The Impact of “Land and Services” Dual-Scale Management on Agricultural Operational Benefit: A Comparison with Land-Scale Management

Abstract

This study aims to explore whether the dual-scale management model, formed by integrating service-scale management with land-scale management, can further break through the benefit limits of single land-scale management and unlock additional profit potential in agricultural scale operations. This study used data from a 2024 questionnaire survey of 2166 farming households in Anhui Province and employed a coupling coordination degree model to measure the level of dual-scale management. Subsequently, we utilized OLS regression and mediation effect models to empirically examine the impact of dual-scale management on agricultural operational benefit and their underlying mechanisms. We find that dual-scale management significantly improves agricultural operational benefit. Our measurements show that dual-scale management not only breaks through the upper limit of the optimal operating area inherent in single land-scale management but also yields a greater improvement in agricultural operational benefit than single land-scale management. Heterogeneity analysis reveals that dual-scale management significantly enhances the agricultural operational benefit of farmers in plain areas and farmers with fully developed high-standard farmland. Mechanism analysis indicates that dual-scale management enhances agricultural operational benefit through an endogenous efficiency improvement mechanism and an exogenous risk-burden-sharing mechanism. These findings suggest that fostering a synergistic development system for land-scale management and service-scale management is conducive to improving the economic returns for land scale operators and unlocking new dividend spaces for agricultural scale operation in China’s post-land transfer era.

1. Introduction

Scale management is a necessary condition for the development of modern agriculture. Over the past decade, the form of agricultural scale operation has evolved iteratively. Relying solely on the transfer of land management rights to form land-scale management has, in practice, gradually exhibited diminishing returns in growth potential. Moreover, as operational scale increases, the rising total cost of input factors, heightened risk agglomeration, and greater uncertainty of economic output led to increased volatility in the factor input–output ratio, which is a key factor contributing to the instability of large-scale agricultural land operation [1]. Therefore, in recent years, service-scale management, developed through agricultural socialized services, has emerged, and has played a positive role in solving the problem of arable land abandonment and promoting the optimal allocation of factors. Since service-scale management is somewhat dependent on the transfer of land management rights, land-scale management right transfer can promote the efficiency of service-scale management, but purely relying on a single scale of operation to enhance total factor productivity has certain limitations; China’s agricultural scale operation should be turned into a “land + service” scale Synergistic path [2]. Thus, China’s agricultural modernization strategy is currently undergoing a profound shift, moving from a primary focus on land transfer alone towards emphasizing agricultural socialized services, thereby transitioning to diversified modes of moderate-scale operation, including service-led models [3]. In the diversified scale operation system, land-scale management and service-scale management present the dual attributes of coupling and synergy. From the perspective of coupling, land-scale management achieves centralized allocation of production factors through the transfer of land management rights, thereby providing the physical space necessary for mechanization and standardized production; service scale-management rely on an agricultural socialized service system, utilizing organizations for trusteeship and specialized division of labor to compensate for technical and efficiency shortcomings in land scale operation. From the synergistic perspective, the continuous operation formed by land-scale management reduces the transaction cost of the service supply, and the embedded specialized service feeds back, enhancing the efficiency of land-scale management by enhancing total factor productivity [4].

Land-scale management refers to an operational approach that achieves concentrated and contiguous land use through the transfer of land management rights, thereby achieving internal economies of scale. Land-scale management enhances agricultural operational benefits by promoting intensive factor utilization through horizontal division of labor. Its advantage lies in increasing agricultural income by expanding land management scale, with the income-boosting effect being more pronounced in large-scale transfers [5]. Land transfers are often accompanied by improvements in technical efficiency, with transferee households generally exhibiting higher production efficiency than transferor households [6]. However, land-scale management also presents notable limitations. On one hand, their expansion is constrained by land resource endowments and transaction costs, imposing boundaries on an optimal scale. Excessive concentration may elevate organizational management costs [7]. On the other hand, as scale increases, the agglomeration effects of risks such as natural disasters and market fluctuations are amplified, posing challenges to operational stability [8].

Service-scale management leverages agricultural socialized service systems to achieve vertical specialization across different stages of crop production through entrusted management organizations and specialized division of labor, thereby capturing external economies of scale. This model helps farmers overcome constraints imposed by household resource endowments, thereby achieving land management scale [9]. Simultaneously, it enhances technical efficiency by optimizing factor allocation [10] and promotes the adoption of green technologies, leading to improvements in green agricultural total factor productivity [11]. However, its effectiveness is constrained by factors such as topographical conditions, crop types, service market maturity, and farmer management capabilities [12]. For instance, in mountainous regions with fragmented plots, the economies of scale for mechanized services are difficult to fully realize.

While scholars have recognized the respective advantages of land-scale management and service-scale management, most existing studies examine their impacts on agricultural operational benefit in isolation, with limited systematic investigation into their synergistic effects and the underlying mechanisms [13]. However, given the complementary nature of land-scale management and service-scale management in terms of division of labor, risk sharing, and factor integration, their coordination can form a “land and service” dual-scale management model. This model not only provides a demand base for service-scale management through land concentration but also enhances land productivity and technology spillovers through service embedding [4]. This approach overcomes the limitations of single-scale management, achieving coupled efficiency gains through “internal economies of scale + external economies of scale”. In rural China, land transfers and purchasing agricultural socialized services represent the primary pathways for farmers to engage in land-scale management and service-scale management. Some land scale operators will purchase agricultural socialized services to meet operational needs after commencing large-scale management on transferred land, which is a model termed dual-scale management in this study [3,14].

A few scholars have also noted the complementarity between land-scale management and service-scale management, suggesting that synergistic effects can arise through division of labor and integration [9]. However, they have not conducted in-depth research on the economic benefits and comparative advantages of dual-scale management. To address these gaps, this study utilizes 2024 survey data from 2,166 farming households across 309 villages in 16 prefecture-level cities of Anhui Province. It aims to investigate three core questions: First, can dual-scale management enhance agricultural operational benefit? Second, what comparative advantages does dual-scale management offer over land-scale management alone? Third, what are the mechanisms through which dual-scale management improves agricultural operational benefit?

This study may contribute to the following aspects. First, the existing literature often treats land-scale management and service-scale management as two parallel paths, examining their respective impacts separately. This study innovatively introduces the core concept of dual-scale management at the theoretical level, elucidating the synergistic principle whereby both achieve coupled “internal economies of scale + external economies of scale” through horizontal division of labor + vertical division of labor. This provides new theoretical support for understanding the complex pathways of agricultural modernization in China and countries with similar national conditions. Second, this study not only validates the positive effects of dual-scale management but also, through rigorous comparative research design, empirically quantifies for the first time the incremental benefits of scale operations compared to single land-scale management. This finding confirms that the synergistic “land and services” pathway is not only feasible but also superior, providing policymakers and agricultural operators with critical decision-making guidance for scale management choices and avoiding the pitfall of solely expanding land scale. Finally, existing research has been relatively weak in examining the mechanisms through which dual-scale management enhances operational benefit. By incorporating endogenous benefit enhancement and exogenous risk-bearing mechanisms into the empirical analysis framework, this study offers new insights and references for empirical research on how dual-scale management impacts agricultural operational benefit.

