The Spatio-Temporal Patterns and Influencing Factors of Different New Agricultural Business Entities in China—Based on POI Data from 2012 to 2021

Abstract

The high-quality development of new agricultural business entities (NABEs) is an important driving force for realizing rural revitalization and accelerating the modernization of agriculture and rural areas. The main purpose of the study is to investigate the spatial distribution pattern, aggregation scales, development mechanism, and internal differences of various types of NABEs in different regions. It provides targeted ideas for alleviating regional differences in the development of NABEs in different agricultural regions. Kernel density estimation, nearest neighbor distance analysis, Tyson’s polygon coefficient of variation, and Ripley’s K function are used to study the spatial and temporal evolution, spatial aggregation, and scale divergence of various types of NABEs, and Pearson correlation analysis is incorporated to explore the specific factors affecting the development of various types of NABEs. The study results: First, family farms are the most widely distributed, and agricultural enterprises are the most sparsely distributed, being distributed “more in the southeast and less in the northwest” in all three categories. Second, the strongest aggregation scales of different NABEs are increasing, and the strongest aggregation scales of agricultural enterprises are larger than those of family farms and cooperatives in all agricultural areas. Third, the development of specialized farmers’ cooperatives (abbreviated as ‘cooperatives’) is more constrained by traditional agricultural inputs and is a kind of agricultural input-oriented development. Family farms are more constrained by the living standards of rural residents in the region and are a kind of rural economy-oriented development. Agricultural enterprises are more subject to the economic level of the region, which is a kind of market economy-oriented development. Finally, in the process of developing NABEs, regional differences should be emphasized, and a small number of agriculturally leading enterprises and model cooperatives should drive a large number of small-scale family farms and smallholder farmers in order to become a characteristic path for China’s agricultural development.

1. Introduction

For a long time, traditional agriculture in China has been dominated by small farms with low productivity and a small farming radius. With the shift of labor to non-agricultural sectors and the progress of agricultural production technology, the traditional smallholder farmers’ self-cultivation business model is gradually emerging. The contradiction between the scale and modernization of agricultural production and further improvement of agricultural productivity and smallholder production and operation is becoming more and more prominent. Due to the constraints of geography and history, as well as development and institutional choices, China cannot follow the Western modernization path at the expense of the disappearance of smallholder farmers [1,2,3], which has become an insurmountable objective existence in agricultural modernization [4,5]. As a result, the Chinese government has proposed a three-dimensional and complex modern agricultural management system based on the development of family businesses, linked by cooperative associations and supported by socialized services [6]. The cultivation of new agricultural business entities, or NABEs, is represented by family farms, specialized farmers’ cooperatives (abbreviated as ‘cooperatives’), large professional households, and leading agricultural enterprises [7] (Figure 1). NABEs could gradually drive and improve the efficiency of individual farmers’ arable land utilization, realize the realistic goals of the centralized and contiguous utilization of arable land, increase agricultural production and farmers’ incomes, and integrate traditionally scattered smallholder farmers into the development track of agricultural modernization [8,9,10].

NABEs represent new productive forces and are the development and innovation of traditional small-scale agricultural production [7]. Since 2012, when China first proposed to cultivate and develop NABEs, the Central Government’s No. 1 document for ten consecutive years and various agricultural policies have explored and improved the development of NABEs (Figure 1). At present, NABEs have flourished everywhere. On the one hand, the NABEs realize the scale, intensification, and industrialization of modern agricultural operations, improve the utilization rate of land for all agricultural production, and bring new opportunities for the ordinary farmers of traditional agriculture in China [8,11]. On the other hand, NABEs solve the real dilemma of the large number of small farms, serious land fragmentation, and resource loss and substitution in China, and put forward new requirements for current agricultural production [11,12].

In terms of national development, the main goal of China’s economic and social development is to move from comprehensive poverty eradication to rural revitalization, to improve the comprehensive national power, and to gradually modernize agriculture and rural areas [13]. This series of development requires the participation of NABEs, which has led to the employment of farmers and prevented the phenomenon of returning to poverty after the comprehensive poverty eradication campaign [14]. At the same time, the capital of some agricultural enterprises has gone to the countryside, injecting new vitality into rural economic development and contributing to the further realization of rural revitalization. As far as national food security is concerned, the capacity of small farmers to produce food is weak, and their tendency towards non-food production has increased [15]. The NABEs have the advantage of scale, which can attenuate the threat to national food security posed by the trend of non-food production by small farmers, thus guaranteeing the proportion of food planted at the macro level [16].

As NABEs have become the focus of policy support, preliminary research on NABEs has been conducted in academia. In terms of operational efficiency, although research confirms that NABEs have more advantageous arable land use efficiency compared to individual farmers [17], the differences in the operational efficiency of NABEs in different regions deserve attention. Taking Fujian family farms as an example, the contribution of pure technical efficiency to the overall efficiency of family farms in the region has exceeded the impact of land-scale expansion in 2015 [18]. However, the pure technical efficiency in the Xinjiang Autonomous Region is still insufficient to support the increase in the overall efficiency of farmland [19]. In contrast, a study of cooperatives in three northeastern provinces showed that the distribution of the combined efficiency of their arable land is characterized by “less at the ends and more in the middle”, with only less than 10% of the cooperatives reaching a combined efficiency of one and exhibiting an increasing scale [20]. The example of the three northeastern provinces suggests that there is still a need to rely on land area expansion to improve efficiency.

In terms of influencing factors, the study shows that the development of NABEs is not only subject to endogenous factors such as production and operation characteristics and individual characteristics, but also external factors, such as agricultural subsidies, credit support, and technical support, which impact them more significantly [21]. Specifically, NABEs in the Middle–Lower Yangtze Plain benefit from the external economic environment, their products are more commercialized, and NABEs show a greater advantage in number compared to other regions in the country [22]. In addition, taking the Loess Plateau region as an example, the cooperation mode between NABEs and farmers has a differentiated impact on farmers’ income-generating benefits, among which the “cooperative + farmers” mode has the most obvious income-generating benefits, followed by the “leading enterprises + farmers” mode and the “government-led + farmers” mode. The “government-led + farmer” model, although not obvious in terms of income benefits, makes up for the ecological and social benefits and complements the other models [23]. On the contrary, Shanxi and the Tibet Plateau region have been slower to develop their NABEs. The former is limited by external factors such as credit funds, policy support, and technological innovation, while the latter is constrained by external factors such as plateau topography, climate, and organizational support [24,25].

However, while vigorously developing NABEs and promoting agricultural modernization, it must be seriously recognized that China is vast, and agricultural production in different regions has large geographical differences due to the influence of regional natural and economic development. First, from the perspective of climate change and land resource pattern, with a 400 mm precipitation line as the boundary, the eastern region is dominated by plains with sufficient light and heat conditions, with a wide inland coastline as well as a dense river network, constituting the vast majority of China’s arable land and cropping areas [26]. However, the western region is dominated by plateaus, mountains, and basins, with poor water, heat, and soil coordination, as well as small and scattered agriculture [27]. Agricultural development from east to west shows a stepwise weakening trend. Second, from the perspective of the economic development pattern, the developed economy has enabled the eastern region to realize the non-farm employment of rural labor earlier [28], and the transfer of agricultural labor has created the conditions for large-scale land management [28]. At the same time, local governments in the eastern region have high fiscal revenues and abundant agricultural subsidies, which provide strong financial guarantees for the development of NABEs. In contrast, the western region has a weak economic base and lacks cohesion in economic development [29,30]. Coupled with an over-dependence on energy, the lack of investment in the agricultural field in the western region makes it difficult to engage in large-scale agricultural production, further widening the gap between agricultural development in the eastern and western regions [31]. The differences in the resource endowment and economic development of different regions show the obvious geographical differentiation of agricultural production in space, thus becoming an important reason to hinder the development of NABEs.

