Evaluation of Farmers’ Livelihood Vulnerability in Border Rural Tourism Destination and Its Influencing Factors—Take Tumen City, Yanbian Korean Autonomous Prefecture, Jilin Province, as an Example

Abstract

Rural tourism is one of the important measures used to realize the rural revitalization strategy in China. At the same time, the “action of prospering the border and enriching the people” is a crucial link in driving the economic development of border areas. With the continuous improvement of the G331 section of the Jilin Line, the development of rural tourism along the border has been accelerated. Therefore, reducing the livelihood vulnerability of farmers is conducive to promoting the rural revitalization strategy in China and consolidating the achievements of poverty alleviation and difficulties in tackling them. This paper takes Tumen City, Yanbian Korean Autonomous Prefecture, Jilin Province, as the sample area; takes Mapai Village, Bailong Village, Hexi Village, Liangshui Village, and Tingyan Village as the research objects; obtains survey data for 224 households through on-the-spot investigation; constructs an evaluation index system for farmers’ livelihood vulnerability in border rural tourist destinations based on the sustainable livelihood theory and the analysis framework of “exposure–sensitivity–adaptability”; calculates farmers’ livelihood vulnerability in this research area by the entropy method and the comprehensive index method; and classifies different villages’ and farmers’ livelihood vulnerability by the natural breakpoint method. By using the multiple linear regression method, this paper analyzes the factors influencing farmers’ livelihood vulnerability in border rural tourist destinations. The results show the following: (1) The overall livelihood vulnerability of farmers is negative, indicating that farmers have a certain ability to resist external risks, but the livelihood vulnerability of some investigated villages is positive, indicating that farmers’ ability to resist external risks is weak, which is closely related to village geographical environment, tourism market environment, family characteristics, and other factors. (2) The livelihood vulnerability of villages with relatively perfect tourism formats is low, which shows that tourism can effectively reduce the livelihood vulnerability of farmers to a certain extent. (3) The household head’s education level, the total price of agricultural machinery, annual income, the ability to borrow money, family size, the number of disabled people, and other factors have a significant influence on the livelihood vulnerability of farmers.

1. Introduction

The report of the 20th National Congress of the Communist Party of China emphasizes that building a modern socialist country in all respects remains an arduous and complex task, particularly in rural areas. It is essential to consolidate and expand the achievements of poverty alleviation as a fundamental imperative, while enhancing endogenous drivers for development in formerly impoverished regions and among those who have escaped poverty. Meanwhile, under the framework of the “Action to Revitalize Border Areas and Enrich Residents’ Lives” implemented by the State Council, efforts have been made to improve infrastructure in border regions, develop distinctive local industries, and implement targeted poverty alleviation initiatives. These measures aim to accelerate socioeconomic development in these areas, raise living standards for all ethnic groups, and achieve the goals of enriching the people, strengthening the border, bolstering national strength, and fostering good-neighborly relations. Rural tourism influences village development through a multidimensional lens that encompasses economic, social, and ecological dimensions, and it has demonstrated positive effects in terms of increasing local revenues, optimizing industrial structures, and raising farmers’ incomes [1]. The development of the G331 route (one of China’s three national border highways, which has a total length of 1437 km, passing through 10 counties and cities, including Tonghua, Baishan, the Changbai Mountain Protection and Development Zone, and Yanbian Prefecture, and is characterized by its “ecological paradise, historical corridor, winter sports destination, heroic heritage, and Sino–North Korean border” resources, making it a popular self-driving tourism route) has been progressing vigorously [2], driving the growth of rural tourism along the border. Households in these border rural tourism destinations participate in tourism activities to mitigate external vulnerabilities and sustain their family economies. However, due to the inherent vulnerability of the tourism industry, households that rely heavily on tourism for their livelihoods face certain livelihood risks.

Although extensive research has been conducted by scholars worldwide on the conceptual dimensions, influencing factors, analytical frameworks, and evaluation methods of livelihood vulnerability [3,4,5,6,7,8,9,10], empirical quantitative studies that integrate rural tourism destinations in border regions with household livelihood vulnerability remain limited. Existing work on household livelihood vulnerability has established a robust foundation for this research, which seeks to contribute to the literature on vulnerability among rural households in distinctive geographical and socioeconomic contexts. Using survey data collected from 224 households across five border tourism villages in Tumen City, Yanbian Prefecture, this study constructs a comprehensive livelihood vulnerability assessment framework that incorporates natural, social, and economic dimensions. The entropy method and natural breaks classification are employed to analyze the data. The study examines the spatial distribution of livelihood vulnerability and identifies key household characteristics associated with varying vulnerability levels. Additionally, a multiple linear regression model is utilized to identify factors influencing household-level and village-level vulnerability outcomes. The results aim to clarify the role of multiple factors in determining livelihood vulnerability, thereby providing a clearer picture of the livelihood conditions faced by households in border-area tourism settings. Based on these findings, targeted policy measures are proposed to reduce vulnerability, support household well-being, and promote the sustainable development of rural tourism in border regions.

1.1. Literature Review

1.1.1. Theoretical Basis of Livelihood Vulnerability

Vulnerability research originated within the context of natural disasters [11] and was initially applied in the geological sciences. It has since evolved into a key focus in the study of global environmental change and sustainable development, serving as a fundamental analytical tool [12]. Timerman P first introduced the concept of vulnerability, positing that the severity of adverse effects from disaster events on various systems depends on their level of vulnerability, while the inherent resilience of these systems can help mitigate such impacts [13]. Subsequently, in 2001, the World Bank defined vulnerability as the probability that individuals or households would fall below the local poverty line as a result of external risk shocks [14]. The Intergovernmental Panel on Climate Change (IPCC) conceptualizes vulnerability as a function of three core components: exposure, sensitivity, and adaptive capacity. It describes the degree to which a system is adversely affected by climate change impacts and its capacity to cope with or adapt to such effects [15,16]. Luo Chengping et al. were among the first to identify environmental sensitivity and degradation trends as two fundamental characteristics of vulnerable ecosystems, conducting foundational research on the ecological environment in northern China’s agro-pastoral transition zone [17]. Building on the R-H (Risk-Hazard) model, Burton defined vulnerability as a comprehensive manifestation of the interactions among hazard factors, exposure of vulnerable units, and sensitivity. In the context of disaster events, it specifically reflects the resistance and adaptive capacities of social systems [18]. Cutter further advanced the field by proposing the HOP (Hazards-of-Place Model of Vulnerability) framework for analyzing place-based disaster vulnerability. She argued that regional vulnerability arises from the interplay between social and natural dimensions of vulnerability, thereby shifting research attention toward the coupling of natural and human systems. This model represented a breakthrough in vulnerability science [19]. As research continued to deepen, the concept of vulnerability extended into socio-economic studies, enriching the body of vulnerability literature from multiple perspectives. In this vein, Lankao et al. highlighted that urban vulnerability is influenced not only by natural disaster risks but also by factors such as urban infrastructure and human activities [20].

The issue of livelihoods began to attract scholarly attention in the 1950s. The concept of livelihood proposed by Chambers has gained widespread acceptance, defining it as a means of making a living based on capabilities, assets, and activities, where assets include both tangible possessions and intangible resources such as rights [21]. As an integrated concept combining livelihood and vulnerability, livelihood vulnerability does not yet have a universally agreed-upon definition. Prowse defines livelihood vulnerability as the shock experienced by individuals or households when their ability to access resources is compromised due to external disturbances [22]. According to the UK Department for International Development (DFID), livelihood vulnerability refers to the unstable condition of individuals’ or households’ livelihoods under the influence of external shocks, stresses, trends, and seasonal changes, as well as their high sensitivity to environmental variations [23]. Although definitions of livelihood vulnerability vary, their common essence refers to the capacity of rural households to withstand and cope with risks when confronted with external shocks and changes. Given that tourism exerts multifaceted impacts on the economic, social, and cultural dimensions of destinations and serves as a critical driver in transforming residents’ livelihoods, integrating “tourism” and “livelihood” into the discourse can effectively address livelihood challenges faced by local communities. This approach also enriches the understanding of livelihood capital and livelihood strategies.

1.1.2. Theoretical Foundation of Livelihood Vulnerability Framework

Scholars commonly employ analytical frameworks such as the sustainable livelihoods framework and the “exposure–sensitivity–adaptive capacity” framework to assess livelihood vulnerability. The sustainable livelihoods framework classifies livelihood capital into five categories: human, social, natural, physical, and financial capital. It proposes that households operating within specific vulnerability contexts draw on combinations of these capital assets and adopt various livelihood strategies to cope with external changes. Through these means, they aim to improve their living conditions and achieve livelihood objectives [24]. This framework serves as an analytical tool for elucidating the formation process and mechanisms underlying livelihood vulnerability. The Intergovernmental Panel on Climate Change (IPCC) identifies exposure, sensitivity, and adaptive capacity as the three core dimensions of vulnerability. Polsky developed the Vulnerability Scoping Diagram (VSD), also known as the exposure–sensitivity–adaptive capacity (ESA) framework, which conceptualizes vulnerability as a function of these three components. This model has established a universal structure for vulnerability assessment systems and is widely applied in academic research [25]. Within this framework, exposure refers to the degree to which households are subjected to external risk shocks; sensitivity indicates the extent of adverse effects experienced by household livelihoods; and adaptive capacity describes a household’s ability to withstand external disturbances, respond continuously to changes, and ensure stable future development. Adaptive capacity is commonly evaluated using indicators related to household livelihood capital [16]. Overall, the ESA framework offers a robust analytical structure for quantifying vulnerability levels.

