Public Service Delivery and the Livelihood Adaptive Capacity of Farmers and Herders: The Mediating Effect of Livelihood Capital

Abstract

Improving the provision and equitable accessibility of public services is an important aspect of poverty alleviation and necessary for rural revitalization. Farmers’ livelihood adaptive capacity is an important foundation for rural revitalization, and the mechanism by which it is influenced by public service delivery requires clarification. In this study, we construct a theoretical framework of public service delivery–livelihood capital–livelihood adaptive capacitiy of farmers and herders to measure the level and quality of public service delivery, based on the level of satisfaction of 334 farmers and herders we surveyed. The relationship between the three elements was verified using mediated effects analysis, which helped to accurately identify the key shortcomings of public service delivery and improve the delivery of public services to enhance the sustainable livelihood adaptive capacity of farmers and herders. We found that public service delivery has a significant positive impact on the livelihood adaptive capacity of farmers and herders, while livelihood capital plays an important mediating role, enhancing the direct impact of public service delivery. Thus, the government should pay attention to this transfer mechanism and actively cultivate farmers’ and herders’ livelihood capital to optimize the effects of rural public service delivery.

1. Introduction

Public service delivery is a basic requirement and core indicator of rural poverty eradication. In February 2021, China announced that public service delivery enabling rural poor people to escape poverty, that is, the “two no worries and three guarantees” policy, was fully realized [1]. “Two No Worries” means to stably realize that the rural poor people do not have to worry about food and clothing; “Three Guarantees” means to guarantee their compulsory education, basic medical care and housing security. Following the completion of poverty eradication, public service delivery remained the core requirement in the comprehensive implementation of the rural revitalization strategy that commenced immediately afterwards.Based on the Outline of the 14th Five-Year Plan for National Economic and Social Development of the People’s Republic of China, 20 government departments, including the National Development and Reform Commission of China, the Central Propaganda Department, the Ministry of Education and the Ministry of Public Security, have jointly issued the Outline of the 14th Five-Year Plan for Public Services. Thus, from the common perspective of poverty eradication and rural revitalization, public service provision represents a critical step toward prosperity for all citizens [2], including rural farmers, who need sound infrastructure and public service support [3].

Farmers and herders are micro-level subjects, and practical subjects of sustainable regional development. It is generally accepted that equal delivery of public services is an important prerequisite for improving the viability of individuals and promoting their development and economic stability [4], which is particularly important in ecologically fragile regions. However, research on the mechanisms through which public service equalization affects the adaptive livelihoods of farming and herding households is still relatively rare. Regarding the impact of public service equalization, studies have been conducted on narrowing the regional economic development gap [5], optimizing the regional investment environment [6], driving the development of related regional industries [7], and improving social welfare in the regions [8]. Few studies have been conducted on the impact of public services on livelihood adaptive capacity in china. Wu and Fan argued that public services in relation to education and health have become a central theme in anti-poverty research, drawing attention to poor people and their development [9]. Cao pointed out that the currently inadequate level of state support in terms of social security, such as rural cooperative medical care, and inadequate systems and institutions are the main reasons for farmers’ unsustainable livelihoods [10]. Yang and He pointed out that the construction of a social security system for dislocated farmers should start with a pension security system, medical security system, employment security system, and social assistance security system to facilitate improvements in farmers’ living standards [11]. Tai and Yang argued that cultural reconstruction is the driving force behind livelihood reconstruction, and cultural public services are the main reason why migrants create different types of livelihoods. To accomplish their relocation goals in the areas where migrants settle, governments should focus on the cultural integration and livelihood reconstruction of migrant households [12]. Most previous studies have analyzed the impact of the external environment on the sustainable livelihood of farming households from a single perspective, such as institutional provision, environmental changes, policy reforms, and cultural shocks, but there is a lack of in-depth analysis of the mechanism underlying the impact of the public service delivery on farming households’ livelihoods.

Adaptive Capacity is the ability of a system to adapt to climate change to reduce or reduce possible losses [13]. The term current adaptive capacity refers to the underlying characteristics of an individual, community or socio-ecological system to respond to threats or opportunities. It is defined as the ability to deal with disturbances and take advantage of new opportunities [14].The livelihood capital of farming and herding households is an important prerequisite for their viable capacity and development, and significant differences in the livelihood adaptive capacity of farmers’ and herders’ households are mainly the result of heterogeneity in terms of livelihood capital. Pasture area, credit situation, household income, and number of livestock are important elements of farming and herding households’ livelihood capital. When social and human capital are low, even abundant natural and financial capital are insufficient for the adoption of optimal livelihood strategies [15].

The concept of sustainable livelihoods discusses how to adopt different livelihood strategies under different livelihood capital conditions to pursue ideal livelihood outcomes. Among them, sustainable livelihoods are used as an analytical framework to study the livelihood status of rural households and consider consumer goods, the geographical location of the family, and the education level of family members to be the main factors in the poverty of farmers [16]. There is also research on tourism from the perspective of sustainable livelihoods, and it is believed that tourism is an industry that consumes fewer resources; it can reduce of poverty in many ways and enable individuals to develop long-term livelihood adaptability [17]. This shows the variable logic of “change in external environment—change in livelihood capital—adjustment of livelihood strategy—optimization of livelihood results”. This system has been widely used in studies on farmers’ livelihood adaptability, resilience, and livelihood capital.

