Assessment of Rural Industry Integration Development, Spatiotemporal Evolution Characteristics, and Regional Disparities in Ethnic Regions: A Case Study of Inner Mongolia Autonomous Region Counties

Abstract

Ethnic regions in China primarily focus on the development of agricultural and animal husbandry economies, which are relatively underdeveloped. Rural industry integration development (RIID) is considered the foundation and guarantee for ethnic regions to achieve high-quality modernization of agriculture. The purpose of this article is to measure the level of rural industrial integration in ethnic minority areas, analyze the spatial evolution and regional differences, and explore the actual situation of RIID in these regions. The aim is to provide a decision-making basis for local governments to effectively promote the development of rural industrial integration. Based on the improvement of the evaluation index system for rural industrial integration development, this paper takes the counties of the Inner Mongolia Autonomous Region as the research area. Utilizing panel data from the statistical yearbooks of 68 banners and counties in Inner Mongolia from 2011 to 2020, the panel entropy weight TOPSIS method is employed to assess the average level of rural industrial integration in the research area. The ArcGIS natural breakpoint method is employed to classify the level of RIID in county areas. Exploratory Spatial Data Analysis (ESDA) and GeoDa are utilized to analyze the spatial distribution characteristics of RIID. Finally, the Theil index is employed to analyze the regional differences in the level of RIID. The results show the following: (1) The overall level of RIID in ethnic regions is relatively low, with the contributions of the four dimensions in the evaluation index system as follows: integration path > integration foundation > integration sustainability > integration effect. The level of RIID in the study area is as follows: western region > eastern region > central region. (2) Spatially, there are positive correlations and significant spatial clustering in the level of RIID, with the spatial clustering effect of RIID weakening. (3) There are regional differences in the level of RIID, which are expanding. The inter-regional differences are decreasing, while the intra-regional differences are increasing. (4) The construction of agricultural processing facilities, financial investment, financial support, and talent policies are important influencing factors for the current stage of RIID in ethnic regions. Therefore, in the low-level development stage of RIID in ethnic regions, it is necessary to fully utilize the advantages of resource endowment, increase investment in rural infrastructure, and strengthen the guidance of talent flow into rural revitalization construction.

1. Introduction

Being the largest developing nation globally, the main conflict in Chinese society has transitioned to the increasing desire of the populace for an improved quality of life with uneven and inadequate development. Data from the latest census in China indicate that the rural populace is nearing 500 million, with the average disposable income of farmers still falling below the middle-income bracket. Therefore, the most arduous and burdensome task of achieving high-quality development of Chinese-style modernization and common prosperity still lies in the countryside [1,2]. Like most developing countries and developed nations, like the United States, counties in China also face issues of population outflow, aging, and insufficient economic vitality [3]. The Communist Party of China’s 20th National Congress report put forward the initiative to comprehensively revitalize rural regions, emphasize the importance of prioritizing agriculture and rural development, promote the integrated progress of urban and rural areas, and ensure the smooth flow of resources between urban and rural zones [4,5]. To revitalize the countryside, industries must first be invigorated, integrating rural industries is the fundamental path to revitalizing China’s rural industries [6,7]. Central Document No. 1 for both the years 2022 and 2023 stressed the importance of advancing the integration of primary, secondary, and tertiary rural industries to drive high-quality development and cultivate new industries and business models in rural areas. Acting as a crucial connection between urban and rural zones, the county region encompasses more than 90% of China’s land, over 70% of its population, and more than 50% of its economic output, making it the cornerstone for achieving equitable regional growth [8]. The “Opinions on Promoting Urbanization Development Centered on County Towns”, published by the Central Committee of the Communist Party of China and the State Council Office in 2022, highlights the significance of county towns within China’s urban framework [9]. The term “county region” refers to the scope of county-level administrative divisions, including county towns, townships, and rural areas. Compared to China, other countries have conducted more extensive research on the classification methods and standards for rural areas. The classification methods mainly include geographic, structural, and composite approaches, as well as the use of population density indicators to categorize rural areas [10]. In contrast, the classification standards for rural areas in China are primarily based on administrative division standards. The territory is divided into four administrative levels: provincial, prefectural, county, and township. The broad definition of rural areas in China generally refers to areas below the county level, including townships and lower-level administrative villages and natural villages. The narrow definition, similar to the economic function-based classification used in other countries, refers to areas predominantly focused on agricultural economies [11]. Research on the comprehensive development level and disparities of rural industries is primarily concentrated on the overall context of China, as well as on areas such as the Yangtze River Economic Belt and the eastern coastal regions. There is comparatively less research concerning counties, particularly regarding northern frontier ethnic minority regions, where such studies are almost nonexistent. Following more than five years of implementing the rural revitalization strategy, what is the current status of RIID in the county regions of China’s northern frontier ethnic minority areas? What are the characteristics and evolutionary trends of this development? Are there regional disparities in the level of RIID at the county level, and if so, what are the underlying causes of these disparities? These issues urgently need to be addressed, as these responses not only assess the extent of RIID in the county regions of China’s northern frontier ethnic minority areas but also provide a foundation for formulating policies on the sustainable development of rural industries in ethnic regions of China. Therefore, this study holds substantial theoretical and practical significance.

Research on the integration of rural industries is primarily grounded in the theories of development economics. The dual-sector model, represented by W.A. Lewis, posits that developing countries exhibit a coexistence of modern industry and traditional agriculture, advocating for a development strategy that prioritizes urban industrial growth [12]. From the perspective of regional economic theory, P. Krugman’s core periphery model suggests that industrial sectors and urban areas occupy the core regions, while agricultural sectors and rural areas are on the periphery [13]. This creates an unbalanced urban rural development relationship where the core dominates the periphery, contributing to rural poverty [14]. Rural industrial integration is considered an effective pathway to stimulate rural economic growth [15]. It involves the merging of agriculture with secondary and tertiary industries. County regions serve as critical zones where industrial and agricultural economies coexist, making them significant areas for rural industrial integration [16]. F. Perroux’s growth pole theory asserts that economic growth in county regions cannot simultaneously and uniformly occur across all areas or industries but will first emerge in regions with advantageous location conditions or in leading driving sectors [17]. Consequently, rural industrial integration will also exhibit regional disparities. Studying the evolving characteristics of rural industrial integration helps identify the underlying causes of these regional differences.

Current academic research on integrated rural industrial development primarily focuses on five key aspects: Firstly, it explores the definition, driving factors, and pathways of RIID from a theoretical perspective. While scholars may have varying interpretations of RIID, there is a consensus on its core meaning [9]. A representative definition is that it involves “the integration, permeation, and cross-reorganization between the primary, secondary, and tertiary industries in rural areas, which is specifically manifested in the extension of the agricultural industry chain, the expansion of the industrial scope, and the transformation of industrial functions, this leads to the emergence of new technologies, new business forms, and new models, thereby achieving an integrated and optimized reorganization of resources, factors, and technology in rural areas” [18,19]. Farmers and new types of business entities are the main participants in this development [20,21]. Driven by the intrinsic motivation of the participants’ interests and the external impetus of technological innovation, as well as the upgraded consumption demands of rural residents, the gradual integration of rural factors, and the support of reforms in the rural management system and financial and agricultural policies, rural industrial integration continues to deepen [22,23,24]. The RIID presents a variety of integration methods [25,26], with land, labor, capital, technology, and information being crucial productive elements in rural industrial integration [27,28]. The pathway to RIID hinges on the establishment of rural property rights and interests, the enhancement of interest linkage mechanisms, and the implementation of innovative mechanisms in rural services [29,30]. The second aspect involves constructing an index system through various dimensions of RIID, such as the endogenous and exogenous dimensions of RIID, the dynamics and trends of integration, the environmental dimension, etc. [31,32,33]. This system evaluates the overall integrated development of rural industries by assessing indicators like the agricultural industry chain extension, agricultural multifunctionality expansion, agricultural services integration, the growth of farmers’ incomes, and the blending of urban and rural areas [31,34,35]. It also further explores the impact of the RIID on urbanization, common prosperity, county-level economic growth, and carbon emissions, among other environmental factors [36,37,38,39]. On a micro level, it discusses the income effects, poverty reduction effects, and ecological protection effects of RIID on micro-entities [40,41,42]. Key objectives of the unified advancement of rural industries include boosting farmers’ income, creating job opportunities, and fostering urban rural integration [43]. The third aspect involves the study of the spatiotemporal distribution of RIID, focusing primarily on the overall region of China, the Yangtze River Economic Belt, and the southern frontier region. The research indicates that the level of RIID in China is generally low, with regional disparities showing a trend of narrowing [41,44,45]. The fourth aspect focuses on research related to practical experiences of RIID, both domestically and internationally. This includes examining the additive and multiplicative effects of Japan’s “Sixth Sector Industrialization”. South Korea’s promotion of the “Agri-Industrial Commercial + Government Academia Research” cooperative model and supportive policies for rural industrial integration provides a theoretical basis for developing rural industrial integration pathways in China [46,47]. Australia, which operates primarily on family farms as a mode of ecological agriculture, has achieved scaled agricultural development through tax incentive policies, addressing issues such as the lag in the development of rural industrial organizational models, small enterprise scale, and short agricultural industry chains [48]. Case studies from areas such as Qinba Mountains and Pilin Village in Guizhou analyze the mechanisms of RIID in China from four aspects: integration models, driving forces, value enhancement, and integration effects [21,49]. The fifth key area of research is assessing the extent of integration between rural industries. Scholars, from various perspectives, have applied different methods, such as the Herfindahl Hirschman Index (HHI) [50], input output method [51], grey relational analysis [52], coupling coordination model [53], and comprehensive evaluation method [54]. By comprehending the core of RIID, they developed an evaluative indicator system and then calculated the degree of RIID.