This paper adopts the “exposure–sensitivity–adaptive capacity” framework as its analytical foundation, as it enables the construction of an indicator system from multiple perspectives encompassing natural, economic, social, and cultural dimensions. The three components of this framework comprehensively reflect the relationships between different dimensions and livelihood vulnerability, allow for clearer quantification of the impacts of various indicators on households, and facilitate targeted assessment of livelihood vulnerability. In recent years, influenced by the perspective that “vulnerability originates from humans themselves,” vulnerability research has increasingly incorporated social dimensions. The focus has shifted from singular studies of natural ecosystems to coupled human–environment systems, social–ecological systems, and human–land systems, with human and societal adaptive capacity becoming a core issue in vulnerability assessment. This approach has emerged as an effective tool for analyzing human–environment relationships and sustainable development. The framework has been extensively applied in social–ecological systems (SESs). For instance, Batenga et al. examined the impacts of multiple stressors, such as resource scarcity and political conflict, on the livelihood vulnerability of agro-pastoral households in the Kilombero Valley, Tanzania [26]. Against the backdrop of rural revitalization and the development of rural tourism, the issue of livelihood vulnerability among rural households has gradually become a focus of scholarly attention. Households are not only key participants in the development of rural tourism but also direct drivers of accelerated development and transformation in rural tourism destinations. Ashley pioneered the use of the sustainable livelihoods framework to investigate the broad effects of tourism on rural community livelihoods in Namibia, highlighting that a livelihoods perspective offers a valuable entry point for understanding tourism development. Scholars such as Jia Yaoyan [27] and Cai Jingjing [28] have further evaluated household livelihood vulnerability from this perspective. In summary, the impact of tourism activities on household livelihood vulnerability has become a critical issue influencing both the development of tourism destinations and the sustainable development of rural households.

1.1.3. Characteristics of Border Regions

Borders represent critical zones of national governance, characterized by their unique geographical positions. They play vital roles in safeguarding national sovereignty and territorial security while possessing distinct natural, cultural, and geopolitical attributes [29]. In geopolitical terms, geopolitics not only influences international relations but also introduces challenges and risks to economic and trade exchanges. Geopolitical cooperation or conflict has become a significant force shaping international trade dynamics. Favorable geopolitical relations can facilitate trade, reduce trade costs, and promote industrial cooperation and development. Conversely, geopolitical conflicts—such as terrorism, political instability, territorial disputes, and military confrontations—heighten instability in cooperation, increase trade costs, and create mechanisms like trade barriers that elevate investment and trade risks. These factors amplify the “barrier effect” of national borders, hindering economic development [30,31]. In demographic and economic aspects, border regions often experience economic marginalization and geopolitical instability, resulting in long-term underdevelopment, low levels of openness, inadequate infrastructure, and diversified yet unstable livelihood strategies. These areas frequently exhibit imbalanced regional development. Severe outmigration, sluggish economic growth, and uncompetitive industries contribute to household hollowing-out and village abandonment in border regions [32,33]. Culturally, border areas are often home to diverse ethnic minorities. These communities have developed distinct characteristics in terms of production methods, lifestyles, customs, and socio-cultural traditions, often including unique languages, writing systems, and religious beliefs [34]. Among these factors, the emergence of geopolitical conflicts poses threats to the livelihoods of households in border areas, destabilizes their income structures, and hinders industrial development. Concurrently, population outflows lead to labor shortages within households, increasing economic pressure and impeding capital accumulation. This reduced capacity to withstand external shocks makes these households more susceptible to livelihood vulnerability. From a cultural perspective, fully leveraging cultural resources to develop distinctive cultural industries and promote tourism activities can diversify livelihood options and increase household income. Such initiatives may, to some extent, reduce livelihood vulnerability among border households. Border tourism encompasses both cross-border tourism and tourism within border areas. Border rural tourism refers to tourism activities organized within the territorial boundaries of a country, capitalizing on the unique peripheral characteristics of these regions. Its social impacts are primarily reflected in employment generation, cultural transformation, and economic development. Research on border rural tourism can provide valuable insights for its planning, development, and spatial layout, thereby enhancing regional competitiveness and core appeal. Promoting border rural tourism can effectively address issues such as inadequate rural infrastructure, facilitate cultural dissemination, and diversify livelihood strategies for local households. To a certain extent, this contributes to reducing livelihood vulnerability in these communities [35,36,37,38].

1.1.4. Factors Influencing Livelihood Vulnerability

Risk factors contributing to vulnerability include natural disasters, economic fluctuations, policy changes, ethnic conflicts, diseases, unemployment, and sudden accidents [39]. As tourism destinations represent complex socio-economic and natural ecosystems influenced by economic, social, environmental, and institutional factors, the interplay of these diverse elements collectively shapes livelihood vulnerability. Research on tourism and vulnerability typically focuses on the vulnerability of economic systems in tourist areas as well as the fragility of tourism environments and resources. Building on the scholarly contributions of domestic and international researchers [40,41,42,43], this paper categorizes vulnerability factors into environmental factors, psychological resilience factors, livelihood capital factors, and policy factors. Environmental factors constitute the most direct influencers, particularly for households relying on natural resources for their livelihoods. TT Deressa et al., through a discrete choice model analysis of households in the Nile Basin of Ethiopia, demonstrated that education level, land scale, and access to climate information are key drivers of adaptive behaviors among Ethiopian farmers [44]. Psychological factors typically encompass social support, social status, and relational networks. Schininá et al. incorporated psychological elements into the study of livelihood vulnerability factors, highlighting the relevance of behavioral economics and humanitarian approaches in vulnerability research [45]. The study of livelihood capital provides a comprehensive framework for assessing households’ capacity to withstand external risks, enabling systematic evaluation of risk resilience across multiple dimensions. Drawing from the sustainable livelihoods framework, Li Xiaoyun et al. analyzed five key types of livelihood capital—natural, human, social, financial, and physical—among impoverished households, revealing the concomitant relationship between livelihood vulnerability and asset vulnerability in poor communities [46]. Policy plays a crucial role in reducing livelihood vulnerability by enhancing livelihood capital and optimizing institutional environments to strengthen risk resistance capabilities. M.F. Gaworek-Michalczenia et al. evaluated the implementation effectiveness of climate adaptation and resilience-building projects under the Global Climate Change Alliance (GCCA+) in Tanzania, thereby assessing the impact of adaptation interventions on vulnerability and livelihood resilience [47].

2. Research Area and Data Sources

2.1. Study Area

Tumen City is a county-level city under the jurisdiction of Yanbian Korean Autonomous Prefecture, Jilin Province (Figure 1). It is located in the east of Jilin Province, along the lower reaches of the Tumen River, at the border of China, North Korea, and Russia, with a total border length of 60.3 km, and is the geographical center of the Northeast Asia Economic Circle. Tumen City benefits from a geographically advantaged location and serves as a pivotal transportation hub. It has been officially designated by the state as a first-class border open city. The city functions as a major node along several key transport routes, including the Hunchun–Ulanhot Expressway, the Changchun–Hunchun High-Speed Railway, the Northeast Eastern Railway, and the Eastern Corridor initiative led by the China–Mongolia–Japan Tripartite Committee. Furthermore, Tumen borders the Democratic People’s Republic of Korea across the Tumen River and is a multi-ethnic area with a predominant Korean ethnic population. It is adjacent to Hunchun City to the east, Longjing City to the southwest, Yanji City to the west, and Wangqing County to the north. The terrain is high in the northwest and low in the southeast, and the Nangang Mountains run through the whole territory in a north–south direction. With a total area of 1142.65 km², there are 4 townships, 50 administrative villages, and 1 provincial economic development zone. Tumen City faces Wencheng County, North Hamgyong Province, North Korea, across the Tumen River, which is the boundary river between the People’s Republic of China and the Democratic People’s Republic of Korea. Tumen City is 100 km away from the Sino-Russian border and 130 km away from the Sea of Japan. It is the first border port city in Jilin Province with border, river, traffic-line, and offshore characteristics, and it is also an important transportation hub connecting China, North Korea, and Russia to radiate Northeast Asia.

By the end of 2023, there was a registered population of 97,242 in Tumen City, including 51,894 Koreans and 42,868 Han individuals. The Korean population is relatively large, and the Korean folk culture is rich, including intangible cultural heritage such as Korean hand-drum dancing, knife dancing, Korean “drama”, and Korean rice sausage-making skills, as well as well-preserved Korean characteristic eating habits. Tumen City enjoys the reputation of “the first city along the Tumen River”, relying on its unique border scenery, long-standing minority culture, and remarkable geographical location, and it promotes the development of border rural tourist destinations. At present, the national AAA-level tourist attractions owned by Tumen City include the Tumen Port Scenic Spot and the Riguangshan Forest Park, folk-culture tourist attractions headed by century-old tribes, and industrial tourist attractions headed by Jilin Tumen River Pharmaceutical Co., Ltd., which are rich in tourism resources. Some villages face North Korea across the river, such as Mapai Village and Bailong Village, attracting a large number of tourists due to their characteristic tourism resources. In 2024, Tumen received 2.49 million tourists and realized tourism income of CNY 1.21 billion, which figures were up by 18.23% and 61.33%, respectively.

