Synergistic Impacts of Dual Agricultural Scale Operations on Mechanical Utilization: Evidence from Rice Production in Jiangsu, China

Abstract

The development of diverse forms of agricultural scale operations is widely recognized as a cornerstone of modern agricultural management. Most existing studies largely examine land-scale or service-scale operations in isolation and pay little attention to their potential synergies in achieving economies of scale. Using survey data on 1026 plots from 865 rice farmers in Jiangsu Province, China, this study employs fixed-effects regression models to investigate how land-scale and service-scale operations jointly promote scale economies through agricultural machinery utilization. The empirical results reveal three key findings: (i) both land-scale and service-scale operations significantly reduce per-mu (1 mu = 0.067 ha) machinery costs, thereby generating scale economies; (ii) their synergy further amplifies these economies, providing strong evidence of synergy rather than substitution; and (iii) village governance significantly moderates this relationship, with stronger governance reinforcing the synergistic effects between land- and service-scale operations. These findings suggest that dual agricultural scale operations are mutually reinforcing in promoting mechanization. Policy should therefore prioritize their synergistic development and recognize the coordinating role of village collectives.

1. Introduction

Across the globe, the question of how to achieve efficient agricultural production while maintaining sustainability has long been debated. Smallholder farming remains the dominant production model in many developing countries, particularly in Asia and sub-Saharan Africa, yet it is often constrained by fragmented landholdings, limited access to technology, and low productivity [1]. At the same time, agricultural mechanization and service outsourcing have emerged as important pathways to overcome these constraints and promote productivity growth [2]. Similar challenges and transformations can also be observed in China. Since the implementation of the Household Responsibility System (HRS) in the late 1970s, which allocated use rights of collectively owned land to farm households and combined household autonomy with collective ownership [3], China’s farmland has been predominantly cultivated by a vast number of smallholders. In recent years, however, the accelerated transfer and concentration of farmland have gradually altered this pattern. According to the National Bureau of Statistics of China (NBSC), the national farmland transfer rate (the proportion of transferred area in contracted farmland) increased from 5.2% to 30.4% between 2007 and 2014, with an average annual growth of 3.6% over seven years, reflecting a rapid upward trend. According to the Third National Agricultural Census, by the end of 2016, the number of scale agricultural operators (including livestock producers) reached 3.98 million, accounting for only 2% of all farming households but managing over 28% of the total cultivated area. This indicates that they have become an increasingly significant force in agricultural production. Nevertheless, although the share of land-scale operation continues to grow, a substantial number of elderly farmers still cultivate contracted farmland. Some scholars predict that smallholder household farming will remain the dominant production pattern until 2050 or even later [4,5].

Agricultural scale operations have long been a central concern in both theoretical and policy research. In 2019, the State Council issued the Opinions on Promoting the Organic Connection between Smallholders and Modern Agriculture, which emphasized the need to “improve support policies for smallholders, strengthen socialized services for them, and integrate smallholders into the track of modern agricultural development”. By the end of 2020, China had 955,000 agricultural socialized service organizations, providing services covering 1.67 billion mu of farmland to more than 78 million smallholder households. Land-scale operation and service-scale operation are two fundamental types of agricultural scale operations. Land-scale operation refers to the consolidation or transfer of farmland that enables households to expand their managed area, thereby achieving economies of scale in production [6]. This pathway emphasizes the direct enlargement of cultivated land under the control of a single operator. In contrast, service-scale operation does not necessarily involve land transfer but instead relies on the provision of agricultural services such as mechanized plowing, planting, harvesting, or pest control through specialized cooperatives, enterprises, or service organizations [7]. This enables smallholders to access scale economies indirectly by outsourcing key production tasks while retaining land-use rights. Recent evidence suggests that service-scale operations have become increasingly important in China’s rural development, as they reduce transaction costs, improve resource allocation, and facilitate technology adoption, particularly in areas where land fragmentation remains a challenge [8,9].

In evaluating the impacts of agricultural scale operations, most existing studies focus on the functions and limitations of developing a single form of scale operation. On the one hand, numerous studies have affirmed the positive impacts of expanding farmland operation scale in terms of improving agricultural labor productivity, raising farmers’ income, enhancing production efficiency, and reducing per-unit production costs. At the same time, concerns have been raised that such expansion may lower land productivity and thus threaten the supply of agricultural products [10,11]. On the other hand, the development of the agricultural machinery service market suggests that by outsourcing certain production stages or processes to service organizations with lower operation costs, farmers can effectively alleviate household endowment constraints, improve production efficiency, and even increase land productivity [8,12]. While existing studies have acknowledged that expanding service-scale operation helps reduce operation costs [13,14], others also highlight that its effectiveness is significantly constrained by small-scale and fragmented land management [15]. Regarding the debate over which form of scale operation is “superior,” the academic community has yet to reach a consensus. Some scholars argue that land-scale operation is the foundation and an essential means to achieve sustained income growth for farmers and enhance agricultural competitiveness [16,17]; others maintain that service-scale operation has greater advantages in promoting grain production [18]. It is important to recognize that the developmental trajectory of agricultural scale operations ultimately depends on the choices of farmers and the objectives set by the central government. In the long run, for achieving agricultural modernization, land-scale operation and service-scale operation are not mutually exclusive [19].

In practice, both land-scale operation and service-scale operation have developed steadily. According to the data released by Ministry of Agriculture and Rural Affairs of China (MARA), the number of households operating more than 50 mu of farmland increased from 2.734 million in 2010 to 4.357 million in 2020, with an average annual growth rate of 6.7%. Meanwhile, from 2019 to 2020, the number of agricultural social service organizations increased from 893,000 to 955,000, and the number of smallholders served expanded from over 60 million to 78 million. Overall, since the end of 2017, the proportion of various forms of agricultural scale operations nationwide has exceeded 40%.

At the policy level, the development of diverse forms of agricultural scale operations has been highlighted as a critical pathway for transforming agriculture from traditional smallholder farming to a modern operating system. Since 2016, successive “No.1 Central Documents” issued by the Chinese government have emphasized the importance of “accelerating the cultivation of new agricultural business entities and service-oriented operators.” These policy directives, along with practical developments, indicate that land-scale and service-scale operations have advanced in parallel. However, the nature of their relationship remains contested in academic research, warranting further investigation.

