Study on the Impact of Rural Tourism Construction Projects on Farmers’ Livelihood Capital and Livelihood Options

Abstract

Rural tourism is a new driving force for achieving rural revitalization and development, promoting rural economic prosperity, and serving as a new industrial approach to achieving the transformation and upgrading of farmers’ livelihoods. This paper focuses on Manas County as the research subject and employs farmer questionnaire interviews and participatory rural appraisal methods to categorize villages into four types: “scenic area-dependent” villages, “industry-dependent” villages, “folklore-dependent” villages, and “suburban-dependent” villages. Based on the sustainable livelihood analysis framework proposed by the Department for International Development, we developed a sustainable livelihood framework for farm households in Manas County and analyzed the effects of rural tourism on farm households’ livelihood capital and livelihood strategies as the watershed of the rural tourism takes shape, taking livelihood capital and livelihood strategies as the entry point. The factors influencing the livelihood capital and livelihood options of farm households in the context of rural tourism are analyzed. The results show that: (1) The overall livelihood capital of farmers engaging in rural tourism is significantly greater than that of farmers not participating in rural tourism. Additionally, variations exist in the livelihood capital of different types of rural tourism within villages. (2) Participation in rural tourism significantly influences farmers’ choices of livelihood strategies. Compared with individuals who do not engage in rural tourism, those involved in rural tourism are more likely to opt for self-management livelihood strategies. The proportion of self-managed farmers in villages based around scenic areas is the highest. (3) The effect of rural tourism on farm households in Manas County is generally positive, and various types of villages yield different impact effects. (4) Participation in rural tourism significantly increases the possibility of farmers choosing self-management livelihood strategies, while it reduces the possibility of farmers choosing agriculture-oriented livelihood strategies. In all villages, except those dependent on industry, the decision to engage in rural tourism significantly impacts farmers’ selection of self-management livelihood strategies.

1. Introduction

The United Nations World Tourism Organization (UNWTO) launched the “Best Tourism Villages” selection activity in 2021, aiming to promote rural prosperity and well-being, as well as protecting unique cultural and natural heritage [1]. By protecting rural cultural heritage and the natural environment of the countryside from destruction, efforts are being made to innovate in all aspects to achieve the Sustainable Development Goals (SDGs). It is widely believed that developing rural tourism not only helps promote balanced economic development, but also plays an important role in alleviating rural population loss and promoting rural revitalization [2,3,4].

As China continues to promote industrialization and urbanization, it is also considering how to address the issue of rural decline. The large rural population base and long transition period make it impossible for China to rely solely on transformational forces to resolve the problem of rural decline [5,6,7]. The specific practice in China of promoting transformation while addressing rural issues lies in the dual efforts of poverty alleviation and rural revitalization [8,9]. In the process of modernization in a Chinese style, rural society should share the benefits of development. The No. 1 Central Document of 2023 emphasizes the development of modern rural service industries, such as rural tourism, to promote the high-quality development of rural industries and support the implementation of the rural revitalization strategy [10]. In the context of rural revitalization, the rural tourism industry is increasingly being positioned as an important engine for optimizing the development of production, life, and ecological spaces in rural areas [4,11].

Currently, rural tourism is considered a significant economic activity in rural communities. Many scholars focus on individual farmers to study changes in their livelihood levels and livelihood transformations. On one hand, scholars generally believe that rural tourism can create more job opportunities for farmers [12], leading to better economic benefits [13]. As a result, farmers are able to shift from traditional single livelihood methods, enhancing their livelihood resilience [14]. This ultimately leads to an overall improvement in the rural economy and a reduction in the urban–rural gap [15]. In addition, some scholars believe that the primary beneficiaries of rural tourism are actually tourism enterprises and related organizations. However, for the farmers who participate in these tourism enterprises, their way of earning a living changes, and their livelihood diversification also increases as a result [16].

On the other hand, as rural tourism develops, investment opportunities, market competition, and cognitive differences [17], it can result in resources becoming concentrated among a few individuals. This can enable specific groups to quickly amass significant wealth, thereby worsening income disparities among farmers [18]. The seasonal (cyclical) nature of rural tourism also results in unstable economic income for farmers [19], reducing personal and family investment opportunities and leaving them unable to cope with sudden events or economic risks, further increasing the vulnerability of farmers’ livelihoods [20]. The process of rural tourism development may lead to the destruction of local natural resources [21], with a noticeable negative impact on the ecological environment [22].

Existing research at home and abroad has provided valuable reference significance for this article. However, based on a comprehensive review of the existing literature [13,23,24,25,26], most studies have only analyzed the role of rural tourism in rural household livelihood strategies from the perspective of livelihood capital, with insufficient exploration of the impact pathways of household factors. Existing research primarily categorizes household livelihood strategies based on the ratio of non-farm income to total household income [27,28,29]. However, non-farm income can only indicates household investment in non-farm livelihood activities [30], which makes it challenging to accurately reflect household investment in a specific livelihood activity. Based on the above, this study chose Manas County for its empirical research. Manas County is situated in the central area of the economic belt on the northern slope of the Tianshan Mountains. The study focused on changes in household livelihood capital and illustrated the transformation path of household livelihoods in Manas County. At the same time, households in the same village who did not participate in rural tourism development were selected for comparison to analyze the differences in livelihood capital and livelihood strategies. By utilizing a propensity score matching model, the study investigated the correlation between rural tourism development and the evolution of household livelihood strategies. Ultimately, it revealed the mechanism through which rural tourism affects changes in household livelihoods, providing a scientific reference for designing a rural revitalization path driven by tourism. Compared to existing research, this study makes contributions in the following aspects:

(1) Modifying the sustainable livelihood framework to categorize factors affecting farmers’ livelihoods into external factors and endogenous forces. Studying households in a typical area of rural tourism development in Manas County can provide valuable insights for rural tourism development in the arid northwest region of China and even across rural China.

(2) This study utilizes household income sources as the foundation for categorizing livelihood strategies and employs a multi-index combined approach to classify rural household livelihood strategies into three types: self-operated, labor-led, agriculture-led, and policy-supported. This approach overcomes the limitations of solely relying on the proportion of agricultural income as the standard for classifying livelihood strategies.

(3) Analyzing the differences in livelihood capital and livelihood strategy choices of households from different types of villages, and further exploring the impact of rural tourism on various types of rural tourism development in villages from the different perspectives of types of household and village.

(4) Utilizing a propensity score matching model and probit model, respectively, to investigate the factors influencing the impact of engaging in rural tourism on farmers’ livelihood capital and their selection of livelihood strategy.

2. Theoretical Framework

The issues of agriculture, rural areas, and farmers are fundamental concerns related to the national economy and people’s livelihoods [31]. The livelihood of farmers refers to the way of making a living based on their abilities, assets, and actions [32]. Farmers can only achieve sustainable development in their livelihood when they have the ability to cope with external pressures, maintain or enhance their assets, and minimize damage to natural resources [33].

Based on the concepts of livelihood and sustainable livelihood, scholars most commonly use the Sustainable Livelihoods Framework proposed by the UK Department for International Development [34]. The Sustainable Livelihoods Framework comprises various elements such as fragile environments, policy structures, livelihood assets, livelihood strategies, and livelihood outcomes. These elements balance and interact with each other, collectively impacting the livelihood levels of households.