Combined with the current exploration of NABEs by scholars, the existing research still needs to be expanded in the following three aspects: first, due to the differences in the nature of different NABEs, their spatial and temporal development characteristics should have their own development logic, and there is a lack of comparative research on the classification of different NABEs; second, the spatio-temporal pattern of different types of NABEs and the comparative analysis of agricultural regions are still unknown, and it is difficult to capture the developmental differences and stage changes of NABEs in the different regions; third, the comparative study of the formation mechanisms of different types of NABEs needs to be explored. The development of different types of NABEs has its own dynamic changes, and the comparative analysis of their respective formation mechanisms will help to discover their developmental shortcomings, which have not yet been explored in the relevant studies.

To fill the current research gap, this paper uses geospatial analysis to explore the spatial distribution patterns, aggregation scales, development mechanisms, and internal differences of NABEs in different regions based on the POI data of three types of NABEs—family farms, farmers’ professional cooperatives, and agricultural enterprises—from 2012 to 2021, which are divided into nine agricultural regions in mainland China. The results of this paper provide important references for the development and cultivation of NABEs, especially for the regional differences in the development of China’s agricultural modernization, and propose targeted solutions that are tailored to local conditions and different from one entity to another. The rest of this article is organized as follows. Section 2 describes the theoretical analysis and research hypotheses about NABEs. Section 3 describes the study area and research methodology as well as the main data sources. Section 4 describes the spatial distribution characteristics of NABEs. Section 5 summarizes the influencing factors of the development of NABEs. Section 6 discusses and analyzes the results generated. Finally, the conclusion of this study is summarized in Section 7.

2. Theoretical Analysis and Research Hypothesis

According to the spatial selection of the industrial location theory, through the interrelationship of transportation, labor, and agglomeration factors, the point with the lowest production cost is identified as the ideal location for industrial enterprises [32]. For NABEs, it is also necessary to consider the comprehensive costs of agricultural resource endowment (land), labor (rural population), and the market (sales location). Specifically, on the one hand, traditional agricultural areas, major grain-producing areas, and other areas with a long history of agricultural development are rich in arable land resources and have a high potential for the large-scale utilization of land parcels, which can provide a better material basis for the large-scale operation of NABEs [7]. On the other hand, regions with abundant labor can meet the seasonal short-term hiring demand of NABEs [33]. Finally, regions with faster economic development have stronger policy support for the development of large-scale and high-efficiency NABEs, a wider sales market, and more flexible pre-production and post-production services, thus giving rise to the emergence of more NABEs [34].

Based on the above analysis, Hypothesis 1 is put forward:

H1. 

There are regional differences in the distribution of NABEs, and their development is more prominent in regions with higher agricultural location advantages.

According to Tobler’s First Law, everything is related to everything else, except that things in close proximity are more closely related [35]. Geographic things are spatially related to each other, and there are different distribution characteristics such as clustering, randomness, and regularity. However, since there are obvious differences in the business models of different NABEs, their spatial distribution characteristics or aggregation scales may be different. Specifically, family farms, with family production as the unit and family labor as the main member, have their threshold of generation only in the requirement of land scale [36]. That is to say, the traditional small farmer only needs to expand the scale of arable land to develop into a family farm, and his spatial location will not be transformed too much. In this process, the income-generating effect of family farms will have a demonstration effect on other farmers in the vicinity, thus prompting more family farms to appear in the vicinity, and ultimately leading to the phenomenon of the short-distance clustering of family farms in rural areas [37]. Unlike family farms, cooperatives are the main body composed of multiple farmers and family farms, and there are socialized services such as agricultural machinery, agricultural technology, and marketing in the production process, and there are multiple opportunities for cooperation among different cooperatives in the pre-production, production, and post-production segments [38]. Therefore, compared with family farms, cooperatives may produce aggregation effects at a greater distance. However, most cooperatives in China are trapped within the geographical boundaries of a single village or several villages; therefore, their ability to radiate and drive similar cooperatives over long distances is relatively weak. Finally, the development of agricultural enterprises is not bound by administrative boundaries, and there is competition among agricultural enterprises in the same area for the supply of raw materials and market share; therefore, the development of agricultural enterprises is highly exclusive, and the degree of the agglomeration of agricultural enterprises within a small area will be low. However, when a linkage of interests is created between agricultural enterprises, cooperations, and associations among them may lead to the emergence of aggregation on a larger scale [39].

Based on the above analysis, Hypothesis 2 is put forward:

H2: 

The spatial aggregation characteristics of different types of NABEs are different, with agricultural enterprises and cooperatives being likely to have larger spatial aggregation scales than family farms.

Because of the different production characteristics of various types of NABEs, there are differences in their respective growth patterns. Specifically, although family farms have the advantage of the arable land scale compared with traditional small farmers, their production is still not detached from the characteristics of product specialization, and, thus, has a higher dependence on terrain, water, and soil resources [34]. In other words, the larger the area of arable land and the better the fertility of the land, the more it can support the development of family farms. On the contrary, an agricultural enterprise has the typical characteristics of capitalized operation, is the integration of advanced production factors, can cover the entire chain of agricultural products, such as planting, processing, storage, sales, etc., and has a high degree of scientific research organization and specialization [40]; therefore, agricultural enterprises are more dependent on the economic environment. Finally, cooperatives are in the middle of the two, and they are concentrated in the rural areas of traditional agricultural production links but also have the advantage of land scale [41]; therefore, the development of cooperatives may rely more on economies of scale. Specifically, cooperatives may be more advantaged in areas where higher economies of scale are achieved through mechanization and the coordination of inputs and outputs [42].

Based on the above analysis, Hypothesis 3 is put forward:

H3. 

The influencing factors favored by the development of different NABEs vary, especially in their dependence on soil and water resources, the economic environment, and the efficiency of production scale.

3. Research Methodology and Data Sources

3.1. Research Area and Data Sources

In view of the different conditions, characteristics, and development directions of agricultural production in different regions, China can be divided into nine agricultural regions by combining key measures, major problems, and the integrity of administrative units in each region (Figure 2). They are the Northeast China Plain (NE), the Huang-Huai-Hai Plain (HHH), the Middle–Lower Yangtze Plain (MLY), Southern China (SC), the Loess Plateau (LP), the Yunnan–Guizhou Plateau (YG), the Northern Arid and Semi-Arid Region (NAS), the Sichuan Basin and surrounding regions (SBS), and the Tibet Plateau Region (TP) [43]. Therefore, this study takes nine agricultural regions in China as the study area to explore the development patterns and influence mechanisms of NABEs in different regions. The research data mainly include: (1) POI data, which, compared with traditional data, have better objectivity and validity, which helps to conduct fine regional research [44]. In this study, we select the POI data of the Gaode Map as the basic data of various NABEs, including name, category, latitude and longitude, and other information. The points of interest of the corresponding years are obtained using self-programmed web crawling tools, and the vector data are converted to spatial projection as well as coordinate correction, cleaning, screening, and de-weighting before finally generating the point Shp layers of various NABEs; (2) The statistical yearbook data, which include the relevant socio-economic, agricultural production, and natural environment index data, as obtained from the statistical yearbook data, including the China Statistical Yearbook, China Rural Statistical Yearbook, China Environmental Statistical Yearbook, and China Labor Statistical Yearbook.