The relationship between land-scale and service-scale operations encompasses both competitive and synergistic dimensions. On the competitive side, land-scale operation increases farmers’ ownership of machinery, thereby reducing their reliance on outsourced services [20,21], while service-scale operation alleviates resource constraints for smallholders, thus diminishing their incentives for land transfer [22]. However, greater scholarly attention has been devoted to the synergistic side: land-scale operation can reduce the cost of machinery services and stimulate demand [23,24]. Similarly, service-scale operation helps larger-scale households overcome resource constraints, thereby facilitating land inflows [25]. Importantly, the realization of economies of scale and the alleviation of resource constraints in large-scale operations will increasingly rely on service outsourcing, while the growing maturity of the agricultural machinery service market depends on land-scale operations to overcome barriers caused by land fragmentation. Consequently, the synergy between land-scale and service-scale operations is expected to become more pronounced [25,26]. At the same time, since the implementation of the national “Rural Revitalization” strategy, the collective dimension of China’s dual household–collective agricultural management system has been reinforced, with village governance assuming a growing role in the organization of agricultural production. Against this backdrop, two important questions arise: to what extent are land-scale operation and service-scale operation characterized by synergy rather than competition, and how might effective village governance contribute to shaping this relationship?

Realizing economies of scale is a fundamental requirement for agricultural scale operation, and the utilization of agricultural machinery represents its primary source. As multiple forms of agricultural scale operation continue to evolve, the scale economy characteristics of machinery utilization are also undergoing transformation. Building on this, the present study examines the interaction between land-scale operation and service-scale operation from the perspective of machinery utilization. The underlying logic is straightforward: if both forms of operation contribute to achieving economies of scale in machinery utilization, they should be viewed as synergistic rather than competitive in advancing agricultural mechanization.

This study makes three main contributions. First, it not only examines land-scale operation and service-scale operation in isolation, but also assesses their joint effects on machinery utilization. The results demonstrate that their interaction significantly reduces mechanical operation costs, indicating that the two forms of agricultural scale operation are more complementary than competitive. Second, it incorporates the role of village governance into the analysis of scale economies, showing that effective village governance strengthens the synergy between land-scale operation and service-scale operation. This provides new evidence on the relationship between grassroots governance and the process of agricultural scaling-up. Third, using micro-survey data of rice farmers in Jiangsu Province, the study provides direct evidence of the economic significance of synergistic effects in agricultural scale operation, thereby strengthening the practical relevance of the findings and complementing existing macro-level studies.

The remainder of this paper is structured as follows. Section 2 presents the theoretical analysis, elaborating on the scale economy characteristics of machinery utilization and the interaction mechanisms between land-scale operation and service-scale operation. Section 3 introduces the data, empirical model, and descriptive analysis. Section 4 reports the estimation results. Section 5 provides discussion, and Section 6 concludes with policy implications.

2. Theoretical Analysis and Research Hypothesis

2.1. Economies of Scale in Machinery Utilization—Concept and Logical Analysis

According to The New Palgrave Dictionary of Economics, under given technological conditions, if the unit cost of production decreases (or increases) within a certain range of scale, economies of scale (or diseconomies of scale) exist within that range [26]. In this context, “scale” refers to the size of the production unit, which typically means firms, but in agriculture mainly refers to farm households or farms. Economies of scale arise primarily from the ability of larger-scale production to promote greater labor specialization and more efficient utilization of tools and equipment [27]. Diseconomies of scale, in contrast, generally stem from two main sources. The first is factor inaccessibility: when input supply is limited in the short run, large-scale farming may experience input shortages at current prices or even face rising input costs. The second is internal management difficulties: as operational scale expands, increasing supervision costs, misaligned incentives, and agency problems between owners and workers may undermine production efficiency [28].

In this framework, economies of scale refer specifically to the internal economies of production units. By contrast, external economies arise when average total costs decline as the overall industry expands, even if the scale of individual producers remains unchanged. Industrial agglomeration is an important source of such external economies, as it enables the development and sharing of specialized suppliers, facilitates labor market pooling, and promotes knowledge spillovers [29]. Conversely, external diseconomies occur when excessive clustering generates congestion, intensifies competition for scarce inputs, and drives up factor prices.

Data envelopment analysis (DEA) provides a systematic approach to evaluating these scale effects. By decomposing overall technical efficiency into pure technical efficiency and scale efficiency, DEA makes it possible to identify whether production units operate under increasing, constant, or decreasing returns to scale. A substantial body of empirical research applying DEA to agriculture has demonstrated that efficiency generally improves with moderate scale expansion, but that such gains gradually diminish and may even reverse once farms exceed an optimal threshold [30,31]. This evidence suggests that economies of scale are neither linear nor unbounded, and that the relationship between farm size and efficiency is contingent upon factor endowments and institutional environments.

Building on this framework, economies of scale in machinery utilization, as discussed in this paper, refer to the cost-reducing effects of expanding the operational scale for a given mechanical task in agricultural production. When scale refers to the land managed by a household, it corresponds to the internal economies of land-scale operation. When scale refers to the operational size of machinery service providers, it reflects the external economies of service-scale operation. This perspective underscores that even if the operational scale of individual households remains constant, an expansion of the aggregate operational area within a region can still reduce average machinery costs.

2.2. Internal Economies of Land-Scale Operation in Machinery Utilization

In China, farm households and plots are generally small, which gives rise to internal economies of scale in machinery utilization through land-scale operation. Because machinery purchase and machinery services are partially decoupled, these internal economies can be divided into two dimensions.

First, at the household level, machinery purchase generates internal economies. These economies mainly stem from the adoption of more efficient machinery types. Machinery design and manufacturing inherently display economies of scale: doubling machine capacity (e.g., working width) does not necessarily double manufacturing or operating costs. Consequently, when properly utilized, medium-scale and large-scale machinery generally achieve lower average operating costs than small-scale machinery, and may even deliver higher work quality (e.g., deeper and more uniform tillage) [32]. Expansion of land-scale operation allows households to spread machinery purchase costs over a larger cultivated area, thereby reducing per-unit operating costs. The magnitude of this internal economy largely depends on the development of the machinery service market. If machinery services are primarily provided by specialized service organizations, the role of household machinery purchases in generating internal economies diminishes.

Second, at the plot level, internal economies arise from more efficient machinery use during field operations [33,34]. Specifically: (1) Due to scale economies in agricultural production and marketing, machinery manufacturers tend to mass-produce a limited range of standardized models with fixed working widths and turning times. Larger plots reduce time lost in border-following and frequent turning, thereby improving effective machine utilization and operational efficiency. By contrast, very small plots are inefficient and costly for machinery use, and in some cases may not even be serviceable under prevailing charging standards. (2) Transfer costs, including the time required to move machinery to a plot, are largely fixed. Larger plots allow these fixed costs to be spread more effectively.

Overall, expansion of land-scale operation is typically accompanied by increases in plot size, which further enhances internal economies of machinery utilization.

2.3. External Economies of Service-Scale Operation in Machinery Utilization

With the rise of agricultural specialization, machinery utilization also benefits from external economies of scale through service-scale operation. These external economies are mainly reflected in three aspects.