Due to the rapid development of border rural tourism in Tumen City, the construction of border villages has been in depth. In addition to traditional ways of livelihood, such as working and farming, farmers have also found a way of gaining a livelihood by participating in tourism. Some farmers have enriched their livelihood types by building characteristic homestays, operating farmhouses and picking gardens, and operating restaurants. We selected sample villages by analyzing the current development status of the villages and related village data, taking into account factors such as existing specialized industries, geographical location, unique tourist attractions, and the current state of tourism development. Five villages were selected as research sites: Ma Pai Village, Bai Long Village, Hexi Village, Liangshui Village, and Tingyan Village. Ma Pai Village and Bai Long Village are located within Yueqing Town and are Korean ethnic minority villages with relatively rich cultural tourism resources, facing North Korea across the river. Hexi Village, Liangshui Village, and Tingyan Village are located in Liangshui Town, which is a border minority township with a 4 km border with North Korea. It has provincial key leading enterprises in agricultural industrialization and diversified industries, including agriculture, aquaculture, the tourism industry, etc. The tourism industry has a certain foundation, and it constantly builds a new type of tourism industry with development potential. All the above five villages belong to border villages, which are representative with their own characteristics.

2.2. Data Sources

Based on the principle of typicality and representativeness, this paper selected five border villages in Tumen City as the investigation area. Typicality and representativeness refer to the selection of villages that are representative in terms of geographical location, population composition, livelihood patterns, and economic activities. The chosen case villages also possess a long historical continuity, well-preserved traditional customs, and distinct tourism development models, leading to diverse household livelihood patterns within them. The study employed random sampling and questionnaire surveys for data collection. Tumen City comprises four towns: Yueqing, Shixian, Liangshui, and Chang’an. Based on the number of border villages and the current state of tourism development, Yueqing Town and Liangshui Town were selected as the study areas. Subsequently, characteristic villages within these towns were identified according to the following criteria: location along the border, capacity to support tourism activities, ethnic typicality, and rich ethnic cultural heritage. Ultimately, five villages were chosen as research sites: Ma Pai Village and Bai Long Village in Yueqing Town and He Xi Village, Liangshui Village, and Tingyan Village in Liangshui Town. A one-to-one questionnaire survey was conducted to fully understand the tourism development and livelihood of farmers in border villages. A total of 245 questionnaires were distributed in this field survey, and 21 data points with extreme values and extreme outliers were eliminated. The effective rate of questionnaires was 91.43%. The Cronbach’s α coefficient was 0.780. Meanwhile, the Bartlett Sphere Test and the KMO value were used to evaluate the validity of the questionnaire. The results show that the KMO value was 0.824 and that the Bartlett Sphere Test probability significance was 0.000, indicating that the reliability of the questionnaire is good.

3. Research Framework and Research Methods

3.1. Research Framework

Livelihood vulnerability is affected by exposure sensitivity and adaptive capacity. That is, the higher the exposure sensitivity, the higher the livelihood vulnerability, the stronger the adaptive capacity, and the lower the livelihood vulnerability. The exposure of farmers in border rural tourist destinations is affected by social risks, farmers’ family conditions, changes in the external natural environment, and other factors. Sensitivity is usually measured by economy, environment, and perception. Adaptability is measured by selecting relevant indicators from the dimension of livelihood capital. This paper explores the intensity of adaptability. Farmers’ livelihoods are vulnerable to the impact of external risks, which has an important impact on farmers’ livelihood capital and livelihood strategies. With the continuous development of rural tourism, farmers’ livelihoods will change accordingly. Tourism brings benefits to farmers, but it also has some drawbacks. At the same time, tourism is easily disturbed by the external environment, which leads to the vulnerability of farmers who rely too much on tourism. Therefore, according to the livelihood vulnerability framework and sustainable livelihood framework combined with the development status of the study area, an analysis framework (Figure 2) was constructed to evaluate the livelihood vulnerability of farmers and identify the influencing factors so as to provide effective measures for farmers to reduce their livelihood vulnerability.

3.2. Construction of Livelihood Vulnerability Index System

Livelihood vulnerability assessment is based on the vulnerability analysis framework proposed by the IPCC, and the three dimensions in the analysis framework are the main elements used to measure livelihood vulnerability. This study focuses on farmers in rural tourism areas located along the border, with a particular emphasis on factors directly related to the farmers themselves. The study is based on the current situation of the research area. However, due to the sensitivity and difficulty in obtaining data related to geopolitical factors, such as border policies and cross-border movements, these indicators have not been included in the current research framework. In the future, as more data become available, we will gradually enrich and refine this research. By combining the current situation and the characteristics of farmers’ livelihood vulnerability in border rural tourism destinations, fully considering the social and economic pressures faced by farmers in the study area, combining the livelihood capital content in the sustainable livelihood analysis framework based on the IPCC’s “exposure–sensitivity–adaptability” analysis framework, and referring to the existing research results on livelihood vulnerability, the livelihood vulnerability index system [16,24,47,48,49] is constructed, thus constructing the livelihood vulnerability evaluation system for farmers in border rural tourism destinations (as shown in

Table 1. Evaluation system and weight of farmers’ livelihood vulnerability in border rural tourist destinations.

Dimension	Sub-Dimension	Indicator Layer	Indicator Definition
Exposure (E)	Natural risk	The frequency of natural disasters every year (0.106)	Frequency of natural disasters: almost infrequent = 1, low frequency = 2, general = 3, relatively frequent = 4, very frequent = 5
		Market competition (0.032)	The degree of competition among villagers in running the tourism industry is very weak = 1, not strong = 2, strong = 3, very strong = 4
	Market risk	Market price fluctuation (0.145)	Whether the change in market price in the off-season of tourism will affect families: no = 1, yes = 2
		Influence of peak season and off-season tourism on family income (0.098)	Barely affected = 1, not serious = 2, average = 3, serious = 4, very serious = 5
		Difficulty degree of rural tourism industry operation (0.018)	Difficult = 1, more difficult = 2, general = 3, easier = 4, very easy = 5
	Family risk	Influence of medical expenditure on family economy (0.046)	No influence = 1, relatively small influence = 2, general influence = 3, relatively large influence = 4, very large influence = 5
		Number of disabled family members (0.294)	0 people = 1, 1 people = 2, 2 people = 3, 3 people and above = 4
		Human expenditure (0.159)	CNY 0~2000 = 1, CNY 2000–3000 = 2, CNY 3000–4000 = 3, CNY 4000 and above = 4
		Number of family members with chronic diseases (0.102)	0 people = 1, 1 people = 2, 2 people = 3, 3 people and above = 4
Sensitivity (S)	Economic sensitivity	Proportion of tourism income (0.024)	The proportion of tourism income to family income is 0% = 1, 0%~25% = 2, 26%~50% = 3, 51%~75% = 4, over 75% = 5
		Proportion of practitioners in the tourism industry (0.054)	The proportion of people engaged in the tourism industry relative to the total number of people is 0% = 1, 0%~25% = 2, 26%~50% = 3, 51%~75% = 4, over 75% = 5
		Household debt (0.254)	Does the household have a loan? No = 1, yes 2
		Liability impact (0.261)	The impact of debt on households is completely non-effective = 1, slightly effective = 2, averagely effective = 3, greatly effective = 4, very greatly effective = 5
		Sales of featured products that account for the proportion of total revenue (0.022)	Below 25% = 1, 26%~50% = 2, 51%~75% = 3, above 75% = 4
	Environmental sensitivity	Improvement of tourism infrastructure (0.036)	Perfect = 1, relatively perfect = 2, general = 3, relatively imperfect = 4, imperfect = 5
		Drinking water quality (0.211)	Tap water = 1, mountain spring water = 2, well water = 3, water directly obtained from rivers, lakes, reservoirs/pits, and ponds = 4
		Distance to nearest medical institution (0.108)	The time from home to the nearest medical institution is perfect = 1, relatively perfect = 2, average = 3, relatively imperfect = 4, and highly imperfect = 5
	Perceptual sensitivity	Impact of tourism policy on livelihood (0.028)	No influence = 1, little influence = 2, general influence = 3, great influence = 4, serious influence = 5
Adaptability	Human capital	Education level (0.008)	Illiteracy = 1; primary school = 2; junior high school, technical secondary school = 3; high school, junior college, and technical school = 4; bachelor’s degree or above = 5
Adaptability		Family size (0.042)	Number of people in the family: 0 = 1, 1 = 2, 2 = 3, 3 and above = 4
		Number of migrant workers (0.122)	0 people = 1, 1 people = 2, 2 people = 3, 3 people and above = 4
		Amount of skills training (0.048)	Participated in skills training organized by villages and towns 0 times = 1, 1 time = 2, 2 times = 3, 3 times and above = 4
		Number of cadre members/public officials in the family (0.141)	Number of cadre members/public officials in the family: 0 = 1, 1 = 2, 2 = 3, 3 and above = 4
	Material capital	Family housing area (0.032)	Below 70 m2 = 1, 71 m2~80 m2 = 2, 81 m2~90 m2 = 3, 91 m2~100 m2 = 4, 100 m2 and above = 5
		Quantity of durable consumer goods (0.015)	Number of motor vehicles or items of agricultural machinery: none = 1, 1 item = 2, 2 items = 3, 3 items and above = 4
		Number of homestays operated (0.122)	Number of family-run homestays: none = 1, 1 = 2, 2 = 3, 3 and above = 4
		Price of agricultural machinery (0.036)	Total price of agricultural machinery (CNY): NO = 1, less than CNY 5000 = 2, CNY 5000~10,000 = 3, CNY 10,000~20,000 = 4, more than CNY 20,000 = 5
	Natural capital	Cultivated land area (0.040)	The actual value of cultivated land area owned by farmers/hm2: 1 = 0 mu~5 mu, 2 = 5 mu~10 mu, 3 = 10 mu~15 mu, 4 = 15 mu~20 mu, 5 = 20 mu and above
		Distance from residence to surrounding core scenic spots/scenic spots (0.011)	More than 5 km = 1, 4 km~5 km = 2, 3 km~4 km = 3, 2 km~3 km = 4, less than 2 km = 5
	Financial capital	Annual household income (0.015)	Total household income in the past year, unit (CNY)
		Income diversity (0.024)	Family tourism industry, agriculture, migrant workers, and other diverse income sources (CNY)
Adaptability		Access to loans/lending methods (0.101)	Relatives or friends = 1, banks or credit unions = 2, others = 3
	Social capital	Government support for the tourism industry (0.030)	Very small = 1, small = 2, general = 3, large = 4, very large = 5
		The degree of government support for farmers’ employment (0.034)	Very helpful = 5, helpful = 4, unclear = 3, unhelpful = 2 completely, unhelpful = 1
		Sources of information (0.024)	The channels for obtaining relevant policy information in the village are chatting with villagers = 1, watching media platforms = 2, publicizing policies in the village = 3, watching news = 4
		Participation in subsistence allowances (0.095)	Do family members have a minimum living allowance? No = 1, yes = 2
		Is there a Class A scenic spot around (0.059)	No = 1, yes = 2
		Accessibility of tourist attractions (0.011)	Poor = 1, relatively poor = 2, average = 3, relatively good = 4, good = 5