3.2. Research Methodology

3.2.1. The Kernel Density Estimation Method

The kernel density estimation method is a visual–spatial data analysis tool that can visually reflect the degree of aggregation and distance decay effect of geographical elements in space [45] and express the overall density distribution of NABEs and the spatial aggregation and dispersion degree of different types of NABEs in a close way. The formula is as follows:

$$f\left(x\right)={\displaystyle \sum }_{i=1}^n\frac{1}{nh}k\left(\frac{x-x_i}{h}\right)$$

(1)

where $f\left(x\right)$ is the kernel density estimate; $k\left(\frac{x-x_i}{h}\right)$ is the kernel function; $h$ is the bandwidth; $n$ is the number of NABEs; $k$ is the default spatial weight kernel function; $\left(x-x_i\right)$ denotes the distance from the valuation point to the event $x_i$. This paper uses the kernel density analysis module in ArcGIS to analyze the spatial clustering pattern of different NABEs. The larger the kernel density estimate, the denser the spatial distribution of NABEs.

3.2.2. Nearest Neighbor Distance Analysis

Nearest Neighbor Analysis is an effective spatial measurement method to quantitatively describe the proximity of spatial point-like elements and determine their spatial pattern characteristics and distribution patterns [46]. The nearest neighbor index (NNI) is used to determine the spatial aggregation of “points”. The formula is as follows:

$$NNI=\frac{{{\displaystyle \sum }}_{i=1}^n{d}_i}{n}{\displaystyle /}\frac{1}{\sqrt[2]{n/A}}$$

(2)

where $d_i$is the distance between the ith grid and its nearest neighboring grid, n is the number of various types of new NABEs, and A is the area of the study area. In this paper, the NNI index tool of the ArcMap platform is used to analyze the spatial and temporal distribution characteristics of the point elements of NABEs on the whole. When the NNI is less than 1, the elements are clustered; when the NNI is greater than 1, NABEs are uniformly distributed.

3.2.3. The Coefficient of Variation

A Tyson polygon, also known as a Voronoi diagram, is a set of continuous polygons consisting of perpendicular bisectors of line segments connecting two neighboring points. The distance from any point in a Tyson polygon to the control points of that polygon is smaller than the distance to the control points of the other polygons [47]. Voronoi is a dissection of the space plane that is characterized by the fact that any location within a polygon is closest to the sample point of that polygon (such as a residential point) and furthest from the sample point within an adjacent polygon, and that each polygon contains only one sample point. These sample points are the generators of Voronoi diagrams, which can be line sets, surface sets, or other complex sets of shapes, in addition to point sets [48].

Since the area and shape of each agricultural region in China is different, the NNI index cannot be compared horizontally in each agricultural region. Moreover, the criteria for defining cluster, random, and uniform distribution in determining the type of spatial distribution of point-like elements in the nearest neighbor index are not uniform. Therefore, this paper introduces the method of Tyson’s polygon coefficient of variation for retesting [49]. Specifically, this paper takes the points of interest of three types of NABEs as sample points and divides the set of polygons formed by continuous space based on the nearest target principle, with each point corresponding to a polygon. Since the area of the Voronoi polygons corresponding to the spatial distribution of sample points is different, the standard deviation and the mean of the polygon areas are compared. By calculating the coefficient of the variation of polygons and the NNI, the degree of the spatial variation of point-like elements can be further illustrated to determine the spatial distribution characteristics of NABEs in each period [50]. The formula is as follows:

$$R=\sqrt{\frac{{\displaystyle \sum }{\left(S_i-S\right)}^2}{n}}$$

(3)

$$CV=\frac{R}{S}$$

(4)

where R is the variance; $S_i$ is the area of the i th polygon (i = 1, 2, …, n); S is the mean value of polygon area; n is the number of polygon areas; CV is the coefficient of variation. Usually, the polygon area variation is positively correlated with the coefficient of variation (CV). When the point set type is uniformly distributed, the polygon variation is single and the area is similar; therefore, the CV value is relatively small. When the point set is aggregated, the variation of polygon area is larger, and the CV value is also larger. In this paper, the spatial distribution of point elements is classified according to the three suggested values proposed by Duyckaerts et al. [51]: when CV < 33%, the point set is “uniform” in its distribution; when 33% ≤ CV ≤ 64%, the point set is “random” in its distribution; when CV > 64%, the point set is “clustered” in its distribution.

3.2.4. Ripley’s K Function

Ripley’s K function is one of the common methods for spatial point element analysis, which is used to explore the clustering and diffusion of point elements in different distance ranges and is suitable for multi-scale spatial pattern analysis [52]. In this paper, we use the multi-distance spatial clustering analysis function of ArcMap10.2 software to calculate the observed K and expected K of spatial distances between three NABE sample points. The formula is as follows:

$$K\left(d\right)=\sqrt{\frac{A{{\displaystyle \sum }}_{i=1}^n{{\displaystyle \sum }}_{j=1}^{n}k\left(i,j\right)}{\pi n\left(n-1\right)}}$$

(5)

where n represents the number of NABEs; A represents the area of the study area; $k\left(i,j\right)$represents the actual distance between NABEs i and j in the range of d; K(d) is the observed K; the distance d is the expected K. When the observed K is greater than the expected K, it represents that the point group of the NABEs of this type is a clustered pattern; when the observed K value is less than the expected K, it represents that the point group of the NABEs of this type is a dispersed pattern. When the observed K is greater than the higher confidence envelope, it indicates that the spatial clustering of the distance is statistically significant; when the observed K is less than the lower confidence envelope, it indicates that the spatial dispersion of the distance is statistically significant. The degree of clustering is highest when the difference between observed K and expected K is the largest, and the corresponding distance at this time is the strongest aggregation scale (Figure 3).

3.2.5. The Pearson Correlation Analysis

There is no mature evaluation index system that can be used for research on the factors influencing the development of NABEs. The factors affecting the development of NABEs may be different because of the different production characteristics of each type of NABE. Family farms are affected to a higher degree by topography and soil and water resources [36]. The development of cooperatives is more scaled up relative to family farms and may be more affected by the level of mechanization and inputs and outputs [40,41]. Agricultural enterprises are highly specialized and organized and may be more affected by the level of economic development [42]. According to the above characteristics, and considering the availability of data and the representativeness of indicator selection, this article selects 22 influencing factors from three aspects, namely, socio-economic factors, agricultural production factors, and natural environment factors, to construct an evaluation index system for the development of NABEs (

Table 1. Evaluation index system for the development of new agricultural business entities.

Categories	Indices	Introduction
Socio-economic	Urbanization level (X1)	Non-farm population/total population
	Road area (X2)	Road area of each agricultural region (10,000 square meters)
	Gross domestic product (X3)	Gross product of each agricultural region (100 million yuan)
	Primary sector of the economy (X4)	Agricultural area primary industry output/total output value of each agricultural region (%)
	Agricultural output per capita (X5)	Agricultural output/rural population
	Consumption expenditure of rural residents (X6)	Consumption expenditure of rural residents in each agricultural region (yuan)
	Per capita disposable income (X7)	Disposable income per capita in each agricultural region (yuan/person)
	Producer price index for agricultural products (X8)	Producer price index for agricultural products by agricultural region
Agricultural production	Total agricultural machinery power (X9)	Total power of all agricultural machinery power in each agricultural region (kW)
	Area sown for crops (X10)	Area of crops sown in each agricultural region (hectares)
	Family contract farmland transfer area (X11)	Total area of family contracted arable land transferred in each agricultural region (hectares)
	Water for agriculture (X12)	Total agricultural water use by agricultural regions (100 million cubic meters)
	Fertilizer (fold the pure quantity) (X13)	Folded pure application of chemical fertilizer in each agricultural region (t)
	Amount of agricultural film used (X14)	Amount of agricultural film used in each agricultural region (t)
	Pesticides (X15)	The amount of pesticides applied in each agricultural region (t)
	Diesel fuel (X16)	Application of agricultural diesel in each agricultural region (t)
	Rural electricity consumption (X17)	Total rural electricity consumption in each agricultural region (million kWh)
Natural Environment	Area removed from waterlogging (X18)	Total area of waterlogging removal in agricultural region (thousand hectares)
	Soil erosion area (X19)	Total area of soil erosion in each agricultural region (thousand hectares)
	Sunshine hours (X20)	Average annual sunshine hours for each agricultural region representative city (h)
	Average annual temperature (X21)	Average annual temperature in representative cities of each agricultural region (°C)
	Precipitation (X22)	Sum of precipitation in each agricultural region (mm)