In summary, while existing research outcomes on the integration of rural industries are already quite substantial, there are still some shortcomings: On one hand, the existing evaluation indicator system for rural industrial integration development lacks consideration of sustainability indicators. On the other hand, in terms of research regions, there are few studies on the RIID in economically underdeveloped frontier minority areas. The concept of ethnic areas is not specific to China; other countries also have similar regions, commonly referred to as “ethnic areas” or “indigenous areas.” Examples include Indian reservations in the United States, First Nations areas in Canada, and Aboriginal communities in Australia. These areas may have their own languages, cultures, and traditions and enjoy certain degrees of political, economic, and social autonomy. Ethnic areas are similar to other minority or indigenous areas in some respects, such as cultural diversity, reliance on natural resources, and facing economic and social development challenges. However, each country’s ethnic areas have their unique historical, political, and cultural backgrounds, which shape their distinct characteristics. China is a multi-ethnic country with 56 ethnic groups. Ethnic areas refer to regions where ethnic minorities are concentrated, usually located in borderland areas adjacent to other countries, characterized by harsh natural conditions, lower economic levels, lagging social development, and relative poverty. Administratively, ethnic areas usually refer to autonomous regions that enjoy a certain degree of autonomy. China has five autonomous regions, covering 64% of the country’s total area. Therefore, the development of rural industrial integration in frontier ethnic areas is of great significance for promoting rural revitalization in China. Additionally, there is a lack of systematic analysis of the spatial and temporal evolution characteristics and regional differences in rural industrial integration development, as well as decomposition studies of regional differences. So, what is the level of RIID in the Inner Mongolia Autonomous Region? How have these levels and characteristics evolved over the past decade? What regional differences exist in the rural industrial integration of county-level areas in Inner Mongolia Autonomous Region? This paper conducted research on these issues.

The marginal contributions of this paper are as follows: (1) introducing the dimension of sustainable development to improve the evaluation indicator system for rural industrial integration development; and (2) analyzing the development level, evolution characteristics, and regional differences of rural industrial integration in frontier ethnic areas, thereby filling the gap in theoretical research on ethnic areas and providing a basis for policy-making in these regions.

Based on this, this study focuses on the counties of Inner Mongolia Autonomous Region, utilizing macro panel data from 2010 to 2020 and performing the following: (1) establishing a comprehensive assessment framework for county-level RIID from multiple dimensions to evaluate the overall level of RIID in ethnic areas at the county level; (2) attempting to analyze the development level of rural industrial integration, using the panel entropy-weighted TOPSIS method; (3) combining Geographic Information System (GIS) technology and ESDA methods to explore the extemporization evolution of the level of rural industrial integration at the county level; and (4) using the Theil index and its decomposition method to analyze regional disparities in the level of RIID and their sources. This research will provide a basis for policy formulation for the sustainable development of RIID in ethnic regions.

2. Materials and Methods

2.1. Conceptual Framework

2.1.1. Rural Development

In this paper, rural development refers to well-developed rural infrastructure, good ecological environment quality, clean and tidy villages, civilized rural customs, and prosperous living conditions.

2.1.2. Rural Area

The broad definition of rural areas in this paper refers to areas below the county level, including townships and lower-level administrative villages and natural villages.

2.2. Research Logic

Drawing on a comprehensive comprehension of the core concept of rural industrial integration and a survey of the existing literature, this paper defines rural industrial integration as the process where, guided by the government, stakeholders such as farmers, new agricultural business entities, and village collective economic organizations make full use of rural resource elements. Driven by the market, technology, and policy factors, this integration extends the agricultural industry chain, enhances the multifunctionality of agriculture, and integrates agricultural services. Such an approach facilitates resource mobilization between urban and rural regions, streamlining resource distribution in industrial progress and subsequently fostering urban rural integration and the sustainable advancement of rural industries. This research dissects the subject matter across four focal points: the groundwork for integration, the trajectory of integration, the repercussions of integration, and the perpetuity of integration, as illustrated in Figure 1.

2.2.1. Integration Foundation

Rural development depends on national policies (rural, regional, social, etc.) and factors affecting heterogeneity, mainly originating from resource factors [55,56]. RIID relies on agriculture, the primary sector supported by natural and social resources. Land, water resources, and climate provide the material foundation. Social resources, including human and financial resources, can enhance the momentum of rural industrial integration development. This study assesses integration levels based on per capita land area, water resources, secondary and tertiary industry employment, and general budgetary revenue.

2.2.2. Integration Path

The integration path refers to the specific implementation method of rural industrial integration, involving how natural and social resources flow and integrate across rural areas. This directly affects the effectiveness of RIID. In China, the integration of rural industries within rural regions hinges on leveraging local agricultural bases and exploring diversified integration paths tailored to local resource advantages. A critical component of this is the investment in infrastructure and facility agriculture [57]. This study refers to the indicators in the “Pilot Demonstration Implementation Plan for the Development of Rural Industrial Integration” issued in 2016 and the existing literature to divide the integration paths into three categories: extension integration of the agricultural industry chain, integration of agricultural multifunctionality, and integration of agricultural services, taking into account the actual situation of counties and the availability of data [32,34,58].

2.2.3. Integration Effects

The impact of integration is the consequence of implementing rural industrial integration and serves as a key measure for assessing the objectives of agricultural modernization and shared prosperity. It also reflects the role of social welfare in promoting the development of rural Economy [59]. The integration of industries in rural areas promotes the development of new rural formats, optimizes the agricultural industry structure guided by the market, increases the added value and competitiveness of agriculture through the extension of the value chain [60], promotes the diversification of rural economies, and improves agricultural production and living conditions. It also creates more employment opportunities and expands channels for increasing farmers’ income [31]. Therefore, this paper evaluates the effect of RIID from the perspective of urban rural integration.

2.2.4. Integration Sustainability

RIID essentially involves collaboration cross-sector to integrate various sectors that are agriculture-based in terms of space, technology, capital, and management. Urban rural integration helps to achieve the integration and mutual permeability of resource elements such as science and technology, funds, and human capital. Particularly, the role of human resources is crucial, and the continuous increase in professional talents can generate strong driving forces for the sustainability of RIID. This paper evaluates the sustainability of RIID from three aspects: technological contribution, financial contribution, and talent contribution.

2.2.5. Research Process

The main research process of this paper is based on a review of domestic and international studies related to the integrated development of rural industries. By constructing an evaluation index system for the integrated development of rural industries, the research process is formed through the measurement of the level of RIID, analysis of spatial evolution characteristics, and regional disparities. The research framework is shown in Figure 2.

2.3. Construction of Measurement Index System

This article conducts a comprehensive evaluation of the level of RIID in the counties of Inner Mongolia using the entropy weight TOPSIS method. Drawing from a set of guidelines and policies released by China to advance the integration of farming, manufacturing, and services in rural and pastoral regions, it outlines the overall focus for selecting indicators, which is to boost rural rejuvenation. Leveraging previous studies on gauging the level of RIID [61,62,63], this paper integrates highly consensual and representative research findings, which are not only reliable in data but also clear in specifics. Based on this, a rural industrial integration evaluation index system was constructed, including four subsystems: the foundation of rural industrial integration, pathways, effects, and sustainability, along with 9 primary indices and 18 secondary indices (

Table 1. Evaluation index system for the RIID in county areas.

