The Evolutionary Game in Regulating Non-Agricultural Farmland Use within the Integrated Development of Rural Primary, Secondary, and Tertiary Industries

Abstract

Food security is a cornerstone of national development, and farmland protection and rationale are crucial for its assurance. However, integrating primary, secondary, and tertiary industries (IPSTI) in rural areas places significant pressure on farmland, threatening food security. This paper employs an evolutionary game model to explore how effective IPSTI can be achieved through stakeholder cooperation, ensuring rational farmland use under strict protection policies. The results reveal eight local equilibrium points in the game model, which can achieve asymptotic stability under varying conditions. Additionally, the behavioral parameters of local governments significantly impact their regulatory strategies for the non-agricultural use of farmland. When benefits increase, or penalties are strengthened, local government regulatory initiatives increase, constraining the behaviors of other participants. The village committee’s support strategy is influenced by benefits and costs, requiring a balance between promoting local economic development and minimizing non-agricultural use of farmland by enterprises. Enterprises’ decision-making primarily depends on the trade-off between the benefits and costs of non-agricultural use of farmland. The core findings of this study provide a crucial theoretical basis and decision support for formulating farmland protection policies and promoting industrial integration.

1. Introduction

In the context of globalization, food security has emerged as a major global concern. According to the Food and Agriculture Organization of the United Nations (FAO), the global population is projected to rise from 6.8 billion today to 9.1 billion by 2050, resulting in a one-third increase in the number of people requiring food. The majority of this population growth will occur in developing countries, increasing the demand for food, which relies on an adequate supply of farmland [1,2]. However, climate change, soil degradation and water scarcity threaten global food production stability [3,4,5]. In this context, protecting farmland is crucial as it forms the foundation of global food security [6,7]. However, the balance between farmland conservation and economic development has become a complex issue [8]. In regions with limited farmland, reducing non-agricultural land use to meet growing food demands without harming the ecological environment has become a key issue for global sustainable development goals [9,10,11].

As the most populous country in the world, China faces significant pressure to protect its farmland. China’s farmland accounts for only about 10 percent of its total land area, but it has managed to produce a quarter of the world’s food and solve the problem of feeding one-fifth of the world’s population [12]. Recognizing the importance of protecting farmland, the Chinese government has implemented strict policies, including the “red line of farmland” system, to ensure that the quantity and quality of bare farmland are preserved [13,14,15]. To promote overall rural economic development, the government has advocated for integrating primary, secondary, and tertiary industries (IPSTI), developing regional brands and expanding reproduction. This strategy encourages the deep integration of agriculture with industrial and service sectors to enhance agricultural added value and efficiency [16,17,18]. However, the conflict between the supply and demand for land resources in the IPSTI has become a critical issue.

In the IPSTI in rural areas, increased land use or non-agricultural use of farmland is often required. This includes practices such as land hardening, which are destructive to farmland; for example, constructing agricultural product processing factories and developing rural tourism projects may require occupying significant amounts of farmland or changing its nature. This conflicts with the policy of strictly protecting farmland [13]. Inadequate management of rural agricultural land may lead to irrational and irregular expansion of land use [19,20], posing threats to food security, sustainable agricultural development, and the overall well-being of the rural environment. Thus, striking a balance between these two aspects has become a pressing issue that demands immediate attention.

In the IPSTI, the utilization of non-agricultural land involves multiple stakeholders: local governments, village committees, and enterprises [21,22,23]. Local governments enforce farmland protection policies for food security and ecological balance but may relax restrictions to attract investments and foster industrial growth when faced with economic and fiscal pressures. Village committees, recognizing the potential benefits of industrial integration, may endorse non-agricultural land use, yet concerns about environmental harm, land value fluctuations, and villagers’ rights may prioritize agricultural use. Enterprises, seeing non-agricultural use as enhancing competitiveness and profitability, seek collaboration to secure land rights and invest in industrial chain extension and innovation. However, considerations such as input–output ratios, policy unpredictability, or market competition may lead them to focus on traditional agriculture, mitigating risks and ensuring a stable income. The interplay of these diverse interests varies across regions, reflecting distinct stages of evolution. A thorough analysis of the dynamics among these three stakeholder groups is vital for improving governance and fostering effective management.

In recent years, research on the IPSTI in rural areas has proliferated, yielding notable achievements [16,24,25]. Scholars have examined its modes, factors, and rural economic impact. Some studies emphasize the economic benefits, boosting farmer incomes and cooperative performance [26]. Others highlight the importance of agricultural cooperatives and enterprise clusters in promoting rural revitalization [24,27]. Agricultural tourism, a key part of rural economic diversification, provides extra income and reduces dependence on traditional farming [28,29]. Research also focuses on agricultural industrialization, strengthening production and distribution coordination [30,31]. Consequently, many regions are advancing the integrated development of agriculture with industry and services, but this inevitably leads to competition for spatial resources, particularly the improper use or change in agricultural land use [19,32]. Numerous studies have also shown that government policies play a pivotal role in promoting agricultural industrialization and agricultural tourism [31,33] and regulating land use [20].

However, despite this progress, a notable gap remains in the literature regarding the conflict between the supply and demand of land resources and its management in the IPSTI, especially under strict farmland protection policies. The significance of this issue is evident in many practices in China, where excessive land use has led to the reduction and destruction of farmland. Furthermore, the methodology incorporated case studies and in-depth interviews to facilitate numerical simulations. Therefore, exploring the discrepancies above and intricacies in land resource management amidst rural industrial integration and conducting a thorough analysis of the interaction dynamics among these stakeholders and their consequences for farmland preservation holds substantial value and novelty.

