Improving the Livelihood Resilience of Poverty-Stricken Population under Rural Revitalization: A Case Study of Chongqing M Reservoir Area

Abstract

With the victory in poverty alleviation, China’s “Three Rural Issues” effort is shifting its attention to the execution of the rural revitalization strategy. To consolidate the poverty eradication gains and boost the resilience of the livelihoods of people who have been lifted out of poverty, we must implement several different strategies. Improving the livelihood resilience of the poverty-stricken population is the main objective of the long-term mechanism of promoting rural revitalization. Based on the theory of sustainable livelihoods, this paper creates an analysis framework for livelihood resilience of the poverty-stricken population. On this basis, we use principal component regression to measure the livelihood resilience of poverty-stricken population in Chongqing M Reservoir Area. We use the seemingly unrelated regression model to identify important variables influencing the stability of farm households emerging from poverty and propose policy to optimize resilience of the livelihoods of people lifted out of poverty. The results show that the population living in poverty around the Chongqing M Reservoir has a steadily rising livelihood resilience index. Among them, policy support has a significant positive effect on the livelihood resilience of poverty-stricken population. Similarly, regional endowments have a significant positive effect on the livelihood resilience of poverty-stricken population. However, livelihood risk has a significant negative effect on the livelihood resilience of poverty-stricken population. These findings provide a basis for the subsequent enhancement of livelihood resilience.

1. Introduction

With the general victory in the struggle against poverty, the execution of the rural revitalization plan is the focus of worldwide attention in order to boost the sustainable development of the countryside [1]. National leaders have repeatedly stressed that the comprehensive implementation of the rural revitalization strategy is of great significance to China’s modernization drive and social development. China attaches great importance to rural vitalization and has issued a series of important opinions and major plans on the implementation of the rural vitalization strategy, thus making solid progress in rural vitalization. The general requirements of the village revitalization strategy are “ecological livable, civilized village customs, thriving industry, effective governance and rich life” [2]. The primary objective of rural regeneration is to achieve shared prosperity. Increasing the people’s capital for livelihood and strengthening their ability to resist poverty via livelihood is a crucial objective in the rural revitalization process [3]. The aim is to prevent people lifted out of poverty from returning to poverty and to ensure the stability and sustainability of their livelihoods.

The majority of academic research focuses on underprivileged farmers, immigrants, and other groups, but there are few studies on how the poverty-stricken population resists changes in their means of subsistence [4,5,6]. Although this segment of the population has been lifted out of absolute poverty and the basic issues of food and clothing have been resolved, they are still relatively poor vulnerable groups, with weak resilience, risks, and they are difficult to maintain. Therefore, it is necessary to pay attention to the sustainable livelihood of the people and find effective ways to improve their livelihood resilience. Even though all the impoverished counties in the Chongqing M Reservoir Area have been taken out of poverty, more government policy support as well as other strategies are necessary to improve the standard of living for the populace. The livelihood stress of the poverty-stricken population has become a major issue. It is necessary to deeply analyze the factors affecting their livelihood stress and formulate targeted development paths to promote their sustainable development. The complexity of the issue would increase if livelihood resilience is not improved, which is detrimental to the peace and stability of rural communities. Based on these considerations, the goal of this study is to examine the problem of improving the livelihood resilience of the poor population within the framework of the strategy of rural revitalization in the modern era. Additionally, the study points out the different influencing variables, which limit the growth of poverty-stricken people’s ability to support themselves and proposes focused optimization strategies to support rural revitalization and sustainable development.

The term “livelihood” has multiple meanings; it can refer to both a process and its result. Habib et al. [7] viewed livelihood as a series of living methods adopted by individuals or families in order to survive, including activities, assets, and skills. On this basis, Habib et al. [7] proposed that livelihood diversification is crucial for lifting poor households out of poverty. The concept of resilience has its roots in the study of mechanical mechanics and engineering [8], and it has since been used in studies on ecology [9], social–ecological systems [10], natural disasters [11], and other areas. However, the concept of resilience is inconsistent due to different perspectives and scientific backgrounds. Although scholars have diverse definitions of resilience, there are three aspects, which are generally agreed on when faced with risks: the skills and resources, the method of risk avoidance, and the beneficial results of responding [12,13,14,15,16]. In general, the ability of individuals or organizations to actively fend off risks and recover from calamities can be thought of as resilience. According to some academics, the ability of a person or group to retain and improve sustainable livelihood possibilities in the face of external challenges is the definition of livelihood resilience [17,18,19]. In addition, some academics have identified livelihood resilience as a crucial element in farmers’ capacity to tolerate danger and escape poverty [20,21].

According to this paper, there are still two gaps in the theoretical studies of living resilience. First of all, the existing studies on this topic do not sufficiently take into account the livelihood resilience of population reduction in poverty. Second, while livelihood resilience has been presented, it is difficult for it to become a focused dimension of livelihood resilience research in the study of certain groups due to its limited theoretical expansion. Therefore, it is necessary to further extend the dimension of livelihood resilience assessment. In order to examine how to increase the livelihood resilience of poverty-stricken population, this study uses poverty-stricken population as its research object and builds a model for evaluating livelihood resilience. This paper intends to address the following three research questions:

(1)

What is the current trend in poverty-stricken population’s livelihood resilience index in the vicinity of the Chongqing M Reservoir Area?

(2)

Are there differences in livelihood elasticity of different types of poor households?

(3)

What factors influence the livelihood resilience of poverty-stricken population?

To answer the aforementioned questions, this study creates an analysis framework for the livelihood resilience of poverty-stricken population based on the sustainable livelihood theory. On this basis, we use principal component regression to measure the livelihood resilience of poverty-stricken population in the Chongqing M Reservoir Area. We use the seemingly unrelated regression model to identify important variables influencing the stability of farm households emerging from poverty and propose policy to optimize resilience of the livelihoods of people taken out of poverty. The results show that the population living in poverty around the Chongqing M Reservoir has a steadily rising livelihood resilience index. Among them, policy support has a significant positive effect on the livelihood resilience of poverty-stricken population. Similarly, regional endowments have a significant positive effect on the livelihood resilience of poverty-stricken population. However, livelihood risk has a significant negative effect on the livelihood resilience of poverty-stricken population.