	Subsystem	First Grade Indexes	Second Indexes	Calculation Method and Explanation	Index Type
Rural industry integration level	Integration foundation (A1)	Natural resource production factors (B1)	Per capita land area (hectares) (C1)	Administrative area land area/end-of-year total population	Positive
			Water resources (mm) (C2)	Annual precipitation	Positive
		Social resource production factors (B2)	Number of people employed in secondary and tertiary industries (people) (C3)	The sum of people employed in the secondary and tertiary industries	Positive
			Fiscal budget revenue (ten thousand yuan) (C4)	Fiscal general budget revenue	Positive
	Integration path (A2)	Agricultural industry chain extension (B3)	Per capita agricultural, forestry, animal husbandry, and fishery output value (ten thousand yuan) (C5)	Total output value of agriculture, forestry, animal husbandry, and fishery/rural employment-population	Positive
			Per capita output of raw agricultural materials (tons/person) (C6)	(total grain output + total meat output + total oil crop output + total milk output)/rural employment-population	Positive
			Facility agriculture land area (hectares) (C7)	Facility agriculture land area	Positive
			Number of Large-Scale Industrial Enterprises (C8)	Number of large-scale industrial enterprises	Positive
		Agricultural multifunctionality (B4)	Per capita possession of grain, oil, and meat (tons) (C9)	(Total grain output + total meat output + total oil crop output + total milk output)/end-of-year population	Positive
			Per capita green space area (square meters/person) (C10)	Green space area/end-of-year total population	Positive
			Road mileage (kilometers) (C11)	Road mileage	Positive
		Agricultural service industry integration (B5)	Per capital total value of agricultural, forestry, animal husbandry, and fishery service industry (ten thousand yuan) (C12)	Total output value of the service industry for agriculture, forestry, animal husbandry, and fishery/rural employment-population	Positive
			Total social fixed-asset investment (ten thousand yuan) (C13)	Total social fixed-asset investment	Positive
			Per capita consumer expenditure (yuan) (C14)	Total retail sales of consumer goods/end-of-year total population	Positive
	Integration effects (A3)	Urban–rural integration (B6)	Urbanization rate (%) (C15)	Urban population/end-of-year total population	Positive
			The ratio of disposable income between urban and rural residents (C16)	Rural residents’ per capita disposable income/urban residents’ per capita disposable income	Negative
	Integration sustainability (A4)	Technical contribution (B7)	Information level (C17)	Fixed, mobile phones, and internet broadband access users	Positive
		Financial contribution (B8)	Per capita loan balance (C18)	Year-end balance of financial institutions’ loans/rural employment-population	Positive
		Talent contribution (B9)	Professional and technical personnel (C19)	Number of professional and technical personnel (people)	Positive

). To minimize the economic disparities among counties, per capita values or ratio values that can best reflect the degree of integration were chosen as indices.

2.4. Research Methodology

In the first step, this paper implements a combination of the Entropy Weight Method and the TOPSIS method to perform a comprehensive evaluation of the level of RIID in Inner Mongolia’s counties. The entropy weight method employs the concept of information entropy to gauge the amount of information within indicators primarily for determining each indicator’s weight, thus calculating the proportionate weight of each indicator in the overall assessment. TOPSIS is a method used for decision-making that determines the ranking and grading evaluation of each scheme by calculating the relative closeness to the optimal and worst schemes. Compared to other multi-attribute decision-making methods, TOPSIS also does not require assumptions about the functional relationships between attributes, offering greater flexibility and practicality. The assessment process is as follows:

(1)

Let us assume that we have selected a dataset comprising h years, n banners/counties, and m evaluation indicators. Here, X_λij denotes the value of the j-th evaluation indicator for the i-th banner counties in the λ-th year (λ = 1, 2, …, h; i = 1, 2, …, m; j = 1, 2, …, n). To mitigate the impact of inconsistent measurement units, we apply the range normalization method to the original sample data, carrying out positive and negative indicator transformations. Considering the possibility of zero values post-normalization, which could render the data non-informative, we follow the sample data processing approach, as suggested by Zhang Lin [45], by performing a shift operation on the data, specifically adding 0.0001 to the entire dataset. Distinct formulas are applied to positive and negative indicators as follows:

The formula for positive indicators is as follows:

$${\mathrm{Y}}_{\mathsf{\lambda }\mathrm{i}\mathrm{j}}={\displaystyle \frac{({\mathrm{X}}_{\mathsf{\lambda }\mathrm{i}\mathrm{j}}-{\mathrm{X}}_{\mathrm{m}\mathrm{i}\mathrm{n}})}{({\mathrm{X}}_{\mathrm{m}\mathrm{a}\mathrm{x}}-{\mathrm{X}}_{\mathrm{m}\mathrm{i}\mathrm{n}})}}+0.0001$$

(1)

The formula for negative indicators is as follows:

$${\mathrm{Y}}_{\mathsf{\lambda }\mathrm{i}\mathrm{j}}={\displaystyle \frac{({\mathrm{X}}_{\mathrm{m}\mathrm{a}\mathrm{x}}-{\mathrm{X}}_{\mathsf{\lambda }\mathrm{i}\mathrm{j}})}{({\mathrm{X}}_{\mathrm{m}\mathrm{a}\mathrm{x}}-{\mathrm{X}}_{\mathrm{m}\mathrm{i}\mathrm{n}})}}+0.0001$$

(2)

In this context, Y_λij represents the value of the j-th indicator for the i-th county in the λ-th year. X_λij is the original data, while X_max and X_min are the maximum and minimum values of the j-th indicator across all evaluation subjects and all years respectively.

(2)

Normalize the indicators, typically denoted as P_λij:

$${\mathrm{P}}_{\mathsf{\lambda }\mathrm{ij}}={\displaystyle \frac{{\mathrm{Y}}_{\mathsf{\lambda }\mathrm{ij}}}{{\displaystyle \sum _{\mathsf{\lambda }=1}^{\mathrm{h}}{\displaystyle \sum _{\mathrm{i}=1}^{\mathrm{m}}{\mathrm{Y}}_{\mathsf{\lambda }\mathrm{ij}}}}}}$$

(3)

(3)

Calculate the entropy value E_j of the j-th indicator:

$${\mathrm{E}}_{\mathrm{j}}=-\mathrm{k}{\displaystyle \sum _{\mathsf{\lambda }=1}^{\mathrm{h}}{\displaystyle \sum _{\mathrm{i}=1}^{\mathrm{m}}{\mathrm{P}}_{\mathsf{\lambda }\mathrm{ij}}}}{\mathrm{InP}}_{\mathsf{\lambda }\mathrm{ij}}, 0\le {\mathrm{E}}_{\mathrm{j}}\le 1,$$

(4)

The constant k > 0, $\mathrm{k}={\displaystyle \frac{1}{\mathrm{In}(\mathrm{h}\times \mathrm{m})}}$.

(4)

Calculate the information efficacy value D_j for the j-th indicator:

D_j = 1 − E_j

(5)

(5)

Calculate the weights of each indicator, W_j:

$${\mathrm{W}}_{\mathrm{j}}={\displaystyle \frac{{\mathrm{D}}_{\mathrm{j}}}{{\displaystyle \sum _{\mathrm{j}=1}^{\mathrm{n}}{\mathrm{D}}_{\mathrm{j}}}}}$$

(6)

(6)

Calculate the weighted decision matrix, V_ij, for the measurement index of the λ-th year:

$${\mathrm{V}}_{\mathrm{i}\mathrm{j}}={({\mathrm{v}}_{\mathrm{i}\mathrm{j}})}_{\mathrm{m}\times \mathrm{n}},{\mathrm{v}}_{\mathrm{i}\mathrm{j}}={\mathrm{P}}_{\mathrm{i}\mathrm{j}}\times {\mathrm{W}}_{\mathrm{j}}$$

(7)

(7)

Based on the weighted matrix, V_ij, determine the most positive ideal solution, C_j+, and the most negative ideal solution, C_j−, for the λ-th year:

$${\mathrm{C}}_{\mathrm{j}}+=(\mathrm{m}\mathrm{a}\mathrm{x}{\mathrm{v}}_{\mathrm{i}1},\mathrm{m}\mathrm{a}\mathrm{x}{\mathrm{v}}_{\mathrm{i}2},\dots ,\mathrm{m}\mathrm{a}\mathrm{x}{\mathrm{v}}_{\mathrm{i}\mathrm{n}}), \mathrm{i}=1,2,3,\dots ,\mathrm{m}$$

