Research on Satisfaction ofexsitu Relocated Households

College of Economics and Management, Huazhong Agricultural University, Wuhan 430070, China

Abstract Based on the survey data ofexsiturelocated households in Hubei and Hunan provinces in 2017, the specific satisfaction levels of the relocated farmers in the two provinces and the influencing factors were analyzed in this paper. It is found that theexsiturelocated farmers in Hubei and Hunan provinces have higher levels of satisfaction. The satisfaction of farmers who have been relocated to poverty alleviation is obviously affected by the characteristics of relocation and support policies. Even after controlling the farmer individual characteristic variables and family characteristic variables, the relocation characteristics and support policies still have a significant impact on the satisfaction of the farmers in the anti-poverty relocation, indicating that the relocated farmers with different backgrounds have a similar understanding of the satisfaction of theexsituanti-poverty relocation.

Key words Exsituanti-poverty relocation, Satisfaction, Hierarchy process analysis, Ordered logit model

1 Introduction

Theexsitupoverty alleviation project is not just simple migration of people or change of residence. Instead, it is a systemic project of economic restructuring and social reconstruction. Needless to say, after the relocation, the housing conditions of the farmers have been improved. However, there are also problems such as insufficient participation of poor households, too tight schedules for relocation work, and imperfect follow-up supporting industries, which may lead to the phenomenon that relocated households are "unwilling to move in" or "willing to move out after relocation". In the survey, it is found that after some poor households in the mountainous areas of western Hubei were relocated, they were far from the original production area. Due to lack of follow-up industry support, some relocated households still choose to ride a motorcycle or walk for more than 10 km to work in the field. Due to weak industrial base, coupled with insufficient workforce skills training, some of the relocated households are still narrow in the income-increasing areas, and their subsequent development is difficult. The infrastructure construction in the relocation areas lags behind, and the investment in environment sanitation, education and medical care is insufficient. These factors not only affect the relocation enthusiasm of the relocated households. They are more related to the expected policy effect of the relocation project. What is the satisfaction of the relocated households on theexsituanti-poverty relocation policy? Does the policy supply meet the actual needs of the relocated households? Has the relocation project reached the policy objective of "relocating one household making one household get rid of poverty"? How to improve the relocation policy afterwards? In addition to moving in, how to ensure that the relocated households are stable, secured and able to get rich? These issues are subject to further study.

Farmer’s adaptation to the natural and social environment of the resettlement site is the basis for their integration into local life. The satisfaction of the relocated farmers can reflect the impact of the entire relocation project on their lives and the overall affect of the project[1]. Therefore, the satisfaction ofexsituanti-poverty relocated households is a key factor in evaluating the success or failure of resettlement work. By exploring the main factors affecting the relocation satisfaction of farmers, the practical understanding ofexsitupoverty alleviation research will be deepened, the direction of reform and improvement of theexsitupoverty alleviation policy will be clarified. As China’s urbanization process accelerates, a large number of rural populations continue to gather in cities and towns, leading to a phenomenon of hollowing out and empty nesting in rural areas. Theexsituanti-poverty relocation project is to move the poor households in remote mountainous areas to the centralized resettlement communities with relatively good production and living conditions. On the basis of further promoting urbanization construction, it also provides a new research direction for rural population agglomeration.

2 Literature review

TheEleventhFive-YearPlanforex situAnti-PovertyRelocationissued by National Development and Reform Commission in 2006 proposes that the relocation of poverty-stricken areas is an important poverty alleviation measure that the party and the government have explored an implemented in the new era. By relocating the poor living in areas that are not suitable for human survival, the dual goal of eliminating poverty and improving the ecology will be achieved. Chen Minfengetal. believe thatexsituanti-poverty relocation, as one of the important modes of precision poverty alleviation, resettle the poor people living in the area that cannot afford the people to other places with better natural and infrastructure conditions, blocking the intergeneration transmission of poverty and creating conditions for them to get rid of poverty and get rich. It is essentially a poverty alleviation method for systematic relocation through comprehensive consideration of the survival cost, survival mode and living environment of poor groups[2]. From the experience of China’s ex situ anti-poverty relocation, the relocated households can be divided into two categories. One is "survival-oriented" relocated households of which the original living environment is poor and the original living is remote, or the original living areas are threatened by geographical disasters. The other is "development-oriented" relocated households facing poor housing conditions, difficult medical treatment and difficulties in attending school[3]. The current studies on theexsituanti-poverty relocation in China are mostly conducted using theoretical elaboration or theoretical elaboration combined with case analysis, lacking evidence from empirical research.