Evolutionary game theory combines game theory with evolutionary principles, studying the strategic choices of individuals or groups in long-term interactions. It is widely used in stakeholder analysis [34,35,36,37]. This study will adopt the evolutionary game approach to simulate the dynamic interactions among local governments, village committees, and enterprises in farmland use decision-making. The evolutionary game approach reveals how stakeholders adjust their strategies based on feedback, ultimately reaching a stable state or evolutionary stable strategy [38,39]. This is crucial for understanding the dynamic evolution of the conflict between the supply and demand of land resources. Furthermore, to facilitate numerical simulations, case studies and in-depth interviews were incorporated into the methodology. Building upon relevant research [40], extensive field research was conducted in Guangzhou and Urumqi, China, focusing on the investigation of key stakeholders involved in land use conflicts within the IPSTI. By integrating findings from literature reviews and in-depth interviews, the research constructed models and performed numerical simulations, focusing on analyzing the asymptotic stability of the evolutionary game and its influencing factors.

The significance and contribution of this research lies in bridging existing knowledge gaps, encompassing two main aspects: (1) providing a comprehensive understanding of the complex interactions and strategic behaviors among local governments, village committees, and enterprises in managing land resources during the integration process, and (2) offering insights into the mechanisms that can facilitate the reconciliation of economic development and farmland protection goals.

2. Model Design

It can be observed that local governments, village committees, and enterprises constitute the core stakeholders. The interplay among local governments, village committees, and enterprises in regulating the non-agricultural use of farmland is a dynamic game process, where each participant adjusts their actions based on the strategies of the others to maximize their interests. Given the assumptions and problem analysis, further parameterization can be undertaken to delve deeper into their intricate interactions (

Table 1. Model parameters and their meanings.

Parameters	Meaning
x	Probability that a local government chooses a strict farmland use regulatory strategy
y	Probability that a village committee chooses to support non-agricultural use of farmland
z	Probability that an enterprise chooses non-agricultural use of farmland
Cg	Costs to local governments of implementing strict farmland use regulatory strategies
Ug	Benefits to local governments of enforcing strict farmland use regulation strategies
Ac	Rewards given by local governments to village committees for outstanding performance in farmland protection
Pc	Penalties imposed by local governments on village committees for irregularities in supporting non-agricultural use of farmland
Ae	Incentives given by local governments to enterprises to comply with regulations on the protection of farmland
Pe	Penalties imposed by local governments on enterprises for illegal non-agricultural use of farmland
Cc	Costs of village committee support to support enterprises’ non-agricultural use of farmland
Uc	Village committee support gains from non-agricultural use of farmland
Ce	Enterprises’ chosen costs borne by non-agricultural use of farmland
Ue	Enterprises’ chosen operating income received from non-agricultural use of farmland
R	Benefits are given by enterprises to the village committees to gain their support when the local government chooses a strict regulatory strategy
S	Support provided by village committees to enterprises to promote the integration of agricultural secondary and tertiary production
F	Enterprises are penalized by village committees for non-compliance with non-agricultural use of farmland when the local government chooses a strict regulatory strategy

).

Assumption 1:

In reality, local governments, village committees, and enterprises are often influenced by incomplete information, limited cognitive abilities, and environmental complexities when making decisions about farmland use, exhibiting characteristics of bounded rationality. The local government has a probability x of choosing a strict farmland use regulation strategy and a complementary probability of 1 − x for a non-strict regulation. Similarly, the probability of the village committee choosing to support the non-agricultural use of farmland is y, and the probability of choosing not to support the non-agricultural use of farmland is 1 − y. The probability of an enterprise choosing the non-agricultural use of farmland is z, and the probability of choosing the agricultural use of farmland is 1 − z. The bounded rationality assumption is closer to the actual situation and can more accurately reflect the behavior patterns of parties in farmland use decision-making. This bounded rationality assumption better reflects the actual behavior patterns of all parties involved in farmland use decision-making.

Assumption 2:

In implementing the strategy of strict farmland use regulation, local governments need to invest significant human, material, and financial resources in regulation and law enforcement, which constitutes the main cost, denoted as C_g. For example, establishing a comprehensive monitoring system, training professional law enforcement personnel, and conducting publicity and education activities all require substantial resources. However, the benefits of strict regulation are also significant, denoted as U_g, including safeguarding food security, maintaining ecological balance, and obtaining policy support and financial incentives from higher levels of government. Additionally, the local government can use reward and punishment mechanisms to motivate village committees to actively cooperate in protecting farmland. If the village committee performs well in farmland protection, the local government can provide financial subsidies or policy preferences, denoted as A_c. Conversely, if the village committee violates the law by supporting non-agricultural use of farmland, the local government can impose punitive measures such as fines and public criticisms, denoted as P_c. For enterprises, local government rewards and punishments are equally important. If an enterprise complies with farmland protection regulations, the local government can offer incentives such as tax breaks and priority project approval, denoted as A_e. Conversely, if an enterprise violates the law by using farmland for non-agricultural purposes, the local government will impose severe penalties, denoted as P_e, including hefty fines and orders to restore the land to its original state, thereby constraining the enterprise’s behavior.

Assumption 3:

When the village committee chooses to support non-agricultural use of farmland, it needs to provide a lot of support for enterprises to participate in the integration of agricultural secondary and tertiary industries, denoted as C_c, e.g., infrastructure construction, coordination and communication. However, supporting non-agricultural use of farmland may also bring certain benefits, recorded as U_c, such as attracting business investment and promoting village economic development, thus increasing the performance and reputation of the village committee. In addition, the reward and punishment parameters of the village committee being managed by the local government will also have an important impact on its decision-making. If the village committee actively cooperates with the local government’s farmland protection efforts, the local government may provide additional resource support or policy preferences, denoted as A_c. Conversely, if the village committee violates the law, it will face corresponding punishments, denoted as P_c. Additionally, if the village committee supports enterprises’ non-agricultural use of farmland in a planned and regulated way, it can benefit from the village’s economic development, including improved infrastructure and increased employment opportunities.