First, the concentration of machinery service demand within a given geographic area can stimulate the emergence of local service organizations or attract external providers [35]. Cross-regional machinery operations along latitudinal zones are particularly common in China. In regions with large contiguous areas under rice–wheat rotation, machinery costs (purchase and transport) can be spread across multiple regions. By contrast, for crops with scattered planting and insufficient aggregate demand (e.g., maize harvesting in the middle and lower reaches of the Yangtze River), even aggregated demand cannot effectively offset fixed costs. Second, large-scale service operations shorten the distance that machinery and operators must travel between households and plots, thereby reducing transfer costs. Third, scale-based service operations facilitate collective contracting between farmers and machinery providers, which reduces search, bargaining, and other transaction costs [36].

In practice, external economies of service-scale operation can be categorized into three forms:

(1)

Contiguous specialized planting with uncoordinated services: even if households contract machinery services individually, the aggregation of demand in contiguous areas spreads machinery purchase and transport costs, thereby generating external economies.

(2)

Contiguous specialized planting with sequential services: when contiguous plots are outsourced jointly, collective contracting and sequential operations further reduce transfer, search, and bargaining costs.

(3)

Joint farming and planting: in some regions, farmers eliminate plot boundaries and engage in collective production, allowing machinery to operate on an even larger scale and further enhancing utilization efficiency [37].

2.4. Analysis of the Interaction Mechanism Between Land-Scale Operation and Service-Scale Operation

On the one hand, the development of land-scale operation facilitates the realization of external economies of contiguous operation. Farmers’ choices between purchasing machinery and outsourcing services are primarily guided by cost comparisons [38,39]. The cost of machinery acquisition must be spread over the total cultivated area, whereas the market-oriented use of machinery is no longer confined to the land owned by purchasing farmers. When service-scale operation can deliver services at a lower cost without compromising quality, even large-scale farmers may prefer outsourcing [40,41]. However, under conditions of fragmented and small-scale landholdings, the expansion of service-scale operation faces rapidly rising transaction costs [15,42]. As noted in the theoretical analysis above, contiguous specialization can effectively reduce both operational and transaction costs, thereby fostering external economies. However, under the household contract responsibility system, land is fragmented and dispersed among many smallholders, making collective action costly due to high coordination costs [43]. When the benefits of external economies are insufficient to offset these costs, contiguous specialization and operation become difficult to realize. By contrast, lower fragmentation or larger average plot size reduces coordination costs and facilitates contiguous specialization and operation. Once plots reach a sufficiently large size, contiguous operation can be achieved even without collective decision-making, as it becomes internalized into individual decisions. Thus, land-scale operation plays a critical role in reducing coordination costs, thereby promoting contiguous specialization and ultimately expanding external economies.

On the other hand, the development of service-scale operation helps strengthen internal economies at the plot level. As service-scale operation expands, household-level internal economies associated with machinery purchase diminish, while plot-level economies related to mechanized operations are enhanced [44]. Nevertheless, large-scale farmers may encounter diseconomies when relying on purchased services. First, service unavailability may occur. Agricultural production is highly seasonal, creating temporary shortages of key inputs such as machinery services. In regions lacking the conditions for service-scale operation, external providers may be unwilling to bear high transport costs, while local farmers may resist investing in expensive machinery for service provision. In the short term, when service demand created by land-scale operation exceeds supply, problems such as service shortages or inflated prices can arise. Second, service quality is difficult to monitor. In operations such as plant protection, service quality is often unobservable. Moral hazard behaviors, such as shirking or opportunistic pricing, raise the risk of poor-quality services, which is especially detrimental to large-scale farmers [45,46]. The development of service-scale operation helps mitigate these diseconomies. On the one hand, the expansion of service supply enhances availability and reduces costs, curbing short-term price spikes. On the other hand, larger-scale operations attract more farmers into joint contracting and monitoring. As the likelihood of detecting shirking or opportunism increases, the risk of poor-quality services diminishes. In this way, service-scale operation contributes to alleviating diseconomies and reinforces internal economies. However, when homogeneous crops occupy excessively large areas and short-term service demand surges, congestion effects may emerge. Competition for limited machinery may leave some farmers underserved, driving up costs and prices. In such cases, service-scale operation may enter a stage of external diseconomies, further aggravating internal diseconomies.

2.5. The Role of Village Governance in Facilitating Multi-Scale Machinery Utilization

Moreover, the synergy between land-scale and service-scale operation is significantly influenced by the level of village governance. Under China’s dual household–collective management system, farmland remains collectively owned while contracted to individual households. Within this institutional arrangement, village collectives and their leaders play a central role in coordinating land transfer and organizing contiguous operation. Stronger village governance lowers the transaction and coordination costs associated with collective action, thereby facilitating contiguous specialization [47].

In addition, village governments act as intermediaries between farmers and machinery service providers. They can aggregate service demand, coordinate scheduling, and supervise service delivery [48]. By offering logistical support and resolving disputes, village authorities help to alleviate seasonal supply–demand bottlenecks and reduce risks of service quality problems. Such governance capacity creates a more favorable environment for multi-scale operations, complementing natural endowments and physical conditions.

Accordingly, we propose the following hypotheses:

H1. 

Expansion of land-scale operation significantly reduces per-unit machinery service costs.

H2. 

Expansion of service-scale operation significantly reduces per-unit machinery service costs.

H3. 

Land-scale operation and service-scale operation are mutually reinforcing, and their interaction further reduces per-unit machinery service costs.

H4. 

The synergistic effect between land-scale and service-scale operation is stronger in villages with higher levels of governance.

Based on this analysis, the study constructs a theoretical framework for examining the interaction mechanism (Figure 1).

3. Data Sources, Model Construction, and Variable Selection

3.1. Data Sources

The data used in this study come from a household survey conducted by the research team in 2020 across all 13 prefecture-level cities in Jiangsu Province, China. Jiangsu lies in the central part of the Yangtze River Delta and is one of China’s most important agricultural provinces. It has excellent natural endowments for agricultural production: the terrain is flat, with an average elevation of less than 50 m; it has a subtropical monsoon climate with an annual average temperature of 17.5 °C and annual precipitation of about 1055 mm; and the total arable land area reaches 4.58 million hectares, accounting for approximately 0.057 hectares per capita. More than 70% of this land is flat, covering over 7 million hectares, which provides favorable conditions for the adoption of mechanized and large-scale farming practices. At the same time, Jiangsu also exhibits substantial internal heterogeneity. The southern region, adjacent to Shanghai, is deeply integrated into the Yangtze River Delta and has developed a modern industrial system dominated by advanced manufacturing and high-tech industries. By contrast, the northern region lags behind economically but serves as a major national grain base, with vast plains and abundant agricultural resources, playing a strategic role in ensuring national food security. The central region lies in between these two, both geographically and socioeconomically. Given these significant interregional differences in rural development and agricultural structures, Jiangsu provides an ideal setting for studying the structural characteristics of household- and service-scale agricultural operations.