Based on the current research, the influencing factors of the satisfaction of farmers with theexsituanti-poverty relocation are roughly divided into five aspects: individual characteristics, family characteristics, economic factors (Yang Yongmeietal.[4]), social factors (Wang Jintao and Chen Qi) and policy factors. Zhang Yueetal.[5]studied the ecological relocation site in Jingling New Village in Ningxia and found that the three main factors that affect the satisfaction of the relocated groups are environment, ecology and social services. Chen Wenchaoetal.[6]believe that adaptation to the resettlement, the conditions of resettlement, and the cognition of managers are important factors influencing the satisfaction of reservoir resettlers. Taking the characteristics of relocated households, income and expenditure of relocated families, production conditions of relocation site, living conditions of relocation site, social conditions, resource conditions and environmental conditions as independent variables and the satisfaction of reservoir resettlers as depended variable, Qiu Yuanfeng found that the impact of children’s dependency ratio, old-age dependency ratio, per capita net income, per capita arable land area, cultivated land quality, cultivated land radius and irrigation conditions, as well as infrastructure and implementation of relocation polices on the satisfaction of reservoir resettlers is significant. Among the characteristic variables of relocated households, the age of householder, the health status of householder and the population of household have certain influence on the satisfaction of reservoir resettlers, but the impact is not significant. Among the variables of living conditions, the influence of housing material, and per capita housing area on reservoir resettlers’s satisfaction is not significant[7]. Feng Xiaotian studied the resettlers in the Three Gorges. It was found that in the early stage of relocation, the relocation method is the main factor affecting the social adaptation of resettlers. In the middle stage of relocation, the way of relocation, lifestyle differences and government concern are important factors. In the later stage of relocation, resettlers have become accustomed to the lifestyle of resettlement sites over time. Therefore, the relocation method is no longer the main influencing factors. Instead, the factors such as the security situation of the resettlement site and the time of relocation have become the main influencing factors.

3 Data sources,indicator selection and model selection

3.1DatasourceThe data of this paper comes from the survey data of the projects of National Development and Reform Commission (Assessmentofex situAnti-povertyRelocationWorkin2017) and the State Council Leading Group Office of Poverty Alleviation and Development (TracingResearchandImpactAssessmentofex situAnti-povertyRelocationduring13thFive-YearPlanPeriod) hosted by the School of Agriculture and Rural Development of Renmin University of China. The survey is based on the file establishment and card registration big data platform. Tow provinces, Hubei and Hunan provinces with lots of relocation work are selected. Three counties in Hubei Province are selected randomly, Jianshi, Badong and Zigui. Two counties in Hunan Province are selected, Xupu and Pingjiang. A total of 125 households are selected from each county. All sample households must correspond to the poverty alleviation households in the national poverty alleviation and development information system. In the sampling process, the villages and farmers who have been relocated in the past two years are selected as much as possible to facilitate follow-up monitoring after the relocation in the following years.

During the investigation, interviews and questionnaire surveys are conducted on all sample households in Hubei and Hunan provinces (except special circumstances such as the whole family are working in city) to learn more about the current situation of the relocated farmers and the actual demand for the relocation. A total of 558 pieces of valid data are collected. Because this study is for poor households who have already implemented relocation, 339 households that have been relocated are selected from the survey data for research.

3.2Indicatorselection

3.2.1Indicators for the satisfaction ofexsituanti-poverty relocated households. The five-level Likert scale is used to measure the satisfaction of the relocated households with the relocation policies. That is, the evaluation of the relocated households on the relocation work is divided into five levels, very unsatisfied, unsatisfied, general, satisfied, and very satisfied. The corresponding scores of the evaluation levels are 1, 2, 3, 4 and 5. On this basis, this paper constructs a policy evaluation index system for relocated farmers consisting of 4 grade II indicators (living conditions, infrastructure, neighborhood relations and support measures) and 16 grade III indicators. Finally, the weight of each indicator is calculated by the hierarchy process analysis. The importance of every two factors is compared by the professors and doctoral students from the College of Economic Management of Huazhong Agricultural University. After the consistency test, the judgment matrix is obtained. The weight of each indicator is calculated using the yaahp 10.1 software (Table 1). According to the weighed calculation method, the satisfaction of the relocated households is calculated.

