Research Article |
Corresponding author: Wei Xiao ( xiaow@swufe.edu.cn ) © 2023 Non-profit partnership “Voprosy Ekonomiki”.
This is an open access article distributed under the terms of the Creative Commons Attribution License (CC BY-NC-ND 4.0), which permits to copy and distribute the article for non-commercial purposes, provided that the article is not altered or modified and the original author and source are credited.
Citation:
Li Y, Jia H, Xiao W, Naumov AS (2023) Spatial segregation and human capital of impoverished areas in China: Implications for livelihood resilience building. Russian Journal of Economics 9(4): 424-439. https://doi.org/10.32609/j.ruje.9.108719
|
Improving people’s livelihood resilience against risks and challenges plays an important role in consolidating the achievements of poverty reduction. The paper uses 64 poverty-stricken counties in China’s Sichuan province as the study area and explores the link between spatial segregation and human capital. The results show that the proximity (spatial segregation) is significantly and negatively associated with people’s educational attainment and their acquisition of non-farming employment. Residents in villages which are distant from the county center tend to obtain less educational opportunities and are less likely to engage in non-farming jobs than those who are close to the county center. The mediating effect analysis indicates that remoteness mainly reduces the propensity of getting non-farming jobs by reducing the human capital of rural residents. Further analysis shows that the association between proximity, human capital and the probability of acquiring non-farming work is higher in areas with lower economic level and less developed transportation infrastructure. Policy implications for improving people’s livelihood resilience in impoverished areas are proposed in the paper.
spatial segregation, human capital, rural poverty, livelihood resilience, China
Human society has achieved great progress in poverty reduction during the past decades though various challenges still exist which may set back this cause. The World Bank statistics show a reduction of 1.22 billion people out of poverty (below $1.90 a day at 2011 Purchasing Power Parity) in the period 1990–2017.
Being the most populous country in the world, China has been fighting poverty, in tandem with the state promoting its long-term economic growth, since 1949 when P. R. China was founded. In this process, a series of measures were taken towards poverty reduction. In particular, China initiated the targeted poverty alleviation strategy in late 2013 and identified 70.17 million rural impoverished inhabitants who are living below the national poverty line (RMB 2,300 per capita annual net income, equivalent to $314). By the end of 2020, all this impoverished population shook off poverty. In the meantime, a total of 832 nationally designated poor counties and 128,000 impoverished villages rose up from poverty. The complete eradication of extreme poverty in China is 10 years ahead of the schedule to accomplish the UN’s no-poverty goal by 2030. This campaign has led to dramatic changes in the lives of the impoverished and laid the foundations for overall development of poverty-stricken areas in the future.
Decades of research have illustrated the multidimensional fact of poverty which closely relates to people’s status in terms of material deprivation, social isolation, exclusion and powerlessness, and physical and psychological ill-being (
In early 2021, China set a five-year transitional period (2021–2025) to consolidate and expand the achievements of poverty reduction, and raise the overall effect of development in areas that have cast off poverty. Improving poor households’ livelihood resilience against challenges and shocks has been emphasized to reduce the risks of reverting to poverty. Generally, the vulnerable groups are those in China’s west and southwest mountainous areas which are characterized by less developed economy, backward infrastructure, inadequate public services, and social and economic underdevelopment. Particularly, the spatial location of those mountainous poor areas has segregated people from reaching the outside world. Our concern lies in the impact of spatial segregation on people’s access to education, and how human capital of the poor influences their access to jobs that would boost their livelihood resilience and provide a stable route out of poverty.
The paper takes Sichuan province as the study area, a region in the southwest of China that has profound poverty. There were 6.25 million poor people in Sichuan province by the end of 2013, accounting for 7% of the total poor people of China. The poverty incidence of this province was 9.6% which was higher than in many other places in China. The aim of the paper is to investigate the logic between spatial segregation and human capital accumulation as well as livelihood resilience of the poor. The structure of the paper is as follows. Section 2 introduces China’s targeted poverty alleviation and the relationship between spatial segregation and human capital accumulation as well as people’s livelihood resilience. In Section 3, the paper explains in detail the research area, methodology and data sources. Section 4 presents the research findings which are followed by the discussion and conclusion of the paper.