). Considering the influence of the covariance of the selected indicators on the research results, the covariance diagnosis of each indicator was carried out separately, and the results of the covariance test of each indicator in the system met the research requirements; therefore, it was considered that there was no covariance among the indicators. Pearson correlation analysis can be used to quantitatively analyze the intrinsic relationship between the number of NABEs and the influencing factors of three dimensions. The formula is as follows:

$$P=\frac{cov\left(X,Y\right)}{\sqrt{var\left(X\right)}\sqrt{var\left(Y\right)}}$$

(6)

where P is the overall correlation coefficient, cov(X,Y) is the covariance of the random variable X/Y, and var(X) /var(Y) represents the variance of X and Y, respectively. The correlation coefficient between samples is calculated as:

$$r=\frac{{{\displaystyle \sum }}_{i=1}^n\left(x_i-\overline{x}\right)-\left(y_i-\overline{y}\right)}{\sqrt{{{\displaystyle \sum }}_{i=1}^n{\left(x_i-\overline{x}\right)}^2}\sqrt{{{\displaystyle \sum }}_{i=1}^n{\left(y_i-\overline{y}\right)}^2}}$$

(7)

where x and y are the mean values of the two groups of samples, respectively, and the value range of r is [–1, 1] [53]. When the correlation coefficient is positive, it indicates that the influence factor is positively correlated with the development of NABEs and vice versa. An absolute value of r between 0.8 and 1.0 is a very strong correlation, an r between 0.6 and 0.8 shows a strong correlation, an r between 0.4 and 0.6 shows a moderate correlation, an r between 0.2 and 0.4 shows a weak correlation, and an r below 0.2 means no correlation [54].

4. The Analysis of the Spatial Distribution Characteristics of NABEs

4.1. The Spatial and Temporal Change Process of NABEs

From the overall distribution range, all kinds of NABEs show the distribution characteristics of “more in the southeast and less in the northwest”, but the distribution of different NABEs in different agricultural regions is slightly different. In 2012, family farms were widely distributed in all agricultural regions, mainly in the Northeast China Plain, the Huang-Huai-Hai Plain, the Middle–Lower Yangtze Plain, and Southern China. A small reduction in distribution occurred in 2016, with a more pronounced decrease in the Northeast China Plain, the Huang-Huai-Hai Plain, and Southern China. By 2021, the distribution range of family farms increases again, concentrated in the Northeast China Plain, the Huang-Huai-Hai Plain, the Middle–Lower Yangtze Plain, Southern China, and the eastern part of the Sichuan Basin and surrounding regions (Figure 4A–C).

The east–west pattern of cooperatives is more pronounced than that of the family farms. Cooperatives are concentrated in the eastern part of the Huang-Huai-Hai Plain and the Middle–Lower Yangtze Plain in small amounts in 2012, and a small spread occurs in all directions with the two at the center in 2016. By 2021, the distribution range of cooperatives emerged in a similar pattern to that of family farms, mainly concentrated in the Northeast China Plain, the Huang-Huai-Hai Plain, as well as in the northern part of the Middle–Lower Yangtze Plain, Southern China, and the eastern part of the Sichuan Basin and surrounding regions (Figure 4D–F).

Compared with cooperatives, the distribution of agricultural enterprises is more southerly. In 2012, the number of agricultural enterprises was small, forming two small sporadic aggregations in the Yangtze River Delta and the Pearl River Delta. In 2016, the number of distributions in each agricultural region increased, with more obvious increases in the middle and lower reaches of the Middle–Lower Yangtze Plain and the Huang-Huai-Hai Plain. In 2021, the distribution of agricultural enterprises is further expanded, aggregating the Huang-Huai-Hai Plain, the eastern and western parts of the Middle–Lower Yangtze Plain, Southern China, and the eastern part of the Sichuan Basin and surrounding regions (Figure 4G–I).

Overall, the distribution of family farms is the widest and the distribution of agricultural enterprises is the sparsest, except for a small reduction in the distribution of family farms in 2016, and the distribution of cooperatives and agricultural enterprises is expanding. This verifies Hypothesis 1, i.e., that there are regional differences in the spatial distribution of NABEs. The Huang-Huai-Hai Plain, the Middle–Lower Yangtze Plain, Southern China, and the Sichuan Basin and surrounding regions have become “hotbeds” for the development of three types of NABEs.

In terms of quantity changes, the number of various types of NABEs in each agricultural district shows obvious stage differences (

Table 2. Number of new agricultural business entities in each agricultural region of China in 2012, 2016, and 2021.

Type	Year	NC	HHH	MY	SC	LP	SBR	YP	NSR	TP
family farms	2012	1024	716	1383	3343	194	103	802	1063	116
	2016	580	677	1235	1767	147	106	616	631	89
	2021	1730	4215	11,687	4214	607	3672	1859	2675	0
cooperatives	2012	51	235	352	18	41	55	13	36	2
	2016	138	462	460	56	116	91	79	55	2
	2021	870	3393	3249	550	481	686	492	588	96
agricultural enterprises	2012	64	85	146	47	4	29	14	29	2
	2016	109	207	351	115	28	67	106	108	12
	2021	164	1383	2757	970	323	816	560	326	31

). The number of cooperatives and agricultural enterprises has been increasing in all agricultural regions, while family farms show a trend of first decreasing and then increasing. In terms of regional differences, family farms were the most numerous in 2012 and 2016 in Southern China, followed by the Middle–Lower Yangtze Plain; however, in 2021, they surged in the Middle–Lower Yangtze Plain (11,687), far exceeding the Huang-Huai-Hai Plain (4215) and Southern China (4214). The regions with the highest number of cooperatives in 2012, 2016, and 2021 are located in the Middle–Lower Yangtze Plain and the Huang-Huai-Hai Plain, and the number of cooperatives in the two regions is very close in each period. By 2021, cooperatives reach the largest number in the Huang-Huai-Hai Plain (3393), followed by the middle and lower reaches of the Middle–Lower Yangtze Plain (3249). The number of agricultural enterprises was more concentrated in both the Middle–Lower Yangtze Plain and the Huang-Huai-Hai Plain in 2012 and 2016, and the largest in the middle and lower reaches of the Middle–Lower Yangtze Plain (2757), the Huang-Huai-Hai Plain (1383), and Southern China (970) in 2021. Due to the influence of climate and topography, the three types of NABEs in the Tibet Plateau Region and the Loess Plateau have the lowest number. Among them, the number of family farms in the Tibet Plateau Region experienced a decline. Overall, the number of NABEs varies widely among different agricultural regions. The Middle–Lower Yangtze Plain and the Huang-Huai-Hai Plain are the gathering regions with the largest number of family farms, cooperatives, and agricultural enterprises, in addition to cooperatives in the Northeast China Plain and family farms and agricultural enterprises in Southern China.

4.2. The Spatial Pattern of the Kernel Density of NABEs

The results of nuclear density estimation show that the high-density aggregation centers of family farms in 2012 and 2016 are mainly concentrated in the central part of Hainan Island in Southern China. The core range is further expanded in 2021, and the high-density aggregation centers are concentrated in the eastern part of the Middle–Lower Yangtze Plain and the central part of the Sichuan Basin and surrounding regions (Figure 5A–C).