The calculation results show that the overall satisfaction of the relocated farmers in the two provinces reaches a very high level. As shown in Table 2, the average score of evaluation of relocated farmers in the two provinces reaches 4.08 points. Among them, the individuals with evaluation ranging between "very unsatisfied" and "unsatisfied" account for 0%, and the individuals with evaluation ranging between "unsatisfied" and "general" account for 2.37%. The average score of satisfaction of relocated farmers in Hunan Province is slightly higher than that in Hubei Province. There are 10 farmers in Hunan Province but only one farmer in Hubei Province who scores five points for policy satisfaction.

Table1Weightsofsatisfactionindicators

Target layerCriteria layer indicators and weightsMeasure indicators and weights cjCombination weight diPolicy satisfaction yLiving conditions C1(0.393 5)House site selection D1(0.045 2)0.017 8House area D2(0.205 7)0.080 9House quality D3(0.267 3)0.105 2House design D4(0.214 4)0.084 4Water and electricity design D5(0.267 3)0.105 2Infrastructure C2(0.127 8)Road construction D6(0.319 9)0.040 9Garbage disposal D7(0.070 7)0.009 0Kindergarten D8(0.137 6)0.017 6Primary school and junior high school D9(0.168 9)0.021 6Hospital and clinic D10(0.108 4)0.013 8Supermarket and convenience store D11(0.115 6)0.014 8Leisure place D12(0.078 9)0.010 1Neighborhood relationship C3Familiarity D13(0.5)0.039 2(0.078 4)Help each other D14(0.5)0.039 2Support measures C4(0.400 3)Support measures D15(0.5)0.200 2Whether the support is in line with the actual situation D16(0.5)0.200 2

Table2Satisfactionofrelocatedfarmers

ProvinceNumber of peopleSatisfaction mean1≤Y<22≤Y<33≤Y<44≤Y<5Y≥5Hubei2093.9601.90%52.63%44.98%0.49%Hunan1304.2903.08%26.92%62.31%7.69%Total3394.0802.37%42.77%51.62%3.24%

3.2.2Influencing factors. (i) Individual characteristic variables of farmers. The individual characteristic variables of farmers mainly include age, gender, health status, marital status and years of schooling. Most of the relocated population lived in the mountainous areas, and the living environment and living conditions were poor. Young people could obtain more resources for survival and development through relocation, so the satisfaction is higher. The elderly are mainly engaged in farming. After relocation to other places, the cultivated land may be far away, leading to inconvenient production, which will reduce the satisfaction of the elderly with the relocation policies. In addition, young people are more capable of accepting new things than older people, and they are able to integrate into a new living environment faster. Farmers of different genders play different roles in family life and social production. Therefore, there may be differences in the level of attention and evaluation of policies. After the merger of administrative villages or the withdrawal of townships, the medical and health room has become far away from farmers, and seeing a doctor becomes more difficult for the farmers. By relocation, the farmers moved to a central village close to the clinic, so someone that is in poor health, or has a chronic illness or an emergency, may be highly satisfied with theexsituanti-poverty relocation policies. Generally speaking, married people usually have higher housing satisfaction[8], so their relocation satisfaction is not low. The level of education plays a decisive role in the knowledge and ability of workforce. Therefore, farmers with different levels of education may have different views on the relocation policies. Farmers with higher education have strong non-agricultural skills, and they are more inclined to engage in non-agricultural industries and need more employment development opportunities. Therefore, they tend to have higher satisfaction with the relocation policies. It is expected that the two have a proportional relationship[9].