Government-supported increases in the physical capital of farmers and herders contribute to alleviating incompatibility [18]. Most previous studies on livelihood adaptive capability have explored the impact of natural disasters such as drought [19], floods [20], and external shocks such as land loss [21] and ecological degradation [22] on the livelihood stability of farmers under the framework of vulnerability analysis. However, inadequate livelihood capital remains a major factor contributing to farmers’ and herders’ livelihood vulnerability, exposing them to multiple disadvantages in the process of livelihood transition. These include factors such as ecological migration [23], rapid urbanization [24].

In this study, we draw on the model of private proactive adaptation to climate change (MPPACC) proposed by Grothmann et al. to deconstruct farmers’ perceived adaptive capacity in response to external environmental changes into self-efficacy perception, adaptation cost perception, and adaptation efficacy perception [25]. The perceived difference between public expectations and public service delivery performance is reflected through the public’s subjective evaluation of the quality and quantity of public services provided [26]. The relationship between the public service delivery and livelihood adaptive capacity of farmers and herders was verified using mediated effects analysis, which helped to accurately identify the key shortcomings of public service delivery and improve the provision of public services to enhance the sustainable livelihood adaptive capacity of farmers and herders.

2. Research Hypotheses

Effective public service delivery is an important prerequisite for improving the viability of individuals and promoting their development and employment stability. The achievement of sustainable livelihoods for farmers and herders depends on the promotion of a new type of urbanization characterized by equal rights and interests, of which public services are a key component [27]. Employment service networks, public health service systems, and social security safety nets are all necessary to facilitate the transformation of farmers and herders’ livelihoods. The social security system for farmers should comprise a pension security system, medical security system, employment security system, and social assistance security system to facilitate a higher standard of living for farmers [28]. Cultural reconstruction is the driving force behind livelihood reconstruction, and culture-based public services are the main reason why migrants create different types of livelihoods [29]. In general, this means that public service delivery in terms of institutional provision, livelihood improvement, and social welfare is strongly associated with the improvement of farmers’ and herders’ livelihood adaptation capacity. Therefore, we propose the following hypotheses, as shown in (Figure 1).

Hypotheses 1 (H1).

The level of public service delivery has a significant effect on the livelihood adaptive capacity of farming and herding households.

Differences between the livelihood adaptive capacity of different farming and herding households are mainly the result of heterogeneity in livelihood capital. Regardless of whether it is a non-grazing household or an immigrant herding household, there are large differences in terms of livelihood capital acquisition, livelihood strategy expansion capacity, and opportunities as a result of gaps in public service delivery [30]. Many farming and herding households with low levels of livelihood capital have a low sustainable livelihood capacity and are at high risk of returning to poverty [31]. The effect of public service delivery on the livelihood adaptive capacity of farmers and herders is mainly reflected in the process of enhancing livelihood capital and converting livelihood capital into livelihood output. Therefore, we propose the following hypotheses, as shown in (Figure 1).

Hypotheses 2 (H2).

Public service delivery indirectly affects the livelihood adaptive capacity of farm households through livelihood capital.

3. Model, Variables, and Data Specifications

Currently, the measurement of the level of basic public service delivery includes both internal and external measurement. Internal measurement focuses on efficiency, including the efficiency of public service delivery at the regional level, using a method based on data envelopment analysis that measures aggregate country-level inefficiency [32]. Some scholars have also used comprehensive evaluation methods such as multiple indexes based on the evaluation of basic public transportation services, combining the analytic hierarchy process with fuzzy evaluation to build a quantitative multi-index evaluation model [33]. Other scholars have used the entropy method to determine indicator weights for evaluating the quantity and quality of basic urban public services in each city, and then undertaken an in-depth analysis of the relationships between them [34]. This also includes an analysis of the impact of investment in basic public services on the quantity and quality of the labor force from the perspective of government finances [35]. The level of integration of basic public services is measured from three perspectives: supply mode, efficiency, and mechanism of public service delivery [36]. In contrast to the technical rationality of efficiency evaluation from the input–output perspective, external measurement is more concerned with the perception of public services by the recipients and focuses on their overall feelings about the services in comparison with their expectations, which is more reflective of public needs and public values [37]. Measuring citizens’ level of satisfaction with public services is a very important part of public service delivery and feedback [38].

3.1. Measurement of Variables

3.1.1. Explained Variables

Based on the social cognitive model of individuals’ active adaptation to climate change [39], farmers’ perceived adaptive capacity was deconstructed into self-efficacy perception, adaptation cost perception, and adaptation efficacy perception. Details of the explanatory variables are shown in (

Table 1. Descriptions of explanatory variables.

Indicators	Measurement Issues	Indicator Assignment	Average Value	Standard Deviation
Perceived self-efficacy	How is your ability to adapt?	Cannot at all 1; generally 3; definitely can 5	2.32	0.73
Adaptation cost perception	Do you think the budgetary cost of adaptation actions is high?	Very low 1; generally 3; very high 5	3.42	0.83
Adaptation efficacy perception	What are the expected outcomes of your adaptation actions?	No effect 1; general 3; very good effect for 5	3.11	0.82

). The perception information obtained from the questionnaire survey was used to construct the corresponding perception index. Adaptation perception is the perception and expectation of farmers and herders regarding their adaptive capacity, adaptation cost, and adaptation effectiveness, which are the key factors involved in farmers’ and herders’ adaptation activities. Farmers’ perceptions of adaptation are highly subjective and heterogeneous, and are divided into self-efficacy perceptions, adaptation cost perceptions, and adaptation efficacy perceptions based on their predicted ability to take adaptive actions, the expected cost of implementing those actions, and the predicted efficacy of those actions. Only when farmers and herders perceive that their adaptive capacity is strong, their adaptation cost is low, and the adaptation efficacy is clear, will farmers’ and herders’ willingness to adapt be enhanced and their adaptive capacity be improved.