China has a total population of 1.4 billion people who are spatially distributed in a vast territory of diversified socioeconomic and geographical conditions. Besides its weak socioeconomic foundations and uneven territorial development, China has long been bedeviled by poverty. China’s fight against it entered a critical stage in 2012 as the nation endeavored to accomplish its First Centenary Goal to build a moderately prosperous society in all respects by 2020. No single poor area or single poor person should be left behind in achieving this goal—that has been particularly emphasized by the Chinese government. Then, the targeted poverty alleviation strategy was initiated and implemented which highlights the importance of correct poverty identification, appropriate projects arrangement and accurate implementation effect to ensure that the assistance reaches poverty-stricken villages and households (
The contribution of China’s targeted poverty alleviation to rural development is multifaceted (Fig.
Eradicating poverty through education (including professional training) has been highly emphasized in China with the aim of improving poor people’s human capital and their livelihood resilience in the long run. The Chinese governments have invested much in education, and people’s health care in the impoverished areas. As a result, 108,000 primary schools have been renovated since 2013 to strengthen the provision of compulsory education in poor areas, and ensure that all school-age rural children receive kindergarten and elementary education within their own villages. More than 8 million poor households were offered professional education and training. All these efforts and input have helped the poor to obtain greater knowledge and skills, and enabled them to get better paid jobs through improved human capital. Statistics show that the number of poor people who get employment or start their own business increased from 12.27 million in 2015 to 32.43 million in 2020. This has greatly contributed to the increase of impoverished people’s income. The per capita disposable income of the rural poor increased from RMB 6,079 (equivalent $831) in 2013 to RMB 12,588 (equivalent $1,721) in 2020, up by 11.6% per annum on average.
As
The concept of resilience has gained people’s attention in recent decades as human society is facing climatic, economic, and social changes and unforeseen challenges which have caused huge losses in terms of human lives and prosperity (
Generally, resilience indicates the capacity of a system to absorb stresses and disturbances while maintaining or improving essential properties and functions (
This study targets all the poor counties in Sichuan province which is located in the Sichuan Basin in Southwest China (Fig.
In this study, we proceed from the fact that people’s educational attainment is the key way to human capital accumulation which further influences their occupation acquisitions and ways of livelihood. The more education an individual has obtained, the more stable and better-off livelihood he can attain. To examine the effect of spatial segregation of villages on residents’ educational attainment and occupational acquisitions, we estimate the following equation:
Deptivc = α + β 1 Dist_countyvc + β 2 Dist_prefecturevc + θXivc + countyc + εivc, (1)
where the subscript ivc indicates that the variable is for an individual i from village v in county c. The dependent variable Deptivc represents two outcomes. The first outcome is Eduivc, the educational attainment of individual i, including the number of educational years and whether the individual has completed junior high school education or senior high school education. The second outcome is a dummy variable Occupivc that is equal to 1 if individual i was engaging in a non-farming job in 2015. The key explanatory variables Dist_countyvc and Dist_prefecturevc are the logarithmic distance from village v to the center of its county and prefecture, respectively. Xivc represents a group of individual characteristics affecting educational attainments, such as gender, age, age squared, and a dummy Han for the Han ethnic group. The county fixed effects help remove the effects of county-level characteristics, such as the spatial advantage, social conventions, the average level of educational development, the number of schools, and the teacher-student ratio. Adding the county fixed effect in the regression helps us control all these regional factors.
Due to the availability of data, we select 64 poverty-stricken counties as the study area. Our data includes both individual and county level data. Individual information is extracted from the mini-census conducted by the National Bureau of Statistics of China in 2015 when the country was in the process of promoting the targeted poverty alleviation. The 2015 mini-census collected a wealth of demographic information such as gender, age, educational level, and employment status of 1.55% of China’s population. Due to their small population size, some poor counties in Sichuan province were not covered by the 2015 mini-census.
This study aims to explore the association between spatial segregation and educational attainment of rural residents in Sichuan province that includes a large group of impoverished counties. To measure the spatial segregation, we calculate the spatial distance from a village to the center of the county, or prefecture, which is administratively managing the village and county. We only keep rural residents between the age 15 and 60 and exclude individuals who were still in school in 2015. We also exclude villages with less than 20 observations. The GDP of counties is collected from the China Statistical Yearbook (County-level). Some counties are not included in the data due to their small population size. The total amount of the research sample is 13,009 individuals from 653 villages located in 64 impoverished counties in Sichuan province.