This paper focuses on the theoretical innovation of specific population livelihood resilience and the innovation of livelihood resilience evaluation index system. The contributions of this paper can be summarized as follows. First of all, in the existing research, most researchers only focus on the relationship between vulnerability and livelihood resistance of poverty. In order to strengthen livelihood resilience and more thoroughly and methodically characterize the livelihood resilience of the population engaged in poverty alleviation, this study integrates policy support, regional endowment, and livelihood risk. Secondly, by combining the sustainable livelihood theory in this paper, the study quantifies livelihood resistance through human capital, social capital, material capital, financial capital, and natural capital. On this basis, the development of an indicator system for assessing the livelihood resilience of those who have overcome poverty is helpful for the government’s creation of tailored welfare measures.

The rest of the paper is organized as follows. Section 2 briefly reviews the related literature. Section 3 contains a logical analysis of the relevant theories and presents the research hypotheses. Section 4 describes the research area and data sources, the construction of the index system of livelihood resilience measurement, the selection of independent and dependent variables, and the descriptions of the empirical model. Section 5 is the main part of the paper, which mainly contains two aspects. The first aspect is to use the principal component regression method to measure the livelihood resilience of poverty-stricken population in the Chongqing M Reservoir Area. The second aspect is to use the seemingly unrelated regression model to analyze the key influencing factors of the livelihood resilience of people in the M Reservoir Area of Chongqing. Section 6 analyzes the influencing factors of the livelihood of poverty-stricken population in the Chongqing M Reservoir Area and puts forward optimization suggestions to improve the livelihood resilience of poverty-stricken population.

2. Literature Review

Three fields of previous literature are relevant to our current work: the research on poverty, the research on poverty alleviation, and the research on resilience. In the following section, we separately show how our paper relates to these research areas.

2.1. Research on Poverty

Poverty exists in various countries and affects their economic development and social stability. Researchers have put a lot of effort into studying the issue of poverty and have gathered useful theories and experience in the process. Wang et al. [22] built a multidimensional poverty measuring model at the village level and investigated the causes of poverty using the contribution index and linear regression. On this basis, they also used the spatial measurement analysis model to identify the types and differences of poor villages. Yang et al. [23] used the spatial measurement model to study the spatiotemporal evolution relationship between the urban and rural income gap and poverty level in various regions of Chongqing, which showed that the spillover effect of poverty factors varied between poor and non-poor counties. Li et al. [24] suggested that efforts to reduce poverty should ensure that poor families have stable access to food and clothing, education, health care, housing, or they should raise their net income above the national poverty line.

The present study on poverty focuses on its root causes, its geographic distribution, and the obstacles to its alleviation. These studies have partly explored the main factors affecting poverty. Most researchers focus on poverty affecting farmers, immigrants, and other groups, but there are comparatively few studies on the livelihood resilience of poverty-stricken population. Even though this group of people is no longer in utter poverty, it is nonetheless a vulnerable one. Therefore, it is important to focus on the individuals who have been pulled out of poverty’s sustainable livelihood situation.

2.2. Research on Poverty Alleviation

In 2013, the national leaders proposed targeted poverty alleviation, and China’s research on poverty alleviation is gradually increasing. In terms of poverty management, scholars have proposed poverty control measures from different perspectives. Wang et al. [25] discovered that social networks had a beneficial effect on the development of resilience when they investigated the role, which social networks play in the formation of vulnerability in China’s traditional disadvantaged districts. In the face of the problem of how to improve and maintain rural families above the poverty line, Li et al. [24] proposed that the livelihood resilience of families lifted out of poverty is closely related to their ability to use the available resources, learn new knowledge, and use external resources. Guo et al. [26] pointed out that raising rural family income can indirectly improve the sustainability of rural life and achieve the effect of poverty alleviation. Cairns et al. [27] suggested that in order to effectively improve living standards, it would be best to scientifically design targeted poverty alleviation work procedures and to create a comprehensive and effective targeted poverty alleviation policy system. Lundin [28] claimed that in order to implement a dynamic and scientific intervention in the process of reducing poverty, the state should boost its support for health, education, employment, culture, social security, and other factors.

The current research on poverty alleviation focuses on targeted poverty alleviation, with policy support from the family, society, and government. However, in order to fundamentally lift people out of poverty and improve the quality of life, different measures are needed to be taken to enhance the livelihood level of the people and improve the livelihood quality of the poverty-stricken population. This study holds that external disruptions as well as internal structural elements have an impact on the livelihood resilience of poverty-stricken population. Therefore, this paper addresses the issue from three aspects: policy support, regional endowment, and livelihood risk.

2.3. Research on Resilience

Researchers frequently integrate livelihood resilience into the vulnerability research paradigm, where it serves as a crucial tenet along with exposure and sensitivity [29]. Livelihood resilience has the opposite meaning of vulnerability: the higher the vulnerability, the lower the livelihood resilience, and the stronger the livelihood resilience, the lower the vulnerability [30,31].

Su et al. [6] established a regression model to examine the effects of poverty alleviation strategies and the accessibility of different forms of capital on the sustainable livelihoods of farm households. The results demonstrate that poverty alleviation strategies, natural capital, and social capital all have a significant impact on farm household sustainability. Amadu et al. [32] presented a three-dimensional framework for livelihood resilience based on learning, self-organizing, and buffering capacities, and they used structural equation modeling to validate the validity of the framework for interventions aimed at improving fishermen’s livelihood resilience. Shi et al. [5] used logistic regression to examine migrant livelihood resilience issues and discovered that home economic circumstances did not affect how resilient a livelihood was regarded by relocators. In the context of climate change, Yang et al. [33] explored the livelihood vulnerability of farmers in various disaster-prone regions, analyzed it using polynomial logistic regression, and offered specific recommendations to support livelihood resilience among households working to eradicate poverty.

The majority of current research on livelihood resilience has been focused on applying an analytical framework for sustainable livelihoods to quantify the changes in livelihood capital in order to gauge the changes in livelihood stress resistance. These studies partially focus on the relationship between vulnerability and livelihood resilience in poor populations. The difference in this study is that it combines the sustainable livelihood theory, using principal component analysis and seemingly unrelated regression model to measure the livelihood resilience index of poverty alleviation population and to study the key factors affecting livelihood resilience.

3. Theoretical Framework

3.1. Sustainable Livelihoods

Sustainable livelihood refers to the livelihood capacity, assets, and income activities of individuals or families to improve long-term living conditions. The United Nations Conference on Environment and Development in 1992 pushed for guaranteed livelihoods as the main objective of ending poverty. The academic community has developed an analytical framework for sustainable livelihood in response to the increasing concern about the contribution of sustainable livelihood to the struggle against poverty. The sustainable livelihood analysis framework (SLA), developed by the Department for International Development (DFID), is the one, which is most frequently utilized. The framework includes five components: the vulnerability context, livelihood capital, shifts in structure and process, livelihood strategies, and livelihood outcomes.