3.1.2. Explanatory Variables

This study takes into account the diversity of types of public service delivery and the difficulty of quantitative measurement and, based on the level of socioeconomic development and the production and life styles of ethnic areas, draws on the research of Liu et al to establish 15 indicators of farmers’ and herders’ satisfaction with the delivery of public services in five areas: livelihood-based, development-based, culture-based, ecology-based and security-based services [40]. A five-point Likert-type scale was used to measure respondents‘ views, with scores of 1 to 5 representing “Very dissatisfied,” “Dissatisfied,” “Okay,” “Satisfied,” and “Very satisfied,” respectively. Data reliability for all 15 variables was tested and Cronbach’s α was 0.838. Based on the general principle that a Cronbach’s α of between 0.6 and 0.7 indicates acceptable reliability, and that α greater than 0.7 indicates high reliability, the reliability of the data was high. A public service satisfaction index was constructed based on the five categories of public service delivery, as shown in (

Table 2. Public service satisfaction index.

Tier 1 Indicators	Specific Public Service Content	Measurement Issues
Security-based Public service delivery	County/township/village security duty point construction	Are you satisfied with the way the police station works?
	Rural legal aid	Are you satisfied with the county government’s legal aid public welfare activities?
	Natural disasters and sudden crisis management	Are you satisfied with the county government’s disaster relief services and relief measures?
Development-based Public service delivery	Rural public employment guarantee	Are you satisfied with the free non-farm jobs and skills training provided?
	Agricultural science and technology extension and training	Are you satisfied with the government’s agricultural technology training?
	Rural infrastructure development	Are you satisfied with the transportation, postal and commercial services around your home?
Culture-based Public service delivery	Rural public sports facilities and activities	Are you satisfied with the fitness equipment and public sports Competitions in the village square?
	Famous cultural heritage activities	Are you satisfied with the public performance and publicity activities of the county art troupe in the village?
Ecological-based Public service delivery	Uniform treatment of household waste	Are you satisfied with the service of unified domestic waste disposal?
	Sanitary toilets in agricultural and pastoral villages	Are you satisfied with the sanitary toilets built by the government for free?
Livelihood-based Public service delivery	Rural social security	Are you satisfied with the premiums and coverage of the new social insurance?
	Popularization of safe drinking water in farming and herding villages	Are you satisfied with the convenience and quality of your own home drinking water?
	Rural public health and medical care	Are you satisfied with the level of doctors’ treatment and medical facilities in the village health center?
	Rural compulsory education	Are you satisfied with the teaching and student recreational activities in the township schools?
	Administrative villages through hardened roads	Are you satisfied with the quality and accessibility of the hardened roads around your home?

).

3.1.3. Mediating Variables

Based on the sustainable livelihood analysis framework proposed by the Department for International Development, the livelihood capital of farm households was divided into natural capital, human capital, social capital, financial capital, and physical capital [41]. Furthermore, based on the livelihood capital measurement indexes and quantitative values developed by Zhao et al., [42], the values of each index were standardized using the extreme difference standardization method, the weights of each index were determined using the expert consultation method, and the weighted summation method was used to measure the various livelihood capital indices of farmers and herders, as shown in (

Table 3. Definitions of variables and descriptive statistics.

Dependent Variable	Indicator Weights and Assignment		Average Value	Standard Deviation
Livelihood adaptive capacity	Perceived self-efficacy + Perceived cost of adaptation + Adaptation efficacy perception; Strong adaptive capacity OR poor adaptive capacity assign 0 or 1		0.067	0.251
Independent Variable
Satisfaction with Livelihood-based public services	“Very dissatisfied” to “very satisfied” assignment 1–5		3.56	0.90
Satisfaction with Development-based public services	“Very dissatisfied” to “very satisfied” assignment 1–5		3.52	0.92
Satisfaction with Culture-based public services	“Very dissatisfied” to “very satisfied” assignment 1–5		3.72	0.84
Satisfaction with Ecological-based public services	“Very dissatisfied” to “very satisfied” assignment 1–5		3.51	0.90
Satisfaction with Security-based public services	“Very dissatisfied” to “very satisfied” assignment 1–5		3.51	0.94
Intermediate Variables	Indicator Meaning	Weights	Average Value	Standard Deviation
Natural Capital	Arable land area per capita (mu)	0.47	1.53	6.50
	Grassland area per capita (mu)	0.53	11.46	31.35
Human Capital	0 for non-labor force; 0.5 for half labor force; 1.0 for full labor force	0.44	4.12	1.21
	0 for illiterate; 0.25 for elementary school; 0.5 for junior high school and school leavers; 0.75 for high school; 1.0 for college and above	0.56	1.27	0.89
Social Capital	Number of public officials among relatives (persons)	0.40	0.94	0.43
	The number of relatives in the village is 1 for many; 0.75 for more; 0.5 for more; 0.25 for less; and 0 for less	0.60	0.56	0.26
Financial Capital	Household cash income	0.65	5987.34	6213.32
	There is a loan opportunity for 1; none is 0	0.35	0.57	0.51
	Number of livestock	0.68	196.32	332.76
Physical Capital	The number of fixed assets items as a percentage of the number of items listed	0.32	0.34	0.17
Control Variables	Indicator Weights and Assignment		Average Value	Standard Deviation
Age of head of household	Actual observations (years)		42.50	11.36
Family Size	Total family size (persons)		5.50	1.21
Non-farm employment share	Non-farm payroll employment/total household size (%)		0.54	0.26
Traffic conditions	Very poor = 1; poor =2; fair = 3; good = 4; very good = 5		3.14	1.23
Economic Development Level	Very poor = 1; poor = 2; fair = 3; good = 4; very good = 5		1.72	0.94