(8)

$${\mathrm{C}}_{\mathrm{j}}-=(\mathrm{m}\mathrm{i}\mathrm{n}{\mathrm{v}}_{\mathrm{i}1},\mathrm{m}\mathrm{i}\mathrm{n}{\mathrm{v}}_{\mathrm{i}2},\dots ,\mathrm{m}\mathrm{i}\mathrm{n}{\mathrm{v}}_{\mathrm{i}\mathrm{n}}), \mathrm{i}=1,2,3,\dots ,\mathrm{m}$$

(9)

(8)

Calculate the Euclidean distance from each county to the positive and negative ideal solutions for the λ-th year.

$$Z^{+}=\sqrt{{\displaystyle \sum _{\mathrm{j}=1}^n({\mathrm{C}}_{\mathrm{j}}^{+}-{\mathrm{v}}_{\mathrm{ij}}}}{)}^2$$

(10)

$$Z^{-}=\sqrt{{\displaystyle \sum _{\mathrm{j}=1}^n({\mathrm{C}}_{\mathrm{j}}^{-}-{\mathrm{v}}_{\mathrm{ij}}}}{)}^2$$

(11)

(9)

Based on the distance values, calculate the comprehensive scores, C_i, for the positive and negative ideal solutions for each county, where 0 < C_i < 1. Finally, rank them accordingly.

$$C_i={\displaystyle \frac{{\mathrm{Z}}_{\mathrm{i}}^{-}}{{\mathrm{Z}}_{\mathrm{i}}^{+}+{\mathrm{Z}}_{\mathrm{i}}^{-}}}$$

(12)

In the second step, utilize ArcGIS version 10.8 software for the spatial visualization analysis of the four subsystems and apply the natural breaks classification method to grade the level of RIID. The natural breaks method adopts the core idea of cluster analysis: maximizing homogeneity within groups and heterogeneity between groups. Global spatial autocorrelation is a description of the spatial characteristics of related data across the entire range, generally measured by the global Moran’s I index. When Moran’s I index > 0, it indicates a positive autocorrelation; if it is less than 0, it indicates a negative autocorrelation; and when it is equal to 0, it represents the absence of spatial correlation. If global positive spatial autocorrelation exists, it is also necessary to combine Geoda with ArcGIS for a further local spatial analysis. When setting the spatial weight matrix, considering the dispersed geographical locations of the counties and several isolated areas, the matrix is constructed based on the geometric center distances of the county spatial shapes, with bandwidth set to the software system’s default to ensure that each county is adjacent to at least one neighbor. High values of the local Moran’s index indicate the clustering of area units with similar variable values, whereas low values indicate the clustering of area units with dissimilar variable values.

In the third step, this paper employs the Theil index to analyze regional disparities in the level of RIID within the study area. The Theil index can decompose the overall disparities of the study area into intra-regional differences and inter-regional differences. This aids in observing and revealing the direction and magnitude of changes in both internal and external regional disparities, as well as their impact on the total variance. This section of the study calculates the Theil index using Stata version 15.0 software.

2.5. Research Area and Data Sources

This paper takes the counties of the Inner Mongolia Autonomous Region as the study area (Figure 3).

The Inner Mongolia Autonomous Region is located on the northern frontier of China, spanning the Northeast, North, and Northwest regions, and borders eight provinces and territories. It is one of the provincial administrative regions with the most neighboring provinces in China, and it is also one of the five ethnic minority autonomous regions of China, sharing borders with Mongolia and the Russian Federation. The Inner Mongolia Autonomous Region is a crucial grain and livestock production base in China. Annually, its grain output ranks first among the five ethnic autonomous regions. The region is also rich in mineral resources, and its economic growth predominantly relies on natural resources. From the changes in GDP growth rates between 2010 and 2020 (Figure 4), it can be observed that, compared to neighboring provinces, such as Shaanxi, Shanxi, Gansu, and Jiangxi Province in Southeastern China, Inner Mongolia’s annual GDP growth rate shows a declining trend. Moreover, since 2011, the growth rate has been significantly lower than that of the other provinces. This decline is primarily due to China’s implementation of the Western Development Strategy, which has strengthened the protection of natural resources. In addition, like other border ethnic regions, Inner Mongolia faces socio-economic issues such as population mobility, aging, and uneven urban rural development. Therefore, exploring the rural industrial integration development in Inner Mongolia can help formulate strategies to address social changes and economic transformation. There are a total of 103 county-level administrative units in the Inner Mongolia Autonomous Region, including 23 districts, 11 county-level cities, 17 counties, 49 banners, and 3 autonomous banners [64,65]. Due to the administrative centers often being far from rural pastoral areas, this study excluded district-level cities, county-level cities, and administrative units where municipal governments are located when selecting county-level data and instead chose 68 banners and counties as the specific research areas to more accurately reflect the actual situation of RIID. The primary data for literature analysis in this study were derived from several sources, including the “Inner Mongolia Statistical Yearbook”, the “China County Statistical Yearbook”, the “China Rural Statistical Yearbook”, the China Urban and Rural Construction Database, the China Population Census and Sampling Survey Database, statistical yearbooks of various banners and counties, and government work reports of various banners and counties, spanning the years from 2011 to 2020. In case of missing data, linear interpolation was used for supplementation, and dimensionless processing was performed.

3. Results

3.1. Measurement Analysis of RIID Level

3.1.1. Evaluation of RIID Level

This article applies the entropy weight TOPSIS method to measure the level of RIID in 68 banners and counties of Inner Mongolia from 2011 to 2020. The measurement results can be found in

Table 2. Comprehensive index and evaluation of RIID in county areas.

Banner and County Names	Acronym	Comprehensive Index	Rank	Banner and County Names	Acronym	Comprehensive Index	Rank
Ejina Banner	EJNB	0.465	1	Zhalaite Banner	ZLATB	0.11	35
Zhungeer Banner	ZHGB	0.326	2	Xinbaerhuzuo Banner	XBEHZB	0.108	36
Ewenki National Autonomous Banner	EWKB	0.324	3	Toketo County	TKC	0.104	37
Ningcheng County	NCC	0.302	4	Keerqin Youyiqian Banner	KEQYQB	0.103	38
Yijinhuoluo Banner	YJLLB	0.298	5	Balinzuo Banner	BLZB	0.103	39
Etoke Banner	ETKB	0.274	6	Kalaqin Banner	KLQB	0.101	40
Alasanyou Banner	ALSB	0.262	7	Keerqinzuoyihou Banner	KEQZHB	0.101	41
Chenbaerhu Banner	CB	0.240	8	Xinbaerhuyou Banner	XBEHYB	0.1	42
Siziwang Banner	SB	0.232	9	Wulatehou Banner	WLTHB	0.097	43
Dalad Banner	DB	0.222	10	Naiman Banner	NMB	0.096	44
Wushen Banner	WSB	0.213	11	Hangjinhou Banner	HJHB	0.095	45
Wengniute Banner	WNTB	0.198	12	Wuyuan County	WYC	0.095	46
Dongwuzhumuqin Banner	DWZMQB	0.167	13	Keshiketeng Banner	KSKTB	0.093	47
Chahaeryouyihou Banner	CHEYHB	0.165	14	Suniteyou Banner	SNTYB	0.093	48
Xiwuzhumuqin Banner	XWZMQB	0.162	15	Wulatezhong Banner	WLTZB	0.09	49
Helingeer County	HLC	0.157	16	Keerqin Youyizhong Banner	KEQYZB	0.09	50
Chahaeryouyizhong Banner	CHEYZB	0.152	17	Alukeerqin Banner	ALKQB	0.089	51
Tumoteyou Banner	TYB	0.142	18	Tuquan County	TQB	0.088	52
Arong Banner	ARB	0.141	19	Kulun Banner	KLB	0.087	53
Morin Dawa Daur National Autonomous Banner	MDDNAB	0.139	20	Balinyou Banner	BLYB	0.086	54
Etuokeqian Banner	ETKQB	0.137	21	Xinghe County	XHC	0.085	55
Elunchun National Autonomous Banner	ELCB	0.129	22	Qingshuihe County	QSHC	0.075	56
Zhenglan Banner	ZLB	0.128	23	Linxi County	LXC	0.075	57
Sunitezuo Banner	SNTZB	0.126	24	Guyang County	GYC	0.074	58
Abaga Banner	ABGB	0.123	25	Taipusi Banner	TPSB	0.072	59
Aohan Banner	AHB	0.118	26	Liangcheng County	LCC	0.070	60
Hangjin Banner	HJB	0.117	27	Chahaeryouyiqian Banner	CHEYQB	0.068	61
Tumotezuo Banner	TTB	0.117	28	Duolun County	DLC	0.066	62
Keerqinzuoyizhong Banner	KEQZZB	0.116	29	Shangdu County	SDC	0.062	63
Darhan Muminggan United Banner	DMDUB	0.116	30	Xianghuang Banner	XHB	0.060	64
Wulateqian Banner	WLTQB	0.114	31	Huade County	HDC	0.055	65
Zalute Banner	ZLTB	0.111	32	Zhuozi County	ZZC	0.054	66
Dengkou County	DKC	0.111	33	Wuchuan County	WHC	0.052	67
Kailu County	KLC	0.110	34	Zhengxiangbai Banner	ZXBB	0.044	68

. The comprehensive index (mean value) for rural industrial integration at the county level ranges from 0.044 to 0.465. The highest score is 0.465, achieved by Ejin Banner in the western region, while the lowest score is 0.044, attributed to Zhengxiang Bai Banner in the central region.