Starting from a vulnerability perspective, Speranza et al. utilized the concept of resilience to investigate the livelihood recovery of households following significant natural disasters [35]. Shah et al. found that households have low resilience when facing risks and unexpected shocks [36]. Therefore, they developed an indicator system to evaluate the vulnerability of household livelihoods. This system aims to assist households in avoiding livelihood strategies with high vulnerability. Zhao et al. found that under the impact of the outbreak of COVID-19 and strict epidemic prevention policies, farmers faced livelihood shocks higher than those of common diseases, with an overall low level of livelihood resilience [37]. It can be observed that significant health events are not conducive to the sustainable development of household livelihoods.

Starting from the basic livelihood status of farmers, Habib and others reviewed the relevant literature and found that human capital, natural capital, and financial capital are the main determining assets of livelihood diversification strategies [38]. Strengthening livelihood diversification can be achieved by enhancing the acquisition and supply of livelihood assets. Yu and colleagues conducted a study on changes in livelihood capital among communities surrounding Huangshan Park [13]. They discovered that regions with medium to high livelihood capital are predominantly located in the vicinity of Huangshan Park. This suggests that the development of tourism infrastructure can enhance the livelihood assets of farmers to some extent. From the perspective of decision-making and behavior of rural households, Wu et al. found that under policy interventions, rural households are more willing to change their livelihood strategies, and the trend of livelihood transformation is more significant [39]. Rigg et al. found that, although many studies predict that small-scale farmers will exit the historical stage, in most parts of East Asia and Southeast Asia, there has been no shift in farm scale, and small-scale farming economies are still retained [40]. The main reason for this is that small-scale farming economies in East Asia and Southeast Asia are more productive than large-scale farm economies. Countries are increasingly providing agricultural subsidies for small-scale farming. As a result, most small-scale farmers are shifting their focus away from agricultural production and are diversifying their livelihood sources through various channels. At the same time, several scholars have explored the internal and external driving factors of rural livelihood transformation using the Sustainable Livelihoods Framework. These factors include policy interventions [39], environmental influences [41], social development [42], and the economic development levels of the communities [43].

Building on previous research, Figure 1 revises the Sustainable Livelihoods Framework to better illustrate the role of rural tourism in livelihood systems. First, the “vulnerability background” of the original framework is revised in accordance with the rural tourism policy. The scope encompassed by the legal, policy, and institutional aspects within the sustainable livelihood framework is excessively broad [44], and the definition is overly general to evaluate the real impact of government policies. The government plays a fundamental role in people’s livelihoods by providing basic public services [45]. Social security and infrastructure provided by the government form the foundation for individuals to choose their livelihood strategies [46]. Therefore, the analysis framework includes all basic public services provided by the government, such as infrastructure construction, policy promotion, and policy implementation. Second, farmers are a particularly vulnerable population to environmental stresses from the outside world [47]. In this context, the direct impact on farmers is the change in livelihood capital, such as: the construction of tourism infrastructure occupying agricultural land resources; the involvement of external tourists broadening channels through which farmers can obtain information; village committees organizing farmers to participate in tourism-related technical training; farmers’ income increasing, enabling them to purchase more production and living tools, etc. At the same time, the different levels and combinations of these livelihood capitals will directly affect farmers or other individuals in choosing their livelihood strategies [27]. In addition, rural tourism is based on the authenticity and uniqueness of rural culture. The cultural heritage of villages naturally possesses attractive qualities [48]. Therefore, farmers’ identification with and perception and inheritance of local culture are closely related to their livelihood development [49,50]. Furthermore, farmers and landowners both play a role in implementing common agricultural goals [51]. They will comprehensively consider the future economic environment and their own resources, and take proactive measures to adapt to current environmental changes through their own volition [52]. The integrated cultural characteristics of rural households and the internal dynamics of rural household decision-making [53,54] also influence the choice of livelihood strategies. Finally, under the background of rural tourism, the issue of sustainable livelihoods for rural households is transformed into a process of achieving sustainable livelihood goals through adjusting livelihood assets and different livelihood strategies, influenced by both external environmental factors and individual factors of rural households.

In general, this framework mainly includes external environmental factors, livelihood capital, household internal dynamics, and livelihood outputs. Based on the framework, the revised model places livelihood capital as the central variable, emphasizes individual labor factors, examines government roles, and considers environmental policy factors and the direct influence of households on livelihood strategies. This article mainly addresses the following three questions: (1) What changes will occur in the selection of livelihood capital and livelihood strategies by households under the constraints of the external environment and internal dynamics in the context of rural tourism? (2) How does participation in rural tourism activities affect the selection of livelihood strategies by households? (3) What are the differences in livelihood capital and livelihood strategy selection by households under different rural tourism development models? (4) After being influenced by rural tourism, do the adjustments made by households to their livelihood capital and livelihood strategies align with sustainable livelihood development?

3. Materials and Methods

3.1. Overview of the Study Area

Manas County is situated in the central area of the economic belt on the northern slope of the Tianshan Mountains (Figure 2), within the one-hour economic radius of Urumqi City. It is an important part of the “Wuchangshi” urban agglomeration [55]. The total area of the county is 11,000 square kilometers, and the total population is expected to reach 230,000 by 2023. Manas County boasts beautiful ecology, green livability, abundant natural resources, and cultural significance. In 2019, it was designated as a pilot county for the construction of a national rural governance system and, in 2021, it was designated by the autonomous region as a leading county for the demonstration of rural revitalization.

In recent years, Manas County has seized opportunities for constructing the core area of the Silk Road Economic Belt in Xinjiang, rural revitalization, and the tourism development strategy in Xinjiang. Rural tourism, modern agriculture with distinctive characteristics, and the revitalization of rural culture have been identified as key areas for industrial development. New forms of rural tourism have been developed, with towns like Lanzhou Bay, Guangdongdi, Qingshuihe Kazakh Township, and Taxihe Kazakh Township included in government work plans. By offering diverse new forms of rural tourism experiences such as characteristic homestays, farmhouses, and agricultural experiences, it attracts tourists from all over to visit and sightsee, empowering rural revitalization. By 2023, the county had one autonomous region-level key village for rural tourism, four townships at the prefecture level designated as tourism demonstration towns, and four villages at the prefecture level designated as tourism demonstration villages. Additionally, there will be three prefecture-level research and study tourism practice bases, two prefecture-level agricultural sightseeing and leisure tourism demonstration sites, and a total of over 130 rural homestays and guesthouses. These initiatives are expected to create employment opportunities for over 4000 people [56]. Rural tourism, as a novel approach for farmers to diversify their livelihoods, enables farmers in Manas County to leverage the distinctive cultural resources and local customs of ethnic minorities for the development of rural tourism. This helps in broadening livelihood opportunities and fostering sustainable livelihood development within the framework of the “Cultural Enrichment of Xinjiang” project and the “Tourism Development of Xinjiang” strategy.

3.2. Data Source and Sample Description

The “Manas County Comprehensive Tourism Development Plan (2021–2030)” mentions promoting the ‘one village, one landscape’ initiative, selecting a group of key rural tourism development villages, and implementing differentiated support policies. In recent years, more than half of the townships in Manas County have incorporated rural tourism development into their annual plans. According to the statistics on village policies aimed at developing rural tourism, it has been observed that the majority of rural tourism villages in Manas County are concentrated in the central plains and southern mountainous regions (Figure 3). These villages are primarily affiliated with towns and townships including Lanzhouwan, Gudongdi, the Qingshuihe Kazakh Ethnic Township, and Manas. By obtaining points of interest (POI) data of rural homestays, guesthouses, hotels, and other tourism businesses in Manas County, and conducting density analysis, it is observed that high-density areas are concentrated in the central and southern parts of Manas County (Figure 3). Based on the government planning of Manas County, the selection criteria of “the most beautiful Countryside”, and news reports, villages with relatively successful rural tourism construction and remarkable results were selected as the research object.