Assumption 4:

When enterprises choose non-agricultural use of farmland, they must bear the costs of land purchase or lease, land development, possible legal risks, and loss of social reputation, denoted as C_e. For example, illegal use of farmland may result in fines and project stoppages, severely affecting the enterprise’s normal operations and image. Conversely, enterprises can also gain more business benefits, denoted as U_e, such as increased product value and market share through industrial integration. Using farmland for different purposes may mean that the enterprise misses some development opportunities but avoids potential risks and costs. Additionally, the parameters of rewards and penalties for enterprises managed by local governments should not be ignored. Enterprises that comply with farmland protection regulations can gain the trust and support of the local government, benefiting from policy preferences and market access, denoted as A_e. Conversely, enterprises that violate regulations will face severe penalties, denoted as P_e.

Assumption 5:

If the local government adopts a strict regulatory strategy, an enterprise wishing to use farmland for non-agricultural purposes must obtain the support of the village committee, forming a collusive relationship. This requires the enterprise to provide additional benefits to the village committee, such as economic compensation or sharing operating profits, denoted as R. Simultaneously, the village committee will support the enterprise to benefit the IPSTI, denoted as S, including infrastructure construction and collaborative communication. If the local government chooses a strict regulatory strategy, the village committee may also choose a supportive strategy. In this case, if the enterprise insists on non-agricultural use of farmland and fails to comply with regulations, it may be penalized by the village committee, denoted as F.

Based on the above model assumptions, the payment matrix of the evolution game of the multiple-use regulation of farmland for the integrated development of primary, secondary, and tertiary industries in the context of strict protection of farmland can be further obtained (

Table 2. Payment matrix.

	Local Governments Chooses Strict Regulatory Strategies (x)		Local Government Choose Non-Strict Regulatory Strategy (1 − x)
	Village Committee Chooses to Support (y)	Village Committee Chooses not to Support (1 − y)	Village Committee Chooses to Support (y)	Village Committee Chooses Not to Support (1 − y)
Enterprise chooses non-agricultural use farmland (z)	Rg − Cg + Pc + Pe	Rg − Cg + Pe	0	0
	Rc − Cc − Pc + R − S	F	Rc − Cc − S	F
	Re − Ce − Pe − R + S	Re − Ce − Pe − F	Re − Ce + S	Re − Ce − F
Enterprise chooses agricultural use of farmland (1-z)	Rg − Cg + Pc	Rg − Cg	0	0
	Rc − Cc − Pc	0	Rc − Cc	0
	0	0	0	0

).

3. Model Analysis

3.1. Analysis of Replication Dynamics

3.1.1. Local Government

When local governments, village committees, and enterprises are given probabilities x, y, and z, it is assumed that the local government’s expected return from choosing a strict regulatory strategy is U₁₁ and the expected return from choosing a non-strict regulatory strategy is U₁₂. The average return is then U₁:

$$\left\{\begin{array}{l}{U}_{11}=\left(1-y\right) \left(\left(-C_g+R_g\right) \left(1-z\right)+\left(-C_g+P_e+R_g\right) z\right)+\\ \hspace{1em}\hspace{1em}y \left(\left(-C_g+P_c+R_g\right) \left(1-z\right)+\left(-C_g+P_c+P_e+R_g\right) z\right)\\ U_{12}=0\\ U_1=x{U}_{11}+\left(1-x\right)U_{12}\end{array}\right.$$

(1)

Based on the expected returns, the replicator dynamic equations for local governments choosing a strict regulatory strategy can be further derived:

$$F\left(x\right)={\displaystyle \frac{dx}{dt}}=x\left(U_{11}-U_1\right)=-\left(-1+x\right) x \left(-C_g+R_g+P_c y+P_e z\right)$$

(2)

3.1.2. Village Committees

Assuming that the village committee’s expected return from choosing the support strategy is U₂₁ and the expected return from choosing the no-support strategy is U₂₂, its average return is U₂:

$$\left\{\begin{array}{l}{U}_{21}=\left(1-x\right) \left(\left(-C_c+R_c\right) \left(1-z\right)+\left(-C_c+R_c-S\right) z\right)+\\ \hspace{1em}\hspace{1em}x \left(\left(-C_c-P_c+R_c\right) \left(1-z\right)+\left(-C_c-P_c+R+R_c-S\right) z\right)\\ U_{22}=F \left(1-x\right) z+F x z\\ U_2=y{U}_{21}+\left(1-y\right)U_{22}\end{array}\right.$$

(3)

Based on the expected returns, the replicator dynamic equation for the village committee’s choice of support strategy can be further derived:

$$F\left(y\right)={\displaystyle \frac{dy}{dt}}=y\left(U_{21}-U_2\right)=\left(-1+y\right) y \left(C_c-R_c+P_c x+F z+S z-R x z\right)$$

(4)

3.1.3. Enterprises

Assuming that enterprises choose a non-agricultural use strategy with an expected return of U₃₁ and an agricultural use strategy with an expected return of U₃₂, their average return is U₃:

$$\left\{\begin{array}{l}{U}_{31}=\left(\left(-C_e-F+R_e\right) \left(1-x\right)+\left(-C_e-F-P_e+R_e\right) x\right) \left(1-y\right)+\\ \hspace{1em}\hspace{1em}\left(\left(-C_e+R_e+S\right) \left(1-x\right)+\left(-C_e-P_e-R+R_e+S\right) x\right) y\\ U_{32}=0\\ U_3=z{U}_{31}+\left(1-z\right)U_{32}\end{array}\right.$$

