Multiple Driving Paths for Development of Agroforestry Economy: Configuration Analysis Based on fsQCA

Abstract

Amidst global climate warming and increasingly severe food security challenges, the agroforestry economy, a green ecological industry that balances ecological conservation and economic development, has attracted widespread attention. This study constructs a theoretical analytical framework based on the diamond model to systematically identify key factors influencing the development of the agroforestry economy. Using 56 practical cases from the agroforestry economy in China as samples, the study applies Necessary Condition Analysis (NCA) and fuzzy-set Qualitative Comparative Analysis (fsQCA) to further explore the multiple driving paths of agroforestry economic development and their supporting elements. The research findings show that (1) forest resources, technological innovation, market demand, enterprise forms, related industries, and government support do not constitute necessary conditions for the development of the agroforestry economy. The path to the development of the agroforestry economy exhibits complex and concurrent multi-faceted characteristics. (2) Technological innovation has always been at the core of all configurations, and strengthening technological innovation plays a universal role in enhancing the level of agroforestry economic development. The role of government support in the process of the development of the agroforestry economy is limited. (3) The system identified four driving paths, including the endogenous type, characterized by resource technology enterprises; the collaborative type, characterized by a resource technology market with light promotion by the government; the external expansion type, characterized by market technology enterprises; and the linkage type, characterized by market technology enterprises assisted by related industries. The consistency level of the overall solution reached 0.91, and the coverage was 0.54. It reveals the different driving mechanisms with different combinations of elements for the development of the agroforestry economy. Therefore, each region should strengthen scientific and technological research, innovation, and the transformation and application of research outcomes. It should promote the coordinated development of diverse factors, establish tailored regional development models, and explore suitable pathways for developing the agroforestry economy based on its unique resource endowments.

1. Introduction

The agroforestry economy is an ecological economic development model aligned with Chinese national strategy and societal needs and a green production approach in response to global sustainable development trends [1,2]. With the continuous advancement of Chinese ecological civilization, the agroforestry economy has rapidly developed as an eco-industry integrating multiple sectors. According to the National Forestry and Grassland Administration, by the end of 2024, the agroforestry economy in China had reached a large-scale operational area of 623 million mu and exceeded one trillion RMB in output value, making it one of the four forestry sectors with an output of over one trillion RMB. In 2025, the development of the agroforestry economy was first included in the Government Work Report, marking it as a key entry point for expanding farmers’ income channels [3]. However, the development of the agroforestry economy currently faces numerous challenges: within the industry, it is constrained by insufficient technological support capabilities, monotonous operational models, and weak investment intentions for capital elements, while also grappling with issues such as severe product homogenization and lagging brand development [4,5,6]. From an external support perspective, the agroforestry economy faces challenges such as resource and environmental constraints, information asymmetry in markets, and rising labor costs [7,8]. Therefore, an in-depth exploration of the multiple driving mechanisms of the development in the Chinese agroforestry economy can enrich the research dimensions of ecological economic theories and has significant practical value for expanding employment and income channels for mountain farmers and advancing rural revitalization and common prosperity

As global ecological and environmental degradation, food security crises, and frequent natural disasters have become increasingly severe, agroforestry—a forestry development model aimed at alleviating the contradiction between ecological conservation and economic growth—has gradually attracted widespread attention from the international community [3]. In 1978, the International Center for Research in Agroforestry (ICRAF) was formally established. Its headquarters is located in Nairobi, Kenya. It is the only institution that conducts globally significant agroforestry research in all developing tropical regions. Among the multiple regional offices of ICRAF, the China office, based in Yunnan Province, has played a significant role in promoting the development of the agroforestry economy, gradually elevating agroforestry management from production practice to theoretical research. In 1982, the international journal Agroforestry Systems was first published, marking the maturation of international research on agroforestry and the systematization of its theoretical framework [9,10]. In terms of conceptual content, after years of research and discussion, ICRAF defines the agroforestry economy as a composite term for land use systems and engineering application technologies. It involves the purposeful integration of perennial woody plants with agriculture or livestock farming on the same land management unit, using spatial–temporal arrangement methods or short-term alternating management approaches, enabling agriculture and forestry to interact in ecologically and economically integrated ways across different combinations [11,12,13]. In comparison, theoretical research on agroforestry in China began relatively late. In 1994, Li and Lai published the book Agroforestry in China, defining agroforestry as a systematic management approach that organically integrates agriculture and forestry. Its primary characteristics are labor-intensive, intensive production and management activities that fully utilize natural forces [14]. Some scholars translate “agroforestry economy” as “under-forest economy” arguing that beyond traditional under-forest cultivation and animal husbandry models, it encompasses ecotourism, wellness tourism, and other models that fully utilize forest resources, placing greater emphasis on the optimal allocation of forest resources [1,3]. Although definitions of the agroforestry economy vary across different regions and historical periods worldwide, its fundamental principles, core content, and primary characteristics remain consistent. This study posits that the agroforestry economy is an ecologically friendly economic system grounded in ecology, economics, and systems engineering. It relies on forest resources and their ecological environment, with integrated management as its primary characteristic. It constitutes an artificial ecosystem characterized by multiple species, multiple levels, multiple products, multiple benefits, and sustainability [15,16,17].

Regarding the influencing factors and driving mechanisms of the agroforestry economy, the systematic development and sustainable utilization of agroforestry economy are influenced by the interaction of various internal and external factors, and the construction of a development path has become a focal point of academic research. Existing foreign research primarily focuses on the natural ecological mechanisms and theoretical foundations of agroforestry systems, exploring the impact mechanisms of factors such as species introduction [18], light energy distribution [19], soil nutrient cycling [20], atmospheric regulation functions, and carbon sequestration potential [21,22,23] on system operational efficiency. However, empirical studies targeting specific production practices remain scarce. In terms of research perspectives, foreign scholars have mostly focused on benefit analysis of three-dimensional composite management, emphasizing the synergistic roles of forestry cooperatives, industry associations, and rural economic organizations in system operation [24,25,26]. They advocate for promoting the effective alignment of production objectives with market mechanisms through policy interventions, property rights protection, and institutional design [27,28,29,30] to enhance the sustainable development capacity of agroforestry composite systems. Chinese scholars have mainly focused on the refinement and summary of regional production practice models, taking Heilongjiang [31], Jilin [32], Zhejiang [33], and Yunnan [34,35,36] as typical regions for descriptive research on the current status, existing issues, and policy pathways of agroforestry economic development in different regions. Chen [14] emphasized the green, circular, and sustainable characteristics of the agroforestry economy, proposing four development models: agroforestry planting, agroforestry animal husbandry, collection and processing of related products, and utilization of forest landscapes. These models have been gradually recognized and adopted in the Chinese academic community. In terms of research methods, existing literature primarily relies on regional practice survey data and employs quantitative tools such as the logit model [37], factor analysis [38,39], structural equation modeling [40], and multiple linear regression [41,42] to explore the mechanisms of influence of macro-level factors such as natural resources [31], infrastructure [34], operational scale [37], and government investment and policy supply [39] on the development level and efficiency of agroforestry economies. Some scholars have focused on micro-level variables, such as forest farmers’ educational attainment, management philosophy, risk preferences, and behavioral characteristics [40,41,42,43] to reveal the differing driving effects of various factors on the development of agroforestry economies. However, the development of the agroforestry economy is a complex system process involving multiple dimensions and stakeholders [40]. A single-factor perspective is insufficient to fully capture its underlying dynamic structure. Further theoretical modeling and empirical testing are needed to uncover the deep-seated logic and driving mechanisms behind the development of the agroforestry economy.

In summary, existing research has laid an important foundation for exploring the multiple driving mechanisms behind the development of the agroforestry economy. However, several limitations remain: First, the lack of a theoretical analytical framework. Current studies often rely on inductive reasoning from practical cases when identifying drivers of the agroforestry economy’s development. A theoretical model integrating multi-dimensional driving forces has not yet been established; thus, it is difficult to fully explain the complex nonlinear interactions among various factors. Second, limitations in research methods. The development of the agroforestry economy is highly systematic and dynamic. The interaction of multiple factors influences its development. However, most existing studies remain at the level of single-factor linear analysis. There is a significant lack of exploration into complex and concurrent causal relationships, making it difficult to reveal the differences in development paths under various combinations of driving factors. Third, limitations in research samples. Agroforestry economic models vary significantly across regions. Existing studies are mostly case-based. They tend to focus on specific regions or single development modes. Few studies adopt a multi-case approach combined with configuration analysis to uncover the driving mechanisms of the development in the agroforestry economy.