(ii) Family characteristic variables. The family characteristic variables mainly include household per capita income, degree of concurrent employment, grade difference, distance from the relocation site to the recent market,etc. Generally speaking, the higher the income of the family is, the higher the requirements for production and living conditions are. After the relocation, the production and living conditions of the farmers have been improved, so the satisfaction is higher. The labor force of the families with low level of concurrent employment is mainly engaged in non-agricultural industries. Land income is no longer the main source of household income. The original residences are generally remote with congested traffic and poor living conditions. In contrast, the relocation sites are close to towns and have convenient transportation, more labor opportunities, so farmers are more willing to move out[10]. In our society, some groups are on the top floor, while others are on the ground floor. It is divided into 10 grades from the lowest level to the highest level. The relocation grade level minus the original grade level is the difference. The greater the grade difference is, the higher the satisfaction is. The closer the distance to the nearest market is, the more convenient the farmer’s life is, and the higher the satisfaction is.

(iii) Relocation policy variables. The relocation policy variables mainly include resettlement methods, construction methods, traffic distance before and after relocation (that is, the distance of transportation from the original residence to the relocation site), whether the resettlement houses can be rented and remodeled, and relocation policy negotiation. The resettlement methods include centralized resettlement and scattered resettlement. The housing construction methods include unified planning and unified construction, unified planning and self construction, decentralized self construction and purchase of commercial housing. From the data point of view, the survey area only has two construction methods, unified planning and unified construction, and unified planning and self construction, In terms of relocation distance, if the relocation site is too far away from the original residence, it may cause inconvenience to farmers to increase their farming costs, and cause the original social network to be disrupted, forcing relocated farmers to face new social capital construction problems in the relocation site. These problems will lead to a decline in relocation satisfaction[11].Exsituanti-poverty relocation is a complex system project. Poor farmers, as support object, affect the process of resource allocation and the promotion of poverty alleviation[12]. In the process of policy implementation, expanding the participation of farmers in policy implementation and negotiation has a significant positive impact on farmers’ policy satisfaction[13]. Therefore, the number of relocation-related propaganda and mobilization activities that the county, township and village organizations have organized is selected in this study as the proxy variable for investigating the degree of consultation on the anti-poverty relocation policies. In addition, whether the resettlement hosing can be rented, remodeled,etc. may has an impact on the satisfaction of the farmers with theexsituanti-poverty relocation policies.

(iv) Support policy variables. Support policy variables refer to the policy assistance measures currently being implemented by the government. This paper mainly examines the poverty alleviation policies that have a greater impact, such as poverty alleviation microfinance and industrial development projects. Poverty alleviation microfinance is a poverty alleviation loan product tailored to poor households to obtain development funds. It mainly provides poverty alleviation loans to poor households with a loan less than 50 000 yuan, less than three years, no guarantee, no mortgage, benchmark interest rate lending, fiscal interest subsidy and county-level risk compensation. To a certain extent, poverty alleviation microfinance can alleviate the financial constraints of poor farmers’ development[12]. Industrial development projects include development of poverty alleviation industries such as crop farming, aquaculture and rural tourism. Industrial poverty alleviation is a proven initiative to help poor households achieve poverty alleviation.

3.2.3Model selection. Referring to existing research, to avoid the "linear" effects of individual characteristics on the dependent variable in the model and avoid the identification problem[14], the satisfaction evaluation of the relocated farmers involved in theexsituanti-poverty relocation is regarded as an ordered multi-category variable. Combined with the specific calculation results of the overall satisfaction, there is

Therefore, this paper uses the multi-variate ordered logit model to analyze the influencing factors of the relocated farmers’ satisfaction. The basic form of the multi-variate ordered logit model is as follows:

where,prepresents the probability that the farmer is satisfied with theexsituanti-poverty relocation policies;Xirepresents the factor that affects the satisfaction of relocated farmers;βirepresents the extent and direction of the influence of each independent variable on the dependent variable.

According to the above basic model, the empirical model is obtained as follows:

PSi-αXI+μ

where,PSrepresents the policy satisfaction variable of the relocated farmer;Xrepresents the independent variables, mainly including individual characteristic variables of farmers, family characteristic variables, relocation policy variables, and support policy variables. The data is shown in Table 3.