2. Theoretical Analysis and Research Hypotheses

The essence of dual-scale management lies in the coordinated allocation of land and service factors, with its intrinsic logic for enhancing agricultural operational benefit rooted in several classical economic theories. This study constructs an integrated theoretical framework centered on the theory of division of labor and supported by transaction cost theory and economies of scale/scope theory, as shown in Figure 1. First, according to the theory of the economy of division of labor, land-scale management achieves horizontal division of labor and specialization through land transfers, aiming to achieve internal economies of scale. Service-scale management, meanwhile, leverages the divisibility of production processes to achieve vertical division of labor, targeting external economies of scale. Together, they form the complete structure of deepening agricultural division of labor, serving as the fundamental source of efficiency gains. Second, transaction cost theory indicates that concentrated land management reduces the frequency and cost of transactions for service providers seeking clients, while the integration of standardized services lowers negotiation and supervision costs for scale operators in technology and labor management. The “coupling” of these two effectively curbs the potential rise in transaction costs associated with scale expansion. Finally, economies of scale and scope theory reveals that dual-scale management, by restructuring the production function, not only expands the boundaries of economies of scale for factors like capital and technology beyond those achievable under single land-scale management but also realizes resource sharing and economies of scope through the cross-regional operations of service providers. This approach breaks through the efficiency ceiling that is inherent in single-scale models. Based on the above theoretical logic, this study proposes the following hypotheses.

2.1. Impact of Dual-Scale Management on Agricultural Operational Benefit

The deepening of the division of labor is a source of economic growth, land-scale management and service-scale management. For the role of agricultural operational benefit, there are some differences in the mechanism of land-scale management through the transfer of land to achieve horizontal division of labor, to promote the specialization of agricultural production, so as to achieve “internal economies of scale”. Service-scale management relies on the same crop in different production segments of divisibility through the vertical division of labor to form different production segments of vertical specialization, so as to obtain “external economies of scale” [15].

Theoretically, there is a synergistic relationship between the two; on the basis of division of labor and integration effect, land-scale management is the premise of service-scale management, and service-scale management is to support the development of land-scale management. On the one hand, the refinement of the division of labor in agricultural production optimizes resource allocation and enhances production efficiency. In dual-scale management, land-scale management reduces the unit production cost and improves the land output rate through the land scale effect, while service-scale management further enhances the agricultural production efficiency through the provision of specialized services (e.g., agricultural machinery services, plant protection services, etc.). The synergy of the two approaches not only reduces transaction costs but also promotes a leap in efficiency through division of labor [16]. On the other hand, the more mature the market of socialized service outsourcing related to crop varieties, the more it can promote the synergy of large-scale growers’ choice of planting varieties, thus reducing the cost of production services. Based on this, more and more new agricultural management subjects choose the model of land transfer and agricultural socialized services to develop large-scale operations.

Synergy is one of the core advantages of dual-scale management. The combination of land-scale management and service-scale management can realize the complementary and optimal allocation of production factors. Land-scale management expands the market demand for service-scale management, while service-scale management provides technical support and management guarantee for land-scale management. This synergistic effect not only enhances the total factor productivity of agricultural production but also strengthens the risk dispersion ability of the agricultural management’s main body and reduces production instability [1]. In addition, dual-scale management further promotes the development of agricultural modernization through technological progress and efficiency improvement [17]. In terms of economic benefits, dual-scale management positively affects household income from grain cultivation, and improves land output rate and labor productivity by optimizing household resource allocation [3]. In terms of production stability, the dual-scale management also has a strong risk-resistant ability, which can alleviate the uncertainty of agricultural production to a certain extent [13]. In summary, dual-scale management can alleviate the dilemma of “big country, small farm” and land fragmentation, and promote the smooth and orderly development of agricultural scale operation. Therefore, Hypothesis 1 is proposed as follows:

Hypothesis 1.

Dual-scale management can improve agricultural operational benefit.

2.2. Comparative Advantages of Dual-Scale Management Versus Land-Scale Management in Enhancing Agricultural Operational Benefit

Dual-scale management breaks through the upper limit of the optimal operating area of traditional single land-scale management by integrating the allocation of production factors and the deepening of the division of labor. The efficiency boundary of traditional land-scale management is limited by the marginal constraints of factors such as labor, capital and technology, and its optimal operating area is determined by the dynamic equilibrium of farming capacity and non-farm opportunity cost. While the intervention of agricultural socialized services reconstructs the production function through specialized division of labor, on the one hand, socialized supply such as agricultural machinery services and technology extension replaces rigid inputs of labor and reduces the marginal cost of management per unit area [18]; on the other hand, the scale of services forms the economy of scope through cross-regional resource sharing to alleviate capital constraints of land scale expansion [19]. In this process, land transfer and service supply form complementary effects, and land concentration provides demand scale for socialized services, while service integration enhances land productivity through technological substitution so that the efficiency inflection point of scale operation is shifted outward. Therefore, dual-scale management breaks through the optimal operation boundary of traditional single land-scale management through factor reallocation and technology embedding, and provides a Pareto improvement path for agricultural modernization. Therefore, Hypothesis 2 is proposed as follows:

Hypothesis 2.

Dual-scale management breaks through the upper limit of optimal operation area for traditional single land-scale management.

The synergistic effect of land-scale management and service-scale management can significantly exceed the enhancement effect of single land-scale management on agricultural operational benefit through the complementarity of horizontal and vertical division of labor. Land-scale management is based on horizontal division of labor, through land transfer and continuous operation to achieve economies of scale, its mechanism is to expand the land area of operation, the marginal cost of mechanization and specialization of production decreases, and centralized investment can share the fixed costs and reduce the production cost per unit area. However, land-scale management is limited by land resource endowment and transaction costs. There is a moderate scale boundary; excessive concentration may lead to increased organizational and management costs [20].

The synergistic mechanism between land-scale management and service-scale management stems from the structural complementarity between horizontal and vertical division of labor. Land-scale management forms a spatial agglomeration effect through continuous planting, which significantly improves the transaction density of the service market and thus reduces the transaction cost of vertical division of labor; service-scale management relies on the advantages of specialized division of labor, and makes up for the shortcomings of human capital and technological spillover in land-scale management through factor substitution effect to form the integration of complementary factor chains. Therefore, dual-scale management not only breaks through the constraint of diminishing returns to scale of single land-scale management, but also constructs a positive feedback mechanism of “horizontal specialization - vertical division of labor deepening” through the expansion of market space and the saving of transaction costs [21]. Therefore, Hypothesis 3 is proposed as follows:

Hypothesis 3.

The effect of dual-scale management on agricultural operational benefit is greater than that of land-scale management.