The first part is the vulnerability context, which mainly refers to the insufficient resistance and vulnerability of poverty-stricken population when facing external shocks. Natural catastrophes, financial dangers, and other things are among the effects of the external environment. When people lifted out of poverty are unable to withstand risks, their livelihoods will be affected, and they might even fall back into poverty.

The second part is the livelihood capital. Livelihood capital includes five categories: natural capital, material capital, financial capital, human capital, and social capital. Natural capital refers to the land, plants, and other natural resources owned by poverty-stricken population. Material capital includes the infrastructure to maintain the poverty-stricken population and production equipment. Financial capital refers to the funds, borrowing status, and free assistance obtained by poverty-stricken population. Human capital is the ability of poverty-stricken population to maintain their knowledge, skills, and health. Social capital refers to the network of social relations owned by poverty-stricken population.

The third part is the shifts in structure and process. The poverty-stricken population are part of the society; thus, social policies will have an impact on their way of life. On the one hand, social policies can improve their livelihood capital and create a good environment. However, on the other hand, there will be some defects in policy design, and there will be some obstacles in the implementation process, so the policy effect will be damaged.

The fourth part is the livelihood strategies. Livelihood strategies describe how people use their wealth for their means of subsistence in order to accomplish their means of subsistence goals. The vulnerable population will select their livelihood strategies based on their capital for livelihood, considering the policy environment and the context of their susceptibility.

The fifth part is the livelihood outcomes. Different livelihood strategies would result in various livelihoods. Increasing family income and improving family quality of life are the two main reasons that poor people work for a living. In the DFID sustainable livelihood framework, livelihood capital is the core content, which affects the livelihood strategies and livelihood outcomes of people lifted out of poverty. The livelihood strategy of poverty-stricken population is to use the livelihood capital. The livelihood capital should play a role through the livelihood strategy to affect the livelihood results of poverty-stricken population.

3.2. Analytical Framework

The framework for analyzing livelihood resilience has been a hot topic in academic circles because it is the core component of research on sustainable livelihoods [34]. Livelihood resilience is the ability of farmers to restore their level of livelihood or even increase it in response to the effects of external threats [35,36]. The sustainable livelihood analysis (SLA) framework, which was suggested by the feasible capacity research hypothesis put out by the DFID, is where the concept of livelihood capacity first gained academic recognition. According to the SLA framework’s context, many academics define the five assets as “living capacity” and conduct research on this basis [37,38,39]. The human capital, social capital, material capital, natural capital, and financial capital not only constitute an asset system but also have a mutual transformation relationship between them. Therefore, this paper holds that livelihood resilience can be reflected by the five capital levels of human capital, social capital, material capital, natural capital, and financial capital.

According to the literature research, the scholars’ recommended analysis framework of livelihood resilience is the assessment of livelihood resilience through livelihood capital. This paper believes that livelihood resilience not only emphasizes the stability of the poverty-stricken population’s livelihood system function, but it also emphasizes its potential livelihood resilience. Yet, it does not take into consideration the subject’s adaptability to the external environment where they are placed. However, external disruptions as well as internal structural issues have an impact on the impoverished people’s ability to sustain their ways of life. Therefore, this research builds a paradigm for livelihood resilience analysis based on external disruptions. Figure 1 shows the framework for analyzing the livelihood resilience of poverty-stricken population.

External disturbance is mostly caused by the environmental and social–economic environment, and it is a critical factor in the growth of livelihood resilience. Among them, government-led assistance initiatives have a direct or indirect impact on the availability of various types of capital among the poverty-stricken population. Natural disasters, such as floods, and price variations in agricultural goods represent a severe danger to the long-term growth of farmers’ livelihoods [40,41,42]. External disturbance would not only raise the susceptibility of the livelihoods of individuals pulled out of poverty, but it would also impair their ability to cope with hazards. Perfect infrastructure may help individuals get out of poverty by satisfying their basic requirements and increasing output and revenue [43,44,45]. Generally speaking, livelihood resilience is influenced by policy support, regional endowments, and livelihood risk taken together. The process of getting people out of poverty involves ongoing adjustments to their means of subsistence, an expansion of those means, and ultimately, the achievement of sustainable subsistence.

3.3. Research Hypotheses

Through literature research, this study proposes that there is some influence relationship between policy support, regional endowments, livelihood risk, and livelihood resilience. Therefore, the study hypotheses are as follows.

3.3.1. Policy Support and Livelihood Resilience

Policy support is an important driver for preventing the return to poverty and achieving rural revitalization. Growing leading industries, creating jobs, and enhancing the infrastructure are the major ways to promote industry, employment, and ecological compensation. Policy support may enhance the environment for livelihoods by lowering livelihood risk and enhancing the infrastructure, hence enhancing livelihood resilience. Therefore, policy support has an important impact on the livelihood resilience of people lifted out of poverty. Thulstrup [46] evaluated household livelihood resilience in a study of mountainous people in Vietnam, suggesting that policy support contributes to livelihood resilience. Li et al. [24] argued that the government’s multifaceted efforts to support sustained poverty alleviation and the reduction in family poverty had a significant influence. Zhao et al. [47] discovered that ecological policy is a significant factor impacting the resilience of farmers’ livelihoods based on an examination of farm household survey data. Yan et al. [48] proposed that policy support is one of the main factors affecting the livelihood resilience of animal husbandry farmers in Inner Mongolia. Therefore, the first hypothesis is proposed as follows:

Hypothesis 1 (H1). 

Policy support has a positive impact on the livelihood resilience of poverty-stricken population.

3.3.2. Regional Endowments and Livelihood Resilience

Regional endowments are an important source of support for the economic development of the Chongqing M Reservoir Area. The population’s relationship with the government and its livelihood resilience to improve itself can be strengthened through a good rural development level, a great information network, and accessible transportation. When evaluating the relationship between market towns and social ecological resilience, Perz et al. [42] found that infrastructure has a significant impact on improving the livelihood resilience of farmers. Liu et al. [49] suggested that the convenience of transportation has a substantial impact on the design of the livelihood resilience of poverty-stricken population, improving their livelihood resilience. According to Yan et al. [48], improving social network strength and information gathering can improve the livelihood resilience of farmers. Therefore, the second hypothesis is proposed as follows:

Hypothesis 2 (H2). 