4 Model estimation results and analysis

4.1ModelestimationresultsThe estimation results of the measurement model for the factors affecting the satisfaction of theexsituanti-poverty relocated farmers are shown in Table 4. The regression 1 and regression 2 are the estimated results of the ordered logit model. As this paper focuses on the impact of relocation characteristics and policy characteristics on relocated farmer’s satisfaction, the regression 1 only includes the relocation characteristic variables and policy characteristic variables. The regression 2 adds the individual characteristic variables and family characteristic variables on the basis of regression 1. In contrast, the regression 3 estimates all explanatory variables using the ordinary least squares method. As shown in Table 4, the estimation results of the ordered logit model and ordinary least squares are very close. The sign of each explanatory variable coefficient estimate is basically the same. Before and after the control variable, the saliency of relocation characteristics and policy characteristics is almost unchanged. At the same time, the model passes the multicollinearity test. The variances of maximum expansion factor and minimum tolerance are 1.425 and 0.702, respectively. All of them are within reasonable ranges, indicating that the estimation results of this paper have strong robustness. Taking the results of the ordered logit model regression 2 as an example, the impact of each explanatory variable on the satisfaction of farmers with theexsituanti-poverty relocation is analyzed.

Table3Variabledefinitionanddescriptivestatistics

Variable typeVariable nameVariable definitionMinimumMaximumMeanStandarddeviationFarmer individual AgeActual age (years)158354.5912.28characteristic variablesGenderMale=1, female=0010.660.47Health statusHealthy=1, others=0010.450.50Marital statusMarried=1, others=0010.750.44Years of schoolingDuration of formal education (years)0144.773.23Family characteristic Per capita incomeHousehold per capita income (1 000 yuan)-0.3950.855.915.39variablesDegree of concurrent employmentHousehold agricultural population/(house-hold agricultural population+1)00.80.370.22Grade differenceGrade difference between relocation site and original residence-591.522.04Distance from relocation site to the nearest marketTransportation distance from relocation site to the nearest market (km)0406.276.93Relocation policyvariablesRelocation methodsCentralized resettlement=1, scattered reset-tlement=0010.870.33Construction methodsUnified planning and unified construction=1, unified planning and self construction=0010.740.44Distance from relocation site to original residenceTransportation distance from relocation site to original residence (km)0302.763.26Whether resettlement housing can be rentedYes=1, no=0010.180.38Whether resettlement housing can be remodeledYes=1, no=0010.390.49Number of relocation-related meet-ings organized by county, township and village Number of activities organized by the county, township, and village to promote the mobili-zation and plan collection of relocation0305.413.94Support policyvariablesWhether to participate in poverty alleviation microfinanceYes=1, no=0010.420.49Whether to participate in industrial development project supported by the governmentYes=1, no=0010.420.49

Table4Modelestimationandresultsoffactorsinfluencingsatisfactionofrelocatedfarmers

Independent variablesRegression 1(ordered logit)Regression 2(ordered logit)Regression 3(OLS)Age-0.023(0.012)0.005(0.003)Gender-0.236(0.283)0.041(0.060)Health status-0.252(0.265)0.053(0.058)Marital status-0.375(0.305)0.090(0.066)Years of schooling--0.153(0.045)-0.033(0.010)Per capita income-0.025(0.027)0.003(0.005)Degree of concurrent employment-0.370(0.592)0.067(0.128)Grade difference-0.070(0.076)0.013(0.015)Distance from relocation site to the nearest market-0.020(0.017)0.014(0.008)Resettlement method0.834(0.368)0.762(0.394)0.151(0.087)Construction method-1.018(0.319)-1.042(0.352)-0.215(0.071)Distance from relocation site to original residence0.030(0.037)0.061(0.040)0.004(0.004)Whether resettlement housing can be rented -0.074(0.302)0.273(0.324)0.062(0.073)Whether resettlement housing can be remodeled-0.209(0.242)-0.193(0.257)-0.031(0.056)Number of relocation meetings organized by county, township and village0.092(0.040)0.091(0.043)0.016(0.007)Whether to participate in poverty alleviation microfinance0.287(0.245)0.627(0.273)0.138(0.058)Whether to participate in industrial development project supported by government0.970(0.259)1.010(0.280)0.212(0.060)Constant term--2.207(0.215)Sample size339339339Log likelihood value447434-Quasi R2 or R20.1160.1810.239

Note:*,**and***indicate significant differences at the levels of 10%, 5% and 1%, respectively.