).

3.1.4. Control Variables

To analyze the impact of public service delivery on the livelihood adaptive capacity of farming households through changes in their livelihood capital, other factors that affect the livelihood adaptive capacity of farming households need to be excluded. In this study, these included age of head of household, family size, non-farm employment share, traffic conditions, and level of economic development.

3.2. Model Construction

It is important to analyze the mechanism underlying the A→C→B causal pathway between public service delivery and the livelihood adaptive capacity of farmers and herders. In this study, we focused on the effects of five mediating variables: human capital, social capital, natural capital, physical capital, and financial capital. Prior to model estimation, unit root tests and cointegration analyses were performed, and the results showed that the data used in this study were smooth and had cointegrated relationships. We used the method proposed by Wen and Ye [43] to test whether the level of satisfaction with public service delivery can affect the livelihood adaptive capacity of farmers and herdsmen through livelihood capital. The model is as follows:

$${\mathrm{Y}}_{\mathrm{i}}={\mathrm{a}}_0+{\mathrm{a}}_1{\mathrm{X}}_{\mathrm{i}}+{\mathrm{a}}_2{\mathrm{T}}_{\mathrm{i}}+{\mathrm{e}}_1$$

(1)

$${\mathrm{M}}_{\mathrm{i}}={\mathrm{b}}_0+{\mathrm{b}}_1{\mathrm{X}}_{\mathrm{i}}+{\mathrm{b}}_2{\mathrm{T}}_{\mathrm{i}}+{\mathrm{e}}_2$$

(2)

$${\mathrm{Y}}_{\mathrm{i}}={\mathrm{c}}_0+{\mathrm{c}}_1{\mathrm{X}}_{\mathrm{i}}+{\mathrm{c}}_2{\mathrm{M}}_{\mathrm{i}}+{\mathrm{C}}_3{\mathrm{T}}_{\mathrm{i}}+{\mathrm{e}}_3$$

(3)

In Equations (1)–(3), Y_i denote the livelihood adaptive capacity of farmers and herders, X_i denotes public service delivery, Mi denotes the mediating variable, and T_i denotes the control variable. Here, a₀, b₀, and c₀ are constants, e₁ to e₃ are regression residuals, and a₂, b₂, and c₃ are the regression coefficients of the control variables. In Equation (1), a₁ is the overall effect of public service delivery on the livelihood adaptive capacity of the ith farming household, b₁ in Equation (2) is the effect of public service delivery on the mediating variables, and c₁ and c₂ in Equation (2) are the direct effects of public service delivery and mediating variables, respectively, on the livelihood adaptive capacity of the ith farming household. Substituting Equation (2) into Equation (3) yields the mediating effect of public service delivery b₁c₂, which is the indirect effect of public service delivery on the livelihood adaptive capacity of farming households through the mediating variable.

3.3. Study Area and Data Sources

3.3.1. Study Area

The Gannan Yellow River Important Water Source Ecological Function Area is the first batch of important ecological function areas delineated by China. The Proposal of the Central Committee of the Communist Party of China on Formulating the Fourteenth Five-Year Plan for National Economic and Social Development and the Visionary Targets for 2035 adopted by the Fifth Plenary Session of the 19th CPC Central Committee listed the whole area of Gannan State as a nationally important ecosystem protection and restoration project. The Ecological Protection and Construction Plan for the Important Water Supply Ecological Function Area of the Yellow River in Gannan, Gansu (2006–2020) [44], divides the area into a key protected area, a restoration area, and an economic demonstration area (Figure 2). The key conservation area is where the original state of the ecosystem is well preserved or the ecological function is outstanding, the ecological status is important, and the protection measures are similar. The restoration and treatment area is where the degradation of the ecosystem is more serious and needs urgent restoration and management. The economic demonstration area is smaller, but transportation is more convenient, the population is more concentrated, the economy is relatively developed, and the demonstration effect is strong. We selected Gaxiu Village, Gahai Town, Luqu County, which are in a key conservation area, Anguo Village, Amuquhu Town, Xiahe County, which in a a restoration and treatment area, and Xinsi Village, Zuogai Duoma Town, Hezuo county, which are in an economic demonstration area, as the study areas, as shown in (

Table 4. Division of key ecological function zones of the Yellow River in Gannan.