(5)

Based on the expected returns, the replicator dynamic equation for enterprises’ choice of a non-agricultural use strategy can be further derived:

$$F\left(z\right)={\displaystyle \frac{dz}{dt}}=z\left(U_{31}-U_3\right)=\left(-1+z\right) z \left(C_e+F-R_e+P_e x-F y-S y+R x y\right)$$

(6)

3.2. Evolutionary Equilibrium Analysis

The replicated dynamic equations F(x), F(y), F(z) of the three subjects can be associated to obtain the set of replicated dynamic equations of the three-party evolutionary game, so that dp/dt = 0, dx/dt = 0, dy/dt = 0, dz/dt = 0 in the set of differential equations, and eight local equilibria of the game system can be obtained, E₁(0,0,0), E₂(0,0,1) E₃(0,1,0), E₄(0,1,1), E₅(1,0,0), E₆(1,0,1), E₇(1,1,0), E₈(1,1,1), where E₉(x*,y*,z*) is the solution of the system of equations, and since the stable solution in the multi-group evolution game must be a strict Nash equilibrium solution, while E₉ is not a strict Nash equilibrium solution, only E₁–E₈ are considered.

According to the theory of methods proposed by Friedman [41], the local stability of the equilibrium point can be judged by the characteristics of the system Jacobian matrix. Therefore, the game system Jacobian matrix can be obtained as:

$$J=\left[\begin{array}{ccc}{F}_x\left(x\right)& F_y\left(x\right)& F_z\left(x\right)\\ F_y\left(y\right)& F_y\left(y\right)& F_y\left(y\right)\\ F_z\left(z\right)& F_z\left(z\right)& F_z\left(z\right)\end{array}\right]=\left[\begin{array}{ccc}{\displaystyle \frac{\partial F(x)}{\partial x}}& {\displaystyle \frac{\partial F(x)}{\partial y}}& {\displaystyle \frac{\partial F(x)}{\partial z}}\\ {\displaystyle \frac{\partial F(y)}{\partial x}}& {\displaystyle \frac{\partial F(y)}{\partial y}}& {\displaystyle \frac{\partial F(y)}{\partial z}}\\ {\displaystyle \frac{\partial F(z)}{\partial x}}& {\displaystyle \frac{\partial F(z)}{\partial y}}& {\displaystyle \frac{\partial F(z)}{\partial z}}\end{array}\right]$$

(7)

Through Jacobi matrix analysis, the values of the strategy probabilities at different equilibrium points are substituted, and the corresponding matrix eigenvalues are derived. These are summarized in

Table 3. Equilibrium point stability analysis.

Equilibrium	Jacobian Matrix Eigenvalues	Stability and Its Conditions
E1(0,0,0)	λ1 = −Cg + Rg λ2 = −Cc + Rc λ3 = −Ce − F + Re	When −Cg + Rg < 0, −Cc + Rc < 0, −Ce − F + Re < 0, this point is stable, otherwise it is a saddle or unstable point.
E2(0,0,1)	λ1 = −Cg + Pe + Rg λ2 = −Cc − F + Rc − S λ3 = Ce + F − Re	When −Cg + Pe + Rg < 0, −Cc − F + Rc − S < 0, Ce + F − Re < 0, this point is stable, otherwise it is a saddle or unstable point.
E3(0,1,0)	λ1 = −Cg + Pc + Rg λ2 = Cc − Rc λ3 = −Ce + Re + S	When −Cg + Pc + Rg < 0, Cc − Rc < 0, −Ce + Re + S < 0, this point is stable, otherwise it is a saddle or unstable point.
E4(1,0,0)	λ1 = Cg − Rg λ2 = −Cc − Pc + Rc λ3 = −Ce − F − Pe + Re	When Cg − Rg < 0, −Cc − Pc + Rc < 0, −Ce − F − Pe + Re < 0, this point is stable, otherwise it is a saddle or unstable point.
E5(1,1,0)	λ1 = Cg − Pc − Rg λ2 = Cc + Pc − Rc λ3 = −Ce − Pe − R + Re + S	When Cg − Pc − Rg < 0, Cc + Pc − Rc < 0, −Ce − Pe − R + Re + S < 0, this point is stable, otherwise it is a saddle or unstable point.
E6(1,0,1)	λ1 = Cg − Pe − Rg λ2 = −Cc − F − Pc + R + Rc − S λ3 = Ce + F + Pe − Re	When Cg − Pe − Rg < 0, −Cc − F − Pc + R + Rc − S < 0, Ce + F + Pe − Re < 0, this point is stable, otherwise it is a saddle or unstable point.
E7(0,1,1)	λ1 = −Cg + Pc + Pe + Rg λ2 = Cc + F − Rc + S λ3 = Ce − Re − S	When −Cg + Pc + Pe + Rg < 0, Cc + F − Rc + S < 0, Ce − Re − S < 0, this point is stable, otherwise it is a saddle or unstable point.
E8(1,1,1)	λ1 = Cg − Pc − Pe − Rg λ2 = Cc + F + Pc − R − Rc + S λ3 = Ce + Pe + R − Re − S	When Cg − Pc − Pe − Rg < 0, Cc + F + Pc − R − Rc + S < 0, Ce + Pe + R − Re − S < 0, this point is stable, otherwise it is a saddle or unstable point.

, which shows the eigenvalues of each matrix:

According to the analysis using the Jacobi matrix, the eight evolutionary stable equilibrium strategies can achieve stable equilibrium under certain conditions, necessitating further analysis in eight distinct scenarios.