This paper utilized Stata 15.1 to conduct a descriptive statistical analysis of the level of RIID and its subsystems in county areas (

Table 3. Descriptive statistics of the level of RIID and its subsystems.

Variables	N	Mean	sd	Min	Max
Industry integration level	68	0.134	0.0789	0.0440	0.465
Foundation of integration	68	0.723	0.283	0.252	2.008
Integration path	68	1.580	0.541	0.640	3.091
Integration benefit	68	1.034	0.252	0.608	1.723
Integration sustainability	68	0.382	0.235	0.0644	1.162

). The results reveal that the mean level of RIID in this area is 0.134, with a standard deviation of 0.079. Among the sample, 47 banners and counties exhibit an industrial integration level that falls below the regional average. This suggests that the RIID level is generally low across these counties. This observation is consistent with existing studies, which indicate that the level of RIID in China is predominantly at an initial stage. Furthermore, the greatest disparity lies in the integration paths, followed by the foundation of integration, while the sustainability of integration shows the least variation.

3.1.2. Analysis of Influencing Factors of RIID

By investigating the RIID in the banners and counties from 2011 to 2020 (Figure 5), it was found that the contribution rate changes in the integration foundation, integration paths, and level of sustainable integration all show a fluctuating trend that first rises and then falls. Specifically, the foundation of integration exhibits relatively stable changes, and this is primarily due to the natural-resource elements owned by each banner/county being determined by their geographical location, resulting in minor annual variations. The integration effects show a trend of first declining and then rising, mainly attributable to the advancement of urbanization and the full implementation of the rural revitalization strategy initiated in 2016. During this period, there was an increase in financial investment and infrastructure construction for rural development, which in turn promoted the recovery of integration benefits. Meanwhile, the changes in sustainable integration show a trend that first rises and then falls, possibly due to the outflow of county population leading to a shortage of key personnel, posing a challenge to sustainable development.

When analyzing the weights of indicators composing the evaluation index of the level of RIID (

Table 4. Evaluation indicators and weights of rural industry integration.

	Subsystem Weight	First-Grade Indexes’ Weight	Second Indexes’ Weight
Rural industry integration level	A1 (0.2963)	B1 (0.1860)	C1 (0.1759)
			C2 (0.0100)
		B2 (0.1103)	C3 (0.0270)
			C4 (0.0833)
	A2 (0.4877)	B3 (0.3104)	C5 (0.0604)
			C6 (0.0605)
			C7 (0.1650)
			C8 (0.0244)
		B4 (0.0875)	C9 (0.0254)
			C10 (0.0438)
			C11 (0.0183)
		B5 (0.0898)	C12 (0.0226)
			C13 (0.0440)
			C14 (0.0231)
	A3 (0.0220)	B6 (0.0220)	C15 (0.0196)
			C16 (0.0024)
	A4 (0.1940)	B7 (0.0245)	C17 (0.0245)
		B8 (0.1298)	C18 (0.1298)
		B9 (0.0397)	C19 (0.0397)

), we can see that the contribution rate of the integration path indicator (A2) is the highest, reaching 48.77%, with the highest contribution rate of 31.04% for the extension of the agricultural industry chain (B3). It is evident that enhancing the infrastructure for agricultural product processing plays a vital role in advancing the overall RIID at the county level. The next is the integration basic indicator (A1), with a contribution rate of 29.63%, indicating that natural and social resource production factors still play a role in resource endowment. The contribution rates of integration effects (A3) and integration sustainability (A4) are relatively low, at 2.2% and 19.4%, respectively. Among the primary indicators, the top four indicators are agricultural industry chain extension (B3), natural resource production factors (B1), financial contribution (B8), and social resource production factors (B2), while the influences of urban rural integration (B6) and technological contribution (B7) are minimal. In the subsystem of RIID, in terms of integration sustainability, the contribution of talent is relatively significant.

3.2. Spatial–Temporal Analysis of the Level of RIID

This paper divides the Inner Mongolia Autonomous Region into three areas—eastern region, central region, and western region based on the administrative regional divisions from the official website of the Inner Mongolia Government and the findings of the existing research literature. The eastern region comprises Hulunbuir City, Xing’an League, Tongliao City, and Chifeng City; the central region includes Hohhot City, Ulanqab City, and Xilingol League; the western region encompasses Baotou City, Bayan Nur City, Ordos City, Wuhai City, and Alxa League [66,67]. The specific regional divisions of the 68 banners and counties can be seen in

Table 5. Distribution of banners and counties in the study area.

Region	City	Banner Counties
Eastern region	HulunBuir City	Arong Banner Morin Dawa Daur National Autonomous Banner Elunchun National Autonomous Banner Ewenki National Autonomous Banner Chenbaerhu Banner Xinbaerhuzuo Banner Xinbaerhuyou Banner
	HingganLeague	Keerqin Youyiqian Banner Keerqin Youyizhong Banner Zhalaite Banner Tuquan County
	Tongliao City	Keerqinzuoyizhong Banner Keerqinzuoyihou Banner Kailu County Kulun Banner Naiman Banner Zalute Banner
	Chifeng City	Alukeerqin Banner Balinzuo Banner Balinyou Banner Linxi County Keshiketeng Banner Aohan Banner Wengniute Banner Ningcheng County
Central region	Hohhot City	Tumotezuo Banner Toketo County Helingeer County Qingshuihe County Wuchuan County
	Ulanqab City	Zhuozi County Huade County Shangdu County Xinghe County Liangcheng County Siziwang Banner Chahaeryouyiqian Banner Chahaeryouyizhong Banner Chahaeryouyihou Banner
	XilingolLeague	Abaga Banner Sunitezuo Banner Suniteyou Banner Dongwuzhumuqin Banner Taipusi Banner Xiwuzhumuqin Banner Xianghuang Banner Zhengxiangbai Banner Zhenglan Banner Duolun County
Western region	Baotou City	Tumoteyou Banner Guyang County Darhan Muminggan United Banner
	Erdos City	Dalate Banner Zhungeer Banner Etuokeqian Banner Etoke Banner Hangjin Banner Wushen Banner Yijinhuoluo Banner
	BayanNur City	Wuyuan County Dengkou County Wulateqian Banner Wulatezhong Banner Wulatehou Banner Hangjinhou Banner
	AlxaLeague City	Alasanyou Banner Ejina Banner

.

Based on the comprehensive index of RIID calculated from the previous text for the years 2011 to 2020, an analysis was conducted on the entire county-level area, as well as the eastern, central, and western regions of the study area. For detailed results, refer to

Table 6. The RIID index and its mean values of the county-wide, eastern, central, and western regions from 2011 to 2020.

Region	2011	2012	2013	2014	2015	2016	2017	2018	2019	2020	Mean
Eastern region	0.125	0.129	0.140	0.123	0.123	0.128	0.145	0.139	0.123	0.114	0.129
Central region	0.092	0.099	0.087	0.115	0.116	0.118	0.120	0.089	0.100	0.102	0.104
Western region	0.167	0.180	0.188	0.196	0.185	0.199	0.191	0.166	0.168	0.163	0.180
County-wide	0.125	0.132	0.134	0.139	0.137	0.143	0.149	0.129	0.127	0.123	0.134

. Examining the trend for the entire region reveals that the RIID showed a slow upward trend before a decline that began in 2018. This pattern is consistent with Chen Xiangman’s observations regarding the annual growth rate of rural industry integration in Inner Mongolia, China [68]. Possible reasons for this downward trend include the reduction of large-scale industrial enterprises; the implementation of facility agriculture policies; the decrease in per capita output value in agriculture, forestry, animal husbandry, and fisheries; and the loss of population [69,70,71]. Furthermore, the global pandemic of 2019 may also have had a certain impact on the downward trend in 2020. However, given the limited available data, this influencing factor will be further verified in future research. When comparing the average comprehensive index of RIID across regions, the western region (0.18) > eastern region (0.129) > central region (0.104).