Regional endowments have a positive impact on the livelihood resilience of poverty-stricken population.

3.3.3. Livelihood Risk and Livelihood Resilience

The main livelihood risks faced by the impoverished population in the Chongqing M Reservoir Area are market price variations of agricultural products as well as natural disasters, such as rainstorms and debris flows. Livelihood risks will weaken the ability of people who have overcome poverty to respond to risks, which will lower their ability to withstand risks and make them more vulnerable to them. Kuang et al. [50] concluded that the two main threats to farmers’ livelihoods in agricultural production are natural risk and market risk. Sok and Yu [51] showed that farmers in Cambodia’s lower Mekong River region have low livelihood resistance as a result of frequent natural disasters, such as floods and droughts. Islam [52] discovered that a variety of factors, including social and economic weaknesses, natural disasters, and climate change, had caused locals to relocate. Therefore, the third hypothesis is proposed as follows:

Hypothesis 3 (H3). 

Livelihood risk has a negative impact on the livelihood resilience of poverty-stricken population.

4. Method

4.1. Research Area and Data Source

This study’s data consist of a survey of the living and production conditions of poverty-stricken population in 15 districts and counties of the Chongqing M Reservoir Area. The survey sample spans the years from 2015 to August 2022, with a total of 1242 questionnaires. The survey first includes a pre-survey, and the questionnaire is revised based on the pre-survey, and then, the formal survey is conducted. On this basis, the combination of stratified sampling and random sampling is adopted to ensure the scientific nature and reliability of the survey. The final sample size is 14 villages and 269 households.

4.2. Construction of Livelihood Resilience Measurement Index System

This study builds an index system of the population’s livelihood resilience in the Chongqing M Reservoir Area using the creation process of livelihood resilience and relevant scholars as a base [53,54]. Financial capital, social capital, natural capital, material capital, and human capital are its five categories, along with 16 indicators. The index system for the area around the Chongqing M Reservoir is shown in

Table 1. Index system of livelihood resilience level of poverty-stricken population in Chongqing M Reservoir Area.

Level 1 Sign	Level 2 Sign	Variable	Units and Assignments
Financial capital	Whether to obtain business loans	Z1	1 = Yes, 0 = No
	Whether there is informal lending	Z2	Availability of low-interest or interest-free loans 1 = Yes, 0 = No
	Household existing savings	Z3	CNY/household
Social capital	Degree of neighborhood interaction	Z4	1 = Very close, 0.75 = Close, 0.5 = Average, 0.25 = Less, 0 = No
	Long-term care from relatives or friends	Z5	1 = Yes, 0 = No
	Participation in organizational associations	Z6	Unit/household
	Source of employment information	Z7	1 = Introduce, 0 = Others
Natural capital	Per capita cultivated land area	Z8	m2/person
	Per capita forest land area	Z9	m2/person
Material capital	Housing type	Z10	0.3 = Brick–concrete, 0.2 = Brick and wood, 0.1 = Adobe
	Drinking water conditions	Z11	Whether to use tap water 1 = Yes, 0 = No
	Production machinery conditions	Z12	0.5 = Large equipment, 0.2 = Small equipment
	Household durable consumer goods	Z13	CNY/household
Human capital	Number of family laborers	Z14	Person/household
	The average level of education of the labor force	Z15	0.2 = Illiteracy, 0.4 = Primary school, 0.6 = Middle school, 0.8 = High school, 1 = College
	Degree of health	Z16	With or without serious illness, disability, or chronic disease, 0.2/per person.

.

4.3. Variable Selections

4.3.1. Dependent Variables

The dependent variables are the evaluation indices of financial capital, social capital, natural capital, material capital, and human capital.

4.3.2. Independent Variables

Based on the selection of the survey of poverty-stricken population and prior research, this study selects 11 independent variables from the policy support, regional endowments, and livelihood risk.

The policy support level includes industrial assistance, employment training, ecological compensation, economic organization, and social security. Regional endowment level includes transportation convenience, information network, quality of farmland and related agricultural supporting facilities, and rural development level. Livelihood risk levels include market fluctuation and natural disasters.

Table 2. Description of independent variables’ definitions and assignments.

Variable	Variable Index	Assignment Specification	Mean Value	Standard Deviation
Policy support	Industrial assistance (X1)	1 = Yes, 0 = No	0.74	0.75
	Employment training (X2)	1 = Yes, 0 = No	0.67	1.67
	Ecological compensation (X3)	1 = Yes, 0 = No	0.56	0.23
	Economic organization (X4)	1 = Yes, 0 = No	0.58	0.51
	Social security (X5)	1 = Yes, 0 = No	0.64	1.23
Regional endowment	Transportation convenience (X6)	Distance from village to town/km	3.32	0.43
	Information network (X7)	Network coverage in rural areas/%	0.61	0.32
	Quality of farmland and related agricultural supporting facilities (X8)	5 = Very good, 4 = Good, 3 = Average, 2 = Poor, 1 = Very poor	0.27	0.69
	Rural development level (X9)	Per capita annual income/CNY	20,432	0.56
Livelihood risk	Market fluctuation (X10)	1 = Very small, 2 = Small, 3 = Average, 4 = Large, 5 = Very large	3.99	1.52
	Natural disaster (X11)	1 = Very small, 2 = Small, 3 = Average, 4 = Large, 5 = Very large	2.68	0.44

shows the relevant independent variables’ definitions and assignments.

4.4. Econometric Method

Based on the impact of actual situation and unpredictable factors on the livelihood resilience and five types of livelihood capital of poverty-stricken population at the same time, this study uses the seemingly unrelated regression model for parameter estimation to achieve the result of improving the efficiency of model estimation [55]. The model is constructed as follows:

$$Y_H={\alpha }_1+{\beta }_1{X}_1+{\beta }_2{X}_2+{\beta }_3{X}_3+...+{\beta }_n{X}_n+{\mu }_1,$$

(1)

$$Y_S={\alpha }_2+{\beta }_1{X}_1+{\beta }_2{X}_2+{\beta }_3{X}_3+...+{\beta }_n{X}_n+{\mu }_2,$$

(2)

$$Y_M={\alpha }_3+{\beta }_1{X}_1+{\beta }_2{X}_2+{\beta }_3{X}_3+...+{\beta }_n{X}_n+{\mu }_3,$$

(3)

$$Y_F={\alpha }_4+{\beta }_1{X}_1+{\beta }_2{X}_2+{\beta }_3{X}_3+...+{\beta }_n{X}_n+{\mu }_4,$$

(4)

$$Y_N={\alpha }_5+{\beta }_1{X}_1+{\beta }_2{X}_2+{\beta }_3{X}_3+...+{\beta }_n{X}_n+{\mu }_5,$$

(5)

where $Y_H$, $Y_S$, $Y_M$, $Y_F$, and $Y_N$ refer to the dependent variables, which stand for human capital, social capital, material capital, financial capital, and natural capital, respectively. $X_i$, ${\alpha }_i$, ${\beta }_i$, and ${\mu }_i$ are the independent variables, the intercept terms, the calibrated parameters, and the deviations, respectively.