Cooperatives in the 2012 high-density aggregation center were concentrated in the eastern part of the Middle–Lower Yangtze Plain. In 2016, the high-density agglomeration range, in addition to the eastern part of the Middle–Lower Yangtze Plain and the northern part and central part of the Huang-Huai-Hai Plain of the agglomeration, is more obvious. In 2021, the high-density aggregation centers are further expanded in the Huang-Huai-Hai Plain, while small high-density areas appear in the eastern and western parts of the Middle–Lower Yangtze Plain and the central part of the Sichuan Basin and surrounding regions (Figure 5D–F).

In 2012, the high-density aggregation area of agricultural enterprises was concentrated in the Yangtze River Delta in the Middle–Lower Yangtze Plain and the Pearl River Delta in Southern China. In 2016, the high-density range shifted to the northern part of the Huang-Huai-Hai Plain. In 2021, the scope was further expanded, except for the northern part of the Huang-Huai-Hai Plain and the eastern part of the middle and lower reaches of the Middle–Lower Yangtze Plain, spreading to the western part of the Middle–Lower Yangtze Plain and the southern part of Southern China, as well as the central part of the Sichuan Basin and surrounding regions (Figure 5G–I).

Overall, the high-density areas of family farms, cooperatives, and agricultural enterprises all spread from east to west. However, the high-density areas of family farms and cooperatives are smaller and more concentrated, with the former leaning to the south and the latter to the north. Conversely, the high-density areas of agricultural enterprises are more numerous and scattered, and, compared to the other two NABEs, the high-density areas of the agricultural enterprises are not only to the south but also to the west.

4.3. Changes in the Spatial Aggregation Types of NABEs

The results of the nearest neighbor index analysis (

Table 3. Mean nearest neighbor index and coefficient of the variation of new agricultural business entities in different agricultural regions.

Type	Year	Index	NC	HHH	MY	SC	LP	SBR	YP	NSR	TP
family farms	2012	NNI	0.42	0.50	0.50	0.39	0.59	0.73	0.37	0.20	0.37
		CV (%)	205.10	151.04	201.45	278.60	147.11	158.05	288.78	299.59	239.51
	2016	NNI	0.45	0.53	0.52	0.42	0.55	0.75	0.42	0.20	0.36
		CV (%)	231.77	201.84	203.99	289.74	135.00	241.74	264.16	281.06	118.73
	2021	NNI	0.20	0.21	0.46	0.19	0.29	0.29	0.22	0.29	/
		CV (%)	250.14	283.71	181.39	236.18	161.41	916.58	215.31	423.18
cooperatives	2012	NNI	0.51	0.53	0.49	1.39	0.68	0.68	0.78	0.41	/
		CV (%)	132.42	130.72	254.12	31.34	108.26	206.99	93.66	120.18
	2016	NNI	0.30	0.54	0.52	0.44	0.37	0.53	0.43	0.31	/
		CV (%)	181.04	141.07	145.50	125.24	122.06	245.44	82.12	111.77
	2021	NNI	0.32	0.24	0.55	0.30	0.46	0.55	0.63	0.34	0.44
		CV (%)	254.55	195.47	138.43	185.03	153.57	536.04	125.32	274.01	266.57
agricultural enterprises	2012	NNI	0.46	0.72	0.65	0.87	4.09	0.64	0.93	0.73	/
		CV (%)	102.05	127.39	185.13	210.56	47.92	130.14	86.04	141.24
	2016	NNI	0.36	0.58	0.57	0.53	0.61	0.52	0.42	0.30	0.99
		CV (%)	106.72	149.96	155.31	149.87	102.85	214.28	169.74	167.43	145.94
	2021	NNI	0.53	0.52	0.47	0.41	0.45	0.51	0.47	0.24	0.45
		CV (%)	143.15	193.40	135.50	218.68	178.88	653.10	294.10	255.57	169.34

) show that the NNI of all NABEs in all agricultural regions is less than 1, except for the cooperatives in Southern China and the agricultural enterprises in the Loess Plateau, where the NNI index is greater than 1. This indicates that most of the NABEs show a more stable and significant aggregation in most regions and that only cooperatives in Southern China and the agricultural enterprises in the Loess Plateau experienced a shift from random distribution to significant aggregation. By 2021, the NABEs in all agricultural regions maintained their aggregation distribution characteristics. The results of the Tyson polygon coefficient of variation are all higher than 64%, indicating that all types of NABEs show aggregation in most of the times and regions studied, which confirms the conclusion of the NNI index.

The CV index allows us to compare the degree of aggregation in each agricultural region. The results show that family farms have the highest CV value in the Northern Arid and Semi-Arid Region, and the Yunnan–Guizhou Plateau, in 2012, indicating the greatest aggregation of farms in these regions. In 2016, it shifted to Southern China and the Northern Arid and Semi-Arid Regions, and, in 2021, the most aggregated areas of family farms shifted to the Sichuan Basin and surrounding regions and the Northern Arid and Semi-Arid Region, with the largest increase in the degree of aggregation in the Sichuan Basin and surrounding regions (a 674.84% increase from 2016 to 2021) (Table 3, lines 2–7).

The cooperatives were most aggregated in the middle and lower reaches of the Middle–Lower Yangtze Plain and the Sichuan Basin and surrounding regions in 2012, and this shifted to the Northeast China Plain and the Sichuan Basin and surrounding regions in 2016, with a CV increase of 290.60% in the Sichuan Basin and surrounding regions from 2016 to 2021. On the contrary, the CVs of the cooperatives in the Middle–Lower Yangtze Plain were higher in 2012, while the CVs decreased by 115.69% from 2012 to 2021, indicating that the degree of aggregation of the cooperatives in this region decreased significantly (Table 3, lines 8–13).

Agricultural enterprises gathered to the greatest extent in Southern China, the Middle–Lower Yangtze Plain, and the Northern Arid and Semi-Arid Region in 2012. It shifts to the Sichuan Basin and surrounding regions, as well as the Yunnan–Guizhou Plateau, in 2016 and 2021, with CV increases of up to 438.82% in the Sichuan Basin and surrounding regions from 2016 to 2021, respectively (Table 3, lines 14–19).

In general, the aggregation degree of the three types of NABEs in Southern China, the Northern Arid and Semi-Arid Region, the Sichuan Basin and surrounding regions, the Middle–Lower Yangtze Plain, and the Yunnan–Guizhou Plateau is more significant. It is worth noting that, although the number of the three types of NABEs in the Sichuan Basin and surrounding regions is not high, the growth rate of the spatial aggregation degree is the fastest. On the contrary, except for the low degree of aggregation caused by the low number of some NABEs in the Qinghai-Tibet Plateau, family farms in the Loess Plateau, the cooperatives in the Yunnan–Guizhou Plateau, and agricultural enterprises in the Northeast China Plain show aggregation, but the degree of aggregation is low.

4.4. Different Spatial Aggregation Scales of NABEs

Through the multi-distance spatial clustering analysis of NABEs in different periods, the observed K of all NABEs is larger than the expected K, and all of them are higher than the maximum value of the confidence interval, passing the significance test at a 99% confidence level. The distance where the difference between the observed K and expective K is the largest is the distance where the NABEs gather to the strongest extent, which is the “strongest aggregation scale” (

Table 4. The strongest aggregation scales of new agricultural business entities in different agricultural regions in 2012, 2016, and 2021.

Type	Year	NC	HHH	MY	SC	LP	SBR	YP	NSR	TP
family farms	2012	201.39	164.81	164.28	152.48	107.56	62.09	117.88	335.43	283.88
	2016	71.87	143.78	125.48	114.22	59.51	51.63	178.49	333.96	/
	2021	161.68	223.19	239.88	114.30	222.94	293.13	144.75	421.61	/
cooperatives	2012	111.73	102.21	187.80	/	83.42	76.95	/	110.40	/
	2016	163.21	106.10	95.28	66.21	90.08	89.76	100.14	107.43	/
	2021	209.26	263.03	128.97	113.20	130.80	168.40	142.51	418.80	237.36
agricultural enterprises	2012	38.02	23.00	183.76	78.65	/	65.79	24.22	89.57
	2016	36.51	58.54	184.52	38.41	/	43.94	44.68	73.63	/
	2021	265.26	189.61	99.66	98.10	119.67	112.77	181.12	394.43	116.00

). This “strongest aggregation scale” is an ideal radius, implying that, within this radius, the aggregation of NABEs is the strongest.