Therefore, this study aims to identify the diverse driving forces and key supporting factors in the development of the agroforestry economy. This study adopts the diamond model as its theoretical framework, extends the research perspective to the county level, selects 56 cases of the agroforestry economy’s development in China as research samples, and employs a mixed research method combining NCA, fsQCA, and content analysis. It systematically analyzes the necessary and sufficient conditions of various influencing factors for the development of the agroforestry economy and, from a configurational perspective, delves into the diverse driving pathways for the development of agroforestry economies. The potential contributions of this study are as follows: Firstly, at the theoretical level, this study fully considers unique Chinese national conditions and constructs a theoretical analytical framework that aligns with the actual development of the Chinese agroforestry economy, thereby expanding the application scenarios of the diamond model and enriching the theoretical foundation for the driving mechanisms of the development in the agroforestry economy. Second, at the practical level, this study employs multiple mixed research methods, integrating the advantages of case studies and variable studies. It utilizes configurational thinking to address the complex causal relationships underlying the diverse driving pathways of the development in the agroforestry economy, thereby addressing the shortcomings of current research methods. This provides important theoretical support and practical guidance for developing tailored strategies for the agroforestry economy’s development in different counties across China, thereby contributing to the realization of rural revitalization and common prosperity objectives.

2. Theoretical Analysis

2.1. Theoretical Foundations

Michael Porter’s diamond model, originating from his work The Competitive Advantage of Nations, serves as a classic framework for analyzing an industry’s long-term competitive advantage. With its logical rigor and high explanatory power in practice, it has gained widespread application across multiple disciplines including economics and management [44]. In terms of its constituent elements, the diamond model comprises four core dimensions—factors of production, demand conditions, related and supporting industries, and firm strategy, structure, and competition—along with two auxiliary dimensions: government and opportunities. Notably, the model’s internal dimensions do not exist in isolation. Instead, they form a dynamic system through mutually reinforcing relationships, where changes in any single dimension can trigger chain reactions across others [45]. Regarding the model’s application, scholars generally agree that when analyzing different industries, specific adjustments to relevant dimensions are necessary based on each sector’s unique characteristics to enhance analytical relevance [46,47,48].

Considering the actual conditions of China’s agroforestry economy’s development, primary production factors include elements such as forest resource endowments form the foundation of this economic sector [46]. In contrast, advanced production factors like technological innovation hold greater strategic significance for the agroforestry economy’s development. These factors not only drive the transformation of industrial structures toward deeper and more specialized development but also open up new market domains through business model innovation, cultivating fresh economic growth points for the agroforestry economy. Demand conditions typically encompass aspects like industrial scale expansion and market consumption upgrades. When market demand exhibits rapid growth, it will guide enterprises and governments to increase investment in the agroforestry economy, driving the supply and allocation of advanced and specialized production factors. In the development process of the agroforestry economy, enterprises serve as core participants, and their structural characteristics directly impact the efficiency and quality of industrial development. Associated industries refer to other sectors closely linked to the agroforestry economy at the technological level or within the value chain. When associated industries form industrial clusters, the combined effects of scale and synergy can further incentivize governments and enterprises to increase investments in the agroforestry economy. Governments also play a pivotal role in the agroforestry economy’s development [47], primarily through formulating industrial policies, distributing government subsidies, and investing in educational resources. These measures optimize the allocation of production factors and create a favorable external environment for the agroforestry economy’s growth.

This study adopts the diamond model as its theoretical framework to explore the drivers of the agroforestry economy’s development, based on two core considerations: Firstly, the development process and comprehensive evaluation of the agroforestry economy constitute a typical complex systems engineering project, spanning macro, meso, and micro dimensions. The agroforestry economy involves not only core domains like industrial innovation, resource allocation, and market demand but also diverse stakeholders including government regulators, forestry operators, and consumer groups. Against this backdrop, employing the mature diamond model framework as a research foundation enables a rigorous, comprehensive diagnosis of key influencing factors in the agroforestry economy’s development, holding significant academic value for refining theoretical frameworks in this field. Second, despite extensive practical explorations across China’s regions, the agroforestry economy remains in its nascent stage, lacking a sustainable, virtuous economic cycle mechanism. Facing the national strategic goal of “promoting high-quality development of the agroforestry economy”, achieving innovative breakthroughs in its theoretical framework and deepening practical pathways represent both an active response to national strategic direction and a practical necessity to overcome current industrial bottlenecks. Systematically analyzing the agroforestry economy’s development through the diamond model not only provides robust theoretical explanatory power for deciphering the industry’s intrinsic logic but also offers clear directional guidance for exploring development pathways and advancing high-quality growth, holding significant practical significance.

Therefore, building upon the diamond model, this study further decomposes production factors into two independent dimensions: primary production factors and advanced production factors. This constructs an adaptive diamond model more closely aligned with the realities of China’s agroforestry economy development. The model achieves multi-level integration across macro, meso, and micro dimensions in terms of factor coverage, specifically encompassing key areas such as industrial development, resource allocation, and market supply and demand [48]. Different regions leverage their unique resource endowments and factor variations to form differentiated pathways for driving the agroforestry economy’s development through configuration matching, as illustrated in Figure 1.

2.2. Factor Analysis

First, forest resources constitute the fundamental element for the development of the agroforestry economy. As a typical resource-dependent industry [32], the spatial distribution of the agroforestry economy is significantly influenced by regional variations in ecological conditions, exhibiting pronounced clustering patterns. Its core distribution areas are predominantly concentrated in mountainous regions with favorable site conditions and suitable climatic conditions. Research by Nunes et al. [18] indicates that while technological advancements have partially mitigated the absolute constraints of natural conditions on the agroforestry economy’s development, the fundamental supportive role of these conditions remains unchanged. Against the backdrop of China’s strict ecological conservation policies and logging restrictions, the agroforestry economy has seized rapid development opportunities through its “prosperity without cutting trees” industrial model. According to data from the 2024 China Natural Resources Bulletin, China currently possesses 283.7 million hectares of forest land, including 197 million hectares of tree-dominated forests. This vast forest area provides substantial spatial support for the further development of the agroforestry economy. Moreover, ecological cultivation models perform multiple ecological functions. They not only conserve water resources, prevent soil erosion, and optimize the micro-ecological environment under the forest canopy but also enhance regional vegetation coverage [3]. In building a virtuous-cycle ecosystem, this model effectively promotes the conservation of biodiversity under the forest canopy, positively contributing to the ecological service value of understory cultivation bases [36,49].

Second, technological innovation serves as the core driving force for advancing the agroforestry economy [35]. This manifests in three key aspects: Firstly, production method innovation. Leveraging the enabling role of technological innovation, ecological cultivation techniques such as eco-farming and green pest control have achieved integrated application [42]. Particularly in the field of forest-sourced Chinese medicinal materials, technological innovation simulates wild growth environments and natural developmental states, reconstructing beneficial ecological relationships between plants and their external environment to enhance product quality [3]. Second, upgrading operational models. Technological innovation provides technical support for dynamic tracking of understory resources, ecological risk early warning, and sustainable development through information technologies like real-time meteorological monitoring and intelligent pest forecasting. Simultaneously, innovative practices in multi-tiered cultivation models—such as forest–fungus–poultry and forest–medicinal–bee systems—break the limitations of traditional monoculture, enabling multi-level development of forest land resources and optimized allocation of production factors [1]. Third, enhanced product processing. Technological innovation has accelerated the application of deep processing techniques like freeze-drying and bio-fermentation, driving the development of functional foods and biomedicines. Concurrently, comprehensive quality control systems covering the entire production process have been progressively established and refined. This not only ensures product quality stability but also extends the value chain and enhances the economic value of the agroforestry economy [2].