Table
Variables | N | Mean | Std. dev. | Min | Max |
Village characteristics | |||||
Ln (Distance to the county center, km) | 13,009 | 2.5927 | 1.1491 | 0.0630 | 4.3786 |
Ln (Distance to the prefecture center, km) | 13,009 | 4.1795 | 0.6381 | 1.9473 | 6.0910 |
Educational attainment | |||||
Educational years | 13,009 | 7.7948 | 3.8313 | 0 | 19 |
Junior high school or above | 13,009 | 0.5666 | 0.4956 | 0 | 1 |
Senior high school or above | 13,009 | 0.1786 | 0.3830 | 0 | 1 |
Employment information | |||||
Non-farming job | 10,545 | 0.4677 | 0.4990 | 0 | 1 |
Individual characteristics | |||||
Male | 13,009 | 0.5125 | 0.4999 | 0 | 1 |
Han nationality | 13,009 | 0.7144 | 0.4517 | 0 | 1 |
Age (years) | 13,009 | 38.3405 | 11.6570 | 15 | 60 |
Age squared | 13,009 | 1605.8710 | 895.5428 | 225 | 3600 |
County characteristics | |||||
Ln (GDP, RMB) | 13,009 | 13.2251 | 1.0155 | 10.9164 | 14.7946 |
Highway | 13,009 | 0.5452 | 0.4980 | 0 | 1 |
Railway | 13,009 | 0.2194 | 0.4138 | 0 | 1 |
By estimating Equation (1), Table
We further study the association between the distance and educational outcomes by considering whether residents have completed junior or senior high school education. Table
Results in Tables
Spatial segregation of villages and educational attainment: educational years.
Variable | Educational years | ||||
(1) | (2) | (3) | (4) | (5) | |
Ln (Distance to the county center) | –0.7146*** (0.0859) | –0.6925*** (0.0858) | –0.6912*** (0.0864) | ||
Ln (Distance to the prefecture center) | –0.3422 (0.3084) | –0.2530 (0.2929) | –0.1469 (0.3034) | ||
Male | 0.8887*** (0.0514) | 0.8516*** (0.0523) | 0.8891*** (0.0514) | ||
Han nationality | 1.9683*** (0.2775) | 2.4024*** (0.2818) | 1.9598*** (0.2782) | ||
Age | –0.0181 (0.0185) | –0.0089 (0.0192) | –0.0183 (0.0184) | ||
Age squared | –0.0011*** (0.0002) | –0.0012*** (0.0002) | –0.0011*** (0.0002) | ||
County fixed effects | Y | Y | Y | Y | Y |
N | 13,009 | 13,009 | 13,009 | 13,009 | 13,009 |
R squared | 0.2844 | 0.2482 | 0.4082 | 0.3750 | 0.4083 |
Spatial segregation of villages and educational attainment: junior and senior high school.
Variables | Junior high school or above | Senior high school or above | ||||||
(1) | (2) | (3) | (4) | (5) | (6) | |||
Ln (Distance to the county center) | –0.0678*** (0.0082) | –0.0675*** (0.0082) | –0.0665*** (0.0087) | –0.0665*** (0.0088) | ||||
Ln (Distance to the prefecture center) | –0.0416 (0.0283) | –0.0312 (0.0277) | –0.0129 (0.0342) | –0.0027 (0.0355) | ||||
Individual characteristics | Y | Y | Y | Y | Y | Y | ||
County fixed effects | Y | Y | Y | Y | Y | Y | ||
N | 13,009 | 13,009 | 13,009 | 13,009 | 13,009 | 13,009 | ||
R squared | 0.3538 | 0.3350 | 0.3540 | 0.1510 | 0.1201 | 0.1510 |
This section further explores the association between spatial segregation, human capital accumulation and people’s occupational acquisitions by running the regression of the Equation (1). According to Column (1) of Table
Column (3) focuses on the relationship between residents’ educational attainment and their acquisition of non-farming work. We find that educational years are significantly and positively associated with people’s acquisition of non-farming work. Keeping other things equal, a unit increase in educational years results in a 0.0515 unit increase in people’s acquisition of non-farming work.
Column (4) shows that after adding educational years into the regression, the estimated coefficient of the villages’ distance to the county center becomes smaller than that in Column (1). This indicates that educational attainment is likely to be a mediating factor in the association between spatial segregation and rural residents’ acquisition of non-farming work. This means that improving human capital through enabling long educational years of people in remote villages is an effective way to increase their chances to obtain higher paid non-farming work. This is an effective way to overcome the impact of spatial segregation and improve people’s livelihood resilience in impoverished regions in the long run.
Spatial segregation of villages, educational attainment and acquisition of non-farming work.