4.2Interpretationofestimatedresults

4.2.1Farmer individual characteristics and family characteristics. Among the farmer individual characteristics and family characteristics, only age and years of schooling have a significant impact on the satisfaction of relocated farmers, while other factors have no significant impact on the satisfaction of relocated farmers. The impact of the number of years of education on the satisfaction of relocated farmers is not in line with expectations. That is, as the number of years of education increases, the satisfaction of farmers is reduced. This may be because that the knowledge of farmers with higher education is higher, and they have stricter policy requirements and are not easily satisfied. It also explains the importance of negotiating with farmers during policy implementation. The farmers’ age coefficient is positive, indicating that farmers who are older are more satisfied with theexsituanti-poverty relocation. On one hand, as individuals grow older, they will adjust their evaluation of policies and lower their expectations, so they are more likely to be satisfied with the relocation policies. On the other hand, among the survey samples, the older the farmer is, the lower the education period is. Among farmers with an average age of 55 years and older, the farmers with maximum education period of 5 years account for about 14%, and the uneducated farmers account for about 37%. Therefore, age has a significant positive impact on the satisfaction of farmers withexsituanti-poverty relocation. However, the impact of per capita income on the satisfaction of farmers with relocation is not significant. This may be because that the samples surveyed are all poor households with established files and registered cards, and the per capita income of the households is at a low level. The difference in the social rank of the farmers between original residence and relocation site has no significant impact on the satisfaction of farmers withexsituanti-poverty relocation. The possible explanation is that although theexsituanti-poverty relocation policies have improved the living environment of poor households and improved their production conditions, it cannot improve the social rank of poor households in the short term. This requires the development of follow-up support policies to improve the "hematopoietic" function of the poor households.

4.2.2Relocation policies. The construction methods of resettlers, resettlement methods and the number of relocation meetings organized by the government have a significant impact on the satisfaction of theexsituanti-poverty relocated farmers. Farmers with centralized resettlement have higher satisfaction. This may be due to the fact that compared with scattered settlement, centralized resettlement has relatively complete infrastructure and service facilities, and it can effectively improve the living environment of poor households, meeting the basic needs of the relocated farmers for poverty alleviation and resettlement[15-16]. In terms of the way of resettlement houses are constructed, the farmers under the mode of unified planning and self construction have higher satisfaction with theexsituanti-poverty relocation. The unified planning and self construction mode can ensure the construction of infrastructure and public service facilities in the resettlement area and improve the participation of farmers in theexsituanti-poverty relocation. Therefore, farmers who participate in this type of construction have higher satisfaction. The number of relocation meetings organized by the government has a significant positive impact on the satisfaction of farmers withexsituanti-poverty relocation. When implementing theexsituanti-poverty relocation policies, more village meetings can significantly improve farmers’ satisfaction.

4.2.3Support policies. The estimated coefficients of "whether to participate in poverty alleviation microfinance" and "whether to participate in government-supported industrial development projects" are significantly positive. It indicates that the follow-up support policies have had a significant impact on the satisfaction of farmers withexsituanti-poverty relocation.

5 Conclusions and discussion

On the basis of the survey data of relocated farmers in Hubei and Hunan provinces in 2017, this paper conducts a comprehensive analysis on the specific satisfaction levels of relocated farmers in the two provinces and the influencing factors. The study finds that the satisfaction ofexsituanti-poverty relocated farmers is obviously affected by the relocation polices and support policies. Even after controlling the individual characteristic variables and family characteristic variables, the relocation policies and support policies still have a significant impact on the satisfaction of farmers with theexsituanti-poverty relocation, indicating that the relocated farmers of different backgrounds have a similar understanding on the satisfaction withexsituanti-poverty relocation.

Considering the specific influencing factors of the satisfaction of relocated farmers, in the implementation process of theexsituanti-poverty relocation policies, consultation with farmers from a realistic perspective can effectively improve the policy satisfaction of farmers and improve the quality of policy implementation. On the basis of strict control standards, it is necessary to formulate resettlement housing construction methods according to specific requirements of households. The resettlement methods should be dominated by centralized resettlement, supplemented by scattered resettlement. The sense of participation of the relocated households should be increased. Finally, the follow-up support policies after the relocation should be strengthened. The form of "cooperative+" can be adopted to develop poverty alleviation industries to ensure stable poverty alleviation.