Manas County is mobilizing rural characteristic resources to develop the full industry chain of rural tourism. Various new formats, such as rural tourism picking gardens, leisure farms, rural campsites, and rural homestays, are emerging continuously, creating numerous distinctive models for rural tourism development. Based on core supporting elements, rural tourism in Manas County can be divided into four types: “scenic area-based”, “industry-based”, “suburban-based”, and “folk-based”. Based on these four types, 18 sample villages were selected to explore the differences in livelihood capital and livelihood choices of villages with different development models (

Table 1. Basis for village division and selection of sample villages.

Type of Village	Classification Basis	Sample Village
Scenic area-based village	The geographical distance and transportation connections between villages and surrounding scenic areas are important criteria for division. Tourism-dependent rural villages are typically situated near scenic areas with convenient transportation, facilitating easy travel for tourists between the scenic spots and villages, creating an ideal tourism route.	Hongkeng Village, Tuanzhuangzi Village, Xiaohaizi Village, Bajiahu Village, Yangjiadao Village, Heiliangwan Village
Industry-based village	The primary criteria for classifying rural tourism villages as “industry based” include the scale of dominant industries in the village, as well as the proportion and influence of these industries on the local economy. These dominant industries may include agriculture, forestry, fisheries, handicrafts, etc., which not only provide economic support for the village but also serve as important attractions for rural tourism.	Xiliangzhou household village, Min Ma ecological village, East branch canal village, Shangzhuangzi village, Fritillaria house village
Suburban-based village	The geographical location of a village, especially its distance from major cities or central cities, is a key factor in defining “suburban based” rural tourism villages. These villages are typically located on the outskirts of cities or in urban–rural transition zones, making it convenient for urban residents to visit on weekends or holidays to experience rural charm and enjoy leisure time.	Zhuanglanghu village, Shangergong Village, Shangsangong Village, Tougong Village
Folk-based village	The folk cultural characteristics found in villages are the fundamental basis for classifying rural tourism villages as “folk based”. These characteristics may include unique ethnic customs, traditional festive activities, handicraft production, folklore, and legends, etc. The richness and uniqueness of folk culture can attract tourists and provide them with opportunities to deeply experience rural culture.	Hongshawan Village, Shihuiyaozi Village, Huangtaizi village

).

This article utilizes the participatory rural appraisal (PRA) and questionnaire survey methods to gather first-hand data. In order to ensure the effectiveness and scientific nature of the questionnaire, the research team conducted a pre-survey on the current situation of rural tourism in Manas County in October 2022. They identified issues in questionnaire design and distribution processes and made revisions based on expert opinions. The formal survey utilized a combination of random sampling and stratified sampling methods. Field research was conducted twice, in November 2022 and May 2023, respectively. The research targeted 18 sample villages in the Manas River Basin during both peak and off-peak tourism seasons. To ensure the accuracy of the survey and the reliability of the data, face-to-face interviews lasted an average of over 30 min per person. The survey subjects include farmers, township government workers, scenic area staff, and local tourism enterprise operators. The questionnaire mainly includes the endogenous motivation of farmers, including their willingness to participate in rural tourism and the cultural characteristics of farmers. A total of 1080 questionnaires were distributed, with 1002 valid questionnaires and 78 invalid questionnaires returned, resulting in a questionnaire recovery rate of 92.78%. Although the number of interviewed farmers is relatively small compared to statistical data, the sample can reflect the basic situation of farmers in the research area and has a certain level of representativeness (Figure 4).

3.3. Research Method

3.3.1. Estimation of Farmers’ Livelihood Capital

Research on quantifying the livelihood capital of farmers can provide insight into the basic living conditions and economic foundation of farmers’ households. Referring to existing research both domestically and internationally [27,57,58], a livelihood capital index system suitable for the Manas River Basin area was designed (

Table 2. Livelihood Capital Index System for Farmers.

Capital	Index	Quantitative Method	Weight
Natural Capital	Area of cultivated land (forest land)	Average cultivated land area per rural household	0.34
	Cultivated land quality	Land quality rating: Excellent = 1, Good = 0.75, Fair = 0.5, Poor = 0.25, Very poor = 0	0.37
	Farmland irrigation conditions	Irrigation condition rating: Excellent = 1, Good = 0.75, Fair = 0.5, Poor = 0.25, Very poor = 0	0.29
Human Capital	Population size	Number of people in a farming household	0.26
	Labor capacity	Average age of labor force: 14~17 years old = 0.5, 18~59 years old = 1, 60~70 years old = 0.5	0.07
	Standard of culture	Highest educational attainment of labor force: bachelor’s degree or above = 1.00, high school or college = 0.75, junior high school or technical school = 0.50, primary school = 0.25, illiterate = 0	0.42
	Overall health status	Health rating assignment: Very healthy = 1, Good = 0.75, Fair = 0.5, Poor = 0.25, with poor health condition = 0	0.25
Physical Capital	Housing area	Household housing area for farmers	0.06
	House type	Property type assignment: Concrete = 1.00, Brick = 0.75, Earth = 0.50, Grass = 0.25	0.21
	Number of large agricultural machinery	Quantity of various agricultural machinery	0.73
Financial Capital	Income level	Per capita annual income	0.21
	Income diversity	Number of sources of income for family farmers	0.48
	Credit opportunity	Farmers can access credit opportunities: very easy to access = 1, relatively easy to access = 0.75, average = 0.5, not easy to access = 0.25, very difficult to access = 0	0.32
Social Capital	Trust between neighbors	Level of trust evaluation: high trust = 1, trust = 0.75, neutral = 0.5, distrust = 0.25, high distrust = 0	0.17
	Skills training opportunities	Opportunities for government- or community-provided training: Many opportunities = 1, quite a few opportunities = 0.75, average = 0.5, not many opportunities = 0.25, no opportunities = 0	0.10
	Social participation rate	Degree of participation of social organizations: frequent participation = 1.00, quite frequent participation = 0.75, occasional participation = 0.5, very rare participation = 0.25, no participation = 0	0.22
	Internet media usage	Usage of e-commerce platforms such as Taobao, social apps like WeChat, and short video platforms like Douyin: very frequent use = 1, frequent use = 0.75, average use = 0.5, seldom use = 0.25, do not use = 0	0.22
	Social welfare	Number of social subsidies enjoyed by farmers (such as pension subsidies, medical subsidies, etc.)	0.29

). The system includes five categories: natural capital, human capital, material capital, financial capital, and social capital. Under these five categories, a total of 18 indicators were designed to quantitatively evaluate various types of capital. The technique for order of preference by similarity to ideal solution (TOPSIS) algorithm is a widely used comprehensive evaluation method that is suitable for selecting solutions when there are multiple indicators. Data collected based on the index system are organized, non-dimensionalized, and weighted using the entropy weight method. Subsequently, TOPSIS analysis is conducted to score each indicator, obtain various livelihood capital values, calculate the average, and ultimately determine the final livelihood capital value.