Variables | Non-farming work | Educational years | Non-farming work | Non-farming work |
(1) | (2) | (3) | (4) | |
Ln (Distance to the county center) | –0.0934*** | –0.7498*** | –0.0583*** | |
(0.0120) | (0.0963) | (0.0099) | ||
Ln (Distance to the prefecture center) | 0.0453 | –0.1318 | 0.0515 | |
(0.0428) | (0.3352) | (0.0353) | ||
Educational years | 0.0515*** | 0.0469*** | ||
(0.0019) | (0.0020) | |||
Individual characteristics | Y | Y | Y | Y |
County fixed effects | Y | Y | Y | Y |
N | 10,545 | 10,545 | 10,545 | 10,545 |
R squared | 0.2679 | 0.4261 | 0.3303 | 0.3432 |
Herein, we further explore whether the association between spatial segregation, educational attainment, and acquisition of non-farming work shows any variation among different groups of individuals and counties. Table
Panel B compares the young and old age cohorts by dividing the sample into a younger group (aged 15 to 44) and an elder group (aged 45 to 60). The verified association between spatial segregation, educational attainment and people’s acquisition of non-farming work shows little heterogeneity between the young and old groups.
Table
The construction of transport infrastructure improves the connection of a county with larger markets and may affect the association between the distance to the county center, educational attainment and people’s acquisition of non-farming work. Panel B divides the sample counties into two groups according to whether a county opened railways in 2015. Columns (1) and (5) in Panel B show that the negative association between the distance to the county center and educational years is stronger for counties without a railway. This implies that railways may enlarge the gap of people’s educational attainment between the villages close to the county center and the remote ones. Moreover, the negative association between the distance to the county center and acquisition of non-farming work is mediated by educational attainment in a similar way for counties with and without railways.
Variables | Panel A: Gender group | |||||||||
Male sample | Female sample | |||||||||
Non-farming work | Educational years | Non-farming work | Non-farming work | Non-farming work | Educational years | Non-farming work | Non-farming work | |||
Ln (Distance to the county center) | –0.0895*** (0.0119) | –0.7345*** (0.0972) | –0.0548*** (0.0100) | –0.0956*** (0.0129) | –0.7396*** (0.1005) | –0.0609*** (0.0110) | ||||
Educational years | 0.0518*** (0.0022) | 0.0472*** (0.0024) | 0.0515*** (0.0021) | 0.0469*** (0.0022) | ||||||
Individual characteristics | Y | Y | Y | Y | Y | Y | Y | Y | ||
County fixed effects | Y | Y | Y | Y | Y | Y | Y | Y | ||
N | 5,698 | 5,698 | 5,698 | 5,698 | 4,847 | 4,847 | 4,847 | 4,847 | ||
R squared | 0.2610 | 0.3873 | 0.3205 | 0.3318 | 0.2712 | 0.4715 | 0.3366 | 0.3511 | ||
Variables | Panel B: Age group | |||||||||
Younger sample (age 15 to 44) | Elder sample (age 45 to 60) | |||||||||
Non-farming work | Educational years | Non-farming work | Non-farming work | Non-farming work | Educational years | Non-farming work | Non-farming work | |||
Ln (Distance to the county center) | –0.0871*** (0.0120) | –0.7753*** (0.0970) | –0.0509*** (0.0101) | –0.1084*** (0.0161) | –0.6646*** (0.1232) | –0.0761*** (0.0132) | ||||
Educational years | 0.0508*** (0.0021) | 0.0466*** (0.0022) | 0.0537*** (0.0026) | 0.0486*** (0.0028) | ||||||
Individual characteristics | Y | Y | Y | Y | Y | Y | Y | Y | ||
County fixed effects | Y | Y | Y | Y | Y | Y | Y | Y | ||
N | 7,234 | 7,234 | 7,234 | 7,234 | 3,311 | 3,311 | 3,311 | 3,311 | ||
R squared | 0.2842 | 0.4354 | 0.3491 | 0.3588 | 0.2338 | 0.3895 | 0.2906 | 0.3124 |
Variables | Panel A: per capita GDP in 2015 | |||||||||
Lower per capita GDP sample | Higher per capita GDP sample | |||||||||
Non-farming work | Educational years | Non-farming work | Non-farming work | Non-farming work | Educational years | Non-farming work | Non-farming work | |||
Ln (Distance to the county center) | –0.0911*** (0.0186) | –0.7631*** (0.1443) | –0.0587*** (0.0160) | –0.0957*** (0.0154) | –0.7434*** (0.1231) | –0.0585*** (0.0124) | ||||
Educational years | 0.0469*** (0.0028) | 0.0425*** (0.0031) | 0.0549*** (0.0026) | 0.0499*** (0.0027) | ||||||
Individual characteristics | Y | Y | Y | Y | Y | Y | Y | Y | ||
County fixed effects | Y | Y | Y | Y | Y | Y | Y | Y | ||
N | 5,067 | 5,067 | 5,067 | 5,067 | 5,478 | 5,478 | 5,478 | 5,478 | ||
R squared | 0.2921 | 0.5001 | 0.3408 | 0.3527 | 0.2460 | 0.3436 | 0.3173 | 0.3312 | ||
Variables | Panel B: Railway in 2015 | |||||||||
Railway sample | No-railway sample | |||||||||
Non-farming work | Educational years | Non-farming work | Non-farming work | Non-farming work | Educational years | Non-farming work | Non-farming work | |||
Ln (Distance to the county center) | –0.1272*** (0.0325) | –0.6733*** (0.2194) | –0.0981*** (0.0300) | –0.0900*** (0.0129) | –0.7595*** (0.1046) | –0.0538*** (0.0104) | ||||