Third, market demand serves as the intrinsic driving force behind the development of the agroforestry economy. As the source propelling industrial structure upgrades and product innovation, the scale and quality tier of market demand determine the direction of the agroforestry economy’s development [44]. This manifests in two specific aspects: First, the expansion of the demand market. With the continuous improvement of public health awareness, consumer demand for green, healthy, and organic ecological products shows a sustained upward trend [50]. Taking China’s forest food market as an example, market demand grew from 151 million tons in 2014 to 223 million tons in 2023. During the same period, production increased from 154 million tons to 226 million tons. Guided by economic incentives, the sustained expansion of market demand positively motivates production activities among forest-based economic operators [2]. Second, the quality of the demand market has improved. With rising per capita GDP, consumer demand structures for understory products have shifted noticeably, moving from low-end products toward specialty items. Simultaneously, consumption patterns have evolved from material acquisition-centric models to multifunctional service-oriented consumption patterns [7]. This leap in consumption levels has not only spurred the rapid rise of emerging sectors like forest tourism and forest wellness but also cultivated higher-value-added growth points for the agroforestry economy.

Fourth, enterprise forms serve as the core drivers of the agroforestry economy’s development [45]. Their specific roles manifest in three key aspects: First, cultivating business entities. Currently, China’s agroforestry economy is primarily operated by family-based agricultural households [5]. In contrast, new forestry operators such as forestry enterprises and family forest farms demonstrate significant advantages in the agroforestry economy’s development. These entities possess relatively well-established production, management, and technical systems, enhancing operational efficiency by integrating production capacity resources [3]. Second, optimizing cooperative models. Introducing cooperative models such as production entrustment and commissioned management [43] drives the transition to scaled operations in the agroforestry economy, serving as a key pathway to enhance production stability and economic returns. Establishing multi-party benefit linkage mechanisms helps improve the organizational level of forest farmers [5], providing safeguards for the industry’s sustainable development. Third, the competitive landscape. Intense market competition exerts external pressure on enterprises, compelling them to continuously update products, improve services, and enhance efficiency. In practice, enterprises can extend their deep-processing industrial chains through technologies like bio-breeding and smart equipment. Simultaneously, leveraging ecological product certification and corporate branding can establish market-recognizable product brands [35], further enhancing competitiveness.

Fifth, associated industries serve as auxiliary drivers for the agroforestry economy’s development. No competitively advantaged industry operates in isolation; significant synergistic interactions often exist with upstream, downstream, and supporting sectors [48]. This synergy manifests in two key ways: First, the agroforestry economy constitutes a multi-sector composite industrial system [51], where associated industries can create a more favorable competitive environment through multidimensional collaboration. For instance, agriculture, animal husbandry, and fisheries provide critical technical support for core under-forest activities like cultivation and animal husbandry; Tourism can stimulate market demand through scenario-based design and integration, creating immersive experiential consumption environments [3,50] to attract stable consumer traffic for the agroforestry economy. Logistics development breaks geographical barriers to product sales, ensuring efficient and convenient distribution channels for under-forest products. Second, Porter emphasized the vital role of industrial clusters, where multiple industries form mutually beneficial, interdependent relationships [44]. In the agroforestry economy, this clustering effect manifests as deep cross-sector integration with external industries like education, culture, and health, driving the development of multifunctional industrial parks and demonstration bases [48]. Long-term, a robust associated industrial system must form the critical foundation for achieving deep integration across primary, secondary, and tertiary industries in the agroforestry economy, laying solid groundwork for sustainable industrial development.

Sixth, government support constitutes the institutional safeguard for agroforestry economy development. Local governments play an irreplaceable role in supplying industrial policies and allocating market resources for agroforestry economy development [47], manifested in three specific aspects: First, supporting policy measures. Currently, the agroforestry economy has been established by the Chinese government as a key pillar industry in the modernization of forestry in the new era. The national level has successively issued a series of policy documents, including the “Opinions on Accelerating the Development of the agroforestry economy” [3]. Local governments follow central policy directions while developing differentiated implementation rules based on regional realities, aiming to achieve dual goals of increasing farmer income and ensuring sustainable use of ecological resources. Second, financial support. Local governments have not only established special funds for agroforestry economy to incentivize related activities but also innovatively developed financial products like pledge financing using agroforestry economy operation income rights and forestry insurance. They have established government-bank-insurance-guarantee risk-sharing mechanisms to address short-term capital pressures faced by operators [39]. Third, vocational skills training. Government departments actively advance specialized projects to enhance forest farmers’ professional skills [42]. Through systematic, specialized training, they cultivate a cohort of highly organized farmers possessing both theoretical expertise and practical operational capabilities, providing robust talent support for the sustainable and healthy development of the agroforestry economy.

3. Research Methods and Data Sources

3.1. Research Methods

This study comprehensively applies necessary condition analysis (NCA) to reveal necessary conditions, utilizes fuzzy set qualitative comparative analysis (fsQCA) to explore the multiple combination paths of sufficient conditions, and employs content analysis to assist in the measurement of antecedent variables. By integrating these methods, this study conducts an in-depth analysis of the intrinsic mechanisms driving the development of Chinese agroforestry economy, providing a solid theoretical basis and empirical support for exploring diverse driving pathways and promoting policy improvements. The specific methods are as follows:

First, using the NCA method, we analyze the necessary conditions for the development of the agroforestry economy and reveal which influencing factors are indispensable key elements [52], providing scientific basis for policy-making through quantitative analysis. Before conducting configuration analysis, it is necessary to perform a single-factor necessity analysis. However, fsQCA can only roughly analyze the necessity of a single factor from a qualitative perspective and struggles to quantitatively characterize “at what extent a prerequisite condition becomes a necessary condition for the outcome.” NCA is a specialized method for analyzing the necessity of single factors. It determines whether a factor is a necessary condition for the outcome by comparing effect sizes and conducting Monte Carlo simulation replacement tests and quantitatively assesses the constraint level of the prerequisite conditions on the outcome through bottleneck analysis [53].

Second, using the fsQCA method, which is based on Boolean algebra and set theory, we fully consider the interactions between antecedent conditions to identify the diverse driving paths for achieving the development of the agroforestry economy under different element configurations. This method is a research method that seeks multiple concurrent causal relationships and multiple configurations between result variables and condition variables through comparative analysis of multiple causal conditions in various cases [52]. The reasons for selecting the fsQCA method in this study are as follows: On the one hand, the development of the agroforestry economy is influenced by multiple factors, and there may be multiple equivalent causal chains leading to the same outcome. Although regression analysis can analyze the different influence mechanisms of the selected factors on the agroforestry economy’s development by defining mediating variables or moderating variables, all influencing factors are merely substitute relationships or cumulative relationships, not completely equivalent relationships. In contrast, fsQCA can identify different configurations of factors influencing the development of the agroforestry economy, and each configuration possesses complete equivalence. Second, this method treats variable combinations as an integrated set for analysis, employing configurational thinking to address complex cases with limited data, uncovering complex nonlinear and asymmetric causal relationships between multiple conditional variables and outcome variables. It combines the dual advantages of “case-oriented qualitative analysis” and “variable-oriented quantitative analysis” [54].

Third, content analysis was employed, utilizing Nvivo 12 software to systematically code 56 case texts [55], and summarizing the relevant experiences and practices of the cases. Content analysis is a method based on in-depth analysis of materials to explore the essence of issues, emphasizing objective interpretation and rational construction of materials. This method can extract key information, identify objective patterns, and form theoretical frameworks from a large amount of textual materials. Additionally, repeated application of content analysis across multiple case samples can significantly reduce conclusion biases caused by insufficient data or improper handling, thereby enhancing the accuracy and persuasiveness of the research [56].

3.2. Research Samples

The research sample identified for this study comprises 56 counties (county-level cities) in China. The selection criteria were based on agroforestry economy practice cases publicly released through official channels by the National Forestry and Grassland Administration of China in 2023. The specific details of these cases are presented in

Table 1. Research Sample Information.