The survey was implemented by the China Land Economy Survey (CLES) project team from the College of Economics and Management at Nanjing Agricultural University. A total of 56 trained enumerators were recruited and divided into three field teams responsible for the northern, central, and southern regions of Jiangsu, respectively. The fieldwork lasted for approximately 20 consecutive days during the summer of 2020. To ensure regional representativeness, a multi-stage random sampling strategy was adopted. First, two counties were selected from each prefecture-level city, followed by two townships from each county, and one village from each township. Finally, approximately 50 households were randomly chosen from each village. This procedure yielded a total of 2628 households across 52 villages in 25 counties. Figure 2 presents the spatial distribution of the 32 surveyed villages across Jiangsu Province.

The questionnaire was carefully designed to capture comprehensive information on rural households’ production and institutional environments. Its main modules covered household demographics, landholding and management, agricultural production and mechanization, land transfer and service contracts, rural industries, ecological environment, financial access, poverty alleviation, and village governance. Key variables used in this study—such as land-scale operation, service-scale operation, and machinery service costs—were derived from these modules. Because rice is the dominant staple crop in Jiangsu and the primary production activity of surveyed households, we focus on rice farmers to ensure comparability of production conditions and avoid confounding effects caused by heterogeneity across different crops. For each household, the largest self-owned plot and the largest rented-in plot were identified, and detailed plot-level data on rice production inputs and outputs in the previous year were recorded (households not cultivating rice were excluded from this section). After excluding households with missing values for key variables and removing extreme outliers, the final dataset used for analysis consists of 1026 plots from 865 households across 32 villages in 19 counties. This dataset provides a representative snapshot of the diversity and dynamics of smallholder and service-scale agricultural operations in one of China’s most advanced agricultural regions.

3.2. Model Setting

3.2.1. Measurement Model for the Economies of Scale in Agricultural Machinery Use

The starting point of our analysis is the classical theory of scale economies, which suggests that average production costs decline when certain inputs are indivisible [26]. In agriculture, machinery is a typical indivisible input: larger farmers can amortize the fixed costs of ownership, while smaller farmers may rely on service providers, where external economies of scale emerge through specialization and collective demand [33,49].

To capture these internal and external scale effects, we specify the following econometric model:

$$\ln M{c}_{lik}={\alpha }_1\ln LS{O}_{lik}+{\alpha }_{2}SS{O}_l+\lambda X_{lik}+\delta +\epsilon$$

(1)

In Equation (1), $i$ denotes the household, $l$ denotes the region, and $k$ denotes the plot. $\ln M{c}_{lik}$ represents the total mechanization cost per mu for rice production on plot $k$ (log-transformed in the estimation). $\ln LS{O}_{lik}$ captures the characteristics of land-scale operation, measured by the household’s rice planting area (log-transformed). ${\alpha }_1<0$ denotes a negative relationship between land scale and per-mu machinery cost, implying the existence of internal economies of scale in land-scale operation. $SS{O}_l$ measures the characteristics of service-scale operation, proxied by the degree of service provision in the village or county (excluding household $i$), calculated as the share of serviced area in total arable land. A higher specialization rate indicates a denser service market, lowering the transaction costs of machinery services and reflecting external economies of scale [50]. ${\alpha }_2<0$ denotes a negative effect, indicating external economies of scale arising from service-scale operation. $X_{lik}$ denotes a set of control variables that may influence machinery use costs, while $\delta$ represents city-level fixed effects to account for regional differences in factor prices, production technology, and other unobserved heterogeneity. $\epsilon$ is the error term. This formulation follows the logic of average cost models widely used in the land fragmentation literature, but explicitly distinguishes between internal and external scale effects.

Farmers rely on either self-owned machinery or purchased machinery services at each stage of rice cultivation. When services are purchased, operation costs can be directly observed from service fees. In contrast, the use of self-owned machinery requires imputing expenses such as depreciation and interest, labor for operation, and fuel or power costs. Summing the costs across all stages yields the total machinery cost per plot. A potential concern, however, is that the estimation results may be subject to upward bias. This arises because the expansion of scale often accelerates the substitution of machinery for labor, thereby increasing machinery inputs. Yet this bias merely leads to an underestimation of scale economies. Once adjusted, the true magnitude of economies of scale would likely be even greater, leaving the main conclusion intact.

3.2.2. Model for Measuring the Synergy Effect

To test the potential synergy effect between land-scale operation and service-scale operation, we specify the following econometric model:

$$\ln M{c}_{lik}={\alpha }_1\ln LS{O}_{lik}+{\alpha }_{2}SS{O}_l+\beta \ln LS{O}_{lik}\times SS{O}_l+\lambda X_{lik}+\delta +\epsilon$$

(2)

In Equation (2), the interaction term $\ln LS{O}_{lik}\times SS{O}_l$ captures the interplay between land-scale operation and service-scale operation. $\beta <0$ indicates that the expansion of both forms of operation together reduces machinery costs, providing evidence of a synergy effect. Conversely, a positive coefficient suggests a competition effect. The definitions and interpretations of the remaining variables are consistent with Equation (1). The expected sign of the interaction term is negative.

3.2.3. Model for the Moderating Role of Village Governance

Village governance is introduced as a moderating factor because collective action in land consolidation, coordination of cropping patterns, and the provision of public goods (e.g., field roads, irrigation facilities) may alter the magnitude of scale effects [51,52]. To examine the moderating role of village governance in the synergy (or competition) effect, we estimate the following model:

$$\ln M{c}_{lik}={\alpha }_1\ln LS{O}_{lik}+{\alpha }_{2}SS{O}_l+{\alpha }_{3}G{v}_l+\beta \ln LS{O}_{lik}\times SS{O}_l+\gamma \ln LS{O}_{lik}\times SS{O}_l\times G{v}_l+\lambda X_{lik}+\delta +\epsilon$$

(3)

Equation (3) denotes the level of village governance, measured using two complementary indicators. The first is a subjective indicator, based on survey responses regarding residents’ satisfaction with village governance. To enhance validity and comparability, we construct two binary indicators of village governance. The first is a subjective measure based on residents’ satisfaction. Governance is coded as 1 if at least three-fourths of respondents report being satisfied and 0 otherwise. The second is an objective measure reflecting collective action in land consolidation and contiguous transfers. Specifically, governance is coded as 1 if the proportion of contiguous transferred land in a village exceeds the sample median and 0 otherwise. The interaction term $\ln LS{O}_{lik}\times SS{O}_l\times G{v}_l$ captures the moderating effect. $\gamma <0$ indicates that effective governance promotes the realization of the synergy effect, while a positive coefficient suggests that governance weakens it. Other variable definitions are consistent with Equation (1). The expected sign of the interaction term is negative.