2.3. Analysis of the Mechanisms by Which Dual-Scale Management Affects Agricultural Operational Benefit

2.3.1. Endogenous Efficiency Improvement Mechanism

Through the organic linkage of land-scale management and service-scale management, dual-scale management forms the endogenous efficiency enhancement mechanism of “deepening division of labor - factor integration - technology adoption - two-way organization - bargaining advantage”, which indirectly enhances agricultural operational benefit. First, from the perspective of division of labor deepening, dual-scale management breaks through the constraint of diminishing marginal returns of a single-scale management, and superimposes the realization of “internal economies of scale” and “external economies of scale”. Secondly, from the perspective of factor integration, dual-scale management promotes the two-way integration of factors through the three-dimensionalization of division of labor by land concentration and technology embedding, which makes up for the inadequacy of single-factor integration in land-scale management and further improves the production efficiency. Third, from the perspective of technology adoption, relative to land-scale operators, dual-scale operators, out of the demand for cost-return control and with the endorsement of the benefits of scale, have the willingness and ability to purchase advanced production technology services to enhance agricultural production efficiency [22]. Fourth, from the perspective of two-way organization, the dual-scale management main body itself is the land-scale management organization body; at the same time, through the purchase of service organization body’s socialized service, they realize the production organization and service organization synergy, so as to improve the production efficiency. Fifth, from the perspective of bargaining advantage, dual-scale operators have a stronger bargaining advantage than small farmers in purchasing socialized services due to the endorsement of scale, which can reduce the service price; at the same time, it can rely on the bargaining advantage of the service body’s agricultural procurement, which not only reduces the economic cost of agricultural inputs, but also saves the time cost of the market bargaining, which in turn improves the efficiency of agricultural production and operation.

2.3.2. Exogenous Risk Burden Mechanism

The synergy of land-scale management and service-scale management, through the threefold path of risk defense, risk transfer, and risk compensation, reduces the natural risk and market risk faced by agricultural production, and then enhances agricultural operational benefit. First, at risk defense mechanism level, dual-scale operators, as land scale operators, tend to be more active through increasing agricultural facilities to reduce the probability of extreme weather damage to a single piece of land, and through the adoption of biochemical technology to reduce the application of biological disaster losses, thereby buffering the impact of natural risks. At the same time, as a buyer of specialized services, they are more able to conveniently obtain standardized technical support and share market information, thus reducing the risk of information asymmetry [9]. Secondly, at the level of the risk transfer mechanism, on the one hand, in order to diversify the production risk and stabilize the expected income, the dual-scale operators are more willing to purchase agricultural insurance, so as to transfer their own production risk to the agricultural insurance company [23]; on the other hand, through the purchase of agricultural socialized services, the dual-scale operators will transfer the risk of the agricultural socialized service operation and the risk of the investment of heavy assets in the production process to the agricultural socialized service main body. Finally, at the level of the risk compensation mechanism, the relevant subsidies to support the land transfer operation improve the income stability expectation of dual-scale operators by reducing the cost of land transfer and the risk of income fluctuation, while the relevant subsidies to support the socialized services promote the specialized division of labor and technology spillover by reducing the cost of socialized services. Dual-scale management enjoys the above two types of subsidies at the same time, which not only reduces the risk of income fluctuation, but also hedges the risk of production cost fluctuation, so as to achieve the effect of front-loaded compensation for the uncertainty risk of large-scale agricultural operation, and promote the improvement of management benefit in both directions. In summary, Hypothesis 4 is proposed as follows:

Hypothesis 4.

Dual-scale management improves agricultural operational benefit through endogenous efficiency improvement mechanism and exogenous risk burden mechanism.

3. Research Design

3.1. Data

With the diverse topography and landscape in the domain of Anhui, China, and the various forms of agricultural scale operations, it is of value for practical reference to study the impact of coordinated scale development on the improvement of agricultural business efficiency in Anhui. The data used in this study were obtained from a survey of 2295 households in 309 villages in 16 prefectural-level cities in Anhui Province, China, in 2024, which resulted in 2166 valid samples, with a validity rate of 94.4%. Of the 2166 households surveyed, 788 households that both transferred land and purchased agricultural socialization services were set up as dual-scale management subjects; 844 households that only transferred land were set up as land-scale management subjects, and the remaining 534 households were small-scale farmers. This study compares dual-scale management with land-scale management to measure whether dual-scale management can improve agricultural operational benefit. This study employs quantitative research methods to quantify the causal relationship and influencing mechanisms between dual-scale management and agricultural operational benefit. This approach is suitable for statistical analysis and hypothesis testing on large sample data, thereby ensuring the objectivity and generalizability of research conclusions. The sample sources and distribution are presented in

Table 1. Sample Sources and Distribution (Households).

City Name	Urban Sample	Smallholder Farmers	Land Scale Management	Dual-Scale Management
Hefei	326	86	128	112
Wuhu	76	7	19	50
Bengbu	99	25	36	38
Hauinan	134	39	52	43
Maanshan	87	21	26	40
Huaibei	186	38	103	45
Tongling	39	4	22	13
Anqing	142	37	49	56
Huangshan	42	12	16	14
Chuhzou	217	68	72	77
Fuyang	118	22	46	50
Suzhou	111	24	53	34
lvan	153	41	61	51
Bozhou	157	37	51	69
Chizhou	117	27	45	45
Xuancheng	162	46	65	51
Total	2166	534	844	788

.

3.2. Model Setting

3.2.1. Benchmark Regression Model

The purpose of this paper is to investigate whether dual-scale management can improve agricultural operational benefit, and the OLS model is used to determine the effect of dual-scale management on agricultural operational benefit. In order to compare the difference between the effect of dual-scale management and single-scale management on the improvement of agricultural operational benefit, this paper constructs a multiple regression model. The OLS model and multiple regression model settings are as follows:

$$Y_i={\beta }_0+{\beta }_1{X}_i+{\beta }_2{{\rm Z}}_i+{\epsilon }_i$$

(1)

$$Y_i={\beta }_0+{\beta }_1{D}_{i1}+{\beta }_2{D}_{i2}+{\beta }_4{Z}_i+{\epsilon }_i$$

(2)

In Equation (1), i denotes the farm household, Y_i is the explanatory variable agricultural operational benefit, X_i is the core explanatory variable dual-scale management, Z_i is the set of control variables, including household head characteristics and production characteristics, β₀~β₁ is the parameter to be estimated, and ε_i is the random error term. In Equation (2), D_i1 and D_i2 are dummy variables, D_i1 is whether or not to adopt dual-scale management; if dual-scale management is adopted, it is assigned a value of 1, and otherwise it is 0. D_i2 is whether or not to adopt only land scale operation; if only land scale operation is adopted, it is assigned a value of 1, and otherwise it is 0, and the meanings of other variables are the same as that in formula Equation (1).