Region	Eastern Region	Central Region	Western Region	Northeastern Region
Research Sample	Daxing District of Beijing, Jinghai District of Tianjin, Fei County of Shandong, Guangning County of Guangdong, Ninghua County of Fujian, Youxi County of Fujian, Zherong County of Fujian, Wuyishan City of Fujian, Zhaoan County of Fujian, Wuping County of Fujian, Changting County of Fujian, Nanjing County of Fujian, Xiuying District of Hainan, Wanning City of Hainan, Wenchang City of Hainan, Dongtai City of Jiangsu, Yixing City of Jiangsu, Chunan County of Zhejiang, Songyang County of Zhejiang, Qingyuan County of Zhejiang.	Huguan County of Shanxi, Yunxi County of Hubei, Zhuxi County of Hubei, Guichi District of Anhui, Jingde County of Anhui, Nanqiao District of Anhui, Qianshan City of Anhui, Shexian County of Anhui, Lingbao City of Henan, Lushi County of Henan, Wolong District of Henan, Shaoyang County of Hunan, Shimen County of Hunan, Xiangtan County of Hunan, Fengxin County of Jiangxi, Huichang County of Jiangxi, Poyang County of Jiangxi, and Xiajiang County of Jiangxi.	Lueyang County of Shaanxi, Jinping County of Guizhou, Taijiang County of Guizhou, Xingren City of Guizhou, Yongjing County of Gansu, Kang County of Gansu, Qingcheng County of Gansu, Fangcheng District of Guangxi, Mashan County of Guangxi, Rongshui Miao Autonomous County of Guangxi, Alxa Left Banner of Neimenggu, Baoxing County of Sichuan, Qingchuan County of Sichuan, Shiping County of Yunnan, and Yiliang County of Yunnan.	Tieli City of Heilongjiang, Jiagedaqi District of Heilongjiang, Tonghua County of Jilin.
Main Characteristics	The economic development level is relatively high, market demand is strong, forest land resources are abundant, the industrial chain of agroforestry economy is relatively complete.	The economic development level is moderate, and the forest land resources are relatively good, but the deep processing links of the agroforestry economy are lacking.	The economic development level is low, market demand is weak, and the industrial structure of the agroforestry economy is underdeveloped.	The economic development level is moderate. Restricted by ecological protection, the agroforestry economy is in its infancy.

. In terms of spatial coverage, the research sample spans 23 provinces across China’s four major regions—Eastern, Central, Western, and Northeast—encompassing diverse development models such as product collection and processing, as well as forest landscape utilization. While ensuring overall homogeneity of the sample, this study maintains comparability among cases to guarantee the logical consistency of the configurational analysis. Concurrently, the analysis specifically highlights differences between cases—incorporating both high-level positive examples and lower-level comparative cases—to reveal mechanisms and pathways under varying conditions, thereby enhancing the scientific rigor and generalizability of the findings.

3.3. Data Sources

Given data availability and variations in the commencement dates of agroforestry economy activities across case areas, this study adopts 2023 cross-sectional data as the analysis timeframe. For regions involving townships and village-level administrative units, county-level data is used as a substitute. On the one hand, data regarding forest food production, forest coverage, agroforestry economy output value, agroforestry economy utilization area, per capita GDP, urban and rural disposable income, output value of agriculture, forestry, animal husbandry, and fishery, as well as tourism output value for the research samples were sourced from the Statistical Yearbook of Forestry and Grassland in China, the China Statistical Yearbook, the Statistical Yearbook of China’s Population and Employment, and the Statistical Yearbooks and National Economic and Social Development Bulletins of the administrative regions where the case studies are located. On the other hand, the textual content of the case studies is sourced from the official agroforestry economy practice cases published by the National Forestry and Grassland Administration of China, “https://www.forestry.gov.cn/c/www/gglxjjcy.jhtml” (accessed on 14 September 2025). Furthermore, to gain a comprehensive understanding of the actual development of agroforestry economy in the case regions, between March and April 2025, the research team conducted a sampling survey. Research team members conducted interviews with relevant administrative personnel from the forestry authorities in the administrative regions where the cases were located. Discussions primarily focused on four aspects: the background of the agroforestry economy development cases, key approaches, achievements, and existing challenges. The objective was to gain a comprehensive understanding of the typical practices in the agroforestry economy’s development across different regions.

3.4. Variable Measurement

3.4.1. Outcome Variables

Based on the existing research consensus, the total output value of the agroforestry economy is a comprehensive indicator of all dimensions of agroforestry economic activities, which can intuitively represent the overall scale of agroforestry economic development [38], However, it is difficult to reflect the quality of development. In contrast, output value per unit area, by eliminating area differences, can more accurately reflect the resource utilization efficiency and output capacity of the agroforestry economy, thus effectively identifying efficient intensive and extensive expansion development models in horizontal comparison [43]. Therefore, this study uses the total output value of the agroforestry economy and output value per unit area to represent the “quantity” and “quality” of agroforestry economic development levels, respectively, and employs the entropy weight method to measure the development level of agroforestry economy in different regions.

3.4.2. Causal Conditions

This study is based on the diamond model theory and identifies six key causal conditions influencing the development of the agroforestry economy in China. These factors interact through complex relationships, forming a driving system for the development of the agroforestry economy. First, forest resources serve as the material foundation for the development of the agroforestry economy, and their resource endowment level determines the potential for achieving ecological and economic benefits in the agroforestry economy. Therefore, this study selected available per capita agroforestry economic area and forest coverage rates as quantitative indicators [35] and employed the entropy weight method to assess the status of agricultural and forestry resources, thereby reflecting both the resource stock of forest land in the region and the stability of the ecosystem. Secondly, market demand is the intrinsic driving force behind the development of the agroforestry economy, and its quality and quantity directly influence the direction of its development. Therefore, this study selected per capita disposable income and per capita living expenditure as indicators [40] to characterize the intensity of market demand and employed the entropy weight method to measure market demand conditions. Third, related industries serve as auxiliary driving forces for the development of the agroforestry economy. The leading role of related industries such as agriculture, tourism, logistics, and e-commerce facilitates the transformation of the agroforestry economy toward multifunctional development. By leveraging industrial agglomeration advantages, diversified industrial clusters or bases can be formed to drive the development of the agroforestry economy. Therefore, this study selected the output value of agriculture, forestry, animal husbandry, fisheries, total tourism revenue, and logistics output value for each region, and utilized the entropy weight method to calculate the driving capacity of related industries [35,37].

In addition, this study employs content analysis to quantify these indicators based on three variables: technological innovation, enterprise structure, and government support. The process consists of five steps: First, select the research texts. This study is based on 56 case study texts (51 case study texts were used for coding, and 5 case study texts were used for theoretical saturation testing). Second, establish analytical units. Based on the organization of the texts, “sentences” were selected as the smallest analytical units in conjunction with the research questions. Third, develop a category system. This study employed data coding to generate a category system. Fourth, text content coding. Through three-level coding, core elements were extracted and their interrelationships clarified. The coding results were then discussed and revised using cross-validation and peer review methods. Fifth, saturation testing. To verify the theoretical saturation of the coding results, the five randomly reserved policy documents were subjected to the same three-level coding process, and no new concepts or categories were identified. The constituent elements and typical entries of technological innovation, enterprise forms, and government support are shown in

Table 2. Coding results of selected antecedent condition.