Formula for calculating livelihood capital:

$$LC=\frac{N+H+P+F+S}{5}$$

(1)

In the formula: LC represents the livelihood capital value of farmers, N represents natural capital, H represents human capital, P represents physical capital, F represents financial capital, and S represents social capital.

3.3.2. Partition of Farmer Household Types

The criteria for categorizing livelihood strategies should be grounded in objective facts. Factors that influence the choice of livelihood strategies, such as household assets, willingness to change strategies, and government policy guidance, should not be the sole criteria for categorizing livelihood strategies. Therefore, this article refers to existing research and categorizes livelihood strategies based on the overall income structure of households and their occupational types. By utilizing data from eighteen sample villages before and after the implementation of rural tourism, the K-means clustering analysis method is employed to categorize household livelihood strategies. Referring to the existing literature [59,60], this article examines the proportion of various income sources in rural households. These sources include agricultural income (from sales of agricultural products), operating income (from activities like agricultural processing, rural tourism, sales of specialty handicrafts, and other agricultural-related businesses), wage income (from non-agricultural work in cities or nearby areas), property income (from rental and appreciation of assets like land, property, and equipment owned by households), and transfer income (from government subsidies, relief funds, pensions, etc.), as a percentage of total household income. The proportion of people in the household engaged in agricultural production activities, non-agricultural employment activities, business and entrepreneurial activities, and those enjoying social security benefits as a percentage of the total household population is used as input indicators for cluster analysis. Based on the final cluster center values of the indicators (

Table 3. The final clustering center value.

Index	Type I: Self-Managed Type	Type II: Worker-Oriented Type	Type III: Farming-Oriented Type	Type VI: Policy Support Type
Proportion of agricultural income	0.30779	0.24320	1.13787	0.17367
Proportion of operating income	0.89515	0.21979	0.45208	0.66589
Proportion of wage income	0.51515	1.29156	0.54909	0.03248
Proportion of property income	0.10473	0.14616	0.05651	0.33769
Proportion of transfer income	0.26020	0.41585	0.06427	1.06843
Proportion of people involved in agricultural production activities	0.59306	0.03440	1.06530	0.21440
Proportion of non-agricultural employment activities	0.54926	1.29281	0.46373	0.77368
Proportion of the number of business and entrepreneurial activities	1.37324	0.24667	0.47837	0.20558
Proportion of people receiving social security benefits	0.77432	0.38934	0.45374	0.88742

), and combined with regional characteristics and the actual situation of households, households are classified into self-employed, labor-dominated, agriculture-dominated, and policy-supported types.

The self-managed type emphasizes that individual farmers or families have business production activities, and business income is the main source of income for families. The worker-oriented type is a way of living with migrant workers as the main source of income, and the number of non-agricultural employment activities in this kind of household is relatively high. The economic income of agriculture-oriented farmers mainly comes from agricultural production, such as planting crops, raising livestock and poultry, etc. The number of farmers participating in agricultural production activities is large. Policy-supported farmers are generally limited to specific groups or regions, such as poor farmers and ethnic areas, and such farmers receive more financial subsidies and greater tax incentives.

3.3.3. Propensity Score Matching Method

The participation of farmers in rural tourism activities is spontaneous, determined by the farmers’ willingness, and involves a self-selection bias [59,61]. Furthermore, there are significant differences in asset reserves and distribution among farmers, leading to selective bias in the impact of rural tourism participation on farmers’ livelihood capital. In economic empirical studies, it is not possible to accurately obtain data on farmers who have not participated in rural tourism to compare with those who have, as this would lead to endogeneity. Therefore, this study utilizes the Propensity Score Matching (PSM) method to investigate the influence of farmers’ engagement in rural tourism activities on their livelihood capital. This method assumes scenarios before and after farmers participate in rural tourism, testing whether the livelihood capital of farmers who did not participate in rural tourism is consistent with those who did. By matching or constructing a non-participating farmer for each participating farmer based on characteristics similar to those who did not participate, except for rural tourism participation, this method addresses the self-selection bias issue. By creating a counterfactual control group similar to the treatment group, approximate randomization of the treatment variable is achieved, ensuring that farmers’ participation in rural tourism is close to random. By conducting two different experiments within the same farmer household, we can determine the net effect of participating in rural tourism based on the results of the two samples. The difference in outcome variables between the experiments represents the net effect of rural tourism participation (

Table 4. Definition and descriptive statistics of main variables.

Variable Name	Variable Definition and Description	All Samples	Samples Participating in Rural Tourism	Samples That Did Not Participate in Rural Tourism	Standard Deviation
Participation in rural tourism behavior	Whether farmers are involved in rural tourism (e.g., whether they operate farmhouses; whether to participate in scenic tourism services; whether they run leisure orchards, leisure farms, etc.), participation is 1, non-participation is 0.	0.61	1	0	0.48
External environment	The degree of influence of the external environment: based on the factor analysis method, the score of this element is calculated from the aspects of infrastructure, market environment, whether the village is close to the scenic spot, the degree of government attention, policy implementation, and policy publicity (the kom value is 0.767).	−7.22 × 10−7	0.68	−1.07	1.00
Endogenous power	Willingness to participate in rural tourism: score for farmers willingness to participate in rural tourism: 1–10 points	6.11	7.49	3.79	2.57
	Cultural characteristics score: based on the factor analysis method, the score of this element is calculated from five aspects: village age, number of traditional handicrafts, brand awareness, cultural perception, and cultural identity (the kom value is 0.731).	−5.56 × 10−7	0.38	−0.60	1.00

).

This article will match the treatment group with the control group, explore the impact of participating in rural tourism on farmers’ livelihood capital under the same external conditions, and analyze the following steps.

Firstly, in order to obtain the conditional probability fitting values of farmers participating in rural areas, this paper constructs a logit model for estimation. The expression for the propensity score is as follows:

$$p\left(X_i\right)=E\left(Join=0|X_i\right)=P_r(Join=1|X_i)$$

(2)

In the formula, $p\left(X_i\right)$ represents the propensity score; $Join$ represents participation in rural activities, where when i = 1, it signifies farmers participating in rural tourism, and when i = 0, it signifies farmers not participating in rural tourism; and $X_i$ represents a series of observable control variables.

Second, match the treatment group with the control group. In order to verify the robustness of the model matching results, this study utilizes various matching methods, including K-nearest neighbor matching (1:4), radius matching, and kernel matching. Additionally, common support domain tests and balance tests were conducted for each method. The common support domain test aims to differentiate whether there are any duplicates in the households of the treatment group and the control group by comparing the value ranges of propensity score intervals. The balance test is primarily conducted to ensure the effectiveness of the matching.

After comparing and analyzing the before and after states of the matching (Figure 5 and Figure 6), the standardized deviations of the control group and treatment group were significantly reduced post-matching, with absolute values remaining below 10%. There were no significant differences between the covariates of the two groups post-matching, demonstrating a good balance effect. In addition, an analysis of the common range of propensity scores for the samples revealed that the majority of the propensity score observations for both the control and treatment groups fell within the common range. This further proves the effectiveness of the matching.