).

3.3. Calculation Method of Livelihood Vulnerability Index

3.3.1. Entropy Method

In this paper, the objective weighting entropy method is used to weigh the index. Because the values obtained from the questionnaire had different dimensions, in order to make the indexes comparable, the extreme value standardization method was used to standardize the indexes so that the values were in the range of 0~1, and different positive and negative indexes were standardized by formulas. For example, Formulas (1) and (2) are as follows:

Positive indicators:

$${X^{\prime }}_{ij}={\displaystyle \frac{X_{ij}-X_{j,min}}{X_{max}-X_{min}}}$$

(1)

Negative indicators:

$${X^{\prime }}_{ij}={\displaystyle \frac{X_{j,max}-X_{ij}}{X_{max}-X_{min}}}$$

(2)

where ${X^{\prime }}_{ij}$ is standardized data, $X_{ij}$ is the original value of the j-th index of the i-th survey object, $X_{j,max}$ is the maximum value of item j, and $X_{j,min}$ is the minimum value of item j.

On this basis, the index is weighted, and the weight is written as follows: $W_j$.

First of all, we should calculate the proportion and formula for each index of each farmer:

$$P_{ij}={\displaystyle \frac{{X^{\prime }}_{ij}}{{\sum }_i^n{X}_{ij}}}$$

(3)

where $P_{ij}$ indicates the proportion of index item j of the i-th survey object.

Secondly, $E_j$ is calculated as the entropy value of item j.

$$E_j=\left(-\frac{1}{\mathit{ln}n}\right)\times {\sum }_{i=1}^n{P}_{ij}{ \mathit{ln} P}_{ij}$$

(4)

where $E_j$ is the entropy value of item j.

Then, the index difference coefficient is calculated:

$$G_j=1-E_j$$

(5)

where $G_j$ is the standard difference coefficient of item j.

Finally, the index weight is calculated using the calculation result obtained by Formula (4):

$$W_j={\displaystyle \frac{G_j}{{\sum }_{j = 1}^m{G}_j}}$$

(6)

where $W_j$ is the weight of item j.

3.3.2. Calculation of Livelihood Vulnerability Index

The livelihood vulnerability index (LVI) is calculated according to the method proposed by Hahn [50] et al., and the LVI is defined as a function of “exposure–sensitivity–adaptability”, and its value ranges from −1 to 1. After calculating the weight of each index, calculate the index of each dimension of livelihood vulnerability. The specific calculation formula is as follows:

$$E={\omega }_{ej}\times X_{deij}$$

(7)

In Equation (7), E is the exposure index of farmers, ${\omega }_{ej}$ is its corresponding weight, and $X_{deij}$ indicates its corresponding exposure level.

$$S={\omega }_{sj}\times X_{sij}$$

(8)

In Formula (8), S is the sensitivity index of farmers, ${\omega }_{sj}$ is its corresponding weight, and $X_{sij}$ is its corresponding sensitivity.

$$A={\omega }_{aj}\times X_{aij}$$

(9)

In Equation (9), A is the adaptability index of farmers, ${\omega }_{aj}$ is its corresponding weight, and $X_{aij}$ for its corresponding degree of adaptability.

$$LVI=\left(E-A\right)\times S$$

(10)

In Equation (10), LVI is the livelihood vulnerability index of farmers, E is the exposure index of farmers, A is the adaptability index of farmers, and S is the sensitivity index of farmers.

3.4. Natural Breakpoint Method

The Jenks natural breakpoint method is a method used to classify data based on the similarities and differences within a group of data and divides the data into multiple continuous categories according to requirements [51]. In this paper, the natural breakpoint method was used to classify the livelihood vulnerability index. The natural breakpoint method can minimize the average difference with the mean of other classes and maximize the average difference with the mean of other classes, such that it can better identify the livelihood vulnerability level of farmers. With the help of ArcGIS 10.6 software, according to the calculated livelihood exposure, sensitivity, adaptability, and livelihood vulnerability indexes of farmers, the farmers were graded from low to high by the natural breakpoint method and divided into three grades: low, medium, and high, and the results were statistically analyzed.

3.5. Multivariate Linear Regression Model

In order to explore the factors influencing farmers’ livelihood vulnerability in border rural tourist destinations, a multiple linear regression model was used. The model formula is as follows:

$$Y_j={\beta }_0+{\beta }_1{X}_{q1}+{\beta }_2{X}_{q2}+\dots +{\beta }_p{X}_{qp}+{ϵ}_p$$

(11)

In Equation (11), the dependent variable, Y_j, is the livelihood vulnerability index of the j-th farmer; the independent variables $X_{q1}$, $X_{q2}$, $X_{q3}$, …, $X_{qp}$ are the factors influencing Q farmers’ livelihood vulnerability; ${\beta }_1$, ${\beta }_2$, ${\beta }_3$, …, ${\beta }_p$ are the estimated parameters of the corresponding variables; ${\beta }_0$ is a constant term; and ${ϵ}_p$ is the random error.

4. Results and Analysis

4.1. Characteristics of Farmers’ Livelihood Vulnerability

Farmers’ livelihood vulnerability refers to farmers’ ability to resist risks when facing external environmental disturbance, which is the result of the joint action of farmers’ livelihood and the external environment and can fully reflect farmers’ ability to cope with external changes. If farmers’ existing adaptability cannot be restored when they are impacted, farmers’ livelihood will be in an unstable state. Therefore, the livelihood vulnerability index can reflect the trend of farmers’ livelihood. If the livelihood vulnerability is positive, the exposure degree and sensitivity degree are higher, and the greater the index, the greater the risk faced by farmers, indicating that their livelihood is more fragile; if the livelihood vulnerability is negative, this indicates that the stronger the adaptability of farmers, the smaller the negative value, indicating that the livelihood of farmers is less fragile [52].

According to the above calculation formula, as shown in

Table 2. Livelihood vulnerability index of farmers in border rural tourism destinations.

	Exposure	Sensitivity	Adaptability	Livelihood Vulnerability
Holistic	0.2688	0.3345	0.3129	−0.0147
Bailong Village	0.3306	0.4128	0.3347	−0.0017
Mapai Village	0.3369	0.3498	0.3111	0.0090
Hexi Village	0.3108	0.3140	0.2997	0.0035
LiangShui Village	0.3667	0.3174	0.2810	0.0271
TingYan Village	0.3756	0.3289	0.3070	0.2257

, it can be concluded that the value of the farmers’ livelihood vulnerability index in the study area is −0.0147. Generally speaking, the value range of the farmers’ livelihood vulnerability index in the study area is [−0.1853, 0.3118], which is in the range of −1~1, so the farmers’ livelihood situation in this area has a certain vulnerability. Because the farmers’ livelihood vulnerability is close to the critical point of 0, taking effective measures could help the farmers resist external environmental risks. Through Figure 3, we can see the values for the adaptability, sensitivity, and exposure of farmers in different villages and more intuitively see the villages with prominent values in each dimension.