In terms of family farms, in 2012 and 2016, the strongest aggregation scale is the smallest in the Sichuan Basin and surrounding regions (62 km and 51 km, respectively), and the strongest aggregation scale is the largest in the Northern Arid and Semi-Arid Region (335 km and 333 km, respectively). In 2021, the strongest aggregation scale of family farms in Southern China is the smallest (114 km), while the Sichuan Basin and surrounding regions have the largest change in aggregation scale, with the aggregation range increasing from 51 km to 293 km (Table 4, lines 2–4).

In terms of the cooperatives, their strongest aggregation scales in 2012 are the smallest in the Sichuan Basin and surrounding regions and the Loess Plateau (76 km and 83 km, respectively) and the largest in the Middle–Lower Yangtze Plain (187 km). In 2016 and 2021, the strongest aggregation scales of the cooperatives are the smallest in Southern China (66 km and 113 km, respectively). On the contrary, the range of the strongest aggregation scale of the cooperatives in the Northern Arid and Semi-Arid Region surges from 107 km to 418 km (Table 4, lines 5–7).

In terms of agricultural enterprises, the strongest aggregation scale in all agricultural regions tends to increase. In 2012, the strongest aggregation scale is the smallest in the Huang-Huai-Hai Plain (23 km), and, in 2016, the strongest aggregation scale is the smallest in the Northeast China Plain (36 km), while, in 2021, it is the smallest in Southern China (98 km). The Tibet Plateau Region and the Northern Arid and Semi-Arid Region maintained a larger aggregation scale (Table 4, lines 8–10).

Overall, the strongest aggregation scales of the different types of NABEs are increasing. The minimum value of the strongest aggregation scale for agricultural business is smaller than that for family farms and cooperatives in all years, thus disproving Hypothesis 2. This may imply that there are large differences in the levels of the economic development of different regions and that agricultural enterprises are preferentially distributed in regions with a higher level of economic development due to market distance and accessibility. Therefore, at the time of the highest level of agricultural enterprises’ agglomeration, agricultural enterprises are distributed in the smallest agglomeration areas. It is worth noting that, although the number of the three types of entities in South China is not the largest, its strongest aggregation scale is the smallest, which confirms that the aggregation degree of NABEs in the region is the most obvious.

5. The Factors Influencing the Development of Different NABEs

The analysis of the influencing factors on the number of NABEs shows that different types of influencing factors affect the three types of NABEs differently. The strong influencing factors for the development of family farms are X2, X3, X4, X6, X7, and X17. The strong influencing factors of specialized farmers’ cooperatives development are X2, X9, X10, X15, and X16. The strong influencing factors for agricultural enterprises’ development are X2, X3, X4, X17, and X22 (Figure 6) (please refer to Table 1 for the specific meanings of the variables.). Thus, Hypothesis 3 was tested, and the influencing factors favored by the development of different NABEs differed, especially in the degree of dependence on soil and water resources, the economic environment, and the efficiency of the scale of production.

In terms of family farms, all influencing factors have a positive relationship with the development of family farms, with the main category being socioeconomic factors. In addition, rural electricity consumption is more prominent, and the degree of influence of traditional agricultural inputs such as pesticides and fertilizers, as well as agricultural water use, on family farms, is gradually weaker. It is evident that the development of family farms is more limited by the living standard of farmers and the accessibility of transportation in the region. When traditional agricultural inputs are more moderate, the larger the scale of the agricultural industry, and the higher the consumption level of rural residents, the greater the number of family farms. On the contrary, the lower the per-capita agricultural output value, the smaller the number of family farms and the slower the development. Combined with the characteristics of family farms with a single family-based operation model, the larger the family farm and the easier it is to obtain more financial subsidies and technical support. The higher the per-capita consumption level, the more the farm owners can afford to bear certain financing risks, which, to a certain extent, lays the foundation for the development of family farms.

In terms of the cooperatives, the agricultural output value per capita was negatively correlated with the development of cooperatives, while the rest of the factors showed positive correlations. The main influencing factors are agricultural production factors, in addition to the road area being also more sensitive to the development of cooperatives. It is evident that the development of cooperatives is more limited by agricultural inputs, such as pesticides, diesel, sowing areas, and the level of mechanization, which is related to the nature of the joint operation of cooperatives themselves. The joint operation among farmers is more capable of achieving economies of scale and is more conducive to the mechanized farming of large areas, saving resources, and increasing benefits [55]. In addition, this “grouping” effect among farmers and cooperatives is more likely to occur in areas with low agricultural output per capita. When farmers’ agricultural output is generally low, the desire for cooperative production to improve efficiency is stronger, while, in areas with higher agricultural output, the cooperative production model is less driven by farmers.

Similarly, among the factors influencing the development of agricultural enterprises, agricultural output per capita is negatively correlated with the development of cooperatives, while the rest of the factors show positive correlations. The main influencing factors are socio-economic factors, in addition to rural electricity consumption and precipitation, which are also sensitive to the development of agricultural enterprises. It is evident that the development of agricultural enterprises is more limited by the level of economization of the region. The higher the accessibility, the more developed the economy, and the larger the scale of the agricultural industry, the higher the number of agricultural enterprises; moreover, the lower the overall level of development of the region, the lower the number of agricultural enterprises. This may be related to the dependence of agricultural enterprises on economic and market development. The overall economic development of the region will make it easier for agricultural enterprises to obtain financial support and higher-end market resource allocation, thus providing sufficient impetus for product sales and logistics. In addition, similar to the cooperatives, the higher the agricultural output per capita, the lower the desire of farmers to join agricultural enterprises, and, conversely, when agricultural production is less efficient, the interest of farmers to join agricultural enterprises is stronger.

6. Discussion

6.1. Differences in the Types of NABEs

Taking family farms as an example, their main characteristics are a wide distribution, large number, large aggregation degree, the strongest aggregation scale is small, and sensitivity to comprehensive rural development conditions. This is due to the lower threshold for the establishment of family farms and the often more homogeneous business model. In addition, they have fewer restrictions on land size, technology, and personnel; therefore, the influence of locality on family farms is relatively weak, and they are growing more rapidly nationwide. It is worth noting that family farms are influenced by indicators such as the proportion of primary industry and rural residents’ consumption expenditure, and they are NABEs of the rural development-oriented type. This is similar to the findings that the development of family farms in hilly mountainous areas is influenced by the level of rural affluence and that the level of urbanization is an external factor in the generation of family farms [56,57]. Therefore, traditional agricultural regions with better basic conditions for agricultural production, as well as transportation economy, are the regions where family farms grow the most rapidly (

Table 5. Development characteristics of different new agricultural business entities.

Categories	Family Farms	Cooperatives	Agricultural Enterprises
Spatial and temporal change process	Most widespread, most numerous, and decreasing then rapidly increasing	The distribution of east–west pattern is obvious, and the number has been increasing and shows local leadership	Most sparsely distributed overall to the south, least numerous but increasing all the time
Spatial aggregation scales	The strongest aggregation scale is increasing, and the minimum value is the smallest among the three types of NABEs, but the increase is the largest	The strongest aggregation scale keeps increasing, and the minimum value is moderate among the three types of NABEs with the smallest increase	The strongest aggregation scale is increasing, and the minimum is the largest and moderately increases among the three types of NABEs
Inter-regional differences	Has been significantly aggregated, with the greatest degree of aggregation in the Sichuan Basin and surrounding regions and the least degree of aggregation in the Loess Plateau, with the greatest difference in CV values	Discrete first and then aggregated, greatest aggregation in the Sichuan Basin and surrounding regions, least aggregation in the Yunnan–Guizhou Plateau, moderate difference in CV values	Discrete first and then aggregated, with maximum aggregation at the Sichuan Basin and surrounding regions, minimum aggregation at the Loess Plateau, and minimum difference in CV values
Influencing factors	Rural economy-oriented development	Agricultural input-oriented development	Market economy-oriented development

).