3.2.4. Mechanism Analysis Model

The theoretical framework suggests that the synergy effect primarily arises through two channels: (i) the expansion of economies of scale at the plot level (internal economies), and (ii) the enhancement of contiguous operations (external economies). To empirically examine these mechanisms, we replace them in Equations (1)–(3) with the logarithm of plot size, which captures the characteristics of plot-scale operation. Meanwhile, $SS{O}_l$ remains unchanged, reflecting the degree of contiguous operations. The rationale is that external economies at the plot level often result from the contiguous cultivation of the same crop within a locality. The rationale is that larger and more contiguous plots reduce per-unit costs of transport, machinery turning, and boundary maintenance [53]. Thus, a higher level of regional specialization implies a greater likelihood of contiguous specialization and, consequently, more contiguous mechanized operations.

3.3. Variable Selection

3.3.1. Dependent Variable

The dependent variable is the total mechanized operation cost per mu of rice plots, expressed in logarithmic form. This variable captures farmers’ expenditures on machinery services and serves as a direct indicator of production efficiency. Similar measures have been widely employed in research on agricultural mechanization and land management to evaluate production decisions and cost efficiency [54,55]. The logarithmic transformation mitigates heteroscedasticity and facilitates the interpretation of coefficients in terms of elasticities [56]. Accordingly, this variable is well suited to analyze how land-scale and service-scale operations shape production costs and economies of scale.

3.3.2. Core Explanatory Variables

The core explanatory variables measure land-scale operation and service-scale operation.

Land-scale operation is proxied by the rice planting area of each household, expressed in logarithmic form. This variable reflects internal economies of scale: as farm size expands, mechanization costs per mu are expected to decline through more efficient utilization of machinery [44].

Service-scale operation is proxied by the degree of rice specialization in the village or county, calculated as the share of serviced area in total arable land after excluding the household itself. This variable reflects external economies of scale: higher local specialization facilitates the provision of machinery services, lowers unit costs, and enhances operational efficiency [8].

3.3.3. Moderating Variable

In this study, village governance is introduced as a moderating variable to examine its role in shaping the interaction between land-scale operation and service-scale operation. To ensure validity, governance is measured using two indicators. The first is a subjective indicator, derived from household survey responses on satisfaction with village governance. Governance is coded as 1 if at least three-fourths of respondents in a village report being satisfied and 0 otherwise. The second is an objective indicator, reflecting the extent of collective action in land consolidation and contiguous land transfers. Specifically, governance is coded as 1 if the proportion of contiguous transferred land in a village exceeds the sample median and 0 otherwise.

3.3.4. Control Variables

To account for heterogeneity, a set of control variables is included at both the plot and household levels. Plot-level controls include slope, distance to the village committee, distance to paved roads, and soil type. Previous studies show that flat terrain is more conducive to mechanized farming, whereas sloping land increases operation difficulty and costs [57]. Greater distance from administrative centers or paved roads raises the time costs of transporting machinery and switching operation sites, thereby reducing the likelihood of mechanization adoption [58]. Soil type also matters: loamy soils are generally more suitable for machinery-based plowing compared with sandy or clay soils [59]. Household-level controls include the number of agricultural laborers, and the household’s degree of rice specialization (measured by the share of rice sown area in total cropped area). Prior studies indicate that family labor and mechanization are substitutable inputs; households with fewer farm laborers are more likely to adopt machinery [60]. Part-time household production captures the extent to which household members participate in non-agricultural employment, which reduces available labor for farming while increasing non-farm income, thereby raising the shadow price of household labor and strengthening incentives for mechanization [61,62,63]. Greater specialization in rice production increases demand for specialized inputs and machinery tailored to rice cultivation [64]. In addition, financial capacity is considered, since financial constraints may restrict households’ ability to purchase or access machinery services and invest in complementary inputs [65].

The variable definitions and descriptive statistics are shown in

Table 1. Variable definitions and descriptive statistics.

Variable Name	Variable Definition	Mean	S.D.
Dependent Variable
Mechanization cost	Total mechanization cost per mu of rice plots (CNY/mu)	168.30	87.66
Core Explanatory Variables
LSO_h	Land-scale operation, household rice planting area (mu)	48.07	143.50
LSO_p	Land-scale operation, rice plot area (mu)	8.53	35.64
SSO_v	Service scale operation, rice specialization at the village level, excluding the household (share of serviced area)	0.78	0.24
SSO_c	Service scale operation, rice specialization at the county level, excluding the household (share of serviced area)	0.76	0.20
Moderating Variable: Village governance
Village governance_s	The subjective indicator, majority of villagers (over 75%) satisfied with local governance: 1 = yes; 0 = no	0.69	0.26
Village governance_o	The objective indicator, above-median share of contiguous transferred land in the village: 1 = yes; 0 = no	0.82	0.38
Plot characteristics
Flat	Slope, flat land (reference = others): 1 = yes; 0 = no	0.92	0.27
Distance to village committee	Distance from the plot to the village committee (li)	2.20	1.57
Distance to nearest paved road	Distance from the plot to the nearest paved road (li)	0.61	0.98
Sandy	Soil type, whether the plot is sandy: 1 = yes; 0 = no	0.23	0.42
Loam	Soil type: whether the plot is loam: 1 = yes; 0 = no	0.11	0.31
Clay	Soil type: whether the plot is clay: 1 = yes; 0 = no	0.64	0.48
Household characteristics
Farm laborers	Number of household members engaged in farming	1.81	0.88
part-time household production	Whether the household has members engaged in non-agricultural work	0.80	0.40
Rice specialization	Household rice specialization (share of rice area)	0.86	0.27
Financial Capacity	Whether the household experienced financial shortage in 2020 due to agricultural operation or daily consumption: 1 = yes; 0 = no	0.28	0.45

Note: Data source, compiled by the authors based on micro-level survey data collected by the research team in Jiangsu Province, China, and the following table is the same. 1 mu = 667 m² or 0.667 ha.

.

As shown in Table 1, the average cost of mechanized operations for rice plots was 168.30 yuan per mu. Regarding land-scale operations, the mean household rice planting area was 48.07 mu, while the average plot size was 8.53 mu. For service-scale operations, the degree of rice specialization at the village level, excluding the household itself, was 0.78, and the corresponding specialization level at the county level was 0.76. With respect to village governance, 69% of households reported being satisfied with local governance effectiveness, and in 82% of villages, the proportion of contiguous transferred land exceeded the sample median. As for plot characteristics, 92% of rice plots were located on flat land, with an average distance of 2.20 li from the village committee and 0.61 li from the nearest hardened cement road. In terms of soil type, 23% of plots were sandy, 11% were loamy, and 64% were clay. Turning to household characteristics, the average number of agricultural laborers per household was 1.81. Approximately 80% of sample households have members engaged in non-agricultural employment, indicating that part-time household production is a prevalent livelihood strategy. The mean household rice specialization ratio, measured by the proportion of rice planting area in total cultivated land, was 0.86. In addition, around 28% of households experienced financial shortages in 2020, either due to agricultural operations or daily consumption needs.