Categories	Dimensions	Representative Examples
Technological Innovation	Innovation in Oroduction Methods	(1) The local area has achieved an annual output of nearly 100 million seedlings for agroforestry economy through tissue culture, and developed “wild-type cultivation technology,” resulting in a production increase of 2–3 times. (Cite the source: Nanjing County of Fujian) (2) The local area is exploring the development of a green circular model of “grass-fed livestock under forests, strong livestock and trees in forests,” and has established 11 technical standards for Jianmen stone orchid and honeysuckle. (Cite the source: Jinping County of Guizhou)
	Advancement in Processing and Manufacturing	(1) The county is promoting the deep processing of agroforestry economic products, developing 11 major series of over 110 types of deep-processed products such as broken-cell wall Ganoderma lucidum spore powder and supercritical extraction spore oil, achieving revenue of 110 million RMB. (Cite the source: Wuping County of Fujian) (2) In collaboration with the research team of the National “863 Program” Golden Flower Tea Project, we have developed a series of health and functional products tailored for individuals with sub-health conditions, elevating Golden Flower Tea from a traditional agricultural by-product to a modern industrial product. (Cite the source: Fangcheng District of Guangxi)
	Enhancement in Marketing and Distribution	(1) Leveraging the advantage of being the “Southern China Traditional Chinese Medicine Valley,” the local area has integrated blockchain technology into the entire industrial chain of local medicinal herbs such as poria cocos and polygonatum, enabling precise management from cultivation to sales. (Cite the source: Chunan County of Zhejiang) (2) They use VR technology to demonstrate the process of growing edible mushrooms under forest canopies and explore innovative models such as “adopt-a-farm” agriculture. Visitors can adopt farms through an online platform. (Cite the source: Xiangtan County of Hunan)
Enterprise Structure	Cultivation of Business Entities	(1) The local government has vigorously supported and strengthened leading enterprises, establishing eight municipal-level demonstration bases for agroforestry economy and 52 standardized and scaled enterprises specializing in Jinxianlian and Dendrobium officinale. (Cite the source:Zhaoan County of Fujian) (2) The local government has introduced two well-known domestic spice processing companies, supported and nurtured three technology-based processing enterprises, and expanded the scale of agroforestry economy operations. (Cite the source: Wanning City of Hainan)
	Optimization of Cooperative Models	(1) They have adopted a cooperative model involving companies, cooperatives, and farmers, which has driven the development of over 10 small-scale Polygonatum officinale processing plants and nearly 300 farmer-run businesses in the surrounding area, thereby expanding the scale of agroforestry economic operations. (Cite the source: Zherong County of Fujian) (2) Establish interest linkage mechanisms such as “interest guarantees” and “profit returns.” The proportion of farmers joining specialized cooperatives has reached over 40%, leading to stable income growth for farmers and enhancing their risk-bearing capacity. (Cite the source: Yunxi County of Hubei)
	Reconstructing Competitive Model	(1) They have changed their mindset and promoted the transformation of the agroforestry economy from extensive to intensive, and from products to commodities, creating four geographical indication products such as “Poria cocos” and “Mushrooms.” (Cite the source: Qingchuan County of Sichuan) (2) Vigorously cultivate the “Lu Shi Forsythia” and “Ling Bao Eucommia” brands, forming a competitive strategy centered on the brand value of Chinese herbal medicines. The “Lu Shi Forsythia” brand is valued at 401 million RMB. (Cite the source: Lushi County of Henan)
Government Support	Supporting Policy Measures	(1) The local government attaches great importance to the development of the agroforestry economy and has issued a series of policy documents, including the “Agroforestry Economy Development Plan (2021–2025)” and the “Guidelines for the Construction of Agroforestry Economy Demonstration Bases.” (Cite the source: Songyang County of Zhejiang) (2) They established a special task force to investigate the types and scope of forest land suitable for the development of the agroforestry economy in the entire district and formulated the “Agroforestry Economy Development Plan (2021–2025).” (Cite the source:Ninghua County of Fujian)
	Fiscal and Financial Support	(1) The government places great importance on the “no-cutting agroforestry economic development model,” allocating 3 million RMB annually to specifically subsidize the cultivation, processing, and research and development of Poria cocos. (Cite the source: Qingyuan County of Zhejiang) (2) A total of 110 million RMB has been invested in the development of the agroforestry economy locally, with the government contributing 81.79 million RMB. (Cite the source: Fei County of Shandong)
	Vocational Skill Training	(1) They provide comprehensive technical guidance and services to forest farmers for the development of agroforestry economies through various means such as bringing technology to rural areas, organizing specialized lectures, conducting training programs, and distributing promotional materials. (Cite the source: Shaoyang County of Hunan) (2) The local government has established a science and technology responsibility promotion system and built a professional technical team of 150 people to solve the “last mile” problem of forestry technical services for forest farmers. (Cite the source: Shimen County of Hunan)

.

3.5. Data Calibration

When conducting fuzzy qualitative comparative analysis, data calibration is a prerequisite step. Calibration involves assigning membership attributes to selected cases and standardizing all case data to a range between 0 and 1 to determine the orientation of cases after assignment [54]. Currently, there is no unified standard for selecting calibration anchors in the academic community, and researchers typically need to set them independently based on the research topic and case distribution characteristics. Existing research practices indicate that calibration using complete membership (fuzzy score = 0.95), the intersection point (fuzzy score = 0.50), and complete non-membership (fuzzy score = 0.05) can achieve good results [57]. Based on the actual case circumstances and variable value distributions, this study employs direct calibration for structural calibration, setting 95%, 50%, and 5% as the anchor points for complete membership, cross-points, and complete non-membership, respectively [55,56,57]. To address the issue of unclear configuration membership caused by an affiliation degree of 0.5 in the antecedent conditions, the affiliation degree of 0.5 was adjusted to 0.499 to clarify the case’s membership [55]. Additionally, based on existing theories and case materials [55,56], the three antecedent variables—technological innovation, enterprise form, and government support—were calibrated using a four-valued fuzzy set assignment rule. This involves assigning values based on the frequency of occurrence of each element dimension in the case samples. When an element appears once or less, twice, three times, or four times across the four dimensions, it is assigned values of 0, 0.33, 0.67, and 1, respectively. This assignment method converts variables into membership degrees between 0 and 1, with values closer to 1 indicating higher membership degrees [56]. The variable calibration rules and results are shown in

Table 3. Rules for variable calibration.

Variable Type	Variable Name	Indicator Design	Variable Assignment
Outcome Variable	the Development Level of Agroforestry Economy	Total output value and output value per unit area of agroforestry economy in each region, with entropy weight method used to assign weights to each indicator (the median value is chosen as the crossover point, with three qualitative anchor points for calibration: complete membership at 0.3818, crossover point at 0.0985, and complete non-membership at 0.0241, corresponding to membership degrees of 0.95, 0.5, and 0.05, respectively)	0–1
Causal Variables	Agroforestry Resources	The per capita available area of agroforestry economy and forest coverage rate in each region, and the entropy weight method is used to assign weights to each index (the median value is chosen as the crossover point, with three qualitative anchor points for calibration: complete membership at 82.1540, crossover point at 62.4100, and complete non-membership at 23.7980, corresponding to membership degrees of 0.95, 0.5, and 0.05, respectively)	0–1
	Market Demand	The per capita disposable income and per capita living consumption expenditure in each region, with the entropy weight method used to assign weights to each indicator (the median value is chosen as the crossover point, with three qualitative anchor points for calibration: complete membership at 0.5338, crossover point at 0.2391, and complete non-membership at 0.0681, corresponding to membership degrees of 0.95, 0.5, and 0.05, respectively)	0–1
	Related Industries	Agricultural, forestry, animal husbandry, and fishery output value, total tourism income, and logistics output value in each region, with the entropy weight method used to assign weights to each indicator (the median value is chosen as the crossover point, with three qualitative anchor points for calibration: complete membership at 0.5043, crossover point at 0.2179, and complete non-membership at 0.0569, corresponding to membership degrees of 0.95, 0.5, and 0.05, respectively)	0–1
	Technological Innovation	Innovation in production methods	If all three conditions are met, the assignment is 1; if two conditions are met, the assignment is 0.67; if one condition is met, the assignment is 0.33; if none of the three conditions are met, the assignment is 0.
		Advancement in processing and manufacturing
		Enhancement in marketing and distribution
	Enterprise Form	Cultivation of Business Entities
		Optimization of Cooperation Models; Differentiated Competitive Strategies
		Reconstructing Competitive Model
	Government Support	Supporting Policy Support
		Financial Support
		Vocational Skills Training

.