Third, calculate the average treatment effect on the treated of farmers participating in rural tourism. Measure the difference in livelihood capital between the treatment group and the control group, which reflects the variance between the actual outcomes of farmers engaging in rural tourism and their counterfactual results. This study aims to assess the impact of engaging in rural tourism on farmers’ livelihood capital. The expression for average treatment effect average turnaround time (ATT) is as follows:

$$ATT=E\left(Y_{1i}|R_i=1\right)-E\left(Y_{0i}|R_i=1\right)=E(Y_{1i}-Y_{0i}|R_i=1)$$

(3)

In the formula, ATT represents the average treatment effect of farmers participating in rural tourism. $Y_{1i}$ stands for the livelihood capital of the treatment group, while $Y_{0i}$ represents the livelihood capital of the control group. $E\left(Y_{1i}|R_i=1\right)$ signifies the average treatment effect of participating in rural tourism, which is directly observable. However, $E\left(Y_{0i}|R_i=1\right)$, representing the counterfactual result, cannot be directly observed and can be constructed using the propensity score matching method to generate corresponding alternative indicators.

3.3.4. Probit Model

In this article, farmers’ choices of various livelihood strategies are all binary response variables, so the probit model is used for estimation. The specific model is as follows:

$$P_i\left(y_i=1|X_i\right)=\alpha +{\beta }_1{RT}_i+{\beta }_1{x}_1+{\beta }_2{x}_2+···+{\beta }_n{x}_n+{\mu }_i$$

(4)

where $P_i$ is the probability of farmers choosing the i-th livelihood strategy; ${RT}_i$ represents farmers’ participation in rural tourism activities; $x_1$, $x_2$, …, $x_n$ are explanatory variables, including the external environment, farmers’ endogenous forces, and livelihood capital; $\alpha$ is the constant term, ${\beta }_1$, ${\beta }_2$, …, ${\beta }_n$ are coefficients, and μ_i is the random error term.

4. Results

4.1. Analysis of the Heterogeneity of Farmers’ Livelihood Capital

The difference in livelihood capital between rural households participating in rural tourism and those not participating is significant (Figure 7). Longitudinal comparison reveals that the overall livelihood capital of rural households engaged in rural tourism in each sample village is significantly higher than that of those not participating. This further confirms the conclusion that rural tourism boosts the enhancement of household livelihood capital. However, it is worth noting that the difference between participating and non-participating households is smallest in suburban based villages (0.0288). These villages have a unique geographical location and convenient transportation conditions, enabling both participating and non-participating households to benefit from the infrastructure improvements and regional economic development facilitated by rural tourism. This helps to narrow the gap in livelihood capital between the two types of households. In terms of horizontal comparison of different types of villages, the total livelihood capital is higher in industry-dependent villages (0.4804) than in suburban-based villages (0.4794) > scenic area-based villages (0.4685) > folk-based villages (0.4680). Industry-based villages benefit from industrial agglomeration, which brings economies of scale and cost advantages. Suburban-dependent households benefit from the expansion of urban infrastructure and economic development, which forms the basis for enhancing household livelihoods. Scenic area-based villages may have households overly reliant on income from tourism. Fluctuations or off-seasons in tourism can greatly affect villagers’ income, leading to an increase in the wealth gap between households. Folk-based villages face serious homogenization issues, with incomplete development of local cultural characteristics and weak economic driving effects. This results in an overall unsatisfactory livelihood situation for households in these two types of villages.

There are significant differences in livelihood capital among the various types of farmers (Figure 8). In terms of total livelihood assets, the self-managed type (0.5001) > worker-oriented type (0.4835) > farming-oriented type (0.4634) > policy support type (0.4419). The farming-oriented type and policy support type are significantly lower than the overall average (0.4722). In terms of different types of livelihood capital, the self-managed type has the highest human capital (0.5028), while the farming-oriented type has the lowest (0.4391). Conversely, in terms of natural capital, the farming-oriented type (0.5068) is much higher than the self-employed type (0.4587). Through research, it was found that the education level of farmers primarily engaged in agriculture is generally low, with a majority having only completed primary school education or below. For these farmers, land serves as the fundamental means of survival. They generally own larger areas of land of better quality. Self-employed farmers tend to have larger families and primarily engage in high-profit production activities. They may also utilize some of their land for non-agricultural purposes or lease it to other operators, which can diminish their natural capital. The lowest value of financial capital belongs to policy support-type farmers (0.4572). Some of the income sources for policy-supported farmers depend on farmer subsidies, social welfare, social relief, etc., and direct assistance from policies to farmers’ financial reserves may be limited. Financial institutions also do not provide adequate credit support to these farmers. The smallest gap between material capitals (S2 = 0.0008) is mainly attributed to the government’s planning and restrictions on rural residential land area, building height, building style, etc., aimed at ensuring the standardization and uniformity of rural construction. There is a high level of consistency among households in terms of house size and type. The social capital of self-employed farmers is the highest (0.5087), followed by labor-dominated farmers (0.4835). These two types of farmers have larger social networks, more sources of information, and opportunities for resource sharing. They are also relatively proactive in participating in skills training. The farming-oriented-type (0.4323) and policy support-type farmers (0.4136) have the lowest social capital. These two types of farmers exhibit smaller production and living scope, relatively poor timeliness in obtaining information, and some people may have biases or misunderstandings towards policy-supported farmers, which can affect their social reputation. In terms of livelihood capital allocation, labor-dominated capital allocation is relatively even, resembling a “regular hexagon” in the image. The farming-oriented type’s livelihood capital allocation is the most uneven, with all types of capital except natural capital generally below average levels. In contrast, the self-managed farmers’ main weakness is a relatively low level of natural capital. Policy support farmers have a good allocation of financial capital but their total livelihood amount is lower, leading to the lowest level of livelihood.

4.2. Farmers’ Livelihood Strategy Choice against the Background of Rural Tourism

The proportion of farmers who engage in rural tourism and opt for self-management livelihoods is significantly higher than that of those who do not participate in rural tourism. The self-management strategy, with a participation rate of 61.82%, has become the primary livelihood strategy for farmers engaged in rural tourism, demonstrating a higher level of livelihood diversification. Among the farmers participating in rural tourism, the proportion of those choosing to work as their main source of income (23.64%) is higher than those relying on farming as their main source (7.27%) (Figure 9). In rural tourism development villages, farmers tend to engage in livelihood activities such as working off-farm or starting their own businesses. On the one hand, farmers, especially young and middle-aged farmers, are aware that agriculture is a vulnerable industry heavily reliant on natural conditions. Relying solely on agricultural income is insufficient to deal with natural disasters, epidemics, and other adverse factors. Therefore, they take advantage of the convenience of rural tourism villages located near cities or major transportation arteries to seek more job opportunities in towns. On the other hand, higher levels of education, social networks, and economic foundations enable individuals to invest in profitable independent entrepreneurship or management. Furthermore, the development of scenic and suburban areas offers farmers the opportunity to participate in commercial activities. Villages such as Bajiahu Village, Xiaohaizi Village, and Hongkeng Village, which rely on scenic areas, are primarily situated in areas with favorable living and production conditions conducive to activities like agritourism, agricultural product processing, and transportation. Farmers in industry-dependent villages like Xiliangzhouhu village have achieved economies of scale and increased economic benefits by joining cooperatives, such as the Golden Countryside Agricultural Professional Cooperative, to reduce production costs. The primary livelihood strategy for rural households that do not engage in rural tourism is wage labor-oriented (45.71%), with the majority of their livelihood structure characterized by a semi-agricultural and semi-labor model. The second livelihood choice for rural households not participating in rural tourism is farming-oriented (25.71%) (Figure 9). Research has found that factors such as natural capital, material capital, and household endogenous dynamics promote the selection of a farming-oriented livelihood strategy by households. Since the implementation of the forest vegetation protection and restoration project on the northern slope of the Tianshan Mountains in Manas County, a green ecological barrier has been established, consisting of “valley governance + afforestation on barren slopes.” The quality of farmland in villages like Hongshawan Village and Hongkeng Village has significantly improved. The abundant natural resources make them more willing to engage in farming activities. Influenced by a strong attachment to their homeland, households firmly believe that land provides stable income and security. With the emergence of efficient agricultural machinery and branded agriculture, the farming costs for households have decreased. Farmers who engage in rural tourism have the lowest proportion of livelihood models supported by policies. The farmers who choose this type of livelihood model are mainly elderly individuals with lower levels of education, lack of skills, and basic health conditions, resulting in a relatively small overall population base.

