance.

Additionally, the supplementary role of technical regulation tools should be strengthened, and a “technical standard-certification system-promotion platform” should be established as a comprehensive support system. The agricultural environmental monitoring network should be improved by expanding the deployment of data monitoring stations in agricultural areas to track air quality, heavy metal contamination in farmland, and nitrogen–oxygen pollution. This will help establish a closed-loop mechanism of “pollution source identification–tracking–response.” Then, the supervision and guarantee system for implementing regulatory tools will be improved, and a new model of “digital supervision + social co-governance” will be innovated. The integration of multiple regulatory plans and the establishment of cross-regional joint prevention and control mechanisms should be promoted. River basins should be used as the basic unit to coordinate the management of air, water, and soil pollution, and key regions such as the Beijing–Tianjin–Hebei area and the Yangtze River Delta should be supported in building collaborative agricultural environmental governance systems, thereby achieving regionally coordinated and green agricultural development. Ultimately, this will enhance the international regulatory system and promote a system-oriented approach to openness, addressing distortions in the green innovation efficiency of multinational enterprises.

The third step is to prioritize regional differences and implement a precise governance strategy, including zoning, classification, and grading, to overcome the constraints imposed by the heterogeneity of agricultural economic resilience. Local governments at all levels should implement differentiated environmental regulatory policies based on the heterogeneous characteristics of environmental pollution intensity and the resilience of the agricultural economy in different regions. On the one hand, areas with low agricultural economic resilience, such as inland cities, need to increase their investment in environmental governance. Pursuing a high-quality green economic development path in urban development and achieving a virtuous cycle of harmonious coexistence between humans and nature are crucial ways to enhance the competitiveness of cities [46]. On the other hand, regions with higher resilience in the agricultural economy, such as coastal cities, should focus on maintaining existing environmental regulatory policies and improving the systems of taxation, subsidies, and environmental property rights trading. The ecological and fiscal decentralization system requires innovation to develop green industry clusters. The excellent synergistic growth of the green economy should be encouraged, and the leading and radiating role of inland cities should be reinforced. Lastly, in light of the diversity of reactions to the three resilience dimensions—risk resistance, flexible adaptation, and reconstruction and innovation—we should prioritize soil remediation and infrastructure strengthening projects in areas with low resilience. In middle- and high-resilience areas, we advocate for the deep integration of digital technology and the realization of the system through directional breakthroughs in resilience shortcomings. At the same time, agricultural systems are being restructured, including the introduction of more resilient crop varieties and changes in cropping patterns, to maintain agricultural productivity. Furthermore, this is achieved through the dynamic adaptation of policy tools to resolve the “low resilience–high pollution” dilemma.