Functional Area Name	Construction Tasks	Ecological Construction and Public Services	Specific Name	Administrative Area (County)
Key Protected Areas	No grazing, hunting, logging; all development and utilization activities are prohibited in the area; forest and grass vegetation is restored through measures such as closure and management	Comprehensive management of degraded grasslands; Wetland Conservation and Restoration Grassland and Forest Conservation Biodiversity Conservation Research	Taohe forest ecosystem protection District	Hezuo, ZhuoNi, LinTan
			Shouqu Wetland Ecosystem Conservation District	Maqu
			Gahai-Zecha National Nature Reserve	Luqu
			Animaqing Mountain Grassland Ecosystem Conservation District	Maqu
			Daxia River Headwaters Grassland Ecosystem Conservation District	Xiahe
Restoration of treatment areas	The region implements grass-based livestock, grazing rest and rotation, focusing on the implementation of “three” grass management, forest and wetland and wildlife protection measures, and at the same time, the implementation of concentrated settlement of herdsmen to reduce the pressure of grass carrying	Comprehensive management of degraded grasslands; livestock warming shed construction; barley production base and other ecological agriculture and animal husbandry construction; pastoralist training in applicable technology	Ganjia-Zuogai Manma grassland restoration and management District	Xiahe, Hezuo
			Animaqing Mountain Grassland Ecological Restoration and Management District	Maqu
			Amuquhu grassland ecological restoration and management District	Xiahe
			Jiamaogong-Taoyan Ecological Restoration and Management District	Hezuo, ZhuoNi, LinTan
			Xiqingshan Ecological Restoration and Management District	Luqu
			Zuogai-duoma wetland ecological restoration and management District	Hezuo
Economic Demonstration areas	The region is centered on the county town and linked by national and provincial highways. Development of cattle and sheep fattening, breeding and tourism industries with livestock products processing as the main focus	Nomadic settlement construction; dairy farming; cattle and sheep fattening; pastoralist training in applicable technology; biogas digester construction; human and animal water diversion	Maqu County Economic Model District	Maqu
			Xiahe-Cooperative Economy Model District	Xiahe, Hezuo
			Luqu Economic Model District	Luqu
			Lintan-Zhuoni Economic Model District	Lintan, Zhuoni

).

3.3.2. Data Sources

The data were obtained from a questionnaire survey on satisfaction with basic public service delivery and a questionnaire survey on farmers’ and herders’ perceptions of their livelihood adaptive capacity conducted from January to March 2021 and from March to May 2022, respectively. Using a combination of stratified random sampling and semi-structured interviews, 375 households were randomly selected from the three abovementioned villages. Of these, 113 come from the key conservation area, 118 from the restoration and treatment area, and 103 from the economic demonstration area; 334 valid responses were received, a response rate of 89.1%. The survey questionnaire consisted of three parts: basic demographic information; level of satisfaction with livelihood-based, development-based, culture-based, ecology-based, and security-based public services; and perceptions of self-efficacy, adaptation costs, and adaptation efficacy.

4. Results

4.1. The Effect of Public Service Delivery on the Livelihood Adaptive Capacity of Farmers and Herders

It can be seen from (

Table 5. Overall effect of public service delivery on the livelihood adaptive capacity of farmers and herders.

Variable Name	Farmers’ and Herders’ Livelihoods Adaptive Capacity
Constants	−0.990 ** (0.052)
Satisfaction with livelihood-based public service delivery	0.124 ** (0.020)
Satisfaction with development-based public service delivery	0.133 ** (0.026)
Satisfaction with Culture-based Public Service delivery	0.109 ** (0.017)
Satisfaction with ecological-based public service delivery	0.114 ** (0.014)
Satisfaction with security-based public service delivery	0.108 (0.015)
Age of head of household	0.004 ** (0.001)
Family size	0.007 ** (0.013)
Non-farm employment share	0.057 ** (0.020)
Traffic conditions	0.023 ** (0.024)
Economic Development Level	0.086 ** (0.022)
Sample size	334
R2	0.813
Adjust R2	0.819
F-value	F(12, 332) = 185. 183, p = 0.000

Note: ** indicate significance at the 5% levels.

) that public service delivery had a significant positive effect on the livelihood adaptive capacity of farming and herding households in general at the 1% level of significance, while the F-statistic is significant at both the 5% and 1% levels, indicating that the overall model estimate is a good fit, and thus is eligible for the next stage of analysis. The results indicate that the higher the level of satisfaction of farmers and herders with public service delivery, the higher their perception of self-efficacy, the lower their perception of adaptation costs, and the higher their perception of adaptation efficacy, all of which have a significant positive effect on their livelihood adaptation capacity. Therefore, hypothesis 1 is supported.

4.2. Testing and Estimating the Mediating Effect of Livelihood Capital

4.2.1. Mediated Effects Model Test

The mediating effects of livelihood capital were estimated based on the model constructed above. The results are shown in

Table 6. Results of the mediation effects model.