Educational years | 0.0485*** (0.0043) | 0.0431*** (0.0041) | 0.0523*** (0.0021) | 0.0477*** (0.0023) | ||||||
Individual characteristics | Y | Y | Y | Y | Y | Y | Y | Y | ||
County fixed effects | Y | Y | Y | Y | Y | Y | Y | Y | ||
N | 2,501 | 2,501 | 2,501 | 2,501 | 8,044 | 8,044 | 8,044 | 8,044 | ||
R squared | 0.1999 | 0.3188 | 0.2232 | 0.2455 | 0.2878 | 0.4399 | 0.3607 | 0.3729 |
Modern China represents a very special case of rural development, maybe unique in the world, due both to the scale and the speed of changes affecting its huge and populous rural areas.
The targeted poverty alleviation strategy in China has lifted 70.17 million rural population out of poverty after years of intensive input and assistance. A big concern and work focus in the post-poverty era are to consolidate the achievements of poverty reduction and avoid people’s return to poverty due to unexpected risks and shocks. In this process, improving poor households’ livelihood resilience against risks and shocks through accumulated human capital plays an important role in strengthening people’s capacity for pursuing better-off livelihood. This is also an important way to shift from external assistance to poor people’s self-driven development through increased endogenetic power.
Our study presents evidence of the links between spatial segregation, human capital and people’s livelihood resilience in the impoverished regions of China. The research findings show that spatial segregation is significantly and negatively associated with people’s educational attainment and their acquisition of non-farming work. Thus, rural residents living in villages farther from the county center tend to accumulate less human capital and are less likely to obtain non-farming work than those who are closer to the county center. The mediating effect analysis further shows that spatial isolation mainly reduces the propensity of people’s access to non-farming jobs by reducing their human capital accumulation. As junior or senior high schools are mainly located in the central areas of counties, the distance from villages to high schools implies the cost of rural people’s educational attainment. The longer the distance to the county center, the less educational attainment people may obtain. And this in turn decreases the chances of people’s acquisition of non-farming employments which are normally located in downtown areas of each county. For the impoverished counties in our sample, the role of spatial segregation in affecting people’s human capital accumulation and access to non-farming jobs could be even more significant if the county has a lower economic level and less developed transportation infrastructure.
In today’s China, economic growth places more emphasis on the input of knowledge and demands more supply of innovative and skilled laborers. As for livelihood resilience of people living in remote areas, the policy implications based on our research findings are three folded. First, the transportation infrastructure between villages and county centers needs to be constructed and improved to decrease spatial segregation and improve people’s access to education, employments and services. Second, more input is needed to guarantee the supply of rural education to remote villages by building schools, improving teaching facilities, and attracting teachers. Third, the county economy in those impoverished regions needs to be further developed to offer people more job opportunities and attract a highly qualified labor force.
This study still has some deficiencies. Due to data availability, we cannot accurately calculate the actual traveling time from a village to the center of a county or prefecture. Using geographical distance as a proxy variable for spatial segregation may bias the estimated results. In addition, the data of rural residents’ income is not available. Finally, although we controlled for the county-level fixed effect in the regression, we do not take into account the geographical characteristics of villages, which might also bias the estimated results.
As the quantitative analysis showed, geographical location plays a very important role in enabling rural development, as proximity to the cities, mainly to the county centers, becomes a crucial factor for adding value to the human capital of impoverished regions through education and providing employment.
This paper was supported by the 2022 Research Project at the Rural Development Institute, Yan’an University, the National Natural Science Foundation of China (grant No. 42171208), the National Social Science Fund of China (grant No. 17ZDA069), and the 111 Project of China (B16040). On the part of Lomonosov Moscow State University, the study was funded by state assignment (project 121051400060-2).