Taking the cooperatives as an example, their main characteristics are regional aggregation (e.g., in the central Sichuan Basin and surrounding regions), a clear east–west distribution pattern in China, and a greater sensitivity to the level of agricultural capital inputs. Unlike the small-scale regional scope of the Beijing–Tianjin–Hebei cooperative, which is subject to mall demand outcomes, the large-scale east–west distribution pattern is distinct and more sensitive to the level of agricultural capital inputs [58]. This is due to the larger scale and high mechanization of the cooperatives and the closer ties among farmers, meaning they develop faster in the economically strong and densely populated eastern regions and more slowly in the poor mountainous western regions, showing a pattern of east–west differences. In addition, cooperatives require close association among farmers and between cooperatives; therefore, their agglomeration development is weaker over long distances. Finally, cooperatives are larger and more efficient than family farms; therefore, the level of productive capital inputs, such as land and mechanized inputs, becomes the dominant factor in the development of cooperatives.

Taking agricultural enterprises as an example, their main characteristics are that they are small and scattered in number, the strongest gathering scale is large, the spatial distribution pattern is partial to South China, and they are influenced by the market economic environment. This is consistent with the conclusion of some scholars that agricultural enterprises are clustered in the east and fragmented in the west [59]. The main reason for this is that agricultural enterprises are no longer limited to a small area of development but instead have begun to expand over longer distances. The core contribution to the development of agricultural enterprises is the socio-economic factors, which present a new type of market economy-oriented NABE. Similar to the finding that the development of agribusiness in Sichuan Province is constrained by the level of the urban economy, it is evident that the development of agricultural enterprises is more subject to the role of market allocation and the overall economic development level of the region, especially in the Middle–Lower Yangtze Plain, the Huang-Huai-Hai Plain, and Southern China, where the natural and economic base of agriculture is better.

6.2. Key Agricultural Regions for the Development of Clustered NABEs

To find regions suitable for the development of NABEs, this paper extracted the regions with the densest distribution, the highest nuclear density, the highest CV of Tyson polygon, and the smallest scale of strongest aggregation, performed spatial superposition, and found regions that satisfy at least one of the above four conditions (Figure 7). The results found that the Huang-Huai-Hai Plain, the Middle–Lower Yangtze Plain, Southern China, and the Sichuan Basin and surrounding regions have become fertile ground for the development of NABEs.

The Huang-Huai-Hai Plain, covering five regions in the Hebei Province, Beijing, Tianjin, Shandong Provinces, and Henan Province, is an advantageous area for the number and high density of NABEs to gather. The reason for this is that the Huang-Huai-Hai Plain has superior natural location conditions and is a densely populated, economically active, and important grain-producing region in China [60]. The superior topographic conditions coupled with the advanced mechanization development have steadily increased the development of NABEs. However, the large population in this agricultural area has resulted in less arable land per capita, and the rapid exploitation of groundwater has caused the regional groundwater level to drop, forming a deep groundwater landing funnel area centered on urban water sources. The serious shortage of water resources has become the biggest obstacle to the sustainable development of agriculture in the Huang-Huai-Hai Plain [61,62]. Therefore, it is advisable for this agricultural area to make full use of the advantages of its natural resources and agricultural base in its future development and to vigorously carry out and innovate water conservation systems, cultivate modern NABEs, and maximize the benefits of agricultural production under limited resource conditions.

The Middle–Lower Yangtze Plain covers the Jiangsu, Anhui, Zhejiang, Hubei, Hunan, and Jiangxi provinces, as well as Shanghai, similar to the Huang-Huai-Hai Plain, which is a high-density area where NABEs gather. This agricultural area mainly consists of alluvial deposits from the Yangtze River and its tributaries, with a dense water network, flat and fertile land, a strong resource and environmental carrying capacity, and excellent conditions for agricultural development [63]. However, the development of NABEs in this region is not caused by a single advantageous condition, but rather by the combination of a dense population and economic activities, a well-developed road network construction, and a broad market, enabling the development of NABEs to be more and more mature, showing its spatial “grouping” phenomenon. However, in recent years, the economic contribution of agricultural development to the region has gradually weakened, and the problems of an aging population and agricultural land abandonment have gradually become more prominent [64]. There is an urgent need to innovate the land transfer system and business model, improve the efficiency and effectiveness of agricultural land utilization, and cultivate new green and ecologically diversified NABEs that integrate pre-production, production, and post-production.

The Southern China region covers Guangdong Province, Hainan Province, and Fujian Province and is the agricultural region with the strongest aggregation scale of NABEs. The region has a long coastline, a better coupling of light, heat, and water resources, and is an important cash crop-growing area in China [65]. The population density is second only to the Huang-Huai-Hai Plain, and its good land resource endowment and superior transportation conditions lay a good foundation for the development of NABEs. This agricultural area is the frontier of China’s economic construction and opening up to the outside world, and its unique location advantage has become a strong supporting and driving force for agricultural and rural development [66]. However, with the development of industrialization and urbanization, the regional agricultural geographical function has declined significantly. Relying on the existing agricultural foundation and location conditions in the region, it is advisable to promote agricultural industrialization, informatization, and marketization in the future, and to vigorously develop outward-oriented, urban, and high-quality modernized, multifunctional NABEs.

The Sichuan Basin and surrounding regions cover the Sichuan Province and Chongqing Municipality, and family farms, cooperatives, and agricultural enterprises have the highest degree of aggregation in this agricultural region. The western part of the region is undulating, spanning several geomorphic units, such as the Tibetan Plateau, the Transverse Ranges, and the Sichuan Basin, with obvious vertical zonation and fragmented and small-scale distribution of arable land, as well as a low degree of agricultural mechanization [67]. However, the bottom area of the Sichuan Basin is relatively gentle, with a well-developed transportation network centered on Chengdu and a variety of agricultural types [68]. The low and flat terrain and good accessibility provide favorable conditions for the development of NABEs, which, to a certain extent, creates a clustering effect of NABEs in the central-eastern part of this agricultural region. Therefore, under the limited natural conditions, the further cultivation of NABEs should pay more attention to the local conditions and rely on the economic base and resource advantages of the region with Chengdu and Chongqing as the twin centers, as well as select modernized NABEs with regional characteristics that can drive industrial development. This not only meets the future urbanization development needs of the agricultural area but also forms an agricultural ecological barrier to provide direction for the future modernization of agriculture. As the topographic features of the Sichuan Basin and surrounding regions are dominated by basins, hills, low mountains, and plains, the development of a large number of NABEs in the basin area will aggravate the surface source pollution in the small area due to the use of pesticides and chemical fertilizers, which will lead to the destruction of the natural environment in the area. Therefore, the development of NABEs in this agricultural area must strengthen the control of green agriculture.

Through the analysis of the above key agricultural regions, Hypothesis 1 is again verified, and the development of NABEs is more prominent in areas with higher agricultural location advantages (abundant labor force, vast arable land, and high level of economic development).