3.2.2. Mediating Effects Test Model

After clarifying whether dual-scale management can improve agricultural operational benefit, the potential intermediate effect mechanism needs to be further investigated. Referring to the approach of Wen and Ye [24], the following model is established to investigate:

$$Y_i={\alpha }_0+{\alpha }_1{X}_i+{\alpha }_2{{\rm Z}}_i+{\epsilon }_i$$

(3)

$$M_i=b_0+b_1{X}_i+b_2{{\rm Z}}_i+{\epsilon }_i$$

(4)

$$Y_i=c_0+c_1{X}_i+c_2{M}_i+c_3{{\rm Z}}_i+{\epsilon }_i$$

(5)

where M_i denotes the endogenous efficiency improvement mechanism and the exogenous risk-burdening mechanism, and the other variables have the same meaning as in Equation (1).

3.3. Variable Measurement and Description

3.3.1. Explained Variable: Agricultural Operational Benefit (AOB)

The measurement of agricultural operational benefit can be reflected by constructing a comprehensive indicator evaluation system and synthesizing an index, or it can be approximated by indicators such as agricultural income and total farm household income [25]. Per-mu yield represents the most intuitive and core indicator for land productivity and management returns. This metric is widely used in relevant empirical studies, offering strong comparability and explanatory power [26]. Although a comprehensive index encompassing multiple dimensions such as economic and ecological aspects would be more thorough, this study’s questionnaire primarily focused on land management practices and outputs, and did not systematically collect all the data required to construct such a comprehensive index. Therefore, this paper uses the per-mu profit from farmland management to represent agricultural operational benefit. This per-mu profit represents the net profit after deducting all costs.

3.3.2. Core Explanatory Variable: Dual-Scale Management (DM)

Since land-scale management and service-scale management are synergistic relationships that support and promote each other, the coupling coordination degree model can reflect the degree of association and synergistic effect of two or more subsystems, and this paper carries out a three-step process to measure the degree of the coupling coordination of dual-scale management.

Step 1: Indicator selection. Dual-scale management main body is also the land-scale management main body; referring to the research of Zheng et al. [27], the indicators of land transfer are decomposed into operation support and operation structure. Operation support aims to assess the physical attributes of the transferred land as well as the land operation potential. Specifically, the number of transferred plots characterizes the scale effect of land integration, the degree of terrain leveling reflects the feasibility of mechanized operations, and the proportion of high-standard farmland maps the ability of infrastructure to support large-scale operations, which together constitute the basic conditions for land contiguity, mechanization and sustainable use. The management structure focuses on the rationality of the structure of the transferred land, and consists of two dimensions: the area of land management and the number of planted crops, of which the area of land management reflects the scale of land management, while the number of planted crops reveals the level of specialization of land management to a certain extent.

In terms of purchasing agricultural socialized services, referring to the research of Song et al. [28], the indicators of purchasing agricultural socialized services are decomposed into two first-level indicators of purchasing willingness and purchasing demand, which are measured from the perspectives of subjective attitudes and objective needs. Purchasing willingness is comprehensively characterized by three binary variables: first, whether to tend to use the service scale operation, reflecting the basic preference of writing synergistic scale households for the service scale model; second, whether to be willing to adopt the service scale operation organization form, reflecting the degree of recognition of the synergistic nature of the service supplying body, for example, the tendency to select the organization form of cooperatives, enterprise trusteeship, etc.; and third, whether agricultural socialized services can enhance the benefits, based on the perceived value theory, measures the synergistic scale households’ expected judgment of the economic gains, production efficiency, or risk aversion brought by the service. Purchase demand is then quantified by the cost of socialization services per mu and the amount of owned farm machinery. The former measures the actual demand for agricultural socialization services by cooperative-scale households in terms of the average mu socialization service expenditure. The latter is due to the substitution effect between the stock of owned farm machinery and the demand for external services; the higher the ownership of farm machinery, the lower the purchasing demand of cooperative scale households for mechanical services. The comprehensive evaluation system of the synergistic degree of scale operation is shown in

Table 2. Comprehensive evaluation system for dual-scale management.

Subject of Evaluation	Primary Indicators	Secondary Indicators	Orientations	Weights
Land transfer	Operation support	Number of operating plots	−	0.009
		Level of land formation (1 = Mountainous; 2 = Hilly; 3 = Plain)	+	0.080
		Percentage of high-standard farmland (1 = All; 2 = Some; 3 = All)	+	0.351
	Operation structure	Land operation area (Mu)	+	0.537
		Number of crop types planted	−	0.023
Purchase of agricultural socialization services	Purchasing willingness	Whether there is a preference for using services on a large scale (1 = Yes; 0 = No)	+	0.372
		Whether to be willing to adopt the service scale operation organization form (1 = Yes; 0 = No)	+	0.136
		Whether agricultural socialization services can enhance benefits (1 = Yes; 0 = No)	+	0.490
	Purchasing demand	Cost of socialized services per acre (Yuan)	−	0.001
		Number of owned farm machinery	−	0.001

.

Step 2: Assignment of weights. In this paper, we refer to the research process of Cheng et al. [29] on the data entropy method to determine the weights of the indicators of land transfer and purchase of agricultural socialization services. It should be noted that the weights determined by the entropy method are data-driven and objective, based on the principle that a higher dispersion of indicator values corresponds to greater weight. The relatively low weights assigned to indicators under the “Purchasing demand” dimension in Table 2 reflect that, within this study’s sample, dual-scale farming households primarily consist of grain-growing entities with highly similar crop structures. Consequently, their demand for socialized services and average socialized service costs per mu exhibit significant homogeneity. Given that the sampled households predominantly operate at scales between 100 and 300 mu, and most own only 1 to 3 units of agricultural machinery, which is insufficient for the full mechanization of agricultural production, they therefore rely heavily on purchased socialized services. Consequently, the two indicators reflecting purchasing demand exhibit minimal variation across samples and high entropy values, resulting in lower assigned weights. This objectively reflects the common characteristics of service demand investment among China’s new agricultural operators.

Step 3: In this paper, based on the calculation process of the coupling coordination degree model, the formula for the two-dimensional coupling degree is defined as follows:

$$C=2\times {\left[{\displaystyle \frac{U_1{U}_2}{{\left(U_1+U_2\right)}^2}}\right]}^{{\displaystyle \frac{1}{2}}}$$

(6)

where C is the coupling degree value of land transfer and purchase of agricultural socialized services, U₁ is the degree of land transfer, and U₂ is the degree of purchase of agricultural socialized services. In order to more accurately measure the degree of coordination between land transfer and purchase of agricultural socialized services, the coupling coordination degree model is introduced for the next calculation. Its formula is as follows:

$$D=\sqrt{C\times T}$$

(7)

$$T=a{U}_1+b{U}_2$$

(8)

where D represents the coupling coordination degree of land transfer and purchase of agricultural socialized services, T represents the synergy degree of land transfer and purchase of agricultural socialized services, a and b are the weight coefficients of the system, and since land transfer and purchase of agricultural socialized services are two systems of equal importance in the process of integration, the two weight coefficients a = b = 0.5 are assigned. Finally, dual-scale management is obtained by calculating in accordance with the above formula as the paper’s core explanatory variables.