Figure 6 illustrates an analysis of the evolving trends in rural industrial integration throughout the eastern, central, and western regions. The overall development trends in both the eastern and western regions can be observed to be generally similar, showing an initial increase followed by a decline. The central region, however, undergoes larger fluctuations, but it generally shows a slow upward trend. In most years, the development trends in the eastern, western, and central regions display opposite patterns.

3.2.1. Regional Hierarchical Classification and Comparative Analysis of RIID

Using ArcGIS software and the “Natural Breaks (Jenks)” classification method, the comprehensive index of RIID for the 68 banners and counties in the study area from 2011 to 2020 was categorized into five levels, from high to low, with the results presented in

Table 7. Regional hierarchical classification of rural industry-integration level.

Type	Eastern Region	Central Region	Western Region
Tier 1 region (0.275–0.465)	Ewenki Autonomous Banner, Ningcheng County	_	Ejin Banner, Jungar Banner, Yijinhuoluo Banner
Tier 2 region (0.168–0.274)	Chenbaerhu Banner, Wengniute Banner	Siziwang Banner	Etoke Banner, Alasanyou Banner, Dalate Banner, Wushen Banner
Tier 3 region (0.119–0.167)	Arong Banner, Morin Dawa Daur National Autonomous Banner, Elunchun National Autonomous Banner	Dongwuzhumuqin Banner, Xiwuzhumuqin Banner, Chahaeryouyizhong Banner, Chahaeryouyihou Banner, Abaga Banner, Sunitezuo Banner, Zhenglan Banner, Helingeer County	Tumoteyou Banner, Etuokeqian Banner
Tier 4 region (0.076–0.118)	Aohan Banner, Balinzuo Banner, Balinyou Banner, Kalaqin Banner, Keerqinzuoyizhong Banner, Zalute Banner Keerqinzuoyihou Banner, Kailu County, Keerqin Youyiqian Banner, Zhalaite Banner, Kulun Banner, Naiman Banner Keshiketeng Banner, Tuquan County, Keerqin Youyizhong Banner, Alukeerqin Banner	Tumotezuo Banner, Toketo County, Xinghe County, Suniteyou Banner	Hangjin Banner, Darhan Muminggan United Banner, Wulateqian Banner, Wulatehou Banner, Hangjinhou Banner, Wuyuan County, Wulatezhong Banner, Dengkou County
Tier 5 region (0.044–0.075)	Linxi County	Taipusi Banner, Liangcheng County, Chahaeryouyiqian Banner, Duolun County, Shangdu County, Xianghuang Banner, Wuchuan County, Zhengxiangbai Banner, Zhuozi County, Huade County, Qingshuihe County	Guyang County

. There are five and seven banner counties in Tier 1 and Tier 2 regions, with the quantity distribution showing the following: western region > eastern region > central region. There are 13 banner counties in both the Tier 3 and Tier 5 regions, with the quantity distribution showing the following: central region > eastern region > western region. The Tier 4 region has the highest number of banner counties, totaling 30, with the eastern region > western region > central region.

Figure 7 presents a visual map of the level of rural industry-integration development for the years 2011, 2015, and 2020, where darker colors indicate higher levels of industry integration in the counties. The level of RIID in the study area from 2011 to 2020 showed a general trend of significant decline in many areas with a slow recovery in a few areas. The central and eastern regions experienced a more pronounced decline, although a gradual recovery can also be observed. In contrast, the western region showed less fluctuation and remained relatively stable. The areas with higher industry integration levels displayed in Figure 7 are largely consistent with the locations of the rural industry-integration demonstration parks announced by the Chinese government in recent years, such as the Dalate Banner and Jungar Banner, which have been selected as national rural industry-integration demonstration parks.

Figure 8 illustrates the distribution of banners and counties in the eastern, central, and western regions of the study area across five rank categories, as well as the trend of quantity changes over time in 2011, 2015, and 2020. The observations indicate that over ten years, the number of Tier 1 and Tier 2 regions in the eastern region has decreased, while the number of third, fourth, and fifth-rank regions has increased. In the central region, the distribution of Tier 1, Tier 2, and Tier 3 regions has diminished, whereas the number of Tier 4 and Tier 5 regions has grown. The western region has seen a reduction in the distribution of all five rank categories.

Regional differences of RIID in the Inner Mongolia Autonomous Region are evident. The research employs ESDA to perform an in-depth examination, with the objective of unveiling the spatial relationships of RIID among different banners and counties.

3.2.2. Analysis of the Spatial Distribution Characteristics of RIID at the County Level

Based on the data obtained in the previous section, this paper further explores the spatial autocorrelation at both the global and local levels for these indexes.

(1)

Global Analysis of Spatial Autocorrelation

By utilizing ArcGIS version 10.8 software and Geoda version 1.14 software, we calculated the global Moran’s I and related indicators for the level of RIID in 68 banners and counties from 2011 to 2020 (

Table 8. Global Moran’s I for the level of RIID in the counties of Inner Mongolia from 2011 to 2020.

Year	2011	2012	2013	2014	2015	2016	2017	2018	2019	2020	Mean
Moran’s I	0.2310	0.2082	0.2339	0.1887	0.2065	0.2290	0.3245	0.2677	0.1875	0.1464	0.279
Z-value	4.2233	3.8112	4.2931	3.5021	3.8634	4.2015	5.9217	5.0611	3.6193	2.7544	5.199
p-value	0.002	0.005	0.006	0.003	0.003	0.003	0.001	0.002	0.004	0.0130	0.001

). During the study period, the global Moran’s I > 0 passed the significance level test at 5% for all years except 2020. This indicates that the level of RIID exhibits a positive spatial correlation, meaning that there is a phenomenon of spatial agglomeration in the level of RIID at the county level (either high or low values cluster together). For the year 2020, although the p-value (probability value) was not very small (greater than 0.01 but less than 0.1), the Z-value (standardized distance) was relatively large (greater than 2.5), which means the Moran’s I for that year passed the significance test, indicating that spatial correlation still exists. Moreover, Moran’s I showed a fluctuating upward trend during the study period, suggesting that spatial autocorrelation is intensifying. The significance test of Moran’s I for the average level of RIID in the 68 banners and counties showed a p-value of 0.001, indicating that spatial autocorrelation is significant at a 99.9% confidence level.

(2)

Local Spatial Autocorrelation Analysis

Based on the analysis results from the previous text, it is known that the level of RIID in the counties of the Inner Mongolia Autonomous Region has a spatial agglomeration effect. To further identify which banner counties exhibit high or low-value aggregation phenomena and illustrate the spatial distribution characteristics of the level of RIID between a county and its neighboring counties, this paper conducts a detailed analysis using Moran scatter plots (Figure 9) and LISA cluster maps (Figure 10) based on the RIID index for the years 2011, 2015, and 2020. Figure 9 illustrates that the high–high cluster distribution (HH area) is represented in the first quadrant, the low–high cluster distribution (LH area) in the second quadrant, the low–low cluster distribution (LL area) in the third quadrant, and the high–low cluster distribution (HL area) in the fourth quadrant. The HH area indicates that counties with high levels of rural industry integration are clustered together. This means that these counties not only have high levels of rural industry integration themselves, but their neighboring counties also exhibit high levels of integration. In the HH area, a positive spatial spillover effect can be observed, where such regions might benefit from better resource allocation, policy support, and market conditions, forming an economic belt with a high degree of integration. The LH area represents counties with low levels of rural industry integration surrounded by counties with high levels of integration. This phenomenon suggests that these counties might face certain barriers to development, despite the high integration levels of their neighboring areas. The LL area indicates that counties with low levels of rural industry integration are clustered together. This means that these counties, as well as their neighboring ones, have low levels of integration, showing a negative spatial spillover effect and forming an economic belt with a low degree of integration. The HL area represents counties with high levels of rural industry integration surrounded by counties with low levels of integration. This phenomenon suggests that the integration levels of these counties are significantly higher than those of the surrounding areas. In the HL area, there may be unique advantages that enable these regions to achieve high levels of integration despite unfavorable surrounding environments. There are significantly more points in the first and third quadrants than in the other two quadrants. This suggests that counties with higher or lower levels of rural industry integration tend to cluster spatially, and these areas exhibit relatively smaller spatial differences.