5. Results

5.1. Description of the Poverty-Stricken Population

The basic situation of the survey of poverty-stricken households is as follows. The average age of the households is between 42.19 and 54.47 years old, and the aging phenomenon of rural areas in the Chongqing M Reservoir Area is very severe. The average educational level of household heads ranges from 1.33 to 2.59, with 71.85% of them having only completed basic school or less. As a whole, household heads often have low levels of education. The household population is generally larger, ranging between 4.00 and 5.51. The average number of laborers per household in the study area ranged from 2.32 to 3.17, with an average household labor force of more than 2 persons, and the average number of workers per household ranged from 0.61 to 1.50, with an average of about 1 person. This shows the diversification of livelihoods of the current poverty-stricken population. The average annual net income of households in the Chongqing M Reservoir Area ranged from CNY 10,219.99 to CNY 46,931.73, with an average income of CNY 28,598.38.

The income and expenditure of the sample township population is described as follows. In terms of annual net income, KLZ has the highest paid population lifted out of poverty, at CNY 64,439 per year. The income of the poverty-stricken population in WQZ is CNY 49,752 per year, and that of the poverty-stricken population in NXZ is CNY 36,994 per year. The net income of the poor families in SJX is the lowest, at CNY 26,391 per year. In the average annual net income from breeding, the income of the poverty-stricken population in KLZ is the highest, reaching CNY 93,943 per year, much higher than the other four towns. WQZ, SJX, and LHZ raise CNY 36,238, CNY 29,706, and CNY 29,520 per year, respectively. NXZ has the lowest income, with only CNY 24,016 per year. In terms of work income, the average annual net income of the poverty-stricken population in NZX is the highest, reaching CNY 49,300 per year, while that in LHZ is the lowest, with only CNY 40,774 per year.

In terms of agriculture expenditure, LHZ has the greatest level of poverty-stricken population spending at CNY 42,904 per year, while NXZ has the lowest level at CNY 5699 per year. LKZ has the highest annual living costs at CNY 27,447 per year, mainly due to the convenient transportation and proximity to shopping malls and supermarkets—hence the highest expenditure. Because of its convenient transit and easy access to information, the net income of the poverty-stricken population in KLZ is much higher than that of other towns. Because KLZ has abundant arable land resources, each person is a larger agricultural acreage than other towns, and the number of livestock raised by families is relatively large, so the farming income in the area is much higher than that of other towns. The development of greenhouse vegetable planting in WQZ has significantly increased the income of the poor population, so the net income of the poverty-stricken population is also higher.

5.2. Measurement and Analysis of Livelihood Resilience Level

5.2.1. Measurement of Livelihood Resilience Level

Because the same sample data span from 2015 to 2022, this paper selects the temporal global principal component analysis method to analyze the livelihood resilience level of the poverty-stricken population, so as to avoid the error in statistical coherence for each year [56].

In this study, 16 indicators are tested with the KMO test and Bartlett spherical test. The KMO value is 0.748. If the value is greater than 0.7, it indicates that these variables can be subjected to a principal component analysis. The Bartlett spherical test gave a concomitant probability of 0.000, less than the significance level of 0.05, indicating suitability for principal component analysis. The results are shown in

Table 3. KMO test and Bartlett spherical test for influencing variables.

Kaiser–Meyer–Olkin Measure of Sampling Adequacy		0.748
Bartlett’s test of sphericity	Approximate chi-square	1236.191
	df	210
	Sig.	0.000

.

Because the raw observed variables range in scale and magnitude, it is necessary for the raw indicators to be dimensionless. Using principal component extraction, after maximum orthogonal rotation, five principal components with eigenvalues greater than 1 were extracted, and the total variance explained was 80.291%. This indicated that the interpretability of the content of the factors was good, and the structural validity of the sample reached an acceptable level. The results are shown in

Table 4. The rate of variance explained.

Principal Component	Initial Eigenvalue	Variance Explained %	Cumulative Variance Explained %
1	2.913	30.128	30.128
2	2.205	17.664	47.792
3	1.686	15.923	63.715
4	1.201	9.705	73.420
5	1.114	6.871	80.291

.

According to analysis of the public factor load array in the livelihood resilience index system of the poverty-stricken population in the Chongqing M Reservoir Area, the public factor scores are shown in

Table 5. Component matrix.

	F1	F2	F3	F4	F5
Whether to obtain business loans (Z1)	0.479	0.382	0.419	0.641	0.208
Whether there is informal lending (Z2)	0.277	0.413	0.024	0.704	0.139
Household existing savings (Z3)	−0.262	0.527	0.378	0.646	0.203
Degree of neighborhood interaction (Z4)	0.308	0.692	0.426	0.389	0.069
Long-term care from relatives or friends (Z5)	0.167	0.660	0.237	0.553	−0.131
Participation in organizational associations (Z6)	0.456	0.639	0.526	0.395	−0.273
Source of employment information (Z7)	−0.082	0.564	0.364	0.466	0.059
Per capita cultivated land area (Z8)	0.165	0.143	0.571	0.230	0.437
Per capita forest land area (Z9)	−0.174	0.046	0.151	0.088	0.536
Housing type (Z10)	−0.026	0.028	0.793	−0.407	−0.349
Drinking water conditions (Z11)	−0.370	0.069	0.610	0.169	−0.247
Production machinery condition (Z12)	−0.256	0.121	0.579	−0.383	0.314
Household durable consumer goods (Z13)	0.295	−0.072	0.782	−0.543	0.163
Number of family laborers (Z14)	0.737	0.425	0.505	0.107	0.365
The average level of education of the labor force (Z15)	0.559	0.327	0.207	0.496	−0.108
Degree of health (Z16)	−0.534	−0.375	−0.236	−0.125	−0.162

.