3.3.3. Mechanism Variables

Referring to the study of Yang et al. [30], this paper takes agricultural socialized service spending as the proxy variable of endogenous efficiency improvement mechanism (EI). Agricultural socialized services have a significant positive correlation with agricultural production efficiency. Agricultural socialized services help large-scale business entities overcome resource constraints and strengthen financial liquidity through the provision of technical support and the leasing of agricultural machinery, thus enhancing production efficiency. The cost of agricultural socialized services directly reflects the input of large-scale operators in the purchase of socialized services. Under the premise of the established price standard of socialized services in each segment, the more money spent on the purchase of agricultural socialized services, the higher the production efficiency caused by the increase in the degree of socialized services. This indicator primarily measures the input-intensity pathway of efficiency improvements.

Exogenous risk burden mechanism (RB). The exogenous risk-bearing mechanism referred to by this research institute primarily addresses external and systemic risks encountered in agricultural production and operations, such as natural hazards and market risks, specifically those risks that are insurable and amenable to insurance relief. First, this paper takes the full-cost insurance payout rate as a proxy variable for the risk transfer mechanism (RT). The full-cost insurance payout rate can directly reflect the effect of transferring natural disasters, market fluctuations, and other risks to the insurance market. At the same time, it closely fits the core of “risk transfer” theory, and can also dynamically monitor the actual solvency of external subjects in disaster. Secondly, the risk compensation mechanism (RC) adopts whether or not to enjoy agricultural insurance subsidies as a proxy variable, and the subsidies directly compensate for the losses of the main body of the operation through the financial transfer payment, the scale and structure of which directly affects the main body of the operation’s risk hedging ability. Finally, the risk defense mechanism is excluded due to the universal character of agricultural facilities.

3.3.4. Control Variables

According to the theory as well as the related literature [27], this paper selects control variables from two dimensions: subject characteristics and production characteristics. Among them, subject characteristics include gender, age, education level and non-farm employment; production characteristics include planting experience, the amount of agricultural production, the degree of production organization, agricultural insurance, and the degree of contract standardization. The meanings of the variables and the results of descriptive statistical analysis are shown in

Table 3. Definition of variables and results of descriptive statistical analysis.

Variables	Variable Definition and Assignment	Mean	S.D.
AOB	Land management revenue per mu (Yuan)	902.815	1051.281
DM	Purchase of socialized services while transferring land (as measured by the coupled coordination degree model)	0.561	0.160
EI	Expenditure on socialized services in the main year (Yuan)	5.054	1.202
RT	Full cost insurance payout ratio (%)	47.234	17.722
RC	Availability of agricultural insurance subsidies (1 = Yes; 0 = No)	0.468	0.499
Gender	Gender of household head (1 = Male; 0 = Female)	0.751	0.433
Age	Age (16–35 = 1, 35–50 = 2, 50–60 = 3, 60+ = 4)	2.152	0.792
Education	Subject’s literacy (Elementary school and below = 1, Middle school = 2, High school = 3, University = 4)	3.127	0.859
Non-farm employment	Whether the subject is also participating in non-farm work (1 = Yes; 0 = No)	0.675	0.469
Planting experience	Years of primary engagement in agricultural operations (Year)	12.520	10.343
Number of producers	Number of household members engaged in agricultural production (Persons)	2.344	1.261
Organization	Organizational Level: Family Farm Only = 1 Farmer Cooperative = 2 Agricultural Company = 3	1.680	0.948
Agricultural insurance	Is the principal party participating in full cost insurance (1 = Yes; 0 = No)	0.725	0.447
Standardization	Whether local governments promote the use of standardized and uniformly developed contracts (1 = Yes; 0 = No)	0.797	0.403
Land area	Actual land operation area (Mu), in logarithms	3.538	2.323

.

4. Results

4.1. Baseline Regression

Table 4. Baseline regression results.

Variables	(1)	(2)	(3)	(4)
	Dual-Scale	Dual-Scale	Dual-Scale	Land-Scale
	AOB	AOB	AOB	AOB
DM	469.246**	441.299 *
	(235.421)	(241.591)
Land area			126.384 *	398.446 ***
			(67.527)	(77.765)
Land area			−11.224 ***	−39.688 ***
			(3.723)	(7.410)
Gender	−152.697 *	−160.949 *	−78.550	89.789
	(89.216)	(89.521)	(90.053)	(111.959)
Age	58.421	29.020	46.787	100.701
	(56.150)	(62.616)	(62.097)	(64.234)
Education	48.378	61.037	46.407	2.181
	(51.936)	(52.344)	(52.026)	(49.077)
Non-farm employment	229.635 ***	236.228 ***	111.369	203.510 **
	(85.638)	(87.390)	(89.118)	(81.430)
Planting experience		5.503	3.436	−9.282 **
		(4.271)	(4.258)	(4.648)
Number of producers		27.952	66.075 **	59.674 *
		(30.671)	(31.591)	(30.578)
Organization		2.835	10.575	154.741 **
		(40.269)	(39.647)	(73.141)
Agricultural insurance		−152.102 *	−99.774	−247.937 ***
		(85.237)	(84.708)	(89.085)
Standardization		−74.375	−53.510	16.075
		(94.836)	(94.120)	(50.290)
_cons	322.349	423.976	595.079 *	−293.177
	(294.110)	(340.390)	(331.569)	(321.020)
Sample Size	788	788	788	844
R2	0.020	0.028	0.038	0.069

Note: ***, **, and *, respectively, represent significant levels at 1%, 5%, and 10% statistical levels. The values in parentheses represent the corresponding robust standard errors.

reports the regression results on the impact of dual-scale management on agricultural operational benefit. Column (1) presents results with only household-level characteristics included, while Column (2) shows regression results controlling for production-level characteristics on top of Column (1). The results in Columns (1) and (2) indicate that dual-scale management consistently and significantly enhances agricultural operational benefit regardless of whether other characteristics are included, validating Hypothesis 1.

Additionally, Column (3) presents the impact of land area under dual-scale management on agricultural operational benefit, while Column (4) shows the effect of land area under land-scale management. The results in columns (3) and (4) reveal that the impact of land management scale on agricultural operational benefit for both dual-scale operators and land-scale operators is not strictly linear but follows an inverted U-shaped relationship. This study’s findings of an inverted U-shaped relationship between farm size and agricultural profitability align closely with prevailing theoretical consensus and numerous empirical conclusions in agricultural economics. For instance, Wang et al. [31], based on a longitudinal survey of farmers in the Jianghan Plain, identified an inverted U-shaped relationship between rice field size and scale efficiency. Similarly, Ji et al. [32], using data from rice growers in Shanghai, confirmed the nonlinear impact of scale on efficiency. The underlying economic logic lies in the trade-off between economies of scale and diseconomies of scale. Moderate-scale operations enhance efficiency by spreading fixed costs and promoting mechanization and specialization. However, excessive scale expansion leads to diminishing marginal returns due to surging management coordination difficulties, rising oversight costs, and resource misallocation.