Variable Name	Human Capital	Social Capital	Natural Capital	Physical Capital	Financial Capital	Adaptive Capability
Constants	3.029 (0.221)	0.188 (0.231)	0.914 (0.516)	0.018 (0.065)	−1.067 (0.229)	−0.515 (0.066)
Public Service delivery	0.232 **	0.313 **	0.152 *	0.228 **	0.230 **	0.056
Human Capital						0.133 **
Social Capital						0.051 **
Natural Capital						0.051 *
Physical Capital						0.044 **
Financial Capital						0.130 **
Sample size	334	334	334	334	334	334
Control variables	Controlled	Controlled	Controlled	Controlled	Controlled	Controlled
Adjust R2	0.813	0.785	0.913	0.774	0.731	0.729
F-value	F(12, 332) = 24.183, p = 0.000	F(12, 332) = 35.113, p = 0.000	F(12, 332) = 30.183, p = 0.000	F(12, 332) = 41.283, p = 0.000	F(12, 332) = 49.083, p = 0.000	F(12, 332) = 285.144, p = 0.000

Note: ** and * indicate significance at the 5% and 10% levels, respectively.

. Table 6 shows the results of the regression of Equations (2) and (3). Equation (2) shows the effect of public service delivery on livelihood capital, where livelihood capital consists of five items: human capital, social capital, natural capital, physical capital, and financial capital. The results presented in Table 6 show that the effects of public service delivery are significantly positive at the 5% level for human capital, social capital, natural capital, physical capital, and financial capital. Equation (3) shows the effect of public service delivery on the livelihood adaptive capacity of farming and herding households after the inclusion of mediating variables. The results presented in column 7 of (Table 6) show that the coefficients of human capital, social capital, natural capital, physical capital, and financial capital pass the significance test, implying that these five indicators have a significant partial mediating effect. Of the factors mediating the effect of public service delivery on the livelihood adaptive capacity of farming households, livelihood capital accounts for 61.92% of the total effect.

Table 7. Size of the mediation effect.

Item	a1 (Total Effect)	b1c2 (Intermediary Effect)	c1 Direct Effect	Effectiveness Ratio Calculation Formula	Percentage of Each Effect/%
Human Capital	0.125	0.031	0.056	ab/c	12.754%
Social Capital	0.125	0.016	0.056	ab/c	24.795%
Natural Capital	0.125	0.008	0.056	ab/c	6.181%
Physical Capital	0.125	0.010	0.056	ab/c	7.981%
Financial Capital	0.125	0.030	0.056	ab/c	10.210%

shows that the mediating effect of social capital accounted for 24.80% of the total and contributed more to the livelihood adaptive capacity of farmers and herders than the other elements of livelihood capital. Social capital refers to the degree of social support available to farmers and herders in relation to their livelihoods, and thus has a significant impact on the adaptive capacity of farmers and herders. In reality, public service delivery relies on synergies between local advantages and external agents. Relevant departments providing public services responded positively to the needs of farmers and herders, which increased the mutual trust between grassroots cadres and farmers and herders. The technical and information socialization services provided by the government to professional cooperatives and village collectives are embedded in the social network of herding households, which helps to improve their willingness to adapt and accumulate social capital and reduces the lack of capital and lack of reward caused by gender and wealth differences. The government also extends the interaction radius of poor farmers and herders by improving transportation facilities and infrastructure at relocation sites, thereby increasing cognitive social capital such as trust, norms, reciprocal cooperation, and values.

The mediating effect of human capital accounted for 12.75% of the total effect. Human capital is the fundamental force driving the sustainable development of farming and herding households in terms of livelihood transformation. In livelihood-oriented public service delivery, training in practical farming and herding techniques, and in handicrafts with ethnic characteristics, and public welfare positions such as cleaning, forest protection, and road protection increase the skills and employability of farming and herding households. The participation of farmers and herdsmen in ecological protection construction projects, the implementation of science and technology construction projects, and the upgrading of the production, supply, and marketing of technical services projects have all increased the accessibility of education and training opportunities for farmers and herdsmen. Access to vocational education, medical care, and government-mandated social security rights are also a manifestation of investment in the human capital of farming and herding households. On the one hand, increasing the ability of farmers and herders to popularize the common language, fostering access to and use of information and technology, improving their perceptions of adaptive efficacy and self-efficacy, and increasing their willingness to adapt enables them to choose better ways to adapt to new livelihoods.

The mediating effects of financial capital and physical capital accounted for 10.21% and 7.98%, respectively, of the total. Public service delivery helps to transform livelihood capital by improving the livelihood environment and reducing barriers to adaptation. The grassland ecological protection subsidy and reward policy implemented in this area had a strong short-term effect, accounting for a large proportion of the financial capital of farmers and herdsmen. Local construction of ecologically civilized, well-off villages relies on upgrading a number of public facilities and services in tourist villages, continually broadening the income sources of farmers and herdsmen, and providing tourism-related employment security services and entrepreneurial guidance services for farmers and herdsmen, thereby increasing the wage income, operating income, and transfer income of farmers and herdsmen. In terms of physical capital, the local government implemented the Comprehensive Capacity Enhancement Project for Agriculture and Herding to ensure that there was adequate farmland and pastureland for farmers and herders, to provide complete infrastructure for the dilapidated houses and temporary resettlement houses where farmers and herders live, and to provide free production equipment, driving lessons, and other services for families with two daughters or with a heavy support burden. The grain-to-feed project and public services such as the construction of warm sheds and fattening bases have also significantly increased the material capital of farmers and herdsmen.