According to the calculation, the variables with a greater impact on principal component F1 include the number of family laborers (0.737), the average level of education of the labor force (0.559), and the degree of health (−0.534). Therefore, principal component F1 can be categorized as the “human capital factor”. The variables with a greater impact on principal component F2 include the degree of neighborhood interaction (0.692), long-term care from relatives or friends (0.660), participation in organizational associations (0.639), and the source of employment information (0.564). Therefore, principal component F2 can be categorized as the “social capital factor”. The variables with a greater impact on principal component F3 include the housing type (0.793), drinking water conditions (0.610), production machinery condition (0.579), and household durable consumer goods (0.782). Therefore, principal component F3 can be categorized as the “material capital factor”. The variables with a greater impact on principal component F4 include whether to obtain business loans (0.641), whether there is informal lending (0.704), and household existing savings (0.646). Therefore, principal component F4 can be categorized as the “financial capital factor”. The variables with a greater impact on principal component F5 include per capita cultivated land area (0.437) and per capita forest land area (0.536). Therefore, principal component F5 can be categorized as the “natural capital factor”. The values of each column in the “component matrix” table are divided by the eigenvalue square separately to derive the eigenvector corresponding to each eigenvalue. The expression of each influencing factor is as follows:

$$Human=0.402\ast Z_{14}+0.396\ast Z_{15}+0.486\ast Z_{16},$$

(6)

$$Social=0.441\ast Z_4+0.439\ast Z_5+0.398\ast Z_6+0.324\ast Z_7,$$

(7)

$$Material=0.357\ast Z_{10}+0.384\ast Z_{11}+0.253\ast Z_{12}+0.427\ast Z_{13},$$

(8)

$$Financial=0.326\ast Z_1+0.485\ast Z_2+0.374\ast Z_3,$$

(9)

$$Natural=0.303\ast Z_8+0.416\ast Z_9.$$

(10)

Each factor has a different weight, which reflects its relative importance. By averaging the proportion of variation explained by the primary components, the weights were determined. Therefore, the livelihood resilience comprehensive index (LRCI) of poverty-stricken population is expressed as follows:

$$\mathrm{LR}\mathrm{C}\mathrm{I}=0.452\ast \mathrm{H}\mathrm{u}\mathrm{m}\mathrm{a}\mathrm{n}+0.269\ast \mathrm{S}\mathrm{o}\mathrm{c}\mathrm{i}\mathrm{a}\mathrm{l}+0.208\ast \mathrm{M}\mathrm{a}\mathrm{t}\mathrm{e}\mathrm{r}\mathrm{i}\mathrm{a}\mathrm{l}+0.186\ast \mathrm{F}\mathrm{i}\mathrm{n}\mathrm{a}\mathrm{n}\mathrm{c}\mathrm{i}\mathrm{a}\mathrm{l}+0.103\ast \mathrm{N}\mathrm{a}\mathrm{t}\mathrm{u}\mathrm{r}\mathrm{a}\mathrm{l}.$$

(11)

5.2.2. Analysis of Livelihood Resilience Level

According to formulae (6)–(10), the annual livelihood resilience index for each factor is calculated, as shown in

Table 6. Livelihood resilience index of poverty-stricken population from 2015 to 2022.

Year	Social Capital Factor	Human Capital Factor	Material Capital Factor	Financial Capital Factor	Natural Capital Factor	Livelihood Resilience Score	Standardized Score	Livelihood Resilience Comprehensive Index
2015	−0.532	−0.032	−0.019	−0.044	−0.037	−0.665	−0.981	0.421
2016	−0.447	−0.015	−0.013	−0.031	0.048	−0.458	−0.655	0.454
2017	−0.413	−0.015	−0.009	−0.023	−0.014	−0.477	−0.692	0.457
2018	−0.304	−0.049	0.002	−0.011	−0.021	−0.385	−0.562	0.463
2019	−0.251	−0.050	0.014	−0.016	0.055	−0.249	−0.364	0.483
2020	0.024	−0.026	0.008	0.064	−0.020	0.050	0.134	0.523
2021	0.124	0.009	0.016	0.033	−0.003	0.679	0.537	0.562
2022	0.315	0.151	0.020	0.021	−0.002	0.806	0.884	0.606

. Based on the calculations, we can observe that the comprehensive index of livelihood resilience of poverty alleviation households shows an increasing trend. Among them, the years 2015–2018 were relatively stable, with an increase after 2018 and a large increase from 2019 to 2022. In addition, the human capital, social capital, material capital, financial capital, and natural capital all noted an increase at different rates: 80.71%, 56.32%, 65.66%, 23.25%, and 5.83%, respectively.

In this study, the K-means cluster analysis was used to divide the livelihood resilience index into three levels: high, medium, and low [57]. As shown in

Table 7. Livelihood resilience level.

Grade	Interval of Value	Proportion
High level	≥0.535	24.37%
Medium level	0.423–0.534	42.52%
Low level	≤0.422	33.11%

, if the livelihood resilience is greater than or equal to 0.535, it is classified as high level; it is classified as medium level at 0.423–0.534, and below 0.422, it is classified as low level. At every class level, 24.37%, 42.52%, and 33.11% of households were raised out of poverty.

In order to comprehensively analyze the livelihood resilience of different poverty-stricken populations, the poverty-stricken populations are designated based on the poverty alleviation time, labor endowment, and dependency ratio. Among them, the poverty alleviation period 2015–2016 represents early poverty alleviation households; the period 2017–2018 represents medium-term poverty alleviation households; and the period 2019–2020 represents late poverty alleviation households. Labor endowment is designated according to the number of laborers per household: households with less than or equal to two people are designated as the shortage type, while those with more than two people are designated as the abundant type. A family dependency ratio of less than 0.5 represents households with a low dependency ratio; a ratio between 0.5 and 1 represents those with a medium dependency ratio; and a ratio of more than 1 represents those with a high dependency ratio. The results are shown in

Table 8. Livelihood resilience index of different types of poverty alleviation households.