Scenario 1:

When −C_g + R_g < 0, −C_c + R_c < 0, −C_e − F + R_e < 0, (0,0,0) is the stable point, i.e., the local government chooses not to implement the strict regulation strategy, the village committee chooses not to support the non-agricultural use of farmland strategy, and the enterprise chooses the single-agricultural-use strategy. This indicates that the costs of implementing strict farmland use regulations by the local government, supporting non-agricultural use of farmland by the village committee, and using farmland for non-agricultural purposes by enterprises (including potential penalties from the village committee) exceed their respective benefits. In this case, all participants tend to choose the no-action strategy because the costs of taking action exceed the potential benefits.

Scenario 2:

When −C_g + P_e + R_g < 0, −C_c − F + R_c − S < 0, C_e + F − R_e < 0, (0,0,1) is a stable point, i.e., the local government chooses the strategy of not imposing strict regulations, the village committee chooses the strategy of not supporting non-agricultural use of farmland, and the enterprise chooses the strategy of using farmland for non-agricultural use. Although the costs of enforcing strict regulations by the local government are higher than the benefits, the business benefits that enterprises gain from choosing non-agricultural use of farmland outweigh the costs and possible penalties from the village committee. At the same time, while village committees gain from supporting non-agricultural use, their total benefits remain low due to the potential loss of local government incentives and the shared benefits from agricultural integration. This reflects that enterprises may ignore regulations and penalties when driven by economic interests, while local governments and village committees choose not to act due to insufficient incentives or high costs.

Scenario 3:

When −C_g + P_c + R_g < 0, C_c − R_c < 0, −C_e + R_e + S < 0, (0,1,0) is the stabilization point, i.e., the local government chooses the less stringent regulation strategy, the village committee chooses the support for farmland non-agricultural use strategy, and the enterprise chooses the single-agricultural use strategy. The costs to the local government of enforcing strict regulations continue to outweigh the benefits, but the benefits to the village committee from supporting nonfarm use outweigh its costs, while the costs to the enterprise of choosing the single-agricultural-use strategy outweigh its gains from maintaining the status quo. This suggests that village committees may provide support because they see the potential benefits of non-agricultural use, while enterprises choose not to change the status quo because they lack sufficient incentives or face higher switching costs.

Scenario 4:

When C_g − R_g < 0, −C_c − P_c + R_c < 0, −C_e – F − P_e + R_e < 0, (1,0,0) is the stabilization point, i.e., the local government chooses the strict regulation strategy, the village committee chooses not to support the farmland non-agricultural use strategy, and the enterprise chooses the single-agricultural use strategy. The cost of enforcing strict regulations is lower than the benefit to the local government, providing the driving force for their actions. At the same time, both village committees and enterprises choose not to support non-agricultural use of farmland due to their respective costs and risk considerations. This reflects that local governments can actively implement farmland protection policies with effective policy incentives, while village committees and enterprises comply with strict regulations due to insufficient economic benefits or the risk of higher penalties.

Scenario 5:

When C_g − P_c − R_g < 0, C_c + P_c − R_c < 0, −C_e − P_e − R + R_e + S < 0, (1,1,0) is a stable point, i.e., the local government chooses the strict regulation strategy, the village committee chooses the strategy of supporting non-agricultural use of farmland, and the enterprise chooses the strategy of single-agricultural use. The local government chooses strict regulation even when it faces penalties because its total benefits are still positive. Village committees opt to support non-agricultural land use because the combined benefits of supporting such use and avoiding penalties exceed their costs. On the other hand, enterprises prefer single-use agricultural land because the costs associated with non-agricultural use, including penalties and lost revenue, are greater than the benefits they would gain from their operations.

Scenario 6:

When C_g − P_e − R_g < 0, −C_c − F − P_c + R + R_c − S < 0, C_e + F + P_e − R_e < 0, (1,0,1) is the stability point, i.e., the local government chooses the strict regulation strategy, the village committee chooses not to support the farmland non-agricultural use strategy, and the enterprise chooses the farmland non-agricultural use strategy. The local government enforces strict regulations, but the benefits for enterprises using farmland for non-agricultural purposes still outweigh the costs, including possible penalties, so enterprises choose to use it. Village committees, on the other hand, choose not to support non-agricultural use because the total benefits, considering penalties and shared benefits, are negative. This reflects that enterprises driven by market forces choose to violate the law despite penalties, while village committees choose not to support non-agricultural use of farmland based on cost-benefit analyses.

Scenario 7:

When −C_g + P_c + P_e + R_g < 0, C_c + F − R_c + S < 0, C_e − R_e − S < 0, (0,1,1) is the stability point, i.e., the local government chooses the strategy of less stringent regulation, the village committee chooses the strategy of supporting farmland for non-agricultural use, and the enterprise chooses the strategy of farmland for non-agricultural use. The local government chooses not to regulate because the costs of doing so outweigh the benefits, while the village committee chooses to support non-agricultural use because the benefits of supporting non-agricultural use outweigh the costs of doing so. Enterprises, on the other hand, choose not to use because the costs of non-agricultural use outweigh the benefits. This suggests that village committees can be a major force in promoting off-farm use under certain conditions, while enterprises and local governments choose not to participate or support it due to their own considerations.

Scenario 8:

When C_g − P_c − P_e − R_g < 0, C_c + F + P_c − R − R_c + S < 0, C_e + P_e + R − R_e − S < 0, (1,1,1) is the stabilization point, i.e., the local government chooses the strategy of strict regulation, the village committee chooses the strategy of supporting farmland for non-agricultural use, and the enterprises choose the strategy of farmland for non-agricultural use. All participants choose the strategy of taking action. The local government receives a net benefit from the strict regulatory policy, the village committee covers its costs by supporting non-agricultural use and receiving shared benefits, and the enterprise chooses non-agricultural use because its operating income remains positive after all costs and penalties are accounted for. This shows that when policy incentives, market forces, and cost-benefit analyses are balanced, all participants can benefit from their respective choices, leading to a stable state of cooperation.