This study applies the Jenks natural breaks classification (JNBC) algorithm. Iterative optimization is employed to minimize intra-class variance while simultaneously maximizing inter-class variance, yielding statistically robust thresholds that faithfully represent the empirical distribution of the composite livelihood–vulnerability index. On this basis, households are stratified into three discrete vulnerability classes: high, medium, and low. The computational workflow is documented in full in the referenced source [51].

It can be seen from

Table 3. Numbers of samples with different livelihood vulnerability levels.

Project	Livelihood Vulnerability
Classification	Low livelihood vulnerability	Moderate livelihood vulnerability	High livelihood vulnerability
Sample size	34	159	31
Proportion	15.18%	70.98%	13.84%

that about 15.18% of farmers have a low livelihood vulnerability [−0.1853, −0.0346), 70.98% have a medium livelihood vulnerability [−0.0346, 0.0696], and 13.84% have a high livelihood vulnerability [0.0696, 0.3118]. Through the distribution map of the farmers’ livelihood vulnerability grades in Figure 4, it can be seen that the overall characteristics of the samples show a normal distribution trend, and most of the livelihood vulnerability indexes are above 0, which indicates a state of certain livelihood vulnerability.

Figure 5 is a mountain map of farmers’ livelihood vulnerability levels in border rural tourist destinations, which is used to explain the distribution of each livelihood vulnerability level. The higher the peak, the more concentrated the data distribution density in this interval; the wider the peak, the wider the data distribution. The results show that the data for the low livelihood vulnerability grade exhibit a multi-peak and left-leaning distribution, which shows that the lower livelihood vulnerability index accounts for a larger proportion in the low livelihood vulnerability grade. The data for the middle livelihood vulnerability grade show a right-sided distribution, which shows that the higher livelihood vulnerability index accounts for a larger proportion in the middle vulnerability grade. The data for the high livelihood vulnerability grade are roughly distributed to the right, the data distribution is relatively wide, and the proportion of farmers with a high livelihood vulnerability index is large. Within each level of livelihood vulnerability, the index of farmers’ livelihood vulnerability is unevenly distributed.

4.2. Livelihood Vulnerability Assessment

From the perspective of the livelihood vulnerability index, the overall livelihood vulnerability index of the study area is −0.0147. This indicates that farmers in this area have a certain level of livelihood vulnerability, with the majority experiencing moderate livelihood vulnerability. The livelihood vulnerability index (LVI) generally reflects a directional trend. A positive LVI value indicates relatively high levels of exposure and sensitivity, with larger positive values signifying greater livelihood vulnerability among households. Conversely, a negative LVI value suggests stronger adaptive capacity, with smaller negative values (i.e., values closer to zero) indicating lower livelihood vulnerability [47,52]. The LVI ranges from −1 to 1, where lower values correspond to reduced vulnerability. A negative index value reflects households’ enhanced capacity to adapt to external stresses and shocks [52]. According to the calculation results, the livelihood vulnerability indexes of the five villages are ranked from low to high, namely, Bailong Village (−0.0017), Hexi Village (0.0035), Mapai Village (0.0090), Liangshui Village (0.0271), and Tingyan Village (0.2257). Figure 6 illustrates the distribution of livelihood vulnerability across different villages. It can be seen that the livelihood vulnerability of Bailong Village, Hexi Village, and Mapai Village is relatively low, while that of Liangshui Village and Tingyan Village is on the high side. Areas with low livelihood vulnerability have superior geographical positions and are all located in border areas. Bailong Village and Mapai Village face North Korea across the river. Most of the villages are Korean, and they are rich in characteristic folk tourism resources, such as century-old tribes, Korean characteristic folk villages and homestays, “Twenty-Four Stone” sites, etc., integrating the superior resources in the villages, driving the development of border rural tourism, with remarkable economic benefits and ensuring farmers’ livelihood. The development of the characteristic agricultural products industry in Hexi Village, such as the construction of an organic-agriculture sightseeing experience park, a strawberry-picking garden, and the Tumen River Pharmaceutical Co., Ltd., has given a strong impetus to local economic growth. With the rapid development of the tourism industry and agriculture, farmers’ livelihood adaptability in this region is strong, and farmers’ livelihood exposure and sensitivity are low, resulting in low livelihood vulnerability for farmers in this region. Liangshui Village is a Korean village combining urban and rural areas, with the idea of building a border tourism characteristic township with organic agriculture as the leader and driving economic growth through the establishment of the Tumen Korean Cultural Park and the Chinese herbal medicine industry. However, its characteristic tourism resources are relatively few, the development momentum of the tourism industry is insufficient, farmers’ income is significantly affected by industry, market price fluctuations are sensitive, exposure and sensitivity are high, and livelihood adaptability is low, which leads to slightly higher livelihood vulnerability among farmers. Tingyan Village is far away from the city center, focusing on agriculture and animal husbandry, with a superior ecological environment, Tingyan Mountain City, Daiwanggou, etc., and a rich cultural heritage. However, due to its long distance, imperfect development of characteristic industries, few folk industries, easy disturbance by external factors, and fluctuations in farmers’ income, its exposure and sensitivity are on the high side, resulting in high livelihood vulnerability for farmers in this region.

4.3. Exposure–Sensitivity–Adaptability Assessment

From the perspective of exposure, by measuring the exposure of farmers from three aspects: natural risk, market risk, and family risk, the overall exposure index of the study area is 0.2688, and the value interval of the exposure index of farmers in the study area is [0.0232, 0.8282]. According to the natural breakpoint method, the exposure level of farmers is divided into three types: high, medium, and low. From Figure 7, we can see the proportion of different levels of exposure, among which 85 farmers are in the interval of low exposure [0.0232, 0.2857]. There are 96 households in the range of moderate exposure [0.2857, 0.4703]. There are 43 households in the range of high exposure [0.4703, 0.8282]. So, the farmers in this study area are mainly in the middle and low exposure levels. Figure 8 is the overall exposure distribution map, which shows the exposure distribution of different villages.

The exposure calculation results of the five villages were sorted from low to high: Hexi Village (0.3108), Bailong Village (0.3306), Mapai Village (0.3369), Liangshui Village (0.3667), and Tingyan Village (0.3756). Figure 8 illustrates the distribution of livelihood vulnerability across different villages. Hexi Village has the lowest exposure index. Hexi Village is mainly engaged in agriculture and the agricultural products industry. Through cooperation with Jilin Tumen River Pharmaceutical Co., Ltd., it produces red ginseng drinks, corn beard tea, and the Hovenia dulcis beverage to drive local characteristic agricultural production. Farmers’ production income is relatively stable, but the products are vulnerable to market price fluctuations, and the industry lacks diversity. The distance from the city center is remote, and the tourism industry’s income is unstable, leading to the village facing low livelihood exposure. Bailong Village and Mapai Village have low livelihood exposure, are close to the city center, and are dominated by agriculture and the tourism industry. Because they are easily affected by the off-season of the tourism market, income from the tourism industry fluctuates, and commodity prices are affected by the market. They are traditional ethnic villages, and there are more worldly expenses, such as festivals, which lead to unstable livelihoods among farmers. The livelihood exposure of Liangshui Village is on the high side. Liangshui Village primarily focuses on planting Chinese herbal medicines, with many farmers engaged in agricultural production, mostly small and medium-sized operations that meet the needs of local families. However, their ability to withstand extreme weather conditions is limited. Tingyan Village, which is far from the city center, has the highest exposure index. Most farmers are mainly engaged in animal husbandry and livestock and poultry breeding, while some farmers are mainly agriculturalists, with limited natural resources and relatively few tourism resources. A few farmers are engaged in the tourism industry, and farmers borrow money to operate their industries and invest more in production activities, so the exposure index is high.

From the perspective of sensitivity, livelihood sensitivity is measured from three aspects: economic sensitivity, environmental sensitivity, and perceptual sensitivity. The sensitivity index of the study area is 0.3345, and the value interval of the sensitivity index of farmers in the study area is [0.0637, 0.2752]. According to the natural breakpoint method, the sensitivity grade of farmers is divided into three types: high, medium, and low. As shown in Figure 9, we can see the proportions of individuals in the different sensitivity grades, with 121 farmers in the interval of low sensitivity [0.0637, 0.2752]. There are 45 households in the range of moderate sensitivity [0.2752, 0.5283]. There are 58 households in the range of high sensitivity [0.5283, 0.9115]. So, the farmers in this study area are mainly at low and high sensitivity levels. The sensitivity distribution of different villages can be seen in Figure 10.