Types	Specific Classification	Livelihood Resilience Index	High	Medium	Low
Poverty alleviation time	Early poverty alleviation (2015–2016)	0.443	28.40	49.86	21.74
	Medium-term poverty alleviation (2017–2018)	0.424	24.99	41.42	33.59
	Late poverty alleviation (2019–2020)	0.414	23.45	45.70	30.85
Labor endowment	High level of education	0.430	28.33	46.87	24.80
	Medium level of education	0.422	25.37	44.65	29.98
	Low level of education	0.418	22.84	44.99	32.17
	Labor abundance	0.424	25.00	45.36	29.64
	Labor shortage	0.414	23.47	41.08	35.45
Dependency ratio	Low dependency ratio	0.425	25.33	45.77	28.90
	Medium dependency ratio	0.411	22.84	37.06	40.10
	High dependency ratio	0.357	0.00	48.58	51.42

.

5.3. Measurement Results’ Analysis

5.3.1. Model Diagnosis

The data in this study are processed using a seemingly unrelated regression model. Whether there is a simultaneous correlation between the perturbation terms of the n equations will directly affect the estimation efficiency of the model. It is necessary for the disturbance terms of the n equations to simultaneously correlate, in accordance with the fundamental premises of the seemingly unrelated regression model. Therefore, before adopting the specific model, we conducted the Breusch–Pagan heteroscedasticity test and tested the correlation of the disturbance term of the n equations of the seemingly unrelated regression model.

Table 9. Model suitability test.

	${\mathit{Y}}_{\mathit{M}}$	${\mathit{Y}}_{\mathit{F}}$	${\mathit{Y}}_{\mathit{H}}$	${\mathit{Y}}_{\mathit{S}}$	${\mathit{Y}}_{\mathit{N}}$
$Y_M$	1
$Y_F$	0.165	1
$Y_H$	0.589	0.274	1
$Y_S$	0.348	0.643	0.497	1
$Y_N$	−0.264	0.486	0.276	−0.342	1
Breusch–Pagan heteroscedastic test	chi2(1) = 4.846, Pr = 0.0245

shows that the p value of the non-concurrent correlation test for the seemingly unrelated regression model is 0.0245, meaning that the “non-concurrent correlation test” for the seemingly unrelated regression model was rejected. The data analysis demonstrates that among the disturbance terms of the n equations under study, it conforms with the basic assumptions of the seemingly uncorrelated regression model proposed at the beginning of the paper.

5.3.2. Results’ Analysis

The results of the regression analysis are shown in

Table 10. Regression results’ analysis.

Variables	${\mathit{Y}}_{\mathit{M}}$ (Model 1)	${\mathit{Y}}_{\mathit{F}}$ (Model 2)	${\mathit{Y}}_{\mathit{H}}$ (Model 3)	${\mathit{Y}}_{\mathit{S}}$ (Model 4)	${\mathit{Y}}_{\mathit{N}}$ (Model 5)
Industrial assistance (X1)	0.079 ** (0.022)	0.108 ** (0.021)	0.094 (0.701)	0.147 (0.374)	0.133 (0.565)
Employment training (X2)	0.149 (0.190)	−0.179 (0.187)	0.170 *** (0.000)	0.116 (0.003)	0.121 (0.108)
Ecological compensation (X3)	0.038 (0.241)	0.035 *(0.078)	0.030(0.353)	0.024(0.441)	0.028 ***(0.000)
Economic organization (X4)	−0.020 (0.260)	0.017 *** (0.004)	0.017 ** (0.02)	0.016 *** (0.002)	−0.015 (0.087)
Social security (X5)	0.061 *** (0.007)	0.098 *** (0.000)	0.011 (0.598)	0.015 (0.304)	0.014 (0.473)
Transportation convenience (X6)	0.078 (0.446)	−0.084 (0.054)	0.068 *** (0.000)	0.056 ** (0.049)	0.053 ** (0.048)
Information network (X7)	0.215 (0.983)	0.216 ** (0.028)	0.189 (0.290)	0.176 ** (0.014)	0.161 (0.274)
Quality of farmland and related agricultural supporting facilities (X8)	0.076 (0.881)	0.074 (0.196)	0.086 (0.133)	0.112 (0.624)	0.112 *** (0.489)
Rural development level (X9)	0.038 *** (0.009)	0.036 (0.000)	−0.019 (0.904)	0.026 (0.186)	0.039 ** (0.031)
Market fluctuation (X10)	−0.022 (0.645)	−0.007 ** (0.012)	−0.017 (0.800)	−0.031 (0.638)	−0.024 ** (0.039)
Natural disaster (X11)	−0.101 (0.377)	0.103 (0.356)	−0.060 (0.601)	−0.103 (0.367)	−0.193 *** (0.004)
Constant	0.851 *** (0.000)	0.891 *** (0.000)	1.008 *** (0.000)	1.012 *** (0.000)	1.043 *** (0.000)
N	269	269	269	269	269
R2	0.276	0.291	0.321	0.371	0.374

Note: *, **, *** indicate significance at the levels of 10%, 5%, and 1%, respectively, with p values in parentheses.

. Policy support has a significant effect on the social, human, material, natural, and financial capital of the poverty-stricken population. In particular, industrial assistance and social security have a significant positive effect on the material capital and financial capital of the poverty-stricken population. Employment training has a significant positive effect on the social capital and human capital of the poverty-stricken population. Ecological compensation has a significant positive effect on the natural capital and financial capital of the poverty-stricken population. Supporting the development of economic organization has a significant positive effect on the social capital, human capital, and financial capital of the poverty-stricken population. In general, the government’s multifaceted policy support measures can increase the poverty-stricken households’ income channels and maximize their mode of subsistence, increasing their resilience to subsistence loss. However, there are obvious differences in the impact of different policies on the livelihood resilience of poverty-stricken population. For instance, direct support programs, such as cash subsidies for industrial aid policies, significantly improve the material well-being of poverty-stricken population, but they have little effect on the level of livelihood. Policy support for agricultural technology learning and employment training contributes to the improvement of livelihood resilience and internal motivation of poverty-stricken population.

From the perspective of regional endowment, transportation convenience has a significant positive effect on the natural capital, social capital, and human capital of the poverty-stricken population. The information network has a significant positive impact on the social capital and financial capital students of the poverty-stricken population. Convenient transportation and perfect information network provide the foundation for economic growth in places of extreme poverty. Poverty-stricken households’ space movement costs are lower and more favorable to raising the level of natural capital and human capital the closer they are to the regional developed center. The expansion of network coverage has increased the communication capabilities of families living in poverty, enabling them to obtain more useful agricultural information and strengthening the resilience of their means of subsistence. The financial capital, material capital, and natural capital of the poverty-stricken population are significantly positively impacted by the villages’ per capita net income. This demonstrates how certain industrial scales have developed in the areas of extreme poverty, encouraging the transfer of land to the people living there. The quality of farmland and related agricultural supporting infrastructure has a significant positive impact on the natural capital of poverty-stricken population. This indicates that significant elements influencing the agricultural production activities of poverty-stricken population include the quality of the cultivated land and the amount of agricultural infrastructure.