4. Numerical Modeling

Building upon related research [42], to assess the accuracy of the model analysis and derive more intuitive management insights, additional simulations were executed utilizing Python 3.8. The evolutionary rules of the simulation are dictated by replicator dynamic equations, where each participant modifies their behavioral strategy at each time step based on their individual and collective gains. In order to attain a thorough comprehension of the land resource supply and demand conflict in the IPSTI in rural areas, alongside the behavioral patterns and interests of each stakeholder, extensive field research was conducted in Guangzhou and Urumqi, China. These regions offer valuable case studies due to their distinct geographical positions and differing economic development levels. The research encompassed the present state of local rural industry development, land utilization, and the efficacy of implemented policies. Face-to-face interviews were conducted with local government representatives, village committee members, enterprise heads, and several farmers to gather insights into their challenges, aspirations, and strategic decisions during the integrated development of farmland preservation and industries (

Table 4. Summary of interviewees and interview findings.

Type of Interviewee	Quorum	Interview Findings
Local government officials	5	1. Facing the challenge of balancing the protection of farmland with the integrated development of industry. 2. Desire for policy-led sustainable development. 3. Difficulties of enforcement in the regulatory process. 4. Actively explore innovative regulatory mechanisms and technical means.
Village committee members	6	1. Concern about the conflict between villagers’ livelihoods and the protection of farmland. 2. Facing a shortage of funds when promoting industrial integration projects. 3. Expect more policy support and technical guidance. 4. To endeavor to reconcile the interests of all parties and to avoid over-exploitation of farmland.
Head of Enterprise	5	1. Seek non-agricultural uses of farmland to improve economic efficiency. 2. Encountering policy and regulatory constraints and land transfer difficulties. 3. Attaching importance to social responsibility and wishing to achieve a win-win situation between enterprises and communities. 4. Expect policy flexibility and a simplified approval process.
peasant household	5	1. Fear of non-agricultural use of farmland affecting food security. 2. Expect to increase their income by participating in industrial integration projects. 3. Facing pressure for technological upgrading and insufficient funding. 4. Desire for training and market information support.

).

Interviews with five local government officials revealed that the pursuit of local economic development necessitates significant land allocation for construction projects, whereas the central government imposes stringent regulations on farmland preservation. This dichotomy compels local officials to strike a balance between the demand for industrial land and the necessity to safeguard farmland. In practice, enforcing farmland protection policies, particularly in regulating the conversion of farmland use, has posed considerable challenges. To overcome these hurdles, local governments are actively seeking innovative regulatory frameworks and technological solutions, including remote sensing monitoring and big data analysis, to enhance regulatory efficacy.

Six village committee members were also interviewed, and they expressed that while fostering local economic growth, they must also take into account the tangible interests of villagers and the protection of farmland. A prevalent concern among village committee members is the scarcity of funds, particularly when initiating projects that facilitate industrial integration. They expressed a desire for increased policy backing and technical assistance from higher governmental tiers to facilitate the harmonization of all stakeholders’ interests and prevent the overexploitation of farmland.

Five business leaders were interviewed, who generally expressed interest in diverting farmland for non-agricultural purposes to boost their economic efficiency. However, in practice, they encounter policy constraints, regulatory hurdles, and difficulties in land acquisition, which hinder their development progress. The business leaders emphasized their commitment to corporate social responsibility and aspired to create a mutually beneficial scenario for both their enterprises and the community. To this end, they anticipate a more flexible policy landscape and streamlined approval procedures from the government, enabling them to swiftly adapt to market demands.

Based on this, the parameter values of the simulation are set as follows: C_g = 4, R_g = 4, P_c = 0.4, P_e = 0.4, C_c = 3, R_c = 4, C_e = 1, R_e = 2, R = 0.6, S = 0.4, and F = 0.5. Taking into account the initial strategy selection tendency and uncertainty, the probability that the local government chooses the strategy of strict farmland utilization regulation is x₀ = 0.5, and the probability that the village committee chooses the strategy of support for farmland is y₀ = 0.6. The probability of enterprises choosing the non-agricultural land use strategy is z₀ = 0.6, and the simulation period is t = 15. The result of the simulation is the evolution curve of the behavior strategy of the local government, village committee, and enterprise, as well as the evolution curve of the income of each participant. It can be used to analyze the land resource supply and demand conflict in the integrated development of primary, secondary and tertiary industries under the background of strict protection of farmland by different factors, as well as the game relationship and coordination mechanism among the participants.

4.1. Impact of Local Government Behavioral Parameters

Figure 1a illustrates the effect of varying the return (R_g) that local governments receive for implementing strict farmland use regulations. As R_g increases from 2.00 to 6.00, local governments become more motivated to enforce strict regulations. This rise in motivation stems from the increased financial benefits associated with stricter oversight. However, this shift results in reduced incentives for village committees and enterprises to pursue non-agricultural farmland use, likely due to the heightened regulatory pressure and potential penalties imposed by local governments. In Figure 1b, the cost (C_g) of implementing strict farmland use regulations is adjusted. As C_g increases from 2.00 to 6.00, local governments’ motivation to enforce strict regulations diminishes. Conversely, village committees and enterprises become more inclined to support and engage in non-agricultural farmland use. This suggests that high regulatory costs may lead local governments to relax their oversight, thereby providing village committees and enterprises with greater opportunities to promote non-agricultural use.