Simultaneously, based on the results in columns (3) and (4), the optimal operating scale for dual-scale management farmers is calculated to be 278 mu, while that for land-scale management farmers is 152 mu. Dual-scale management significantly raises the optimal scale threshold from 152 mu to 278 mu compared to pure land-scale management. The core mechanism lies in effectively offsetting diseconomies of scale through service-scale management. Agricultural socialized services, through specialized division of labor, simultaneously replace costly internal management and supervision investments by operators, and spare large-scale farmers from massive fixed-asset investments in highly specialized heavy agricultural machinery. This significantly slows the rise in marginal costs associated with scale expansion. This implies that dual-scale management, through the synergy of “land-scale management” and “service-scale management”, reconstruct the agricultural production function and extend the boundaries of economies of scale. This provides a viable pathway to overcome the bottleneck of scale operations within China’s fragmented and small-scale agricultural structure. Hypothesis 2 was validated.

Table 5. Per-mu income outcomes for dual-scale management, land-scale management, and smallholder farmers.

Variables	Average Yield per Mu (Yuan)	Sample Size
Dual-scale management	902.8	788
Land-scale management	658.3	844
Mallholder farmers	433.8	534

presents the average per-mu income for dual-scale management, land-scale management, and smallholder farming within the sample. The results indicate that dual-scale management yields an additional ¥244.5 per mu compared to land-scale management and ¥469 per mu compared to smallholder farming. Thus, dual-scale management demonstrates the most significant impact on enhancing agricultural operational benefit.

Additionally, the results of the multiple regression analysis for different scale management approaches are presented in

Table 6. Multiple regression results for different scale management approaches.

Variables	(1)	(2)
	AOB	AOB
Whether to adopt dual-scale management	330.785 ***	321.151 ***
	(64.059)	(64.409)
Whether to adopt land-scale management	306.586 ***	285.159 ***
	(63.330)	(62.721)
_cons	378.108 ***	33.801
	(51.997)	(176.046)
Controls	NO	YES
Sample Size	2166	2166
R2	0.014	0.028

Note: *** respectively, represents significant levels at 1% statistical levels. The values in parentheses represent the corresponding robust standard errors. Column (1) reports estimates from the baseline model without control variables. Column (2) reports estimates from the model including the full set of control variables.

. As shown in Table 5, dual-scale management and land-scale management alone significantly improved agricultural management efficiency at the 1% significance level. However, compared to land-scale management, dual-scale management yielded a larger coefficient, indicating that adopting dual-scale management has a greater effect on enhancing agricultural operational benefit. This validates Hypothesis 3.

4.2. Robustness Test

4.2.1. Instrumental Variables Method

To avoid potential endogeneity issues in the regression analysis of the aforementioned model, this study has endeavored to control for the influence of characteristic variables on the regression results. However, factors affecting agricultural operational benefit cannot be fully controlled, leading to potential omitted variable issues. Additionally, dual-scale management may impact agricultural operational benefit, while an improved benefit could in turn further promote dual-scale management—suggesting a possible bidirectional causal relationship between dual-scale management and agricultural operational benefit. To address this, this paper selects contracted land area as an instrumental variable for dual-scale management. This instrumental variable satisfies both the relevance and exogeneity constraints: On one hand, household contracted land area is a key determinant of farmers’ ability to engage in scaled operations. Larger contracted land areas provide farmers with more resources and space, thereby supporting the implementation of dual-scale management, thus meeting the relevance constraint for instrumental variables. On the other hand, the size of land contract area is largely determined by historical policies and land distribution, endowing it with a degree of exogeneity that limits its direct influence on agricultural operational benefit, thus satisfying the instrument’s exclusivity constraint. The regression results for the instrumental variable are presented in

Table 7. Instrumental variables regression results.

Variables	(1)	(2)
	DM	AOB
Land contract area	0.009 ***
	(0.001)
DM		2164.663 *
		(1303.476)
Controls	YES	YES
Sample Size	788	788
F-test (1st stage)	44.636
LM (p-value)	0.000
R2	0.070	0.070

Note: *** and *, respectively, represent significant levels at 1% and 10% statistical levels. The values in parentheses represent the corresponding robust standard errors.

. To further examine the validity of the instrumental variables, we present a series of formal test results. First, the F-statistic for the first-stage regression is 44.636, strongly rejecting the null hypothesis of “weak instrumental variables”. Second, the p-value for the LM test is 0.000, significantly rejecting the null hypothesis of “insufficient identification of the instrumental variables” at the 1% level. Since the number of instrumental variables equals the number of endogenous variables, an over-identification test cannot be performed. These results indicate that the instrumental variables we selected are valid and reliable. The regression results show a highly positive correlation between the instrumental variable and dual-scale management. In the two-stage regression, the impact of dual-scale management on agricultural operational benefit remains significant, further ensuring the robustness of the existing regression results.

4.2.2. Propensity Score Matching (PSM)

To address the issue of self-selection between dual-scale management and agricultural operational benefit, this study further employs a 1:4 propensity score matching method to construct counterfactual samples for estimation. Specifically, first, a control group (low dual-scale management) highly similar to the treatment group (high dual-scale management) was constructed based on the average dual-scale management; then, corresponding treatment and control groups were defined based on score-based proximity. Propensity score matching estimation utilizes the matched control group to approximate the counterfactual outcome of the treatment group as closely as possible, thereby revealing the difference in agricultural operational benefit between the two groups of farmers. Since propensity score matching employs a common domain sample, and the matched samples from both groups generally achieve balanced covariates, this approach resolves the self-selection issue. When the absolute values of standardized deviations for all variables post-matching were below 10% of empirical values, satisfying the balance assumption, the average treatment effect estimates after matching for both household characteristics and production characteristics are shown in

Table 8. Estimation results of the dual-scale management average processing effect.

Variables	Processing Effect	Control Group	Treatment Group	Difference	S.D.	T-Value
AOB	Unmatched	931.735	868.783	62.952	75.163	0.84
	ATT	933.536	779.434	154.102	78.402	1.97 **

Note: ** respectively, represent significant levels at 5% statistical levels.

. Compared to low-dual-scale farmers, high-dual-scale farmers achieved an additional ¥154.1 per mu in average revenue, with these differences passing significance tests at least at the 5% level. This indicates that even when accounting for self-selection issues in dual-scale management, dual-scale management significantly enhances agricultural operational benefit, thereby reaffirming Hypothesis 1.

4.2.3. Additional Robustness Checks

To ensure the robustness of the regression results above, the following robustness checks were conducted: (1) exclusion of outliers; and (2) replacement of the core explanatory variable with a dummy variable. The results are shown in

Table 9. Additional robustness checks.

Variables	(1)	(2)
	AOB	AOB
DM	408.199 **
	(190.590)
Whether to adopt dual-scale management		129.890 ***
		(47.843)
_cons	520.676 **	129.939
	(226.184)	(172.815)
Controls	YES	YES
Sample Size	734	2166
R2	0.062	0.018

Note: *** and **, respectively, represent significant levels at 1% and 5%, statistical levels. The values in parentheses represent the corresponding robust standard errors.