Among the livelihood risk factors, market fluctuation has a significant negative impact on the financial capital and natural capital of poverty-stricken population. Natural disasters have a significant negative impact on natural capital. Agriculture transaction costs will rise as a result of market fluctuation and livelihood risks, such as natural catastrophes. As a result of this, the short-term income of poverty-stricken households is drastically reduced, which has an impact on their ability to maintain their way of life.

6. Conclusions and Policy Recommendations

6.1. Conclusions

This paper constructs an index system for evaluating the livelihood resilience of the poverty-stricken population in the M Reservoir Area of Chongqing from five aspects: natural capital, material capital, human capital, social capital, and financial capital. Based on this, the principal component regression approach is used to measure the livelihood resilience of households experiencing poverty. Furthermore, we use the seemingly unrelated regression model in order to measure the impacts of industrial assistance, regional endowments, and livelihood risk on the livelihood resilience of poverty-stricken population. The main results are as follows.

In general, the Chongqing M Reservoir Area’s poor population’s livelihood resilience index steadily rises; however, it remains at a medium level overall. Different kinds of poor households have various degrees of livelihood resilience. Poverty-stricken households with early time of poverty alleviation, strong labor resource endowment, and low dependency ratio have a higher livelihood resilience index, and among them, the highest proportion of middle and high-level livelihoods.

We find that policy support has a significant positive effect on the livelihood resilience of poverty-stricken population. Among them, industrial support and social security have a significant positive effect on the material capital and financial capital of poverty-stricken households. Employment training has a significant positive effect on farmers’ social capital and human capital. Ecological compensation has a significant positive effect on the natural capital and financial capital of farmers. Supporting the development of economic organizations has a significant positive effect on the social capital, human capital, and financial capital of poverty households. Therefore, the first hypothesis is validated.

Moreover, the results show that regional endowments have a significant positive effect on the livelihood resilience of poverty-stricken population. Among them, transportation convenience has a significant positive effect on the natural capital, social capital, and human capital of poverty-stricken households. The information network has a significant positive impact on the social capital and financial capital of poverty-stricken households. The per capita net income of the village has a significant positive impact on the natural capital of poverty-stricken households. The quality of farmland and related agricultural supporting facilities has a significant positive effect on the natural capital of poverty-stricken households. Therefore, the second hypothesis is validated.

Finally, we find that the livelihood risk has a significant negative effect on the livelihood resilience of poverty-stricken population. Among the livelihood risk factors, market fluctuation has a significant negative impact on the financial capital and natural capital of poverty-stricken households. Natural disasters have a significant negative impact on natural capital. Therefore, the third hypothesis is validated.

6.2. Policy Recommendations

Improving the livelihood resilience of the poverty-stricken population is the basis for realizing rural revitalization in the Chongqing M Reservoir Area. Based on empirical research, this paper analyzes the livelihood status of the people in the Chongqing M Reservoir Area. Four policy suggestions are provided resulting from this research. First and foremost, relevant departments should strengthen employment skills training. Regular training in manufacturing or occupational skills must be provided by governments and social groups. Young people might receive training in relevant information and skills to increase their level of job passion. For those who have overcome poverty and choose to continue farming, basic agricultural knowledge and skills can be provided. For the middle-aged, elderly, and female labor force, their manual or traditional ethnic minority skills can be trained according to the advantages of local resources.

Second, policies should emphasize the empowerment of characteristic industries. Local government units should make the most of their resource advantages, strongly promote distinctive sectors, and offer avenues for raising people out of poverty through increased income. Entrepreneurs should assume an influential role, intensify their collaboration with agricultural science and technology innovation platforms, and collaboratively develop smart agricultural products. Departments should strengthen policy guidance and improve the interest connection mechanism.

Additionally, the governments should establish a natural disaster risk prevention mechanism. On the one hand, the government must improve the building of infrastructure for the dynamic surveillance of hazards and implement preventative actions beforehand. On the other hand, the relevant departments have to train the underprivileged people in risk awareness and catastrophe avoidance techniques in order to increase their understanding of the necessity for self-defense. Furthermore, the government should also establish a market risk prevention mechanism. Under the direction of the law of market development, poverty-stricken areas should optimize their own capital allocation and use industrial competition to raise the interest in industrial poverty alleviation and development. Through the help of the internet and help from all walks of life, the consumption of tourists’ poverty alleviation should be promoted.

6.3. Limitations and Future Research Directions

While this study provides an optimized development path for the livelihood resilience of poverty-stricken population in the Chongqing M Reservoir Area, it is important to note several limitations, which may affect the interpretation of the results. Firstly, numerous variables influence the poor population’s ability to sustain their livelihoods. This study only addresses three aspects—policy support, regional endowment, and livelihood risk—and may lack a comprehensive evaluation in constructing the index system. Secondly, the study only includes poverty-stricken population in the 15 counties of the Chongqing M Reservoir, which may limit the generalization of the study results to other regions or countries. Lastly, the survey method is a questionnaire survey, and the data resources are relatively limited. Further research could address these limitations by collecting more comprehensive and accurate data, using alternative variable methods, or conducting case studies.

Poverty-stricken populations are constantly exposed to different economic, social, and environmental hazards. People with livelihood vulnerability are extremely vulnerable to poverty. Government agencies at all levels should be aware of the danger, which those who got out of poverty run of falling back into. There are numerous directions for future research. Firstly, this study explores and optimizes the development path of poverty-stricken population from three aspects of policy support, regional endowment, and livelihood risk. However, there are many factors affecting the livelihood resilience of poverty-stricken population. Therefore, future research might examine the ways of enhancing the livelihood resilience of people suffering from poverty from a variety of angles. Secondly, how to achieve the long-term goal of sustainable development is also very important. In the future, the research can further explore how to avoid and resolve the interference and destruction of the stable state of poverty alleviation due to the existence of the risk of returning to poverty. It is hoped that more and more scholars will devote themselves to the research of people’s livelihood and contribute to the construction of a modern socialist country in the future.