The mediating effect of natural capital accounted for 6.18% of the total. On the one hand, the free housing, water and electricity, and policy-related subsidies provided by the local government in recent years under the policies of grass-livestock balance and ecological migration have alleviated maladjustment in terms of land loss and ecological migration of farmers and herdsmen. The construction of artificial forage grasslands and sheds implemented by the local government under the grassland resource protection project has ensured the balance of the grasslands areas, and also ensured that the natural capital of the region is transformed from providing ecological public goods into building sustainable livelihood adaptive capacity. On the other hand, since the implementation of the grassland ecological bonus policy, which is closely related to the adaptation of farmers’ and herders’ livelihoods, the implementation of grazing bans and rotational grazing in specific zones has transformed traditional grassland livestock livelihoods, even though the first round of policy-related subsidies has not achieved comprehensive coverage. In addition, there is a gap between the grassland area that has implemented the grassland ecological reward policy and the actual area that needs to implement the policy. The cross-regional grazing area is included in the scope of policy implementation. This has prevented the natural capital transformation potential of farmers and herdsmen from coming into play, hence the intermediary effect of natural capital is the lowest of all forms of capital.

4.2.2. Endogeneity and Robustness Testing

To check whether there was any bias in terms of the effect of government on the results, we used the number of party members in the household as a control variable to indirectly reflect the importance of the local government and mitigate the effect of omitting important variables, such as the importance of the local government, by excluding the subsample with party members in the household and retaining the subsample without party members in the household for estimation. The results were similar, suggesting that endogeneity was within the acceptable range, as shown in (

Table 8. Endogeneity test results.

Variable Name	Farmers and Herderss Livelihoods Adaptive Capacity
Public Service delivery	0.039 * (4.043)
Human Capital	0.048 ** (1.760)
Social Capital	0.034 * (2.663)
Natural Capital	0.043 ** (2.113)
Physical Capital	0.046 ** (5.663)
Financial Capital	0.124 ** (6.132)
Control variables and constant terms	Controlled
R2	0.956
Adjusting R2	0.953
F-value	F(12, 158) = 283.221, p = 0.000

Note: ** and * indicate significance at the 5% and 10% levels, respectively.

).

The above regression results provide preliminary support for our hypothesis that public service delivery promotes the livelihood adaptive capacity of farming households through livelihood capital. To test the robustness of the results, we replaced the measures of the main variables, removed the control variables, and applied binary logit models. The results are presented in column I in (

Table 9. Robustness test results.

Variable Name	I Livelihood Adaptation Capacity	Variable Name	II Livelihood Adaptation Capacity	Variable Name	Livelihood Adaptability
Adaptive Perception	0.173 ** (0.049)	Public Service delivery	0.142 ** (0.014)	Public Service delivery	5.119 * (2.441)
Human Capital	0.023 ** (0.034)	Human Capital	0.089 ** (0.031)	Human Capital	6.112 * (6.114)
Social Capital	0.039 ** (0.011)	Social Capital	0.039 ** (0.015)	Social Capital	2.781 ** (7.110)
Natural Capital	−0.045 ** (0.015)	Natural Capital	−0.072 (0.011)	Natural Capital	−4.008 ** (−6.213)
Physical Capital	−0.127 ** (0.018)	Physical Capital	−0.168 ** (0.019)	Physical Capital	−8.221 ** (6.213)
Financial Capital	0.042 ** (0.004)	Financial Capital	0.062 ** (0.007)	Financial Capital	2.991 ** (7.891)
Control variables and constant terms	Controlled	Delete control variables	Deleted	Control variables and constant terms	Controlled

Note: ** and * indicate significance at the 5% and 10% levels, respectively.

). Livelihood adaptive capacity is measured by adaptive perception, and the stronger the perception of adaptation, the stronger the livelihood adaptive capacity of farmers and herders. The results presented in column II in (Table 9) followed the deletion of the aforementioned control variables to test the influence of the main variables on the livelihood adaptability of farmers and herdsmen. The results presented in column III in (Table 9) are corroborated by model replacement, because whether farmers’ and herders’ livelihood adaptive capacity is enhanced is measured using a binary dummy variable; thus, the results are corroborated by using a binary logit model to conduct a baseline regression.

5. Discussion

In the context of implementing the rural revitalization strategy, there is still a significant gap between farmers’ and herders’ demands for and the delivery of public services, especially in relation to the use of resources and the provision of basic public services such as employment assistance, medical and health care, and social security. Regional disparities in the provision of public services [45] have led to significant differences in terms of regional residents’ ability to acquire livelihood capital and expand their livelihood strategies [46]. In particular, farmers and herders in some key ecological function areas have a low sustainable livelihood capacity and high risk of returning to poverty. This will undoubtedly increase their livelihood vulnerability, weaken their ability to resist threats to their livelihood, and increase the costs of livelihood transition [47].

Current research on livelihood adaptation in key ecological function areas is focused on two areas. The first is the identification of the adaptation intentions and strategies of farmers and herders facing risks and external shocks. The second is embedding the livelihood adaptation strategies of farmers and herders within the natural and social cultures, taking into account the complex patterns and internal logic of livelihood adaptation of farmers and herders in relation to the natural–economic–social modernization process [48]. However, recent studies measuring the livelihood adaptation capacity from the micro-dimension of farmers and herders have mainly drawn on Pandey and Malakarak et al. [49,50]. Related studies have focused on analyzing the adaptive capacity of farmers facing a single external pressure, such as environmental factors, institutional factors, or subjective factors. Therefore, in this study, we used Gannan Tibetan Autonomous Prefecture, a key ecological function area, as the study area; we evaluated public service delivery using an index system, and farmer evaluations based on the livelihood transformation needs of farmers and herders in the region and the special characteristics of livelihood capital in less-developed minority areas. We here construct a theoretical framework of public service delivery–livelihood capital–livelihood adaptive capacity of farmers and herders, and based on the mediating effect of livelihood capital, we propose some targeted countermeasures using livelihood capital.