Figure 1c explores the impact of varying the penalty (P_c) imposed by local governments on village committees for supporting non-agricultural farmland use. As P_c increases from 0.20 to 0.60, local governments’ motivation to enforce strict regulations rises, while village committees’ and enterprises’ incentives to support or engage in non-agricultural use decrease. Notably, the impact on village committees is more pronounced, indicating that severe penalties have a greater deterrent effect on their behavior. Figure 1d examines the effect of varying the penalty (P_e) imposed by local governments on enterprises for illegal non-agricultural farmland use. As P_e increases from 0.20 to 0.60, local governments’ motivation to enforce strict regulations rises, while village committees’ and enterprises’ incentives to support or engage in non-agricultural use decrease. The impact on enterprises is particularly significant, suggesting that severe penalties have a strong constraining effect on their behavior.

4.2. Impact of Village Committee Behavioral Parameters

Figure 2a simulates the effect of varying the revenue (R_c) gained by village committees from supporting non-agricultural farmland use. As R_c increases from 2.00 to 6.00, village committees become more motivated to support non-agricultural use, which in turn prompts more enterprises to participate. Interestingly, local governments’ motivation to enforce strict regulations also increases, possibly due to their awareness of the potential economic benefits and the need to balance farmland protection goals. In Figure 2b, the cost (C_c) of village committees supporting enterprises’ non-agricultural farmland use is adjusted. As C_c rises from 1.50 to 4.50, the incentives for village committees and enterprises to engage in non-agricultural use decrease, while local governments’ motivation to enforce strict regulations also decreases. High costs may inhibit the actions of village committees and enterprises, while also reducing the regulatory pressure on local governments.

Figure 2c explores the impact of varying the support (S) provided by village committees to enterprises for promoting the integration of agricultural, secondary, and tertiary industries. As S increases from 0.20 to 0.60, enterprises become more motivated to choose non-agricultural farmland use, while village committees’ support decreases. This may be due to village committees’ awareness of the potential destruction of farmland and the risk of penalties from local governments. Figure 2d examines the effect of varying the punishment (F) imposed by village committees on enterprises for violating non-agricultural farmland use regulations. As F increases from 0.20 to 0.60, the influence of local governments’ motivation to enforce strict regulations remains relatively unchanged, while village committees’ and enterprises’ incentives to support or engage in non-agricultural use decrease.

4.3. Impact of Enterprises Behavioral Parameters

Figure 3a simulates the effect of varying the operating revenue (R_e) obtained by enterprises from choosing non-agricultural farmland use. As R_e increases from 1.00 to 3.00, enterprises become more motivated to engage in non-agricultural use. However, this trend also prompts local governments to strengthen regulations, as they are aware of the potential risks associated with high returns. Village committees, on the other hand, become less inclined to support non-agricultural use due to concerns about environmental impacts and community interests.

In Figure 3b, the cost (C_e) borne by enterprises for choosing non-agricultural farmland use is adjusted. As C_e rises from 0.50 to 1.50, enterprises’ interest in non-agricultural use decreases due to the reduced economic attractiveness of the project. Simultaneously, village committees become more inclined to support non-agricultural use in an attempt to maintain local economic vitality, while local governments’ motivation to enforce strict regulations weakens due to the self-regulatory role played by higher costs.

Figure 3c explores the impact of varying the benefits (R) that enterprises give to village committees for their support when local governments choose a strict regulatory strategy. As R increases from 0.30 to 0.90, village committees’ willingness to support non-agricultural use increases. However, this does not significantly affect local governments’ regulatory strategy, as they possess a stronger ability to resist rent-seeking behavior. Enterprises’ incentives to choose non-agricultural farmland use decrease, as high rent-seeking costs reduce the overall profitability of the project.

5. Discussion

5.1. Theoretical Implications

By constructing an evolutionary game model, this study analyzes the behavioral strategies of local governments, village committees, and enterprises in regulating the non-agricultural use of farmland, considering their influencing factors, and yields a series of valuable findings. First, it clarifies that there are multiple local equilibrium points and a unique stable Nash equilibrium solution in the three-party game, revealing the complexity and dynamics of regulating the non-agricultural use of farmland. This finding helps us understand the intertwined interests and mutual constraints of the parties in the context of farmland protection.

Secondly, increasing revenue and penalties for local governments can effectively enhance their regulatory incentives, which in turn can constrain the behavior of village committees and enterprises. This provides a clear direction for optimizing local government farmland management policy, emphasizing the importance of incentives and penalties in farmland protection. Village committees’ benefits and costs significantly impact their supportive attitudes, requiring a balance between supporting development and protecting farmland. This suggests considering the interests of village committees and establishing reasonable incentive and constraint mechanisms when promoting rural industrial integration. The benefits and costs of business operations are key factors in their decision-making. This provides a basis for guiding enterprises to use farmland legally and compliantly to achieve sustainable development and highlights the need to regulate enterprise behavior through market mechanisms and policy regulation.

These core findings provide important theoretical support and a decision-making basis for formulating scientific and reasonable farmland protection policies and promoting the IPSTI.

This study echoes and complements existing studies on the IPSTI and the protection of farmland. On the one hand, existing studies on the integration of one, two, and three industries mainly focus on the influencing factors of rural integration of one, two, and three industries and their economic and social impacts [16,24,25,26,43], and there are also studies focusing on the participatory behavior of individual stakeholders, such as agricultural co-operatives or enterprises [17,44], but there is a lack of multi-stakeholder interactions of farmland non-agricultural use in the integration of one, two and three industries, and its governance mechanism. In contrast, the tripartite game model constructed in this study based on the investigation of real cases in China more comprehensively considers the interactions and strategic choices of each subject. Compared with studies that emphasize the dominant role of local governments in farmland protection [45,46], this study further reveals the knock-on effects of changes in the behavioral parameters of local governments, village committees, enterprises on other subjects, highlighting the holistic and dynamic nature of the system. For studies focusing on the role of village committees in rural development, this study clarifies the behavioral changes of village committees driven by interests and their impacts on farmland use through quantitative analyses. This provides new perspectives for understanding grassroots forces in rural land management.