Different types of villages exhibit heterogeneity in the selection of livelihood strategies by farmers. In villages that depend on scenic areas and folk customs, self-employed farmers make up 53.89% and 50.56% of the population, respectively (Figure 10). However, it is worth noting that in folk-based-type villages, the proportion of farmers whose main occupation is agriculture (29.44%) is higher than those engaged in non-agricultural work (17.78%). This is mainly because villages that rely on folk customs are located in areas with ethnic minorities, and their lower level of education limits their career choices and development abilities. In industry-based-type villages, the majority of farmers choose agriculture as their main source of livelihood, accounting for 40.56%. According to surveys, these villages primarily focus on industries related to agriculture, developing specialty agricultural products such as organic vegetables and fruits, and promoting rural tourism through activities like agricultural sightseeing and farm experiences. High-yield cash crops already fulfill the production and living requirements of farmers, making it unnecessary to invest time and energy in activities such as agritainment and folk customs. In villages that depend on suburban areas, the most common livelihood strategy for farmers is non-agricultural work, accounting for 42.78%. Suburban villages are often situated near towns, benefiting from prime geographical locations that facilitate access to urban information, resources, and markets. Compared to traditional agricultural production, non-agricultural work often offers more stable and substantial income, which attracts more farmers to choose this livelihood strategy. The proportion of farmers choosing policy support livelihood models is the lowest among all types of villages.

4.3. The Impact of Rural Tourism on Farmers’ Livelihood Capital

Rural tourism has a positive effect on the livelihood capital of farmers in Manas County. The total livelihood capital of the treatment group increased by an average of 0.0612 compared to the control group. Different matching methods all indicate that the increase in livelihood capital for farmers participating in rural tourism is more significant compared to those who did not participate, and all results are statistically significant at the 5% level (

Table 5. Effects of different types of villages participating in rural tourism on total livelihood capital.

Rural Tourism Type	Matching Method	Treatment Group	Control Group	ATT	Standard Error	T-Test
Scenic area-based village	k-nearest neighbor matching (1:4)	0.50	0.43	0.07	0.09	4.58 **
	Radius matching (ε = 0.01)	0.49	0.43	0.06	0.09	4.67 **
	Kernel matching	0.49	0.43	0.06	0.09	5.23 **
Industry-based village	k-nearest neighbor matching (1:4)	0.50	0.44	0.06	0.12	3.87 **
	Radius matching (ε = 0.01)	0.50	0.44	0.06	0.12	4.39 **
	Kernel matching	0.50	0.45	0.05	0.12	3.04 **
Folk-based village	k-nearest neighbor matching (1:4)	0.48	0.42	0.06	0.10	4.56 **
	Radius matching (ε = 0.01)	0.48	0.43	0.05	0.10	5.28 **
	Kernel matching	0.48	0.43	0.05	0.10	5.82 **
Suburban-based village	k-nearest neighbor matching (1:4)	0.49	0.46	0.03	0.17	3.89 **
	Radius matching (ε = 0.01)	0.49	0.45	0.04	0.17	4.05 **
	Kernel matching	0.49	0.46	0.03	0.17	5.87 **

Note: ** are significant at the 5% levels, respectively.

). Specifically, the results of nearest-neighbor matching (1:4) show that the livelihood capital of the treatment group increased by 0.0607 compared to the control group. Radius matching (ε = 0.01) and kernel matching further validate the effectiveness of these results. It is evident that rural tourism significantly improves the livelihood capital of farmers in Manas County. The impact of rural tourism participation on livelihood capital varies among different types of villages. When considering the three matching methods collectively, the results are all statistically significant at the 5% level. Among them, the results of nearest neighbor matching (1:4) show that the most significant improvement is seen in scenic area-based groups, with an increase of 0.0708, followed by industry-based (0.0631), folk-based (0.0607), and finally suburban-based (0.0343) groups.

From the perspective of different livelihood capitals, there are still variations in the changes between the treatment group and the control group (

Table 6. The effect of participation in rural tourism on various livelihood capitals.

Index	Matching Method	Treatment Group	Control Group	ATT	Standard Error	T-Test
Livelihood Capital	k-nearest neighbors matching (1:4)	0.50	0.44	0.06	0.08	4.58 **
	Radius matching (ε = 0.01)	0.51	0.44	0.07	0.08	3.41 **
	Kernel matching	0.50	0.43	0.07	0.08	4.44 **
Human Capital	k-nearest neighbors matching (1:4)	0.49	0.45	0.04	0.18	2.56 **
	Radius matching (ε = 0.01)	0.48	0.43	0.05	0.18	3.75 **
	Kernel matching	0.48	0.43	0.05	0.18	3.29 **
Financial Capital	k-nearest neighbors matching (1:4)	0.56	0.45	0.11	0.13	3.88 **
	Radius matching (ε = 0.01)	0.55	0.45	0.10	0.13	2.14 **
	Kernel matching	0.55	0.45	0.10	0.13	2.28 **
Natural Capital	k-nearest neighbors matching (1:4)	0.44	0.51	−0.07	0.17	4.76 **
	Radius matching (ε = 0.01)	0.43	0.51	−0.08	0.16	3.86 **
	Kernel matching	0.43	0.51	−0.08	0.16	3.07 **
Physical Capital	k-nearest neighbors matching (1:4)	0.48	0.41	0.07	0.18	1.76 **
	Radius matching (ε = 0.01)	0.48	0.40	0.08	0.18	2.86 **
	Kernel matching	0.48	0.40	0.08	0.18	3.07 **
Social Capital	k-nearest neighbors matching (1:4)	0.50	0.42	0.08	0.11	1.26 *
	Radius matching (ε = 0.01)	0.50	0.41	0.09	0.10	1.78 **
	Kernel matching	0.50	0.41	0.09	0.10	1.05 **

Note: ** and * are significant at the 5% and 10% levels, respectively.

). Apart from physical capital, the total amount of livelihood capital, human capital, financial capital, and social capital all show significant results at the 5% level. Among them, the nearest-neighbor matching (1:4) method resulted in a 0.04 increase in human capital, with the smallest increase (8.16%), a 0.07 increase in physical capital, and the largest increase (19.64%) in financial capital, with a 0.11 increase. Natural capital is the only capital that decreased, by 0.06. The results of radius matching (ε = 0.01) and kernel matching further validate the effectiveness of the aforementioned findings.

4.4. The Impact of Rural Tourism on Farmers’ Livelihood Strategy Selection

The impact of rural tourism on the livelihood strategy choices of various types of rural households varies (

Table 7. Model estimation results of rural tourism affecting farmers’ livelihood strategy selection.