However, the case analysis in this paper has limitations. This research is focused on farmers and herdsmen in key ecological function areas in China. In future research, the research area, research objects, the regional characteristics and typical group used to carry out representative livelihood sustainability research should be broadened. Secondly, the mediation effect model is used to reveal the lack of depth in the interaction mechanism between public service supply and livelihood adaptation; on this basis, the role of structural and process changes on livelihood adaptation can be further analyzed as paths and mechanisms of action, and clarify important issues such as the pattern–process–mechanism of livelihood adaptation, the nesting–feedback relationship of livelihood systems at different scales, and the formation and regulation mechanisms of livelihood adaptive capacity. In addition, factors affecting livelihood adaptation can be further refined in future research, such as the impact of changes to transportation networks on the livelihoods of migrant workers, the impact of new urbanization on the livelihoods of semi-urban residents, and immigration under multiple constraints.

6. Conclusions

6.1. Research Conclusions

Livelihood capital is the most significant factor influencing farmers and herders in the process of livelihood transformation and determines the efficiency of capital conversion. Public service delivery aims to promote the livelihood capital utility obtained by farmers and herders through livelihood capital transformation, and is designed to clear structural obstacles for the efficient conversion of livelihood capital through precise and effective public service delivery to increase farmers’ and herders’ livelihood adaptation capacity. The research conclusions are as follows:

(1)

Increasing the level of public service delivery enhances the livelihood adaptation capacity of farming and herding households, and all variables are significant at the 1% and 5% levels. On the one hand, improving the level of basic public service delivery can optimize the allocation of the livelihood capital of farming and herding households and enhance their development capacity through capital accumulation. On the other hand, public services create a new livelihood environment to help people to adapt to the market demand for the modernization and transformation of agricultural and rural livelihoods.

(2)

The results presented found that the coefficient of the total effect of public service delivery on the livelihood adaptive capacity of farmers and herders was 0.125 and was significantly positive at the 1% level. The direct effect was 0.056, accounting for 38.08% of the total effect. The indirect effect of the five types of livelihood capital, natural, physical, human, social, and financial, on adaptive capacity accounted for 61.92% of the total effect, with coefficients of 0.232, 0.313, 0.152, 0.228, and 0.230, respectively, all of which were significantly positive at the 5% levels. Transformation of livelihood capital by public service delivery is a key factor in improving livelihood adaptive capacity, and empowering farming and herding households to develop themselves in the process of livelihood transition.

(3)

It was found that perceived self-efficacy and adaptation costs of farmers and herders were significantly correlated with social and financial capital: the greater the social and financial capital, the greater the perceived self-efficacy, the lower the perceived adaptation cost, and the greater the adaptation capacity. Perceived adaptation efficacy was not significantly correlated with livelihood capital, but there was significant regional heterogeneity. This provides further evidence that the level of public service delivery is closely related to the perceived livelihood environment and livelihood risk of farmers and herders, which in turn affects their livelihood adaptation capacity.

6.2. Policy Implications

First, a demand identification mechanism in relation to public services should be established to enhance the accessibility of public services and determine the context and direction of factor flows for the conversion of farmers’ and herdsmen’s livelihood capital so that they can gain easier access to livelihood capital, enhance their human capital and social capital through public services, and more conveniently convert their livelihood capital. On the one hand, we broaden the channels through which farmers and herders can express their demands and give full play to the role of institutionalized channels such as people’s congresses and Chinese People’s Political Consultative Conference meetings in transmitting their preferences for basic public services. On the other hand, the governance model based on big data and the Internet enriches the forms of expression of demand for public services, improves the accessibility of public services, and optimizes the level of livelihood capital.

Second, there should be a focus on stimulating the transformative effect of developmental public services on the human and social capital of farming and herding households. Governments should provide graded skills training and diversified employment channels for local residents, stimulate their endogenous motivation for self-development, guide farmers and herders in making non-farm livelihood decisions while reducing non-farm employment costs and unemployment risks, and help them to adapt to the perceived costs and enhance the perceived self-efficacy and adaptation efficacy in an effort to achieve stable and sustainable livelihoods. They should also provide inclusive non-farm skills training adapted to local industrial characteristics and continuously accumulate reserves of knowledge from farmers and herdsmen to improve their ability to develop stable livelihoods and reduce their livelihood risks.

Third, they should focus on the improvement of the livelihood environment through livelihood-based public services and guide farmers and herders in adopting non-farm livelihood strategies. On the one hand, they should improve roads, electricity supply, drinking water, housing conditions, and the human living environment, and give full play to the provision of public services to enhance farmers’ physical and natural capital. On the other hand, they should further improve rural compulsory education and rural public health and medical services, and establish a new social security system for farming and herding areas focusing on four areas: farming and herding area pension insurance, farming and herding area health insurance, rural assistance, and basic protection for urban migrant workers, in an effort to expand the living space of farmers and herders and improve the level of physical, human, and financial capital through livelihood-based public service delivery.