4. Empirical Results and Analysis

4.1. Sample Description

Table 2. Descriptive statistics of machinery utilization by production stage.

Grouping	Tillage		Planting		Plant Protection		Harvesting		Average Machinery Cost (CNY/mu)
	Machinery Use Ratio	Service Purchase Ratio	Machinery Use Ratio	Service Purchase Ratio	Machinery Use Ratio	Service Purchase Ratio	Machinery Use Ratio	Service Purchase Ratio
By household land-scale operation (mu)
$\le$5	0.74	0.64	0.39	0.35	0.32	0.10	0.83	0.80	172.7
5–10	0.91	0.70	0.42	0.31	0.39	0.06	0.99	0.92	165.9
10–30	0.97	0.66	0.41	0.28	0.43	0.16	0.97	0.86	165.8
$>$30	0.94	0.54	0.57	0.26	0.63	0.24	0.96	0.70	159.6
By plot-scale operation (mu)
$\le$5	0.76	0.64	0.40	0.33	0.32	0.09	0.87	0.83	183.2
1.5–2	0.79	0.61	0.41	0.30	0.34	0.09	0.86	0.79	163.2
2–4	0.87	0.68	0.44	0.34	0.41	0.13	0.91	0.84	166.2
$>$4	0.90	0.59	0.49	0.29	0.57	0.21	0.94	0.75	158.2
By village service-scale operation (share of serviced area)
$\le$0.25	0.84	0.69	0.60	0.49	0.55	0.15	0.89	0.87	236.9
0.25–0.5	0.77	0.56	0.29	0.22	0.23	0.11	0.91	0.82	149.4
0.5–0.75	0.88	0.71	0.40	0.34	0.38	0.15	0.88	0.79	159.6
0.75–1	0.82	0.61	0.45	0.31	0.42	0.12	0.90	0.80	168.2
Total	0.83	0.63	0.43	0.32	0.40	0.12	0.89	0.80	168.4

provides a descriptive overview of mechanization and service adoption across different production stages. Overall, mechanization and service outsourcing are most prevalent in the plowing and harvesting stages, followed by planting, while crop protection shows the lowest levels. This outcome is consistent with previous studies indicating that plant protection mechanization faces greater technical and institutional barriers [12]. Farmers’ concerns regarding the effectiveness of plant protection mechanization and the associated environmental risks further constrain their adoption [66]. Moreover, the situation is exacerbated by the limited availability of pest management service providers in Jiangsu, which restricts farmers’ access to reliable services. In general, households rely heavily on mechanized operations, with outsourcing services serving as the primary channel rather than self-owned machinery.

With respect to the relationship between land operation scale and machinery utilization, three key patterns emerge. First, at both the household and plot levels, operation scale is positively correlated with the probability of machinery use. Larger households or plots are more likely to substitute labor with machinery. Second, the relationship between household scale and the proportion of service outsourcing varies across production stages. In stages where production relies heavily on outsourcing, such as plowing and harvesting, the relationship follows an inverted U-shape. Households initially expand service purchases as their scale increases, but at larger scales, they are more likely to invest in their own machinery, which partially reduces their demand for external services. By contrast, in stages where outsourcing services are less developed, such as planting, household scale is negatively associated with service purchases, whereas in crop protection the correlation is positive. These patterns suggest that the relationship is highly dependent on the maturity of local machinery service markets. At the plot level, however, no consistent relationship is observed between plot size and service outsourcing. Third, at both the household and plot levels, operation scale is negatively associated with per-mu total machinery costs, indicating the presence of internal economies of scale in machinery use. As operation scale expands, the intensity of machinery use increases, but the efficiency gains from large-sized and medium-sized equipment lead to a continuous decline in per-mu machinery costs.

Regarding the relationship between village-level specialization and machinery utilization, two points are noteworthy. First, across different production stages, the share of serviced rice cultivation in village land does not show a consistent association with either machinery use or service outsourcing. Second, the share of serviced rice cultivation is broadly negatively associated with per-mu machinery costs: as the share increases, costs decline sharply at first, then level off, and in some cases rise slightly. Naturally, this bivariate relationship may be shaped by other factors such as terrain, transportation infrastructure, and household characteristics. To address these potential confounding influences, the following section employs model-based analysis.

4.2. Scale Economies in Agricultural Machinery Utilization

Table 3. Estimation of Internal and External Economies of Scale in Agricultural Machinery Use.

Variables	Baseline Results		Mechanism Analysis
	(1)	(2)	(3)	(4)
ln_ LSO_h	−0.021 ***	−0.023 ***
	(0.007)	(0.007)
ln_ LSO_p			−0.030 ***	−0.032 ***
			(0.010)	(0.010)
SSO_v	−0.134 **		−0.123 **
	(0.061)		(0.059)
SSO_c		−0.416 ***		−0.406 ***
		(0.135)		(0.134)
Flat	−0.199 **	−0.198 **	−0.201 **	−0.200 **
	(0.100)	(0.100)	(0.099)	(0.099)
Distance to village committee	0.014	0.012	0.014	0.012
	(0.012)	(0.012)	(0.011)	(0.011)
Distance to nearest paved road	−0.003	−0.003	−0.002	−0.002
	(0.005)	(0.005)	(0.005)	(0.005)
Sandy	−0.090	−0.087	−0.091	−0.086
	(0.066)	(0.066)	(0.064)	(0.063)
Loam	0.062	0.066	0.064	0.070
	(0.068)	(0.068)	(0.066)	(0.066)
Clay	0.028	0.035	0.028	0.036
	(0.063)	(0.063)	(0.061)	(0.061)
Farm laborers	0.017	0.018	0.015	0.013
	(0.012)	(0.012)	(0.011)	(0.011)
part-time household production	0.009	0.011	−0.006	−0.004
	(0.025)	(0.025)	(0.024)	(0.024)
Rice specialization	0.031	0.013	0.001	−0.014
	(0.051)	(0.051)	(0.042)	(0.041)
Financial Capacity	0.027	0.030	0.020	0.022
	(0.023)	(0.023)	(0.022)	(0.021)
City FE	Y	Y	Y	Y
R2	0.045	0.050	0.045	0.049

Note: **, and *** indicate that the coefficients are significant at the 10%, 5%, and 1% confidence levels, respectively. The figures in parentheses are standard errors. “Y” denotes that the variable is controlled.

presents the estimation results on the internal and external economies of scale in agricultural machinery utilization. Columns (1) and (2) show that the coefficients of average machinery costs with respect to farm size are significantly negative at the 1% level, suggesting that larger farm sizes are associated with lower average machinery costs. Columns (3) and (4) provide further evidence through mechanism analysis, where plot size is used as a proxy for within-plot scale operation. The coefficients exhibit significant negative elasticities at the 1% level, confirming that average machinery costs decline as plot areas expand. This result aligns with previous studies [67], which document within-plot economies of scale in mechanized farming under market-oriented agricultural service systems.