The LISA cluster visualization maps (Figure 10a–c) created using Geoda software illustrate the distribution at three different time points. Over the decade from 2011 to 2020, the number of counties in the high–high cluster distribution (HH) remained stable, with an extremely low count of only two counties: Alasanyou Banner and Wushen Banner, representing 2.9% of the total counties examined. This observation highlights that the rural industry integration level is high in these two counties and their surrounding areas, with small spatial differences and significant positive spatial correlation, forming hotspot areas. The number of counties in the low–low cluster distribution (LL) was 9 in 2011, decreased to 6 in 2015, and further decreased to 4 in 2020. These counties, compared to their surrounding areas, have a lower level of rural industry integration, small spatial differences, and also exhibit strong positive spatial correlation, constituting cold spot areas. The number of counties in the low–high cluster distribution (LH) has decreased, indicating a reduction in the number of counties with lower levels of rural industry integration that are surrounded by counties with higher levels. The high–low cluster distribution (HL) also shows a downward trend, suggesting that the dispersing effect of high-level counties is weakening; thus, the agglomeration effect of RIID at the county level is gradually diminishing.

Rural industry revitalization is one of the key pathways to achieving rural revitalization, and the sustainability of RIID is crucial to this process. Therefore, this study conducted a local spatial autocorrelation analysis on the average index of RIID in the counties (Figure 10d). The analysis revealed that the counties of Alasanyou Banner, Etuokeqian Banner, and Wushen Banner are in the high–high cluster distribution (HH), indicating a high level of rural industry integration. Meanwhile, counties such as Chahaeryouyiqian Banner, Huade County, Shangdu County, Xinghe County, Suniteyou Banner, Taipusi Banner, Xianghuang Banner, Zhenglan Banner, and Zhengxiangbai Banner are in the low–low cluster distribution (LL), showing a lower level of rural industry integration. Additionally, counties including Chahaeryouyizhong Banner, Chahaeryouyihou Banner, and Siziwang Banner are in the high–low cluster distribution (HL). It is noteworthy that there are no counties categorized under the low–high cluster distribution (LH).

The LISA cluster analysis maps (Figure 10a–d) reveal that the eastern counties of Inner Mongolia do not exhibit significant spatial clustering characteristics. This indicates that the rural industry integration level in the eastern counties shows a random distribution pattern with little correlation to the development of neighboring counties. In contrast, the western region demonstrates distinct high high cluster distribution features, while the central region has characteristics of both low low and high low cluster distributions.

3.3. Analysis of Regional Disparities in the Rural Industry Integration Development Level of Counties

The Theil index is a crucial indicator for measuring disparities in income between individuals or regions. This study follows the methods of Zhang Lin [63] and Zhu Honggen [72], utilizing Stata software and Excel 2016 (version 16.0) to calculate the Theil index, which measures the regional disparities in the level of RIID in the banners and counties of the Inner Mongolia Autonomous Region. The study not only computed the overall disparities at the county level but also included the disparities within regions, between regions, and their respective contribution rates. The results are displayed in

Table 9. Theil index and contribution rates of RIID for the eastern, central, and western regions and the entire county from 2011 to 2020.

Year	Overall Disparities	Inter-Regional Differences (Contribution Rate)	Intra-Regional Differences (Contribution Rate)
			Overall	Eastern Region	Central Region	Western Region
2011	0.1778	0.0274	0.1504	0.1179	0.1826	0.1621
		(15.39)	(84.61)	(25.49)	(26.73)	(32.39)
2012	0.1702	0.0280	0.1422	0.1090	0.1736	0.1534
		(16.47)	(83.53)	(23.92)	(27.02)	(32.59)
2013	0.1974	0.0443	0.1531	0.1646	0.1047	0.1707
		(22.44)	(77.56)	(33.32)	(12.14)	(32.10)
2014	0.1859	0.0278	0.1581	0.0922	0.2127	0.1755
		(14.96)	(85.04)	(16.77)	(33.18)	(35.09)
2015	0.1581	0.0211	0.1370	0.0678	0.1925	0.1574
		(13.37)	(86.62)	(14.77)	(36.26)	(35.59)
2016	0.1603	0.0254	0.1349	0.0958	0.1737	0.1408
		(15.82)	(84.18)	(20.47)	(31.53)	(32.18)
2017	0.1482	0.0170	0.1311	0.1131	0.1498	0.1353
		(11.51)	(88.49)	(28.49)	(28.87)	(31.13)
2018	0.1606	0.0302	0.1304	0.1336	0.0969	0.1506
		(18.80)	(81.2)	(34.40)	(14.76)	(32.04)
2019	0.1847	0.0215	0.1632	0.1078	0.2188	0.1779
		(11.62)	(88.37)	(21.67)	(32.92)	(33.78)
2020	0.2025	0.0188	0.1836	0.1207	0.2561	0.1869
		(9.30)	(90.7)	(21.24)	(37.11)	(32.35)

Note: The values in parentheses are contribution rates, with the unit being percent.

.

From a macro-perspective, the Theil index measuring the level of RIID in the counties of the Inner Mongolia Autonomous Region fluctuated between 0.1482 and 0.2025 from 2011 to 2020, indicating that regional disparities do indeed exist. As depicted in Figure 11a, the overall disparity saw a fluctuating downward trend from 0.1778 in 2011 to 0.1482 in 2017, showing a gradual reduction in disparities until the year before 2018. However, since 2018, there has been an increasing trend in overall disparity, suggesting that regional distinctions in the level of RIID are widening.

The decomposition results from Figure 11a show that the contribution rate of inter-regional disparities has experienced a fluctuating decline, reaching 9.3% in 2020. This indicates that the inter-regional disparities in the level of RIID within the counties are progressively diminishing. Figure 11b illustrates a slight fluctuating upward trend in the intra-regional Theil index from 2011 to 2020, indicating that regional disparities within the counties are expanding with minor fluctuations. Notably, the disparities in the western region of the counties have shown a pattern of initial reduction followed by expansion. Further evaluation of average intra-regional disparities shows that from 2011 to 2020, the Theil index averages were 0.11 for the eastern regions, 0.18 for the central regions, and 0.16 for the western regions. The regional disparity ranking is eastern region < western region < central region. The analysis in Figure 11c shows that the contribution rate of intra-regional disparities is consistently above 50%, higher than that of inter-regional disparities, suggesting that the overall disparities in the level of RIID are primarily due to differences within the counties. The average contribution rates of the eastern, central, and western regions after the decomposition of intra-regional disparities are 24.05%, 28.05%, and 32.92%. This reveals that the eastern zone exhibits the lowest contribution rate, followed by the central zone, with the western zone registering the highest. This phenomenon may be related to the polarizing effect of the “Hohhot–Baotou–Ordos” administrative center and economic circle, as both the western and central regions are located around it.

4. Discussion

4.1. Discussion on the Construction of Indicator Evaluation System and Evaluation Methods

This article supplements and extends the evaluation indicators for RIID, in line with the actual situation of the counties in the Inner Mongolia Autonomous Region. It adds dimensions and related indicators for measuring the sustainability of RIID, thereby constructing an evaluation system that is more suitable for assessing the level of RIID at the county level. As counties are key platforms for achieving urban–rural integration and promoting integrated urban–rural development is an ongoing process, the establishment of this system is of great significance [73]. At the same time, since rural industry integration continues to evolve with socio-economic development, the assessment framework is formulated based on the present agricultural development status in the regions. With the aging of agriculture, mechanization, and changes in agricultural participants, further exploration of more effective evaluation systems will be necessary in the future.

4.2. Analysis of Spatiotemporal Dynamic Evolution and Distribution Characteristics

Unlike previous studies that focused on the spatial distribution and temporal–spatial evolution of RIID at the national or provincial level in China, this study supplements the lack of spatial visualization analysis within regional banner counties and rarely observes the trend of rural industry integration levels in county areas over time. Using the ArcGIS tool of GIS, this study delves into the development level and spatiotemporal pattern of RIID in the Inner Mongolia Autonomous Region, adding a new perspective to the quantitative research of RIID at the county level. Macroscopically, the overall level of RIID in the study area shows a “reverse U-shaped” trend, with slow growth followed by a decline starting from 2018. In terms of temporal–spatial evolution analysis, the “Natural Breaks” method in ArcGIS software is used to classify the Rural Industrial Integration Index of 68 banner counties, revealing that the integration level in western and eastern banner counties is generally superior compared to the central region. Analyzing the evolution of the eastern, central, and western regions for the years 2011, 2015, and 2020, it is found that the eastern region tends to evolve towards categories three, four, and five; the central region moves towards categories four and five; and the western region moves towards category three. The RIID exhibits significant spatial correlation in the counties of the Inner Mongolia Autonomous Region, with a clear agglomeration of high and low values, mainly concentrated in the central and western regions. The research findings will provide a basis for the spatial arrangement of RIID in the study area.