According to the calculation results, the sensitivity of the five villages was ranked from low to high: Hexi Village (0.3140), Liangshui Village (0.3174), Tingyan Village (0.3289), Mapai Village (0.3498), and Bailong Village (0.4128). Hexi Village has the lowest sensitivity. Most family members in Hexi Village are in good health, and the proportion of people engaged in tourism is small. However, due to the diversification of farmers’ income structures, farmers’ income fluctuates little and is less affected by external risks, which makes them have the lowest sensitivity. Through the investigation of Liangshui Village, it was found that most of the farmers in this village are mainly engaged in planting and agriculture, so they are highly dependent on nature; there are few industrial projects in the village, some tourism resources are in the development stage, the tourism infrastructure needs to be improved, and the industrial structure is single, which is not conducive to improvement in the total income of farmers’ families. High exposure will lead to high sensitivity to a certain extent, and Tingyan Village has the highest exposure, which leads to its high sensitivity. Tingyan Village has the least number of people engaged in tourism; is far away from the administrative center; has poor traffic conditions; has made less effort to develop tourism, focusing on animal husbandry and agriculture; and also has higher requirements for the natural environment. Some family members are in poor condition, and debts increase family economic pressure, resulting in high sensitivity. The cyclical and seasonal fluctuations of tourism make it vulnerable to external shocks, which lead to great changes in income fluctuations and increase the sensitivity of farmers’ livelihood. Mapai Village and Bailong Village have superior geographical positions and are deeply influenced by tourism policies, which accelerate the social and economic exchanges between farmers and the outside world. Mapai Village has high sensitivity. Mapai Village is famous for its characteristic barbecue, which helps the development of tourism. There are a large number of families engaged in tourism, and tourism income accounts for a high proportion of total income. Part of farmers’ income structure comes from the sales of characteristic agricultural products. The number of homestays is increasing year by year, and the problem of land use is prominent, so it has a certain dependence on tourism, which leads to the high sensitivity of farmers in Mapai Village. Bailong Village has the highest sensitivity, and its tourism industry is developing rapidly, focusing on agriculture and tourism. The annual number of tourists is as high as 50,000. The proportion of farmers’ characteristic agricultural products sold ranks first among the five villages, and the number of family members engaged in tourism is the largest, such that the area is highly dependent on tourism. The tourism infrastructure in the village needs to be improved, resulting in its having the highest sensitivity.

From the perspective of adaptability, the adaptability of farmers is measured from five aspects: human capital, material capital, natural capital, financial capital, and social capital. The overall adaptability index of the study area is 0.3129, and the value interval of the adaptability index of farmers in the study area is [0.0975, 0.2498]. According to the natural breakpoint method, the adaptability grades of farmers are divided into three types: high, medium, and low. From Figure 11, we can see the proportions of adaptability of different grades, with 67 farmers in the interval of low adaptability [0.0975, 0.2498]. There are 107 households in the range of medium adaptability [0.2498, 0.3875]. There are 50 households in the range of high adaptability [0.3875, 0.7038]. So, the farmers in this study area are mainly at the middle and low exposure levels. Figure 12 shows the distribution of adaptability in different villages.

The calculation results for adaptability are as follows: Liangshui Village (0.2810), Hexi Village (0.2997), Tingyan Village (0.3070), Mapai Village (0.3111), and Bailong Village (0.3347). The per capita cultivated land area of farmers in Liangshui Village is better than that in other villages, with farmers mainly planting Chinese herbal medicines, most of which are agricultural, and the development of other industries is small. Some farmers’ income is mainly transfer income, and their income is single, and the phenomenon of family members working is common. Agricultural products in Hexi Village are vulnerable to market price fluctuations, far away from core scenic spots, and the development of homestay tourism is slow. However, its industrial resources are relatively abundant, and some factories are used for renting and cooperation, which provides employment opportunities for farmers and has a good chance of obtaining loans. Farmers can effectively participate in village decisions. Tingyan Village is the village farthest from the city center, with a relatively small cultivated land area, but it is rich in agriculture and animal husbandry projects, and has also introduced fry breeding, etc. Durable consumer goods are in good condition and have a certain adaptability to resist external risks. Mapai Village has extensive sales of agricultural products and agricultural and sideline products, superior labor ability compared to other villages, high accessibility to surrounding scenic spots, rapid development of homestay projects and catering services, and strong overall adaptability. Bailong Village is rich in catering projects, with advantages in the development of agriculture and tourism, mastering core tourism resources, good family income, and good social capital accumulation. Most farmers have high trust in villagers, can get help when encountering difficulties, and have strong overall adaptability.

To sum up, we find that the exposure index mainly reflects farmers’ susceptibility to market risks and natural risks and that farmers are vulnerable to market price fluctuations, tourist season, and force majeure factors in the natural environment. The fluctuations in market prices and the off-season of tourism have a staged impact on farmers’ income, and the instability leads to exposure risk regarding farmers’ livelihood. The external risks can be mitigated by human factors to reduce farmers’ exposure. As for sensitivity, it mainly affects farmers economically, and villagers have family debts, which lead to economic pressure. Generally speaking, the number of people engaged in tourism is large, the sales of characteristic products contribute to farmers’ incomes, the sensitivity of economic families and perceptions have a secondary impact, farmers’ management is easily affected by local tourism policies, and tourism infrastructure is in a stage of continuous improvement, thus making farmers’ livelihood sensitive. As far as adaptability is concerned, the farmers’ labor force is in good health, cultivated land resources are abundant, material capital is relatively sufficient, the influence of policies accelerates the development of the tourism industry, and villagers can obtain village policy information in time with high flexibility, which provides conditions for farmers to resist external interference factors.

5. Analysis of the Factors Influencing Farmers’ Livelihood Vulnerability in Border Rural Tourist Destinations

5.1. Analysis of Multivariate Linear Regression Results

With the constant change of the external environment, farmers’ livelihood has a certain fragility in terms of their livelihood activities. Previous studies have confirmed that the factors influencing farmers’ livelihood vulnerability include family size, age of head of household, education level, trust in surroundings, family income, market fluctuations, etc. [26,28,47,53,54,55]. Building upon the aforementioned research findings and a synthesis of the literature concerning livelihood vulnerability among farmers in rural tourism destinations, this study conducts a comprehensive analysis of the prevailing conditions in the surveyed regions. Taking into account local contextual factors, livelihood vulnerability is selected as the dependent variable. Explanatory variables are identified and categorized at three distinct levels: characteristics of the household head, household attributes, and features associated with tourism market risks. Based on the above-mentioned related research results, this paper takes farmers’ livelihood vulnerability as the explanatory variable and selects variables as explanatory variables from three levels: farmers’ head characteristics, family characteristics, and tourism market risk characteristics. A multivariate linear regression model is used to analyze the factors influencing farmers’ livelihood vulnerability in border rural tourism areas (see

Table 4. Regression analysis results for factors influencing farmers’ livelihood vulnerability in border rural tourist destinations.

Livelihood Vulnerability
	Explanatory Variable	Coefficient	T Value
	(Constant)	0.002	3.082
Characteristic factors of the head of household	Gender	0.297	1.045
	Age	0.241	−1.175
	Education level	0.039 **	−2.082
Family characteristic factors	Family size	0.050 *	−1.974
	Number of public officials/cadre members	0.029 **	−2.205
	Number of disabled people	4.4323 × 10−8 ***	5.680
	Total price of agricultural machinery	0.002 **	−3.155
	Annual total household income	0.000316 ***	−3.663
	Household loan method	0.000739 ***	−3.425
Risk characteristics of the tourism market	Sales revenue of specialty goods accounts for a proportion of total revenue	0.276	−1.091
	Impact of market price fluctuation on households	0.000001 ***	5.923
	The influence of tourism in the peak season and off-season on income	0.005 **	1.045
R2	0.744
F statistic	51.102

Note: *** indicates significant at 0.01 level, ** indicates significant at 0.05 level.

). According to its fitting model, the goodness of fit, R₂, is 0.744; the F statistic is 51.102; and the variance expansion factor shows that there is no multicollinearity among independent variables, which shows that the selected variables can explain the factors affecting farmers’ livelihood vulnerability in case areas to a certain extent. According to the results of the regression analysis in Table 4, we can see the relationship between household head characteristic factors, family characteristic factors, and tourism market risk factors, and farmers’ livelihood vulnerability.

5.2. Influence of Household Head Characteristics on Farmers’ Livelihood Vulnerability

According to the regression analysis results (Table 4), regarding the variable of household head characteristics, the education level of the household head is significant at the 5% level and has a significant negative impact on farmers’ livelihood vulnerability, which shows that the higher the education level of farmers, the weaker their livelihood vulnerability. At the same time, the relationship and importance of each element can be seen in Figure 13. Farmers with a higher education level have stronger cognitive ability and learning ability, skillfully use the skills they have learned, can quickly apply the skills they have learned to their lives, are more tolerant of new things than other farmers, and capture current affair hotspots and information accurately and screen information accurately. Therefore, they have greater employment opportunities, and with the increase in age, farmers have relatively rich livelihood experience, strong judgment ability, and can master new skills to improve their survival ability, which leads to their higher adaptability and certain resistance in the face of external risks.