4. Result Analysis

4.1. Necessary Condition Analysis

4.1.1. NCA Single Condition Necessity Analysis

Necessity and sufficiency are two important aspects of complex causal analysis. Necessity refers to the condition that the result will not occur if a specific antecedent condition does not exist, while sufficiency refers to the condition that the antecedent condition (combination) sufficiently leads to the occurrence of the result [58]. First, the NCA package installed in R was used to test whether a single causal condition constitutes a necessary condition using two methods: Ceiling Regression (CR) and Ceiling Envelopment (CE). If the effect size of a condition’s necessity reaches 0.1 or above, and the Monte Carlo simulation permutation test shows significant results (p ≤ 0.01), the condition is considered a necessary condition [57]. According to the results shown in

Table 4. NCA method necessary condition analysis results.

Condition 1	Methods	Accuracy	Ceiling Area	Range	Effect Size (d)	p-Value 2
Agroforestry Resources	CR	86.00%	0.109	0.931	0.117	0.188
	CE	100.00%	0.094	0.931	0.101	0.041
Technological Innovation	CR	98.20%	0.093	0.970	0.096	0.003
	CE	100.00%	0.140	0.970	0.093	0.005
Market Demand	CR	94.70%	0.060	0.931	0.065	0.241
	CE	100.00%	0.066	0.931	0.071	0.082
Enterprise Form	CR	91.20%	0.088	0.970	0.090	0.004
	CE	100.00%	0.109	0.970	0.096	0.008
Related Industries	CR	89.50%	0.065	0.931	0.070	0.139
	CE	100.00%	0.057	0.931	0.062	0.104
Government Support	CR	100.00%	0.030	0.970	0.031	0.255
	CE	100.00%	0.059	0.970	0.061	0.188

Note: ¹ Calibrated membership values were used; ² A permutation test was used with 10,000 repetitions.

, none of the six causal conditions reached the standard, indicating that these causal conditions do not constitute necessary conditions for the development of agroforestry economy.

Furthermore, an analysis of the bottleneck levels for each causal condition was conducted. The bottleneck level represents the minimum threshold of causal conditions required to achieve a certain outcome [57]. As shown in

Table 5. NCA method bottleneck level (%) analysis results ¹.

The Development of Agroforestry Economy	Agroforestry Resources	Technological Innovation	Market Demand	Enterprise Form	Related Industries	Government Support
0	NN 2	NN	NN	NN	NN	NN
10	NN	NN	NN	NN	NN	NN
20	NN	NN	0.1	NN	NN	NN
30	NN	NN	2.1	NN	NN	NN
40	NN	NN	4.1	NN	0.9	NN
50	NN	NN	6.1	NN	4.5	NN
60	3.3	NN	8.1	NN	8.1	NN
70	16.3	NN	10.0	NN	11.7	NN
80	29.2	21.3	12.0	9.0	15.3	NN
90	42.2	42.9	14.0	44.7	18.9	15.2
100	55.1	64.6	16.0	80.3	22.5	33.0

Note: ¹ The analysis method is CR; ² NN means “Not Necessary”.

, when at 20%, the development level of agroforestry economy is reached; market demand reaches the 0.10% bottleneck level first, while other causal conditions do not have bottleneck levels. To achieve a 100% development level of the agroforestry economy, agroforestry resources must reach 55.10%, technological innovation 64.60%, market demand 16.00%, enterprise form 80.30%, related industries 22.50%, and government support 33.0%. This result indicates that the development of the agroforestry economy requires the joint action of multiple influencing factors.

4.1.2. fsQCA Necessary Condition Test

Necessity analysis is used to assess the extent to which a result set is a subset of a given condition set. If a single antecedent condition consistently appears whenever the outcome occurs, it is considered a necessary condition for that outcome [58,59]. Consistency level is a key metric for evaluating necessity, and a threshold of 0.9 is widely accepted in the academic community as the minimum standard for identifying necessary conditions. In this study, necessity analysis of individual antecedent conditions was conducted using fsQCA 3.0 software. As shown in

Table 6. QCA Method Single Condition Necessity Test.

Causal Conditions	The Development of Agroforestry Economy		The Economic Development Outside of Agroforestry Economy
	Consistency	Coverage	Consistency	Coverage
High Agroforestry Resources	0.6667	0.6134	0.6243	0.6466
Low Agroforestry Resources	0.6159	0.5929	0.6267	0.6792
High Technological innovation	0.7688	0.7591	0.5118	0.5689
Low Technological innovation	0.5634	0.5062	0.7833	0.7923
High Market demand	0.6903	0.6731	0.5811	0.6379
Low Market demand	0.6288	0.5714	0.7023	0.7185
High Enterprise form	0.7380	0.7783	0.4847	0.5755
Low Enterprise form	0.5974	0.5073	0.8133	0.7775
High Related industries	0.6641	0.6381	0.6217	0.6725
Low Related industries	0.6592	0.6075	0.6655	0.6904
High Government support	0.6025	0.6102	0.6161	0.7025
Low Government support	0.7063	0.6204	0.6581	0.6509

, the consistency scores of all antecedent conditions fall below the 0.9 threshold, indicating that no single condition qualifies as a necessary antecedent for either the occurrence or non-occurrence of the development in the agroforestry economy. These findings are consistent with the results from the NCA.

4.2. Configuration Analysis

Sufficiency analysis of conditional configurations is central to the fsQCA method. It reveals how combinations of multiple conditions contribute to the outcome variable and serves as a key approach for understanding the complex relationships among variables. Referring to existing research [56,57,58], this study sets the case frequency threshold to 1, the original consistency threshold to 0.80, and the PRI consistency threshold to 0.70, resulting in three outcomes for the development of agroforestry economies: complex solutions, intermediate solutions, and simplified solutions. In the simplified analysis, this study adopts the method used by Du and Jia [54] to identify core conditions using the nested relationship between intermediate and simplified solutions: conditions that appear in both the intermediate and simplified solutions are core conditions of the solution, indicating a strong causal relationship with the outcome variable; conditions that appear only in the intermediate solution are marginal conditions, indicating weaker causal relationships. Coverage and consistency metrics are used to evaluate the explanatory power and effectiveness of the paths. When the overall coverage of the solution exceeds 50% and the consistency exceeds 0.80, the solution is considered optimal, indicating that the above analysis is convincing [60]. As shown in

Table 7. Configuration analysis of the development of the agroforestry economy.

Causal Conditions	Configuration 1	Configuration 2	Configuration 3	Configuration 4
Agroforestry Resources	◉	◉	◯	◯
Technological Innovation	◉	◉	◉	◉
Market Demand	◯	●	◉	◉
Enterprise Structure	◉		◉	◉
Related Industries		◎		●
Government Support	◯	●	◯
Consistency	0.9020	0.9500	0.9671	0.9720
Raw Coverage	0.3653	0.2493	0.2958	0.2809
Unique Coverage	0.1258	0.0731	0.0179	0.0080
Overall Consistency	0.9143
Coverage of the Overall Solution	0.5445

Note: ◉ and ● represent the presence of core and peripheral conditions, respectively, while ◎ and ◯ represent the absence of core and peripheral conditions, respectively. A blank space indicates that the condition is not important to the outcome.

, there are four configuration types for the development of the agroforestry economy, and the consistency between the individual solutions and the overall solution for all configurations is higher than the minimum threshold of 0.80. The overall consistency and coverage of the solutions are 0.91 and 0.54, respectively, indicating that all four configurations can be regarded as sufficient condition combinations for the development of the agroforestry economy. Additionally, based on the configuration theorization process, the configurations identified in this study are named accordingly.

4.2.1. Resource-Technology Enterprise Endogenous Model

Configuration 1 shows that with high agroforestry resources, high technological innovation, and high enterprise form as core conditions, complemented by low market demand and low government support as peripheral conditions, it is possible to drive the development of the agroforestry economy. This configuration explains approximately 36.53% of the sample cases, and about 12.58% can be explained by this configuration. Typical regions include Zhuxi County in Hubei, Jiagedaqi District in Heilongjiang, Zherong County in Fujian, Shexian County in Anhui, etc. These areas are rich in agroforestry resources, with coverage exceeding 70%, and have significant resource advantages. In the absence of sufficient market demand and government support, they have achieved the development of the agroforestry economy through technological innovation and the driving role of enterprise organizations. For example, Zhuxi County in Hubei leverages its resource advantages, such as forest land scale and species diversity, and collaborates deeply with research institutes like Huazhong Agricultural University and Hubei University of Chinese Medicine. They jointly tackle key technologies such as breeding good medicinal plant varieties and deep processing, while also carrying out agroforestry economy farming activities such as innovative breeding and grazing methods. At the same time, they actively cultivate leading forestry enterprises, explore interest linkage mechanisms such as “company + base + farmers” and “forestry cooperatives + farmers,” and have signed medicinal plant planting and purchase contracts with 17 villages and more than 1400 households [38]. This effectively integrates scattered forest land resources and optimizes the internal allocation of organizational resources, enhancing the economies of scale in agroforestry economic planting and breeding.