Village Types	Variable	Self-Managed Type		Worker-Oriented Type		Farming-Oriented Type		Policy Support Type
		Regression Coefficient	Standard Error	Regression Coefficient	Standard Error	Regression Coefficient	Standard Error	Regression Coefficient	Standard Error
Scenic area-based village	Whether to participate in rural tourism	0.207 ***	0.014	0.014	0.558	−0.633 **	0.768	−0.554	1.042
	Human Capital	0.024	0.138	0.215	0.035	−2.614	0.092	0.447	0.225
	Financial Capital	0.233 ***	0.027	0.006	0.051	−0.227	0.558	−0.329	0.289
	Natural Capital	0.246	0.161	−0.198 **	0.016	0.093 **	0.073	−1.085	0.419
	Physical Capital	0.097	0.065	0.008	0.058	0.128	0.075	0.134	0.092
	Social Capital	0.036	0.013	0.150	0.078	0.106	0.165	−1.35	0.074
	External Environment Factors	0.169	0.044	0.112	0.078	0.726	1.026	0.114	0.384
	Endogenous Power	0.004	0.091	−1.239	0.069	0.098	0.177	0.17	0.183
Folk based-village	Whether to participate in rural tourism	0.168 ***	0.011	0.003	0.347	−0.384 **	0.632	−0.078	1.406
	Human Capital	0.027	0.023	0.045	0.021	−0.268	0.079	−0.687	0.047
	Financial Capital	0.089	0.005	0.267	0.108	−0.073	0.547	−0.279 **	0.052
	Natural Capital	0.056	0.037	−0.168 **	0.079	0.178 **	0.445	0.458	0.406
	Physical Capital	0.278	0.027	0.489	0.142	0.228	0.037	0.003	0.234
	Social Capital	0.045	0.104	0.241	0.579	0.078	0.009	0.108	0.342
	External Environment Factors	0.27	0.078	0.076	0.074	0.064	0.189	0.087	0.158
	Endogenous Power	0.087	0.033	0.267	0.009	0.175 *	0.045	0.034	0.278
Industry-based village	Whether to participate in rural tourism	0.007	0.147	0.037	0.217	−0.024	0.423	−0.038	1.456
	Human Capital	0.268	0.006	0.174	0.058	0.038	0.038	−0.021	0.537
	Financial Capital	0.176 *	0.183	0.032	0.063	0.064	0.067	−0.137	0.263
	Natural Capital	0.037	0.488	−0.007 *	0.074	0.379 **	0.406	0.037	0.458
	Physical Capital	0.049	0.263	0.003	0.106	0.137	0.034	−0.079	0.032
	Social Capital	0.137	0.037	0.032	0.235	0.358	0.052	0.431	0.245
	External Environment Factors	0.264	0.258	0.043	0.067	0.134	0.008	0.258	0.369
	Endogenous Power	0.178	0.307	−0.267	0.433	0.459 **	0.375	0.042	0.277
Suburban-based village	Whether to participate in rural tourism	0.232 **	0.032	0.056	0.233	−0.134 *	0.778	−0.027	0.268
	Human Capital	0.168	0.174	0.278 *	0.007	0.036	0.505	0.156	0.034
	Financial Capital	0.026	0.259	0.422	0.221	−0.264	0.024	−0.134 **	0.125
	Natural Capital	−0.057	0.433	−0.423	0.034	0.334 ***	0.156	0.239	0.263
	Physical Capital	0.063	0.028	0.037	0.058	0.007	0.034	0.056	0.047
	Social Capital	0.278	0.056	0.148 **	0.179	0.089	0.158	0.032	0.058
	External Environment Factors	0.057	0.144	0.063	0.433	0.032	0.103	0.113	0.026
	Endogenous Power	0.002	0.258	0.078	0.038	0.268	0.042	0.048	0.573

Note: ***, ** and * are significant at 1%, 5% and 10% levels, respectively.

). Participation in rural tourism increases the likelihood of households in scenic area-based, folk-based, and suburban-based villages opting for self-management livelihood strategies by 20.7%, 16.8%, and 23.2%, respectively. Simultaneously, it decreases the likelihood of choosing farming-oriented livelihood strategies by 63.3%, 38.4%, and 13.4%, respectively. The main factors influencing households in industry-based villages to choose self-management and farming-oriented livelihood strategies are financial capital and household endogenous motivation. Among the four types of villages, natural capital has a negative impact on households that choose worker-oriented livelihood strategies. This implies that households with high natural capital are less likely to engage in labor activities. However, natural capital generally has a positive impact on households that choose agriculture-dominated strategies. Financial capital is negatively correlated with households choosing policy support strategies.

In scenic area-based villages, the increase in financial capital will also encourage farmers to opt for a self-employed livelihood model. In folk-based villages, there is a significant inverse relationship between financial capital and farmers’ choice of policy support strategies. In industry-based villages, an increase in financial capital can raise the likelihood of local farmers opting for a self-management business model by 17.6%. At the same time, factors that encourage farmers to choose a worker-oriented strategy include human capital and social capital. Conversely, factors that drive farmers to opt for a farming-oriented strategy include farmers’ endogenous dynamics, which are linked to their cultural characteristics and aspirations. In suburban-based villages, human capital and social capital play a significant positive role in encouraging farmers to adopt a worker-oriented strategy.

5. Discussion

The participation of farmers in rural tourism can effectively enhance farmers’ livelihood capital, influence the selection of livelihood strategies, and contribute to the sustainability of farmers’ livelihoods. This finding aligns with the results of numerous existing studies in the literature [13,24,25,62,63,64]. The conclusions of this study show that there are significant differences in livelihood capital and capital allocation between farmers who participate in rural tourism and those who do not. Farmers participating in rural tourism have higher livelihood capital and tend to choose self-management livelihood strategies [65]. However, scholars such as He and Hong et al. [66] believe that rural tourism contributes to overall income growth for farmers, but there is no significant difference in income between those who participate in tourism and those who do not. The significant difference lies mainly in the income levels of households that have escaped poverty and those that are still in poverty. In this study, financial capital, social capital, human capital, and endogenous dynamics within households all significantly influence farmers’ choice of livelihood strategies. This finding aligns with research conducted by scholars like Wang Manyu et al. [67] and Habib Nusrat et al. [38]. However, some scholars have found that human capital and material capital have a significant positive impact on livelihood strategies, while financial capital has no significant impact on livelihood strategies [27]. This is mainly due to the differences in research areas and backgrounds, which result in research findings with local characteristics.

When comparing different types of villages, it can be seen that farmers mainly engage in self-management livelihood strategies, which confirms the conclusion that rural tourism promotes farmers to choose independent livelihood strategies [23,68]. However, it is worth noting that this conclusion does not apply to industry-based villages. Although industry-based villages are mostly dominated by farming households, their livelihood capital is the highest. In Manas County, rural tourism is mainly developed through the “agriculture + tourism” model. For example, Shangzhuangzi Village is vigorously developing a courtyard economy under the rural revitalization program, guiding farmers to plant fresh corn in idle courtyards. The village committee guides farmers to sign planting order contracts with e-commerce companies to help farmers sell their products and expand sales channels. With the assistance of the courtyard economy, farmers have achieved satisfactory incomes. In cases where there is insufficient startup capital to engage in independent operations, choosing to continue farming is the best solution. Research by scholars such as Ma Ji-liang shows that commercial bean production increases household farm income [69]. Scholars like Adam have found the importance of breadfruit, a new economic crop, in ensuring the livelihoods of poor families [70]. These findings confirm the significant role of economic crops in increasing farmers’ income. Additionally, the integration of the agricultural tourism industry injects new vitality into traditional agricultural production, promoting the specialization, technological advancement, and modernization of agriculture, as well as driving the construction of new rural areas and the sustainable development of rural economies [71,72].