5.2. Practical Implications

The theoretical discoveries of this study bear direct significance to the practical concern of farmland being utilized for non-agricultural purposes in China, as evidenced by the prevalent phenomena of farmland being used for non-agricultural and non-food activities [47,48,49]. The theoretical findings of this study are closely related to reality. Insufficient regulation or relaxation of regulation by local governments in farmland protection due to cost considerations echoes the poor implementation of farmland protection policies in some areas. The study’s conclusion is that local government regulatory incentives can be strengthened by increasing the benefits of regulation and penalties, which provide solutions for these real-life problems. Village committees may support the non-agricultural use of farmland driven by profit, reflecting the neglect of farmland protection in the pursuit of economic development in some rural areas. This study’s recommendations for balancing support and protection can help guide village committees to actively protect farmland in rural development. The tendency of enterprises to overuse farmland in pursuit of economic benefits is common. The idea of regulating enterprise behavior by adjusting their business benefits and costs, as proposed in this study, is a practical guide to curbing the illegal use of farmland by enterprises.

Furthermore, grounded in the Chinese experience, this study unveils the governance mechanisms and common issues associated with the non-agricultural use of farmland amidst the IPSTI in rural areas. The international policy implications of these findings carry considerable weight, especially for developing nations where agricultural development frequently falls short in effectively tackling poverty and resource wastage. To boost agricultural value-added and production efficiency in the forthcoming era, it is imperative to foster diversified avenues, such as agricultural deep processing and agro-tourism. Within this framework, it is paramount to contemplate strategies for harmonizing and regulating the actions of diverse stakeholders, thereby ensuring the sustainable utilization of farmland and fostering the healthy progression of the agricultural sector.

The following policy insights are proposed: Firstly, as safeguarding land resources necessitates the active participation of local governments, it is essential to enhance the incentive mechanism for these governments to protect farmland. This can be accomplished by augmenting the benefits of implementing rigorous regulatory strategies, such as offering financial support or preferential policies. Secondly, it is crucial to strengthen penalties for non-compliance by village committees (or rural administrative organizations) and enterprises, thereby increasing the cost of violations and creating a formidable deterrent. Thirdly, fortifying training and guidance for village committees is imperative to elevate their consciousness regarding farmland protection and to assist them in achieving a balance between economic interests and farmland preservation amidst rural development. Fourthly, optimizing the environment for enterprise development by diminishing the cost of legal and compliant farmland use while augmenting operating income is vital, thereby motivating enterprises to consciously adhere to farmland protection regulations. Lastly, establishing a multi-party synergy mechanism is essential to promote communication and collaboration among local governments, village committees, and enterprises, fostering a robust collective endeavor towards farmland protection.

6. Conclusions

This research endeavors to delve into the evolving game strategies employed by local governments, village committees, and enterprises in managing the non-agricultural use of farmland amidst stringent farmland protection policies within the context of promoting the IPSTI in rural areas. Furthermore, it explores avenues for achieving a balance between this integrated rural industrial development and farmland preservation through the optimization of these game strategies. By constructing a tripartite evolutionary game model and subsequent simulations, this paper provides an in-depth analysis of how variations in different behavioral parameters influence the game equilibrium. The following are the key findings of this study:

• Under the strict protection of farmland, there are eight partial equilibrium points in the game between local governments, village committees, and enterprises. This indicates that in the regulation of non-agricultural use of farmland, the behavioral strategy choices of each participant are influenced by many factors, resulting in a complex interactive relationship.
• The behavioral parameters of local governments have a notable impact on their choice of regulatory strategies for non-agricultural land use. When the benefits of implementing strict farmland use regulation strategies increase, local governments’ regulatory motivation increases significantly. Conversely, when the costs increase, regulatory motivation decreases. Additionally, severe penalties for non-compliance by village committees and enterprises can effectively enhance the regulatory incentives of local governments and constrain the non-agricultural use of farmland by village committees and enterprises. Therefore, local governments can decide to motivate or avoid penalties associated with increased costs, thereby enabling alternative uses of the land.
• The behavioral parameters of village committees likewise influence their strategic choices in supporting non-agricultural land use. When village committees receive more benefits from supporting non-agricultural use of farmland, their incentive to support it increases significantly; however, rising costs inhibit their supportive behavior. Additionally, the support provided by village committees to enterprises to promote the integration of secondary and tertiary agricultural industries also affects their behavioral choices. Too much support may lead to farmland destruction and punishment, so village committees need to find a balance between support and protection.
• The behavioral parameters of enterprises also influence their choice of farmland non-agricultural use strategy. When the operating income from non-agricultural use of farmland increases, enterprises’ motivation for non-agricultural use of farmland increases; however, rising costs reduce their willingness to use it. Additionally, benefits given by enterprises to obtain support from village committees also affect their behavioral choices, but this does not significantly influence the regulatory strategies of local governments.

In summary, this study not only addresses the two core questions posed: how various entities adjust their strategies for non-agricultural land use under stringent farmland protection policies and how these strategies can be optimized to balance industrial development with farmland preservation, but it also uncovers the impact of changing behavioral parameters on game equilibrium. Despite limitations such as simplified model assumptions, the insights provided hold significant value for future research endeavors. Future studies could consider incorporating additional elements such as more participants, product types, and market environment variables to enhance the model’s complexity and realism. Concurrently, field research and case studies would further enrich the theoretical model and offer a more solid empirical foundation for policy formulation. Through interdisciplinary collaboration and a deeper exploration of the intricate mechanisms between farmland protection and integrated industrial development, it is anticipated that more scientific and comprehensive solutions for the sustainable management of farmland resources will emerge.