4.2.2. Government-Driven Resource-Technological Market Synergy

Configuration 2 points out that agroforestry resources, high technological innovation, and non-highly related industries are core conditions, complemented by high market demand and government support as peripheral conditions, can help develop the agroforestry economy. This configuration explains about 24.93% of the sample cases, and about 7.31% can only be explained by this configuration. Typical cases include Songyang County Qingyuan County in Zhejiang, and Jinping County in Guizhou. For example, the government of Songyang County in Zhejiang has introduced incentive policies such as the “Three-Year Action Plan for agroforestry economy Construction” and the “Special Financial Plan for agroforestry economy (Medicinal Herbs).” It allocates 4.2 million RMB in fiscal support annually. It resolves funding issues for business entities and farmers in the agroforestry economy through diversified financial means such as attracting investment, enterprise investment, and bank loans [7]. At the same time, Songyang County vigorously promotes the construction of digital bases, establishing the “Four-in-One” and “Four-Link Mechanism” for technology promotion systems, formulating and releasing 11 local standards, and training more than 4000 forest farmers. These efforts have improved agroforestry medicinal herbs’ planting standards and production efficiency [39],. However, the agroforestry medicinal herbs in Songyang County are only sold as raw materials to pharmaceutical companies outside the area. There is a lack of related industries such as deep-processing enterprises, logistics and warehousing, and e-commerce services. This results in products remaining in the raw material or primary processing stage, without brand premium or high-end market demand. This indicates that government support plays a limited role in the development process of the agroforestry economy [36], acting more as a light push rather than a leading force.

4.2.3. Market-Technology Enterprise Outward Expansion Model

Configuration 3 shows that with high technological innovation, high market demand, and high enterprise form as core conditions, complemented by non-high agroforestry resources and non-high government support, it is also possible to achieve the development of the agroforestry economy. This configuration explains about 28.58% of the sample cases, with typical cases including Daxing District of Beijing, Qingchuan County of Sichuan, Qianshan City of Anhui and Xiuying District of Hainan. Taking the Daxing District of Beijing as an example, although the forest coverage rate in this area is only 33.48% and lacks relevant policy support, it relies on the high-end consumer market, focusing on organic food and health services. By deepening industry–university–research cooperation, implementing full industry chain management norms for authentic medicinal materials and quality standard improvement demonstration projects, and selecting precious medicinal herbs with high economic value, strong ornamental value, and suitable for local planting, the “China Medicine Valley Production Base” has been established. At the same time, the area explores cooperative business models such as “profit guarantees” and “profit returns” [41], combined with product branding, quality traceability systems, and forest food base certification, forming a clear competitive advantage through differentiation. This has significantly increased market recognition and added value for agroforestry economic products.

4.2.4. Market-Technological Enterprise Linkage Model Assisted by Related Industries

Configuration 4 shows that with high technological innovation, high market demand, and high enterprise form as core conditions, complemented by non-high agroforestry resources and high related industries, the development of the agroforestry economy can be effectively promoted. This configuration covers about 28.09% of the sample cases, with typical examples including Dongtai City of Jiangsu Province, Shiping County of Yunnan, Wenchang City of Hainan, and Yiliang County of Yunnan. For instance, Dongtai City in Jiangsu has a forest coverage rate of only 26.18%. However, through market orientation, enterprise operations, and core-driven technological innovation, supported by industries such as agriculture, animal husbandry, and fisheries, the city has innovatively promoted practices like “long set short,” “high set short,” and “tree set irrigation,” forming six major categories and 32 types of three-dimensional composite development models. The agroforestry economy’s output value has increased by more than 25% on average. At the same time, Dongtai City vigorously develops its tourism industry, receiving 10.03 million domestic and international visitors in 2023, generating a total income of 11.24 billion RMB from tourism. This market-driven approach further supports the agroforestry economy, forming an integrated industry chain that includes agroforestry economic planting and breeding, product processing, and leisure and health services. Additionally, Dongtai City has developed the e-commerce industry for the seedling market, creating platforms such as a seedling exhibition hall and e-commerce reception area, nurturing over 600 seedling brokers, and supporting more than 300 e-commerce companies to engage in online direct sales. This has effectively alleviated market information asymmetry and sales challenges. This path emphasizes the guiding role of related industries [37], which helps promote the formation of industrial clusters and, in turn, drives the development of the agroforestry economy.

4.3. Robustness Test

This study performs robustness tests on three aspects (

Table 8. Robustness test results.

Condition Settings	Overall Consistency	Overall Coverage
Original Results (case frequency threshold = 1, original consistency threshold set to 0.80, PRI consistency = 0.70)	0.9143	0.5445
Adjusted Original Consistency Threshold Results (adjusting the previous 0.80 to 0.85, other steps unchanged)	0.9143	0.5445
Adjusted PRI Consistency Threshold Results (relaxing the previous 0.70 to 0.65, other steps unchanged)	0.8945	0.5886
Adjusted Calibration Anchor Points (setting 85%, 50%, and 15% as complete membership points, crossover points, and complete non-membership points, respectively, with other steps unchanged)	0.9159	0.5143

). First, the original consistency threshold [53,54,55] is adjusted from 0.80 to 0.85, using a stricter consistency threshold condition to identify the empirical results. The results show that the original configuration path and types still hold. Second, the PRI consistency threshold [57] is adjusted from 0.70 to 0.65. The results show that, after adjustment, the overall configuration distribution, and the combinations of core and peripheral conditions can still cover all configurations in Table 7. Third, the calibration anchor points [60] are adjusted, setting 85%, 50%, and 15% as complete membership points, crossover points, and complete non-membership points, respectively, with other steps remaining unchanged. The results show that the consistency and coverage of the overall and individual configuration solutions change little after adjustment, and the configurations remain consistent with the original ones, indicating that the research results are robust.

5. Discussion

5.1. Identification of the Factors Influencing the Development of the Agroforestry Economy

This study finds that a single influencing factor is not a necessary for driving the development of the agroforestry economy; it requires the joint action of multiple influencing factors. Although existing literature indicates that natural resource endowment, infrastructure conditions, and government policy supply have significant positive impacts on the development of agroforestry economy [36,37,38,39], market sales conditions and profitability are key factors affecting its development during the implementation and development of agroforestry economy [37]. Technological progress is necessary for improving the efficiency of the development and industrial transformation in the agroforestry economy [32,39]. However, existing studies mostly focus on case summaries of a single region or empirical analyses based on micro-data from farmers, often neglecting regional differences, and the explanatory power of the theoretical framework is somewhat insufficient. Unlike previous studies, this study constructs a theoretical analysis framework based on the diamond model. It identifies the key factors affecting the development of the agroforestry economy in China, which include agroforestry resources, technological innovation, market demand, enterprise form, related industries, and government support, thus addressing the lack of a theoretical analysis framework in past studies. Additionally, the analysis of necessary causal relationships reveals that a single factor is not necessary for the development of the agroforestry economy, indicating that the internal links within the agroforestry economy are complex, and there may be substitutive effects and interactions between factors [33]. For example, although agroforestry resources are the foundational condition for the development of the agroforestry economy, in areas with relatively scarce resources, the development can still be achieved through the synergistic effect of technological innovation, market demand, and enterprise organization. This result confirms the substitution effect of advanced production factors for primary factors in the diamond model [48]. Furthermore, although existing studies generally recognize the important role of government support in developing agroforestry economy, fostering a good policy environment and institutional guarantees [39,40,41]. However, this study further points out that the role of government support in the development of agroforestry economy is limited, acting more as a “catalyst” rather than a “main engine.” Local governments should avoid blanket subsidies when formulating supportive policies to promote the development of various business entities in the agroforestry economy. Instead, they should develop targeted support policies tailored to local conditions [7]. Finally, the key role of technological innovation is significantly demonstrated. Despite differences across regions in resources, market demand, enterprise form, related industries, and government support, all paths to development place technological innovation at the core [42].