. First, we directly excluded samples at the highest and lowest 2% of the distribution for both agricultural operational benefit and dual-scale management. After re-running the benchmark regression on these two adjusted datasets, the magnitude and significance level of the coefficient for dual-scale management remained highly consistent with the main regression results, indicating that our findings are not driven by extreme values. Second, we use “whether to adopt dual-scale management” as the core explanatory variable and conducted robustness tests on the full sample. The results showed that the core independent variable had a significantly positive impact on agricultural operational benefit, confirming that dual-scale management significantly promotes agricultural operational benefit. These regression results align with the previous findings, further validating the robustness of our conclusions.

4.3. Heterogeneity Analysis

4.3.1. Terrain Heterogeneity

Considering the variations in external conditions such as agricultural production capabilities and mechanization levels across different geographical terrains, this study categorizes the sample into three regions—mountainous, hilly, and plains—and conducts benchmark model regressions for each. As shown in the regression results of columns (1) to (3) in

Table 10. Heterogeneity regression results.

Variables	(1)	(2)	(3)	(4)	(5)	(6)
	Mountainous	Hilly	Plains	All Incomplete	Partially Completed	All Completed
DM	746.683	151.352	596.922 *	−395.293	42.291	1708.819 *
	(791.017)	(357.501)	(353.524)	(416.803)	(356.230)	(996.487)
Controls	YES	YES	YES	YES	YES	YES
_cons	−1434.346	226.497	777.415	885.811	378.716	711.158
	(1155.729)	(536.101)	(472.437)	(555.771)	(397.228)	(1565.429)
Sample Size	84	263	441	293	375	120
R2	0.124	0.077	0.035	0.046	0.047	0.073

Note: *, respectively, represent significant levels at 10% statistical levels. The values in parentheses represent the corresponding robust standard errors.

, the impact of dual-scale management on the agricultural operational benefit of households in the plains region is significant at the 10% level, whereas its effect on agricultural operational benefit in mountainous and hilly regions is not significant. A possible reason is that plains regions feature relatively concentrated and contiguous land resources, making them suitable for dual-scale management. Additionally, higher mechanization levels in these areas can effectively enhance land productivity and labor efficiency, thereby significantly improving agricultural operational benefit. In contrast, mountainous and hilly regions have complex topography, fragmented land parcels, and steep slopes, which constrain the implementation of dual-scale management.

4.3.2. Heterogeneity in Scale Operation Conditions

Significant variations exist in soil quality, infrastructure, and production potential across farmlands, representing critical factors that cannot be overlooked in studies of agricultural operational benefit. This study categorizes all farming households based on the extent of high-standard farmland construction: households that have not undertaken high-standard farmland construction, households that have partially completed high-standard farmland construction, and households that have fully completed high-standard farmland construction. Regression analysis is conducted on these groups. As shown in columns (4) to (6) of Table 10, dual-scale management significantly impacts the agricultural operational benefit of households with fully completed high-standard farmland at the 10% significance level, whereas its effect is insignificant for households with partially completed or no high-standard farmland development. A possible explanation is that high-standard farmland possesses stronger production potential and resource endowments, making it better suited to leverage the technological advances and efficiency gains brought by dual-scale management. In contrast, farmland with partial or no high-standard construction struggles to fully exploit the advantages of dual-scale management due to natural constraints or inadequate infrastructure, resulting in limited improvements in operational benefit.

4.4. Mechanism Analysis

Table 11. Mediation Effect Regression Results.

Variables	(1)	(2)	(3)	(4)	(5)	(6)
	EI	AOB	RT	AOB	RC	AOB
DM	0.523 *	413.833 *	8.088 **	408.348 *	0.464 ***	342.738
	(0.274)	(241.880)	(3.577)	(242.099)	(0.113)	(243.133)
EI		52.492 *
		(31.552)
RT				4.074 *
				(2.420)
RC						212.204 ***
						(76.157)
Controls	YES	YES	YES	YES	YES	YES
_cons	4.127 ***	207.335	39.950 ***	261.227	0.085	405.966
	(0.387)	(364.086)	(5.040)	(353.469)	(0.160)	(338.979)
Sample Size	788	788	788	788	788	788
R2	0.041	0.031	0.237	0.031	0.052	0.038

Note: ***, **, and *, respectively, represent significant levels at 1%, 5%, and 10% statistical levels. The values in parentheses represent the corresponding robust standard errors.

reports the estimation results for the two types of mediating effects of dual-scale management. First, columns (1) and (2) present the regression results for the endogenous efficiency enhancement mechanism. It can be observed that the regression coefficient for dual-scale management in Column (1) and the regression coefficient for endogenous efficiency enhancement in Column (2) both pass the significance test. This indicates that dual-scale management significantly enhances farmers’ production efficiency, thereby improving agricultural operational benefit and validating the existence of the endogenous efficiency enhancement mechanism. Second, the regression results for the risk transfer mechanism in columns (3) and (4) show that the regression coefficient for dual-scale management in Column (3) and the regression coefficient for risk transfer in Column (4) both pass the significance test. Dual-scale management significantly enhances risk transfer capacity, thereby improving agricultural operational benefit and verifying the existence of the risk transfer mechanism. Finally, the regression results for the risk compensation mechanism in Column (5) and Column (6) show that the regression coefficient for dual-scale management in Column (5) and the regression coefficient for risk compensation in Column (6) both pass the significance test. This indicates that the enhanced risk compensation capacity brought about by dual-scale management promotes the improvement of agricultural operational benefit, verifying the existence of the risk compensation mechanism. Therefore, Hypothesis 4 is validated.

5. Conclusions and Recommendations

5.1. Conclusions

Based on 2166 farmer questionnaires from Anhui Province, this study employs a coupling coordination degree model to measure dual-scale management and utilizes multiple regression analysis to empirically examine the impact of dual-scale management on agricultural operational benefit and its underlying mechanisms. The findings reveal the following: First, dual-scale management significantly enhances agricultural operational benefit. This indicates that integrating land-scale management with service-scale management generates added value through synergistic effects, providing an effective pathway to improve agricultural performance. Second, further analysis reveals that dual-scale management substantially raises the optimal scale threshold for pure land-scale operations from 152 mu to 278 mu, with a significantly greater impact on agricultural operational benefit than land-scale management alone. This confirms the superiority of the “land and services” dual-scale management across both scale boundaries and efficiency intensity dimensions, demonstrating how it achieves Pareto improvements in economies of scale through factor restructuring and deepened division of labor. Third, dual-scale management enhances agricultural operational benefit through the combined effects of an endogenous efficiency enhancement mechanism and an exogenous risk-sharing mechanism. This reveals its intrinsic logic for boosting efficiency and stabilizing benefits, integrating the relatively independent research dimensions of efficiency and risk within a unified framework. Fourth, the benefits of dual-scale management are more pronounced in flatland areas, and those farmers have fully completed high-standard farmland. This indicates that the model’s scalability is significantly constrained by topographical conditions and infrastructure completeness, providing empirical support for implementing differentiated regional policies.