6.3. The Road of Chinese Characteristics for the Development of NABEs

The various reasons presented here show that the development of NABEs in China cannot copy the model of large farms in Europe and the United States, and it is also difficult to copy the road of agricultural modernization based on high-density mechanization in developed countries, such as Japan. First, it is difficult to change the pattern of fragmentation of arable land and the small-scale production pattern in China in the short term, and the farmers’ deep attachment to the land, as well as the incomplete land transfer mechanism, limit the expansion of land area by NABEs [5,6]. In addition, although China and India are similar in smallholder production, smallholder production cannot enter the market on a large scale, and the production structure of single organic agriculture in China is not suitable [69,70]. Second, the occupation of arable land by construction land makes the existing arable land and new arable land not contiguous. In addition, due to China’s special situation of smallholder farmers, there are still many mountainous areas of arable land, which makes it impossible for NABEs to develop on a large scale in all areas [71,72]. Third, since the reform of China’s tax sharing system in 1994, the central government has extracted 50% of the local tax revenue, leading local governments to prioritize the investment of funds in areas with fast economic benefits (e.g., real estate and high-tech industries, etc.) in order to generate revenue, reducing the funding for agriculture, which has lower economic benefits [73]. This has caused a shortage of funds for NABEs, leading to difficulties in obtaining loans and insurance, making their development prone to short-sightedness and difficulty in the long term.

Therefore, this paper argues that, in the process of developing NABEs, regional differences should be emphasized, and a small number of leading agricultural enterprises and model cooperatives should drive a large number of small-scale family farms, as well as smallholder farmers, in order to become a characteristic path for China’s agricultural development. The realization path is as follows:

First, promote the differentiated development of NABEs among regions. Aiming at the characteristics of the Sichuan Basin and surrounding regions with the strongest aggregation of NABEs, and combining the topographic characteristics of the region with large slopes, the precise layout should rely on the existing industrial base and develop NABEs that rely on local resource endowment and small-scale boutique specialty industries. Aiming at the characteristics of the strongest aggregation scale, regarding Southern China, with the smallest scale, we combine the superior topography and climatic conditions and the capital advantages of the Pearl River Delta to improve the association of NABEs with other industries, as well as to strengthen scientific and technological support, management support, and industrial planning for NABEs and expand from the coast to the inland to create NABEs with driving industrial development. In the Huang-Huai-Hai Plain and the Middle–Lower Yangtze Plain, although the NABEs are developing rapidly, the problems of arable land and water resources are more serious. Therefore, it is advisable to pay more attention to green development behaviors such as environmental improvement, more efficient use of resources, and the improvement of the production processes and quality of agricultural products to promote the green development of NABEs. The Loess Plateau and the Tibet Plateau Region are ecologically fragile, alpine, and remote; therefore, the government’s leadership and direction should be strengthened to break the industry and regional barriers, make up for the disadvantages of the inherent lack of geographical conditions, and vigorously develop NABEs in the special industries of the plateau region.

Second, the parallel development among different NABEs should be realized. Based on the research results of this paper, we strengthen the cooperation and association among various NABEs and the motivation of family farms, cooperatives, and agricultural enterprises to smallholder farmers, and build a bridge of mutual assistance and commonality among various NABEs and between smallholder farmers and NABEs. This paper concludes that family farms are more subject to rural economic factors and have the widest regional distribution, cooperatives are subject to agricultural production, and agricultural enterprises are subject to the market economic environment; the strongest aggregation scale is the largest, having a long industrial chain, and are the basic entities of agricultural products processing and circulation. Therefore, this can be an opportunity for family farms and cooperatives to focus on pre-production and production, with family farms producing high-quality agricultural products and cooperatives driving family farms to improve the mechanization of agricultural production. Agricultural enterprises focus on post-production and the processing and circulation of agricultural products provided by family farms and cooperatives, as well as solve the problem of production and marketing, broaden the sale channels for agricultural products, and gradually improve the entire agricultural industry chain. At the same time, the bonding between smallholder farmers and NABEs is enhanced, with family farms supporting farmers to improve production specialization and land output rate, cooperatives focusing on organizing farm production and providing modern services, and agricultural enterprises focusing on leading and organizing farmers to enter the market. The formation of the agricultural production model of “small farms + NABEs” and “NABEs + NABEs” is the only way to build agricultural modernization with Chinese characteristics and realize the development of NABEs specifically for China.

6.4. Limitation

First, this study makes a comparative analysis of the changes in the three NABEs, but the competition and cooperation mechanism between different NABEs deserves further research. For example, agricultural enterprises are more closely linked to the government and more flexibly linked to the sales market; thus, to a certain extent, they will encroach on the survival space of family farms. Second, POI data are time-sensitive and cannot reflect the distribution of the three NABEs more objectively, and some field research data will be further supplemented in the future. Finally, this paper lacks research on the actual effect of NABEs on traditional small farmers. This relationship deserves our attention, and further research will be performed in the future.

7. Conclusions

Based on the POI data of family farms, cooperatives, and agricultural enterprises among NABEs in China from 2012 to 2021, this paper explores the spatial distribution patterns and aggregation scale differences of various NABEs in different regions through various geospatial analysis methods and reveals the mechanisms of influence on the development of various NABEs by incorporating Pearson correlation analysis. The results show that:

First, the distribution of different NABEs in different agricultural areas is different and the number varies greatly, but they all show the distribution characteristics of “more in the southeast and less in the northwest”. The distribution of family farms is the widest and the distribution of agricultural enterprises is the sparsest, except for a small reduction in the distribution of family farms in 2016, and the distribution of cooperatives and agricultural enterprises is expanding. The Middle–Lower Yangtze Plain and the Yellow and the Huang-Huai-Hai Plains are the gathering areas with the largest number of family farms, cooperatives, and agricultural enterprises. The Huang-Huai-Hai Plain, the Middle–Lower Yangtze Plain, Southern China, and the Sichuan Basin and surrounding regions have become the key agricultural areas for the development of three types of NABEs.

Second, the high-density areas of family farms, cooperatives, and agricultural enterprises all spread from east to west. The high-density areas of family farms and cooperatives are more concentrated, with the former to the south and the latter to the north. The high-density areas of agricultural enterprises are more and more scattered, and the high-density areas are not only to the south but also to the west.

Third, most NABEs show a more stable and significant aggregation in different agricultural regions, and only cooperatives in Southern China and agricultural enterprises in the Loess Plateau experience a shift from random distribution to significant aggregation. The three types of NABEs in Southern China, the Northern Arid and Semi-Arid Region, the Sichuan Basin and surrounding regions, the Middle–Lower Yangtze Plain, and the Yunnan–Guizhou Plateau show a significant degree of aggregation. The number of the three types of NABEs in the Sichuan Basin and surrounding regions is not high, but the growth rate of the degree of spatial aggregation is the fastest. Family farms in the Loess Plateau, cooperatives in the Yunnan–Guizhou Plateau, and agricultural enterprises in the Northeast China Plain show aggregation, but the degree of aggregation is low.

Fourth, the strongest aggregation scales of the different NABEs have their own strongest aggregation scales between them, and they are increasing. The strongest aggregation scales of agricultural enterprises are smaller than those of family farms and cooperatives in all agricultural regions, which confirms that the aggregation degree of NABEs in the region is the most obvious.

Fifth, the development of NABEs is subject to the combined effects of social economy, agricultural production, and the natural environment. The development of cooperatives is influenced by agricultural production factors, which are more limited by traditional agricultural inputs such as pesticides, diesel fuel, sowing area, and mechanization level. The factors influencing the development of family farms and agricultural enterprises are socio-economic factors. The development of NABEs is more subject to the standard of living of farmers and the accessibility of transportation and is a kind of rural economy-oriented development, while the development of cooperatives is more subject to the level of the economization of the region and is a kind of market economy-oriented development.

Finally, this paper argues that paying attention to regional differences and channeling a large number of small-scale family farms, as well as smallholder farmers, through a small number of leading agricultural enterprises and model cooperatives should become a characteristic path that is in line with China’s national conditions. In addition, the differentiated development among regions and the parallel development among different NABEs are the key means with which to realize this path.