5.3. Influence of Family Characteristics on Farmers’ Livelihood Vulnerability

Among the family characteristic variables, as shown in Table 4, family size, the number of family members who are public officials or belong to cadres and the total price of agricultural machinery are significant at the 5% level, while the number of disabled people, annual total household income, and household borrowing methods are significant at the 1% level. Among them, the number of public officials or members of cadres, the total price of agricultural machinery, annual total household income, and household borrowing methods are negatively correlated with livelihood vulnerability, while the number of disabled people is positively correlated with livelihood vulnerability. The relative importance of each element can be seen intuitively from Figure 14. During the investigation, it was found that the larger the family size, the more disposable income a family has and the smaller the livelihood vulnerability of farmers. Farmers’ families in the survey areas can obtain relevant policy subsidies in time, and their family members have strong working ability, can effectively share economic pressure, and can stabilize their income sources. Some farmers’ family members are public officials or belong to village cadres. These families have strong information acquisition ability and communication ability, and their income is relatively stable. The information acquisition channels are diverse, and farmers can use the obtained information flexibly, which has a certain inhibitory effect on farmers’ livelihood vulnerability. The total price of agricultural machinery, annual household income, and household borrowing methods have a negative impact on farmers’ livelihood vulnerability. There are relatively many types of farmers’ income in the investigated areas, and farmers in this area are engaged in agriculture, animal husbandry, and tourism, as well as related policy subsidies. Therefore, farmers’ income is relatively stable, and the total value of agricultural machinery owned by families is high, which can create more income by reliance on existing assets and effectively reduce farmers’ livelihood vulnerability. Farmers in the survey area have strong credit ability, harmonious neighborhood relations, and live in a harmonious atmosphere, and they have the ability to borrow and obtain funds. In addition, there is a positive correlation between the number of disabled people and livelihood vulnerability. The more disabled people, the greater livelihood vulnerability is. However, most of the farmers in the survey area have large families, mainly middle-aged and old people, few young people, stable income sources, timely access to relevant allowances, relatively low family pressure, and livelihood ability, so they can effectively reduce livelihood vulnerability to a certain extent.

5.4. Impact of Tourism Market Risk on Farmers’ Livelihood Vulnerability

The relative relationship of each factor can be seen intuitively through Figure 15. Regarding the tourism market risk characteristics, as shown in Table 4, the impact of market price fluctuation is significant at the level of 1% and the impact degree of tourism in the off-season is significant at the level of 5%, and these factors are positively correlated with livelihood vulnerability. The survey site is easily affected by seasons, and there is a sharp contrast between the off-season and the peak season. The tourism development of villages such as Bailong Village and Mapai Village is better, which leads to the high dependence of some farmers on the tourism industry. For example, the farmers in the survey site operate in agricultural tourism, homestays, specialty product sales, and other industries, and invest more in the tourism industry, which is seriously affected by seasonality. Some villages are far away from the core scenic spots, and the passenger flow is unstable. On this basis, they are affected by the off-season tourism, and their livelihood means and strategies are limited, which leads to seasonal fluctuations in their income. After investigation, it was found that farmers with diversified livelihood strategies have a strong ability to resist risks when faced with external interference such as off-season tourism and market price fluctuation. There are many kinds of products in the research area, and the market prospects for Chinese herbal medicines, agricultural and animal husbandry products, and tourism characteristic products are broad. In recent years, we have continuously increased investment in them to build characteristic local brands. However, price fluctuation in the tourism market is affected by many factors, such as market environment, tourist destination heat, local policies, natural conditions, etc. Some villages are vulnerable to natural disasters when the precipitation is heavy, which leads to fluctuations in product prices, so farmers face certain livelihood fragility.

6. Conclusions and Recommendations

6.1. Conclusions

Based on field survey data and a vulnerability assessment framework that incorporates exposure sensitivity and adaptability, this paper constructs a vulnerability assessment system for the livelihoods of farmers in rural tourism areas located on the border. The research area includes rural tourism areas located on the border, and the research subjects are farmers in the sample area, who were evaluated for farmer livelihood vulnerability based on questionnaire data and geospatial data. In order to excavate the hierarchical characteristics of different villages and farmers’ livelihood vulnerability, this paper classifies farmers’ livelihood vulnerability by the natural breakpoint method, evaluates different dimensions, and analyzes the factors influencing farmers’ livelihood vulnerability by the multiple linear regression method. Based on the research results, the following conclusions are drawn.

(1)

On the whole, the indexes of exposure, sensitivity, adaptability, and livelihood vulnerability are 0.2688, 0.3345, 0.3129, and −0.0147, respectively, and the overall value of livelihood vulnerability is negative. This shows that farmers’ ability to cope with risks is higher than the exposure risks encountered by farmers, and their livelihood vulnerability is small. Engagement in tourism activities has been shown to significantly mitigate livelihood vulnerability, primarily through bolstering farmers’ adaptive capacities, diversifying their livelihood strategies, and augmenting their stock of livelihood capital to a certain extent. These findings are consistent with the empirical studies conducted by Zhang Jiaqi et al. [54]. This result affirms that tourism-oriented livelihoods constitute a viable livelihood strategy and validates tourism-driven poverty alleviation as an effective model for targeted poverty reduction and sustainable livelihood development [56]. However, it is also essential to acknowledge potential adverse effects associated with tourism participation. For instance, heightened financial investments in family-operated tourism businesses may impose debt pressures on certain households, thereby creating economic strains. While livelihood diversification can effectively reduce vulnerability, overreliance on the tourism sector may conversely heighten exposure to livelihood risks. A balanced approach is thus critical to maximizing the benefits of tourism while minimizing its potential drawbacks.

(2)

There are certain livelihood vulnerabilities in the sample areas, among which the livelihood vulnerabilities of farmers in Bailong Village, Hexi Village, and Mapai Village are relatively low. The livelihood vulnerability of Liangshui Village and Tingyan Village is relatively high. Bailong Village, Hexi Village, and Mapai Village have superior geographical positions, are close to the city center and surrounding core tourist attractions and scenic spots, have a good foundation for the development of the tourism industry, and all have characteristic industrial development, so their livelihood vulnerability is low. Liangshui Village has superior location conditions, low development of tourism resources, and relatively few tourism industries, but other industries develop well and farmers’ income is relatively stable, which leads to certain livelihood fragility. Tingyan Village is rich in cultural heritage, but its location is relatively remote, with agriculture and animal husbandry as the main industries, and the development of the tourism industry is relatively small, which makes its industrial types relatively single. However, through investigation, it is also found that Tingyan Village has been planning and developing tourism industry projects in recent years, which has increased farmers’ income sources to a certain extent.

(3)

From the factors affecting farmers’ livelihood vulnerability in border rural tourist destinations, the education level of household heads, the total price and annual income of agricultural machinery, family members being public officials or belonging to village cadres, borrowing ability, and family size have significant negative effects on farmers’ livelihood vulnerability. The number of disabled household members, the degree of tourism off-peak season, and the degree of market price fluctuation have a significant positive impact on farmers’ livelihood vulnerability. This conclusion aligns with certain findings reported by Liang Wangbing [57] and Wang Huiwen [58]. Meanwhile, the present study integrates the concept of diversified livelihood strategies and comprehensively examines factors such as household composition, health status, and material capital—taking into account influences from the tourism market—to evaluate material capital. Some factors differ due to the influence of local environmental differences and local development conditions, resulting in variations. Future research may incorporate other endogenous and exogenous variables to further enhance the depth and robustness of the analysis.

(4)

This study explores the livelihood vulnerability of farmers in border-area rural tourism destinations, drawing on sustainable livelihood theory, livelihood vulnerability theory, and human–environment interaction theory to analyze both their geographical contexts and livelihood conditions. It contributes to the theoretical foundations for researching livelihood vulnerability among farmers in such borderland tourism settings, while also shedding light on how tourism influences livelihood vulnerability from a geographical location perspective. Furthermore, this research expands the scope of rural tourism studies. While traditional rural tourism attractions often emphasize agritourism and ancient towns, this study incorporates distinctive features of border rural areas—such as Korean ethnic culture and views into North Korea—as integral components of the tourism landscape. Thereby, it broadens the thematic and regional content of rural tourism research.

(5)

Research on household livelihood vulnerability encompasses a broad range of fields and has generated substantial findings. However, there remains a notable scarcity of studies focusing specifically on households in rural tourism areas situated along border regions. This study seeks to address this gap by proposing measures to reduce livelihood vulnerability among these households, enhance their adaptive capacities, and ultimately improve their quality of life. Furthermore, it extends the application of key theoretical frameworks—such as livelihood vulnerability theory and sustainable livelihood theory—by broadening their empirical scope and enriching their subject matter, thereby establishing a robust theoretical foundation for this inquiry. By constructing an integrated “exposure–sensitivity–adaptive capacity” framework and incorporating tourism dimensions, this study not only diversifies the forms of theoretical application but also identifies key factors influencing livelihood vulnerability among farmers in borderland rural tourism areas. In addition to its theoretical contributions, this research supports the practical implementation of initiatives such as the “Prosperity through Border Development” policy. It provides insights for fostering local tourism development, emphasizes the preservation and promotion of traditional cultural features as sources of competitive advantage, and encourages the exploration of innovative tourism models. The findings aim to offer practical experience and managerial implications for policy makers, government officials, and relevant stakeholders.

6.2. Limitations

This study examines Tumen City within the Yanbian Korean Autonomous Prefecture as its research area. However, owing to the limited sample size, only a subset of villages could be included in the analysis. As a result, the findings may not fully capture the comprehensive characteristics of livelihood vulnerability among rural tourism households in border regions. Future studies should aim to broaden the geographical and demographic scope of sampling to enhance the representativeness and generalizability of the results. The factors through which tourism affects farmers’ livelihoods are multifaceted and complex. Subsequent research would benefit from incorporating a wider range of endogenous and exogenous variables to enable more nuanced and multidimensional analysis. Furthermore, livelihood vulnerability is inherently dynamic. Investigating its spatiotemporal evolution and examining associated influencing factors over time could help uncover underlying mechanisms and contribute to a more robust body of knowledge in this field.