In a complex external environment, farmers are more inclined to choose livelihood strategies associated with tourism. On the one hand, the tourism industry can effectively compensate for the drawbacks of agriculture, such as seasonality and long return cycles. Industry-dependent villages like Xiliangzhouhu Village have successfully integrated agriculture and tourism. They have transformed the traditional agricultural production model by emphasizing economic crops and developing theme tourism activities centered around rural landscapes, agricultural activities, and specialty agricultural products to attract tourists. This approach can meet the needs of visitors throughout the crop growth cycle, enhancing the overall benefits of agriculture. On the other hand, the tourism industry has a strong consumption-driving force, which can quickly boost economic income. After a prolonged period of lockdown measures and travel restrictions resulting from the COVID-19 pandemic, individuals are now more inclined towards seeking high-quality green building spaces [73]. They are showing a preference for living in green, tranquil, spacious, and secure environments [74]. Green spaces and rural landscapes are attracting urban residents to move to rural areas [75]. Existing studies have shown that in Portugal and the Czech Republic, the number of outbound travelers increased due to the 2019 coronavirus pandemic, and they are more inclined to engage in various leisure activities in rural areas [76,77]. After the pandemic, farmers are more willing to participate in the tourism industry through activities such as farm stays and specialty homestays to offset the heavy blow to their economic income during the pandemic.

In the advancement of the rural revitalization strategy, farmers, as the main body of rural tourism development, have always been the core concern of our country’s strategic implementation in increasing income and becoming prosperous. However, in 2020, the outbreak of the COVID-19 pandemic dealt a heavy blow to the tourism industry [20], with rural tourism suffering even more severely from the impact of the pandemic, disrupting the livelihood systems of farmers and constraining rural sustainable development. As the pandemic subsides, the work of restoring farmers’ livelihoods becomes particularly urgent, while facing unprecedented challenges. Therefore, we need to deeply study and consider how to effectively promote the recovery and development of farmers’ livelihoods in the post-pandemic era so as to drive sustainable prosperity in rural areas. Firstly, finding a balance between the tourism industry and agriculture is crucial. The tourism industry often leads to a transformation in farmers’ traditional livelihoods, and the new livelihood activities and ways brought by the tourism industry also bring vitality to village culture. Most or even all of farmers’ income relies on tourism-based income. However, if the tourism industry is not thriving, local farmers will find it difficult to sustain their livelihoods [78]. This sudden policy environment change and agricultural modernization may increase the insecurity of livelihoods [77,79]. Therefore, farmers need to find a balance between the tourism industry and agriculture, establish tourism based on farms, develop farm functions in a diversified direction, establish new consumption relationships between farmers and tourists [80], increase rural consumption capacity, and promote the stability and improvement of farmers’ livelihoods. Secondly, integrating new business formats and utilizing rural areas as carriers for the tourism industry require increased emphasis on resource sharing, complementary advantages, and deep collaboration to establish new industrial forms [81]. This involves discovering new models of agricultural industrialization with the involvement of the tourism industry. For example, Shen et al. introduced the experience of protecting the ancient village of Xiamei Village. They found that the development of the tea industry and rural tourism construction can achieve a win–win situation by integrating the tea industry with the tourism industry [9]. This integration brings better income to local tea farmers and promotes rural revitalization development guided by tea culture. Lastly, sustainable livelihoods require diversified livelihood combinations because diversification can help spread livelihood risks [16]. By engaging in various livelihood activities, households can reduce their dependence on a single source of income, thereby decreasing vulnerability caused by the failure of a particular livelihood activity or market fluctuations. At the same time, government intervention must align with the transformation trend and adapt to different livelihood patterns by adjusting and implementing various types of capital.

At the same time, this paper also has the following deficiencies. First of all, the data used in this research were mainly obtained through questionnaires and interviews. Due to the deviation of respondents’ ability to understand and awareness of privacy protection, the data can be compared with secondary statistical data such as yearbooks and government bulletins to increase the credibility of the paper. Secondly, the historical and cultural background and national policy background of different periods create the intergenerational differences between groups, so the different research periods will cause the different research results. In the future, research on the difference in farmers’ livelihood capital at different stages of rural tourism development can also be based on group differences to further clarify the micro-processes and paths of livelihood transformation from the individual scale to the rural household scale and community scale.

6. Conclusions

This study examines a typical rural tourism village in Manas County, focusing on changes in household livelihood capital and choices of livelihood strategies. Using propensity score matching and probit model analysis, the study examined the effects of participating in rural tourism activities and the factors influencing this, and drew the following conclusions:

(1) Participation in rural tourism has led to significant differences in the livelihood capital of farmers. In terms of the total livelihood capital of different types of rural areas, industry-based villages have the highest value, followed by suburban-based villages, then scenic area-based villages, and finally folk-based villages. In terms of the total livelihood assets of different types of farmers, the highest is the self-management type, followed by the wage worker-oriented type, then the farming-oriented type, and finally the policy support type.

(2) Participation in rural tourism significantly influences the livelihood strategies chosen by households. Among households participating in rural tourism, the preference is for the self-managed type > worker-oriented type > farming-oriented > policy support type. Among households not participating in rural tourism, the preference is for worker-oriented type > farming-oriented > self-managed type > policy support type. There are differences in livelihood strategy choices between households in various types of villages. Villages that depend on scenic areas and folk customs primarily consist of self-employed farmers. In suburban areas, villages have more households led by labor, while villages that rely on industry have more households led by agriculture.

(3) The overall effect of rural tourism on the livelihood capital of farmers in Manas County is relatively positive. The total livelihood capital change in the treatment group has increased by an average of 0.0612 compared to the control group, and the impact effects vary between different types of villages. In terms of various types of livelihood capital, there is still heterogeneity in the changes between the treatment group and the control group. Apart from natural capital, human capital, financial capital, material capital, and social capital have all shown improvement.

(4) Participating in rural tourism significantly increases the probability of farmers choosing to operate independently but has a negative impact on farmers choosing to rely on farming as the main source of income. The impact on the probability of farmers choosing to rely on wage labor or policy support as the main source of income is not significant. Participation in rural tourism, financial capital, human capital, social capital, and endogenous dynamics of farmers have significant effects on farmers in various types of villages.

Compared with previous studies on the impact of rural tourism on farmers, this study focuses on the configuration of farmers’ internal characteristics and government factors on the sustainable livelihood of farmers in rural tourism destinations, and compares and analyzes the differences in farmers’ livelihood capital and livelihood choices in different types of rural tourism development. The aim was to arrive at a more scientific and useful conclusion on the complex impact of sustainable livelihood, and try to find a rural tourism development model that conforms to the sustainable livelihood of farmers. This is not only an effective extension and development of the issues raised by Natarajan, N. [82], that sustainable livelihood must take into account both internal and external factors, but also a response to the main idea of “rural revitalization enabled by cultural and tourism integration” mentioned in the Central document of China.