5.2. Discussion

This study empirically verifies the significant enhancement of agricultural operational benefit through dual-scale management combining land and services, while revealing its underlying mechanisms and applicability boundaries. The discussion in this section aims to delve deeper into these findings and engage with the existing literature to clarify the theoretical contributions of this research.

First, the core finding is that dual-scale management is not only effective but represents a superior pathway compared to single land-scale management. This conclusion provides robust empirical support for the current transformation of agricultural scale management in China. It aligns closely with Xu et al.’s argument that agricultural scale management should shift toward a “land and services” collaborative pathway [1], substantiating this view with rigorous empirical evidence. Our research reveals that dual-scale management achieves the superposition of “internal economies of scale” and “external economies of scale” through deepening horizontal and vertical division of labor. Although previous studies have demonstrated that land and services achieve economies of scale through two-way scaling, thereby enabling a division of labor economy [16], they have not examined the impact of dual-scale management on enhancing agricultural operational benefit from the perspective of integrating socialized services based on land transfers. Furthermore, whether this model can significantly boost agricultural operational benefit remains debatable. Therefore, this paper advances the field by empirically quantifying the incremental benefits (a ¥244.5 increase in revenue per mu) from this synergistic effect through a coupled coordination degree model and comparative research design. This confirms that the synergistic path is not only feasible but also superior, highlighting the importance of moving beyond a singular focus on land expansion in the pursuit of agricultural scale operations.

Second, the findings on the optimal operating scale have significant theoretical implications. This study confirms an inverted U-shaped relationship between land scale and agricultural profitability, consistent with prevailing agricultural economics consensus [31,32]. Yet, it deepens this understanding by revealing that dual-scale management can significantly extend the optimal threshold from 152 mu to 278 mu. This finding challenges the conventional view that diseconomies of scale inevitably accelerate with increasing land size. Its underlying logic lies in the embedded agricultural socialized services, which effectively offset diseconomies such as rising management complexity and increased oversight costs through specialized division of labor [20]. Service-scale management reconfigures the production function, enhancing the marginal productivity elasticity of factors like capital and technology, thereby delaying the onset of diminishing returns to scale. This offers a fresh perspective on understanding the dynamic boundaries of economies of scale in modern agriculture.

Third, the heterogeneity analysis reveals the boundary conditions for realizing the benefits of dual-scale management. Findings indicate that the positive effects of dual-scale management are more pronounced in flatland areas and high-quality plots. This aligns with Zheng et al.’s assertion that the effectiveness of socialized services is constrained by topographical conditions, further clarifying the specific scenarios where such benefits materialize [12]. In flatland regions, contiguous land parcels provide fertile ground for mechanized services, significantly reducing transaction costs associated with service provision and fully unlocking the potential of a division-of-labor economy. Meanwhile, the well-developed infrastructure (such as irrigation systems and roads) represented by “high-condition” farmland serves as the foundational platform for leveraging the synergistic effects of “land and services”. This indicates that dual-scale management are not a universal “one-size-fits-all” solution; their benefits are highly dependent on specific resource endowments and infrastructure conditions. This suggests that policy formulation must be tailored to local conditions and adopt differentiated approaches.

Finally, the mechanism analysis not only validates theoretical assumptions but also deepens our understanding of the synergistic essence of dual-scale management. This study confirms two key pathways: endogenous efficiency gains and exogenous risk sharing. This aligns with Yang et al. ’s finding that socialized services enhance technical efficiency [9] and Chen’s emphasis on agricultural insurance’s risk transfer function [8]. The contribution lies in integrating these relatively independent pathways into a unified “dual-scale” framework, elucidating their combined effects. The efficiency enhancement mechanism originates from the land scale creating market demand for the service scale, while the service scale feeds back to improve land productivity through technology spillovers and factor substitution. The risk-sharing mechanism stems from the heightened risk awareness of dual-scale operators, which in turn fosters greater willingness to purchase insurance. This enables the broader dispersion and transfer of natural and market risks. The synergy between these two mechanisms constitutes the fundamental reason why dual-scale management enhances operational stability and sustainable profitability.

5.3. Recommendations

The findings provide significant implications for advancing dual-scale management and agricultural operational benefit. First, policy resources should be prioritized in plain areas to establish models of “whole-industry-chain socialized services”. The government can guide and support the integration of service entities to provide integrated solutions from plowing, planting, management, and harvesting to drying, storage, and marketing for large-scale operators. The main challenge lies in coordinating multiple service providers and establishing benefit-sharing mechanisms. Initial costs are high, requiring government seed funding to build demonstration platforms. However, economies of scale can reduce average costs once operational.

Second, instead of blindly pursuing ultra-large scale, policies should guide new agricultural operators to develop moderately scaled operations. Local governments should improve the land transfer market, with a particular focus on providing subsidies for land consolidation to reduce the transaction costs for operators achieving contiguous land management. The challenge is that land consolidation requires negotiating with numerous smallholders. Fiscal subsidies are needed to cover consolidation costs. Utilizing high-standard farmland construction funds for this purpose can improve investment efficiency.

Third, local governments should change from broad subsidies to targeted ones, providing vouchers or direct subsidies for key production links (e.g., integrated pest management, precision fertilization) within agricultural socialized services, especially for small and medium-sized scale operators, to lower the threshold for adopting advanced services. Design insurance products should be linked to the dual-scale model, for example, by developing “scale management comprehensive income insurance” that covers both natural risks and market price fluctuations. Subsidy ratios should be increased for full-cost insurance for dual-scale management who achieve a certain service coverage level. Subsidizing services requires precise identification and supervision to prevent fraud. Designing new insurance products involves complex actuarial calculations and requires close collaboration between government departments and insurance companies. The government needs to bear most of the premium subsidies, but this can effectively stabilize operator expectations.

Fourth, high-standard farmland construction projects should be prioritized in areas with active land transfer and high potential for scale management. Construction standards must meet the needs of large-scale mechanization, focusing on field regularity, irrigation facilities, and road connectivity. This requires significant and sustained investment. In addition to government financial input, exploring Public–Private Partnership (PPP) models to attract social capital is feasible, but clear benefit-sharing mechanisms (e.g., operators paying appropriate usage fees) must be established to ensure sustainability.

5.4. Research Limitations and Future Work

Although this study has yielded some valuable findings, several limitations remain, which also point to directions for future research. First, the cross-sectional survey data used in this study comes from Anhui Province. While Anhui is a major agricultural province in China with diverse topography and terrain, the generalizability of its conclusions to the national level requires further validation. Future research could collect panel data from more provinces, particularly major grain-producing regions, to conduct broader comparative analyses and test the robustness and regional variations in dual-scale management benefits. Second, while this study validated the two mechanisms of efficiency enhancement and risk sharing, it did not sufficiently explore their specific pathways. For instance, how dual-scale management influences specific management decisions such as technology adoption and crop structure choices to mitigate endogenous risks requires future case studies and in-depth interviews for more nuanced analysis. Finally, the study’s use of per-mu profit represents only net income per unit of land. Future research should refine survey methodologies to record detailed production cost structures, enabling more comprehensive metrics for analyzing agricultural efficiency and profitability.