In addition, across all four model specifications, the coefficients of service-scale operation, which are measured by the degree of rice cultivation specialization at the village and county levels, are significantly negative. This indicates that higher levels of service specialization reduce machinery costs, thus verifying the existence of external economies of scale at the regional level. These findings are consistent with earlier research [21]. The underlying mechanism can be attributed to contiguous farming and coordinated machinery operations, which not only allow equipment and transportation costs to be shared but also minimize machinery losses during plot transfers and enhance farmers’ bargaining power, thereby lowering transaction costs.

Among the control variables, flat terrain has a significantly negative effect on machinery costs, indicating that favorable topography facilitates more efficient machinery use. In contrast, the effects of other control variables are not statistically significant.

4.3. Analysis of the Synergistic Effects

Table 4. Estimation Results of Synergistic Effects.

	Baseline Results		Mechanism Analysis
	(1)	(2)	(3)	(4)
ln_ LSO_h × SSO_v	−0.091 ***
	(0.029)
ln_ LSO_h × SSO_c		−0.107 ***
		(0.036)
ln_ LSO_p × SSO_v			−0.124 ***
			(0.043)
ln_ LSO_p × SSO_c				−0.189 ***
				(0.054)
ln_ LSO_h	Y	Y	N	N
ln_ LSO_p	N	N	Y	Y
SSO_v	Y	N	Y	N
SSO_c	N	Y	N	Y
Control variables	Y	Y	Y	Y
R2	0.051	0.053	0.050	0.055

Note: *** indicate that the coefficients are significant at the 10%, 5%, and 1% confidence levels, respectively. The figures in parentheses are standard errors. “Y” denotes that the variable is controlled; “N” denotes that it is not.

presents the estimation results for the synergistic relationship between land-scale operation and service-scale operation. Columns (1) and (2) show that the coefficients of the interaction term, “log of household land scale × village (or county) specialization,” are negative and statistically significant at the 5% level. This finding indicates that the joint expansion of household land scale and service-scale operation lowers the per-mu cost of mechanization. In other words, the two forms of agricultural scale operation complement each other in achieving economies of scale through agricultural machinery utilization.

As discussed earlier, this synergy can be explained by two mechanisms. First, the expansion of land-scale operation enlarges plot size, which facilitates contiguous machinery services and generates external economies. Second, the strengthening of service-scale operation further enhances the realization of internal economies associated with larger plots. Columns (3) and (4) provide supporting evidence from mechanism analysis, showing that the interaction between plot size and service-scale operation significantly reduces per-mu mechanization costs, further validating the above explanation.

4.4. Analysis of the Moderating Role of Village Governance

Table 5. Estimation results of the moderating role of village governance.

	Baseline Results				Mechanism Analysis
	(1)	(2)	(3)	(4)	(5)	(6)	(7)	(8)
Village governance ×ln_ LSO_h × SSO_v	−0.057 ***				−0.032 **
	(0.013)				(0.016)
Village governance ×ln_ LSO_h × SSO_c		−0.046 ***				−0.036 **
		(0.014)				(0.017)
Village governance ×ln_ LSO_p × SSO_v			−0.074 ***				−0.055 ***
			(0.023)				(0.013)
Village governance ×ln_ LSO_p × SSO_c				−0.051 **				−0.043 ***
				(0.025)				(0.014)
Village governance_s	Y	Y	Y	Y	N	N	N	N
Village governance_o	N	N	N	N	Y	Y	Y	Y
ln_ LSO	Y	Y	Y	Y	Y	Y	Y	Y
SSO	Y	Y	Y	Y	Y	Y	Y	Y
ln_ LSO × SSO	Y	Y	Y	Y	Y	Y	Y	Y
Control variables	Y	Y	Y	Y	Y	Y	Y	Y
R2	0.042	0.035	0.030	0.031	0.027	0.029	0.022	0.030

Note: **, and *** indicate that the coefficients are significant at the 10%, 5%, and 1% confidence levels, respectively. The figures in parentheses are standard errors. “Y” denotes that the variable is controlled; “N” denotes that it is not.

presents the moderating effect of effective village governance on the synergy between land-scale operation and service-scale operation. The results in Columns (1)–(4) indicate that, whether measured by subjective indicators (e.g., majority satisfaction evaluations) or objective indicators (e.g., the proportion of contiguous land transfers), stronger village governance further reduces per-mu machinery costs. Two potential mechanisms underlie the role of village governance. First, effective village governance facilitates service-scale operation by enhancing external economies from contiguous operations. Second, it mitigates the internal diseconomies that households may face when accessing services under land-scale operation. The mechanism analysis in Columns (5)–(8) provides further support for this interpretation.

5. Discussion

This study provides new evidence on the role of multiple forms of agricultural scale operation in promoting agricultural mechanization. Using field survey data from 865 rice households in Jiangsu Province, China, in 2020, we examined the internal economies of land-scale operation and the external economies of service-scale operation, as well as their interaction and the moderating role of village governance.

5.1. Internal and External Economies of Scale in Agricultural Mechanization

Previous studies have primarily focused on the relationship between land transfer and machinery services, either discussing how land transfer influences farmers’ adoption of socialized machinery services [63,68] or examining how machinery services promote land transfer [21]. Building on the perspective of economies of scale in machinery use, this paper investigates the roles of both land-scale operation and service-scale operation. By integrating theoretical analysis with empirical evidence, we demonstrate that the relationship between the two is synergistic rather than competitive, and that their interaction significantly reduces mechanization costs. Rather than being confined to the traditional binary debate of substitution versus complementarity, this finding provides a more nuanced understanding and contributes to the theoretical framework of agricultural scale operations.

Our results provide robust evidence of significant internal economies of scale in land-scale operation. Both total farm size and plot size are significantly and negatively correlated with average mechanization costs, indicating that expanding farm operations and enlarging plots reduces per-unit costs of machine services. This finding is consistent with earlier studies suggesting that farm expansion and plot consolidation facilitate more efficient use of machinery and lower operational costs [7,69].