4.3. Regional Disparities Analysis

This paper reveals the level and evolutionary process of RIID, as well as its spatial and temporal distribution characteristics, through the calculation of the Rural Industrial Integration Index. However, these analyses do not fully identify the specific trajectories of regional disparities within the county. Thus, this paper utilizes the Theil index to analyze regional disparities. The RIID within the counties of Inner Mongolia indeed exhibits certain regional differences and shows a trend of expansion. The formation of these disparities can be attributed to the foundation, path, and sustainability of integration. Specifically, the banners and counties within the research area with lower levels of RIID include Xianghuang Banner, Huade County, Zhuozi County, Wuchuan County, and Zhengxiangbai Banner. These banners and counties are located in the central region, adjacent in geography, and characterized by complex terrain with many mountains, hills, few plains, and proximity to deserts, with relatively scarce energy resources. These factors may be the reasons for the significant internal development disparities within these banners and counties. In addition, the economic development level and digital technology level in these areas are relatively lagging, and there is a shortage of professional talent, which undermines the sustainability of industrial integration.

4.4. Limitations and Future Outlook

Although this study constructed an evaluation system for the level of RIID, it is based on the current economic development situation, and there are still some areas that need to be improved and further researched. Due to the limitations of data availability, the analysis of RIID can utilize only relevant data from various statistical yearbooks, and there may be other indicators that have not been covered, limiting the breadth of research on RIID. The lack of information on industrial–organizational patterns, development timelines, participating entities, etc., limits the depth of the research. Additionally, while we calculated and analyzed the contribution of subsystems affecting RIID through the calculation of indicators in the rural industrial integration development-level indicator system, we did not delved deeper into the factors and driving forces influencing RIID. Therefore, future research can supplement this study’s limitations and explore the factors and driving forces of RIID in Inner Mongolia Autonomous Region counties, as this is an important research direction.

In addition to this, this study focuses on a specific ethnic region in China, and the constructed indicators in the research results reflect four dimensions that are applicable to some extent to the evaluation of integrated rural industrial development in other countries. Through comparison with Western and Asian countries’ research on rural development, the following differences were found:

(1)

Differences in policy drive: The development of rural areas in China is greatly influenced by policy drivers, such as the recent rural revitalization strategy. In contrast, rural development in some Western countries is more driven by market forces.

(2)

Differences in development models: Compared to Western countries, the development model of rural areas in China has its uniqueness. For instance, rural development in Western countries often focuses on agricultural modernization and technological innovation, while China places more emphasis on the comprehensive development of rural areas and the integration of urban and rural areas.

Therefore, the implementation of China’s unique rural revitalization strategy will greatly promote poverty reduction in rural areas, thereby accelerating the achievement of urban–rural integration.

5. Conclusions and Suggestions

This study established a logical framework, selecting 19 indicators from dimensions such as the foundation, pathway, effect, and sustainability of rural industry integration to assess the level of RIID in county-level areas. This paper conducted a detailed examination of the spatiotemporal evolution characteristics, spatial distribution, and regional differences of county-level RIID in Inner Mongolia Autonomous Region counties, with the key findings as follows:

(1)

The level of RIID in the border ethnic areas is generally low. Among the four subsystems of the current evaluation index system, the contribution rate of the integration pathway index is the highest at 48.77%, mainly influenced by the extension of the agricultural industry chain. This is followed by the foundation of integration, contributing 29.64%, indicating that the RIID still largely depends on resources and production factors. The contribution rate of integration sustainability is 19.4%, primarily derived from support in rural finance and human resources, aligning with the views of Yan Huang [74] and Wei Houkai [75]. The smallest contribution comes from the integration effect, at only 2.2%. It can be seen that, at this stage, the RIID in ethnic areas still relies on the advantages of factor endowment. China’s long-standing dual economic structure has made it difficult for production and living elements to flow into rural areas, thus suppressing the revitalization of rural industries. Since 2000, the Chinese government has implemented the Western Development Policy in ethnic minority regions. This policy includes accelerating the construction of infrastructure, strengthening ecological and environmental protection policies, adjusting industrial structures, developing scientific and educational causes, and increasing the intensity of reform and opening. These measures have significantly enhanced the overall development level of Inner Mongolia. However, the overall level of rural industry integration remains low. The study results indicate that, while the policy has placed considerable emphasis on integration pathways and foundational aspects, there is still substantial room for improvement in sustainability and effectiveness.

(2)

Further analysis indicates that the level of RIID in the counties has been on a slow growth trajectory since 2011, with a downturn starting in 2018. The average Rural Industrial Integration Index over the period from 2011 to 2020 indicates a ranking of the western region > eastern region > central region. Both the eastern and western regions follow similar development patterns, showing fluctuations followed by a decrease. In most years, the central region shows an opposite development trend compared to the eastern and western regions. This indicates that the RIID within border ethnic areas is unbalanced, with a situation of mutual displacement due to competition for resources.

(3)

This paper categorizes the average index of county-level RIID in ethnic areas into five levels. By comparing the evolution of the eastern, central, and western regions in the years 2011, 2015, and 2020, it was observed that the level of RIID in the eastern region is progressively advancing towards the third, fourth, and fifth classes (Figure 8a); the central region towards the fourth and fifth classes (Figure 8b); and the western region towards the third class (Figure 8c). These trends corroborate the observed overall decline in the level of integration. Spatially, a positive correlation in the level of RIID across counties, along with significant spatial clustering (agglomeration of high or low values). High-value clusters are predominantly found in the western region, whereas low-value clusters are mainly found in the central region. The eastern region does not exhibit distinct spatial clustering characteristics. Furthermore, the spatial clustering effect of RIID in the counties of the border ethnic areas is weakening, indicating a reduction in regional polarization regarding the level of RIID. The results are consistent with the theory of growth poles, indicating that regions with superior location conditions or leading industrial sectors are more likely to experience economic growth. In comparison to the eastern and central regions, the western regions have a main focus on economic crops and abundant coal resources. Additionally, they are adjacent to the “Hubei–Beijing–Mongolia” economic zone, which provides favorable conditions for economic development.

(4)

The county-level RIID level in the Inner Mongolia Autonomous Region shows significant regional differences, and the overall differences are widening. According to the decomposition of the Theil index, the inter-regional disparities in rural industrial integration at the county level are gradually narrowing; however, intra-regional disparities are showing a slight trend of increasing fluctuations. The comparison of intra-regional differences is as follows: eastern region < western region < central region. Overall, the primary source of disparities in the level of RIID among counties stems from within regions, with the Eastern region contributing the least, followed by the central region and the western region contributing the most. This phenomenon may be due to the proximity of some banners in the western and central regions to the “Hohhot–Baotou–Ordos” administrative center and economic zone, leading to a polarization effect.

Based on the conclusions, the following suggestions are made for RIID at the county level:

(1)

Government departments should encourage the development of local agricultural characteristic industries in each banner and county, especially the rural tourism industry, to increase population mobility and attract talent back to rural areas [76]. Additionally, acknowledging the spatial correlation of RIID is crucial to establishing communication platforms and innovative cooperation mechanisms between banners and counties and to prevent the homogenization of characteristic industries and unhealthy competition between regions.

(2)

Government departments should stimulate the vitality of integrated rural industrial development by increasing investment in public infrastructure, advancing land system reform, introducing digital technology, and guiding talent to rural areas. For banners and counties with high levels of integrated rural industrial development that rely on natural resources such as energy, the industrial structure should be upgraded rapidly to gradually transform the internal driving force of rural industrial development.

(3)

Improve the mechanisms for introducing rural talent and cultivating modern farmers. Policies and systems for the flow of people with the willingness to engage in agricultural production from urban households to rural areas should be perfected and opened, as the revitalization of talent and industry is key to the strategy of rural revitalization. The results are consistent with the theory of growth poles, indicating that regions with superior location conditions or leading industrial sectors are more likely to experience economic growth. In comparison to the eastern and central regions, the western regions have a main focus on economic crops and abundant coal resources. Additionally, they are adjacent to the “Hohhot–Baotou–Ordos” economic zone, which provides favorable conditions for economic development.

(4)

Strengthen the monitoring and collection of data related to agriculture, rural areas, and farmers; and establish a more comprehensive, scientific, and rational evaluation index system for the level of RIID. Improving the digital technology level in rural areas can effectively address the issue of missing agricultural information data, but this is a systematic project involving multiple departments, industries, and fields, requiring continuous optimization of the top-level design by government departments related to agriculture.