6.3. Recommendations

Combined with the development status of each village, farmers with fragile livelihoods are effectively identified, which can reduce exposure and sensitivity in relation to farmers’ livelihoods, enhance farmers’ ability to resist risks, and improve the endogenous motivation of farmers in tourist destinations. Farmers’ livelihood strategies should be optimized, farmers’ excessive dependence on tourism reduced, and the development planning of rural tourism destinations and farmers’ learning ability considered when carrying out industrial activities, and importance should be attached to asset construction and risk training. The livelihood vulnerability of each village has not yet reached the best state, and market competition, market price fluctuation, and tourism market development trends are sources of risks faced by farmers. Similar situations exist in the investigated villages. In order to continuously reduce their livelihood vulnerability, optimization should be carried out in the following aspects.

6.3.1. Improve the Livelihood Capital of Farmers in Tourist Destinations and Strengthen Their Ability to Resist Risks

Livelihood capital plays a decisive role in diversifying farmers’ livelihood strategies and reducing farmers’ livelihood vulnerability. Therefore, it is necessary to continuously improve farmers’ livelihood capital in tourist destinations, which can enhance the buffer capacity of families with livelihood vulnerability to cope with external risks. The local government should improve the livelihood capital level of low-income families so as to improve the adaptability of this group. In terms of human capital, we should pay attention to providing relevant production skills training, improving farmers’ skills, enhancing the service management level of scenic spots, establishing a talent return mechanism, attracting young people to return home for employment, cultivating compound talents to promote urban and rural development and rural revitalization, and promoting the sustainable development of border rural tourist destinations. For example, Mapai Village, Bailong Village, and other villages with relatively good tourism development provide targeted skills training for local farmers. To enhance livelihood capabilities, it is essential to deliver more targeted skills training tailored to the specific conditions of local tourism development and the employment needs of farmers. Efforts should also be made to actively strengthen competencies in tourism management and service delivery, enhance farmers’ awareness of tourism services, and give full play to the advantages of bilingual services. Unified standard management of homestays and the catering industry should be promoted, and scattered resources should be integrated. In terms of material capital, the level of tourism infrastructure construction in tourist destinations should be improved, the quality of homestays strengthened, rural accessibility and service quality improved, and the problem of individual accessibility solved. For example, Liangshui Village and Tingyan Village can guide farmers to change their ideas and encourage farmers to actively explore new production methods. Farmers should be encouraged to leverage their existing livelihood assets and mobilize underutilized household labor toward complementary economic activities. Such diversification helps optimize the coordination between tourism-based livelihoods and other income-generating pursuits, thereby broadening revenue streams and enhancing economic resilience, and it encourages the development of shared homestays, using idle houses to realize the potential of resources. Given that Tingyan Village, Liangshui Village, and Hexi Village are located considerable distances from several tourism sites, measures such as enhancing tourist shuttle services will be implemented to improve connectivity to these attractions. This initiative aims to strengthen linkages between the villages and surrounding tourism destinations, thereby facilitating visitor access and supporting local tourism development. In terms of natural capital, in rural development, rural superior capital can give full play to its role, promote the combination of industry and agriculture and animal husbandry development, develop local characteristic agricultural industries, tap characteristic tourism resources, and strengthen the combination of tourism development and ecological protection. For example, Hexi Village, Tingyan Village, and other villages with ecological advantages should establish a circular mechanism of “ecological protection–tourism development–industrial cultivation–farmers’ income”, make rational use of the land space in the village, and give full play to the function of idle land. The construction of characteristic agricultural products and cultivation of characteristic ecological industries should be vigorously developed. In terms of financial capital, we should formulate financial products according to local conditions and increase financial support from banks, credit cooperatives, and other financial institutions to farmers in tourist destinations. For example, in villages with strong industrial development, such as Liangshui Village and Mapai Village, robust industrial development and considerable potential are demonstrated. The government may formulate targeted industrial support policies tailored to the development stage and model of these rural tourism destinations. Such policies could help alleviate funding shortages, offer financial assistance to local farmers, and incentivize young people to return to their hometowns to engage in entrepreneurship. Furthermore, banks and other financial institutions could introduce relevant guarantee mechanisms—such as price index insurance—and improve social security policies to provide safeguards for farmers involved in rural tourism. These measures would contribute to reducing market risks and enhancing the resilience of local livelihoods. In terms of social capital, we should establish organizations to participate in local tourism development, promote tourism development, increase publicity to tourist destinations, expand the connection network between tourist destinations and the outside world, pay attention to local farmers’ ideas, and encourage local farmers to fully participate in rural tourism construction and benefit distribution. The construction of the digital network should be strengthened, courses such as e-commerce operations should be carried out, the e-commerce platform should be flexibly used, marketing methods should be innovated while carrying out tourism publicity, the sales volumes of agricultural products should be increased, the resources of village characteristic agricultural products should be further integrated, the collective advantage of high-quality agricultural products should be formed, and more tourists should be attracted.

6.3.2. Optimize Industrial Structure and Promote Industrial Innovation and Diversified Development

The five villages examined in this study are poised to derive substantial benefits from the preservation and development of their traditional cultural heritage. These efforts have the potential to establish unique tourism identities, elevate the profile and reputation of the villages, attract greater business investment, reinforce the local economy, and contribute to broader industrial diversification within the region. Expediting the development of complementary service industries—such as catering, accommodation, telecommunications, and education—can foster constructive synergies between the tourism sector and other related industries. Enterprises ought to be guided to invest in rural tourism projects; enhance the added value of characteristic agricultural products; establish a certification system of “tourist landmark agricultural products”; strengthen the development of Yanbian rice; plant and process Chinese herbal medicines; establish ecological picking gardens, etc.; improve policies and systems; gradually expand production scales; attract more excellent industries to invest and build locally; vigorously develop sightseeing agriculture and picking experiences; and turn to the integrated development of “production + processing + sales” to reduce uneven tourism income. We should encourage farmers to put part of their labor into non-agricultural production on the premise of stabilizing their livelihood assets, to coordinate the development of various livelihood modes such as tourism livelihood, give full play to the characteristics of local resources, attract foreign companies to invest and build factories here, set up public welfare posts in the fields of rural public facilities maintenance and cultural heritage protection, and provide employment opportunities for agricultural valleys while developing industries. Bailong Village and Mapai Village set up agricultural music and dance teaching and food production workshops in the original folk-culture scenic spots to strengthen the integration of culture and tourism, avoid homogenization competition, and encourage farmers to design tourism products to form independent brands. The planting industry of Chinese herbal medicines in Liangshui Village has developed strongly, with good output of Chinese herbal medicines, such as Angelica sinensis, red ginseng, and ginseng. Hexi Village is characterized by the picking industry, lotus pond sightseeing industry, and processing of Chinese herbal medicines, with high output and added value, which can create a unique brand. Hexi Village can create a new way of industrial tourism. With visits, Tumen River Pharmaceutical Co., Ltd., can enhance the popularity of enterprises, gradually drive the development of local tourism, and promote the diversified development of farmers’ livelihood strategies. Tingyan Village is weak in the development of tourism resources. Based on actively publicizing the existing tourism resources, we should dig deep into tourism resources and explore the development mode of tourism resources and accumulate experience in developing tourism. We should actively publicize the deep processing of Auricularia auricula and its ancillary products, as well as the black fruit picking project, to help farmers sell agricultural tourism products through an e-commerce platform, provide sufficient jobs for farmers, and enhance the diversity of farmers’ livelihood capital.

6.3.3. Coordinate Ecological Protection and Tourism Development to Improve Farmers’ Livelihood Environment

Local governments should further promote the development of ecological civilization, refine mechanisms for ecological compensation, and enhance the sustainable utilization of tourism resources. The ecological environment plays an important role in farmers’ development. We should strengthen the protection and supervision of the ecological environment, rationally develop and utilize resources, make emergency management measures in time for environmental problems, do a good job in ecological environment publicity and protection, uphold the principle of combining development with protection, and enhance farmers’ awareness of ecological environment protection. External market elements should fully consider the sustainable development path of rural tourism and protect local natural ecological resources and cultural ecological resources, while utilizing tourist resources, and promote the construction of ecological functions and cultural functions of the rural tourism industry [31]. We should accelerate the development of agriculture, animal husbandry, and forestry; carry out large-scale production of products; give full play to the advantages of under-forest products; and build a specialized production base. We should guide farmers to reduce their use of pesticides and fertilizers, build the brand of “green agricultural products”, strengthen sewage treatment, avoid direct discharge of wastewater, enhance farmers’ awareness of environmental protection in tourist destinations, carry out ecological education, strengthen education and publicity on ecological protection, and publicize from various angles such as protection methods and laws and regulations to enhance farmers’ independent protection ability. In both daily operations and development initiatives, local villages should prioritize the enhancement of ecological conservation efforts, particularly within ecologically sensitive and vulnerable regions. Such measures may involve the implementation of carefully planned relocation programs for settlements when necessary. For villages that possess more favorable conditions for development, the focus should shift toward improving the quality of scenic areas and tourist services, conducting renovations in compliance with elevated standards, and further strengthening their attractiveness as tourism destinations.