5.2. Multiple Driving Paths of the Development of Agroforestry Economy

This study finds that the driving mechanism of the development of the agroforestry economy presents multiple concurrent causal relationships, meaning that there are multiple paths to the development of the agroforestry economy rather than a single optimal equilibrium. Existing studies, when discussing the driving mechanisms of agroforestry economy development, mainly use logit models, multiple linear regression models, and principal component analysis, respectively, measuring the extent to which various factors drive the development of the agroforestry economy [37,38,39,40,41,42]. However, in reality, the drivers are not isolated from each other, and these methods cannot measure how these drivers interact or to what extent they mutually influence the development of the agroforestry economy [49]. To address this limitation, this study, based on a configurational analysis perspective, extends the research sample to 56 counties (cities) in China, providing a more systematic understanding of the multiple driving paths for the development of the agroforestry economy and offering new theoretical perspectives and empirical support for understanding the complex driving mechanisms of the agroforestry economy in different regions. Specifically, Configuration 1 shows that in the absence of government support and market traction, relying on agroforestry resources and integrating technological innovation with the optimization of enterprise organization can achieve endogenous growth in the agroforestry economy. This is suitable for developing the agroforestry economy with planting and breeding models such as forest grains, medicinal herbs, poultry, and peaks. Both models use natural resources, such as the agroforestry economy’s open land or forest land space, as core production factors and engage in agroforestry composite operations. This improves the economic income of business entities and contributes to the restoration and stabilization of the forest ecosystem, realizing a virtuous cycle of economic and ecological benefits [15]. Configuration 2 shows that in areas lacking the drive of related industries, the government guides market consumption behavior with a “gentle push,” and collaboratively promotes technological innovation and optimizes factor allocation. This is more suitable for developing agroforestry economy product collection and processing models, such as wild mushrooms, berries, and bamboo shoots. This model utilizes good agroforestry resources and the ecological environments, strengthening technological support and innovation to reduce costs and improve efficiency in the agroforestry economy collection and processing industry [51]. Configuration 3 and Configuration 4 indicate that, under the constraint of insufficient forest resource supply, market demand-driven approaches, through market-oriented enterprise operations and technological innovation, overcome resource limitations, making it more suitable to develop forest health services, forest tourism, and nature education as models for forest landscape utilization. This model aims to meet people’s health, cultural, and ecological experience needs, promoting the diversified utilization of forest ecological functions [5,6,7]. In addition, Configuration 4 further emphasizes the synergistic effect of related industries. Integrating related industries such as tourism, logistics, and e-commerce drives the transformation of the agroforestry economy towards a multifunctional direction, forming a composite ecological economy that integrates forest therapy, medicinal food research and development, ecological tourism, and popular science education, achieving the industrialization of forest ecological resources [37].

However, this study has certain limitations, primarily including the following. First, constrained by data availability, the research primarily focuses on analyzing the overall development level of the forest-based economy, with limited depth in exploring the differences among various development models. Second, the longitudinal case analysis of different driving pathways was limited in depth. Future research could utilize typical regional cases to further validate the effectiveness and adaptability of diverse pathways. Third, this study employed cross-sectional data and did not incorporate time-series dimensions to analyze the dynamic evolution of various development pathways. Subsequent research could build upon this foundation to construct a time-series analysis framework, further examining the evolutionary patterns of different configurations. Despite these limitations, the study rests on a solid theoretical foundation and employs scientifically sound methodologies, which do not compromise the validity of its findings.

6. Conclusions and Implications

6.1. Conclusions

This study constructs a theoretical analysis framework based on the diamond model. After systematically identifying the influencing factors for the development of the agroforestry economy, 56 counties (county-level cities) in China were selected as the research sample. Using fuzzy-set qualitative comparative analysis and necessary condition analysis, the study deeply explores the multiple driving paths for the development of the agroforestry economy and its supporting elements. The conclusions of the research are as follows:

First, the key influencing factors driving the development of agroforestry economy are identified, which mainly include agroforestry resources, technological innovation, market demand, enterprise form, related industries, and government support. Among them, technological innovation always plays a core role in all configurations, and strengthening technological innovation has a universal effect on improving the development level of agroforestry economy. The role of government support in promoting the development of the agroforestry economy is limited; it functions more as a “booster” than a “primary engine.”

Second, necessity analysis indicates that no single factor constitutes a necessary condition for the development of the agroforestry economy. This highlights the internal complexity of the industry and suggests potential substitutability and interaction among various elements. Although agroforestry resources serve as the foundational condition for such development, regions with relatively scarce resources can still achieve the outcomes through the synergistic effects of technological innovation, market demand, and enterprise organization.

Third, the driving mechanisms of the development in the agroforestry economy demonstrate multiple, concurrent causal relationships, indicating the existence of diverse development paths rather than a single optimal equilibrium. Specifically, four distinct configurations have been identified: the resource–technology integration endogenous type, which emphasizes the combined application of agroforestry resources, technological innovation, and enterprise structure; the government-facilitated resource–technology–market synergy type, which highlights the amplifying effect of innovation capabilities on top of natural forest endowments, with the government acting as a secondary support; the market–technology breakthrough outward-expansion type, which is driven primarily by market demand through enterprise-led commercialization and technological outreach; and the related-industry-assisted market–technology–enterprise linkage type, which reflects the enabling role of associated industries in catalyzing development. In summary, there is no universal optimal solution for the development of the agroforestry economy. The key lies in forming an “optimal combination of conditions” through the strong linkage of core conditions and flexible configuration of peripheral conditions, thereby providing theoretical support and practical guidance for promoting the development of the agroforestry economy in different regions in accordance with local conditions.

6.2. Implications

Based on the research conclusions, the following policy recommendations are proposed:

First, strengthen technological innovation and R&D to enhance the scientific content of the agroforestry economy. On the one hand, establish industry–academia–research collaboration mechanisms to solicit innovative technical solutions from research institutes, universities, and enterprises, focusing on overcoming technical bottlenecks such as under-forest cultivation techniques, refined management models for animal husbandry, and deep processing technologies for products. On the other hand, improve the forestry science and technology service system, deepen the integration of basic and applied research in scientific institutions, accelerate the transformation and application of scientific achievements into agroforestry economy production practices, and achieve dual improvements in overall production efficiency and product quality.

Second, promote the synergy of multiple factors to optimize the adaptive structure of agroforestry economy development. Initially, break through the traditional mindset of relying on a single factor to drive high-quality agroforestry economy development, shifting toward a systematic path of multi-factor synergy. Comprehensively consider regional resource endowments, market demand, enterprise capabilities, and government support to enhance agroforestry economy outcomes through tailored combinations of factors. Simultaneously, establish dynamic feedback mechanisms to leverage the government’s pivotal role in coordinating factors. Through policy guidance, platform development, and interest coordination, facilitate cross-sectoral flow and efficient allocation of key elements, thereby strengthening the adaptability and resilience of the agroforestry economy system.

Third, develop regional development paradigms tailored to specific areas to explore suitable agroforestry economy pathways. In resource-rich areas, prioritize the resource-technology integration endogenous model or the government-light resource–technology–market synergy model. In market-oriented regions with relatively scarce resources, adopt the market–technology breakthrough outward expansion model to achieve outward growth through technological innovation and market development. In areas with industrial agglomeration foundations, select the market–technology–enterprise linkage model supported by related industries. Simultaneously, strengthen the summarization and refinement of experiences from exemplary regions, replicate successful models, and implement pilot projects for early testing. This will establish a development cascade of “pilot-driven—model-promoted—system-enhanced” propelling the agroforestry economy toward diversified development and comprehensive advancement.