We also find strong evidence supporting the presence of external economies of scale in service-scale operation. A higher degree of specialization at the village or county level is significantly associated with lower per-unit mechanization costs. This supports the view that the expansion of agricultural machinery services generates community-level cost advantages by spreading fixed costs, reducing machinery downtime, and lowering transaction costs through collective bargaining [8,21,70].

5.2. Synergistic Effects and the Role of Village Governance

One of the key contributions of this study lies in identifying the synergistic effect between land-scale operation and service-scale operation. The interaction terms demonstrate that the simultaneous expansion of farm size and service scale leads to additional reductions in mechanization costs. This indicates that, in terms of machinery use, the development of both forms of agricultural scale operation enables them to reinforce each other in capturing economies of scale. Farm expansion facilitates contiguous service operations [60,71], while the growth of service operations promotes more effective plot consolidation [21], jointly enhancing mechanization efficiency. This finding partly aligns with earlier studies emphasizing the independent benefits of either land-scale or service-scale operation. For example, Deininger and Jin (2005) and Foster and Rosenzweig (2011) highlighted the productivity and cost advantages of larger farm sizes [72,73], whereas Wang et al. (2016) underscored the role of service scale in reducing transaction and machinery costs [74]. Unlike these studies that examine land or service economies in isolation, our results reveal that their combination produces an additional synergistic effect, a dimension that has received limited attention in the existing literature.

Furthermore, while earlier research noted potential tensions between land consolidation and collective service provision due to coordination difficulties [75], our findings indicate that higher levels of village governance, assessed through both subjective and objective indicators, are essential for enhancing the synergy between land-scale operation and service-scale operation. Rather than competing, the two forms of agricultural scale operation reinforce each other. Villages with stronger governance capacity ensure more effective collective coordination and institutional enforcement, which lower transaction costs in both service use and land transfer, mitigate problems of free riding and contract breaches, and thereby enhance the feasibility of inter-household cooperation [47,76]. At the same time, robust village governance fosters trust and social capital, which enhances farmers’ acceptance of and reliance on service-scale operations [77]. Moreover, the ability of village collectives to coordinate the construction of basic infrastructure, such as machine roads and irrigation systems [78], significantly reduces entry barriers and operating costs for mechanization services, further amplifying the synergy between land- and service-scale operations. These findings underscore the critical role of grassroots governance in promoting agricultural economies of scale.

Nevertheless, this study has several limitations. First, the CLES survey collects detailed information only for the largest plot of each household. While this plot usually constitutes the main production unit and reflects the primary site of mechanization, it may not capture the full heterogeneity across all plots managed by the household. Second, it relies on data from Jiangsu Province, where mechanization is relatively advanced and service markets are well developed. The generalizability of the findings to other regions, particularly mountainous or remote areas, therefore warrants further validation. Third, the measurement of economies of scale focuses primarily on average mechanization costs per mu and does not capture other relevant dimensions such as service quality, timeliness, and technological advancement. Fourth, the cross-sectional nature of the data prevents us from examining the dynamic evolution of synergy over time, including possible lagged effects. Future research could extend this work in several directions. One avenue is to employ multi-regional and multi-year panel data combined with spatial econometric methods to identify heterogeneity across space and time. Another is to incorporate non-price indicators of service quality and timeliness to deepen understanding of the underlying mechanisms. Comparative analyses of different organizational models of service provision, such as cooperatives, enterprise providers, and cross-regional teams, could also shed light on pathways to strengthen complementarities. Finally, exploring the long-term effects of village governance, land tenure reform, and factor market development on synergy mechanisms would provide richer policy insights for fostering the coordinated development of multiple forms of agricultural scale operation.

5.3. Policy Recommendations

The findings of this study yield several practical implications for promoting coordinated agricultural scale operations in China. Since land-scale and service-scale operations jointly enhance mechanization efficiency, policy efforts should focus on facilitating their integration rather than treating them as separate domains.

First, institutional and market barriers in the primary factor markets—land, labor, and capital—should be gradually removed. Well-functioning markets allow smallholders either to expand or to exit farming efficiently, facilitating land consolidation and optimal resource allocation. Government efforts should focus on improving land registration, standardizing land transfer contracts, and ensuring transparent market transactions to reduce transaction costs. Second, in areas where external economies of machinery service scale are evident, public investment should prioritize essential infrastructure such as machine-accessible roads, irrigation systems, and storage facilities. Local governments and village collectives should work together to coordinate contiguous land transfers and promote cooperative or joint-service models to expand operational efficiency. Third, village collectives play a pivotal role as intermediaries between smallholders and service providers. Strengthening their organizational capacity and governance transparency can help coordinate production decisions across households, reduce free-riding behavior, and enhance trust among participants. Training and incentive programs should be developed to improve local governance competence, especially in areas with weak institutional foundations. Finally, future agricultural policies should encourage innovation in organizational forms of agricultural service provision. Promoting diverse service models—such as cooperative service platforms, cross-regional service alliances, and digitalized machinery-sharing networks—can further reduce costs and foster synergy between land-scale and service-scale operations. Pilot projects combining digital agriculture with collective governance could provide valuable insights for scaling up successful experiences nationwide.

6. Conclusions

This study investigates the interrelationship between land-scale operation and service-scale operation from the perspective of agricultural machinery utilization. The analysis first develops a theoretical framework of internal and external economies of scale in machinery use, highlighting the potential synergy between the two forms of agricultural scale operation. Using survey data from Jiangsu Province, the study empirically verifies the existence of economies of scale, quantifies the synergistic effects, and further evaluates the moderating role of effective village governance. The theoretical analysis suggests that the decoupling of machinery purchase and use has reduced the dependence of economies of scale on expanding household farm size, while the significance of enlarging operational scales at the plot and regional levels has increased with the development of the machinery service market. Consequently, economies of scale in machinery utilization are realized both through the internal gains of land-scale operation and the external gains of service-scale operation.

Within the fragmented smallholder structure, expanding landholding size facilitates the external economies of scale generated by contiguous operations, while the expansion of service-scale operation promotes the realization of internal economies of scale at the plot level. Moreover, the synergy between the two forms of operation is enhanced when village governance is more effective. Empirical evidence shows that the expansion of both land-scale operation and service-scale operation, as well as their interaction, significantly reduces per-unit machinery operation costs, and this effect is even more pronounced in villages with higher governance capacity.

The findings indicate that land-scale operation and service-scale operation are synergistic rather than competitive in advancing agricultural mechanization. They are mutually dependent and reinforcing, enabling each to achieve higher levels of economies of scale while simultaneously promoting one another. Conversely, farmers with small, scattered plots or those cultivating crops distinct from others in the same village are disadvantaged in machinery utilization. With accelerated labor migration and rising labor costs, these farmers are more likely to face difficulties in achieving cost-effective substitution of machinery for labor, which may lead to land abandonment or extensive cultivation, thereby posing risks to food security.