Household characteristics impact on rural to urban migration the case of quang ngai province

NATIONAL UNIVERSITY OF HCM UNIVERSITY OF ECONOMICS VIETNAM INSTITUTE OF SOCIAL STUDIES THE HAGUE THE NETHERLANDS VIETNAM- THE NETHERLANDS PROJECT FOR M.A IN DEVELOPMENT ECONOMICS ()()() HOUSEHOLD CHARACTERISTICS IMPACT ON RURAL TO URBAN MIGRATION THE CASE OF QUANG NGAI PROVINCE ;_ THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENT FOR THE DEGREE OF MASTER OF ARTS IN DEVELOPMENT ECONOMICS BY DUONG DINH QUOC BQ.GIAO DUe VA DAo TA TRUONG 8H K/NH TE TP.HgM cI_!/- 6y5g THU'VIEN Academic Supervisor: Dr LE THI THANH LOAN HO CHI MINH CITY, OCTOBER 2007 CERTIFICATION "I certify that the substance of this dissertation has not already been submitted for any degree and is not currently submitted for any other degree I certify that to the best of my knowledge and help received in preparing this dissertation and all sources used, have been acknowledged in this dissertation." DUONG DINH QUOC ACKNOWLEDGEMENTS Fist of all, let me express my thanks all organizations and member those found Vietnam - Netherland project, all lectures and staffs of the project for useful lessons and latest knowledge I would like to express my special thanks to my academic supervisor Dr Le Thi Thanh Loan for his helpful guidance and assistance during the writing of this thesis I also deeply thanks to Dr Nguyen Trang Hoai and Dr Nguyen Hoang Bao for their valuable suggestions and advices to research this topic Following, I send my thanks to Librarian, Ms.Chi, who helps help me in finding necessary document and reference books Thank to my friends in Quang Ngai - the interviewers help me in collecting surveyed data in Quang Ngai province Many thanks sent to my classmates, who shared with me joys and obstacles during more two years attending this program My thanks send to my wife and sons who shared and encouraged me during my ·- learning and doing this study ABBREVIATIONS HCM: HoChiMinh GSO: General Statistics Office NEZ: New Economic Zone DAN: Development Analysis Network VLSS: Vietnam Living Standard Survey USD: United State Dollar (American Currency) VND: Vietnam Dong (Vietnamese Currency) TABLE OF CONTAINS CHAPTER 1: INTRODUCTION 1.1 Problem statement 1.2 Research objectives 1.3 Research questions 1.4 Methodology 1.5 Organization of the thesis CHAPTER 2: LITERATE REVIEW 2.1 Overview of migration typical theories 2.2 Lewis model 2.3 Push and pull factor theory 2.4 Harris -Todaro model 2.5 New Economics of Labour Migration theory 11 2.6 Comments on the theories 12 CHAPTER 3: MIGRATION IN VIETNAM 3.1 Overview of migration historical in Vietnam 14 3.1.1 Migration in feudal states 14 -·;, 3.1.2 Migration under the French colonial period 15 3.1.3 Migration in the war against US army 16 ~ 3.1.3 Migration after reunification 1975 18 3.2 The factors speed up rural to urban migration 20 3.2.1 The poverty and inequity 20 3.2.2 Population dist!ibution and density 21 3.3 Migration data and tendency 23 3.4 Migration to HCM city 25 3.4.1 The Socio-economic situation of HCM city 25 3.4.2 Overview of migration to HCM city 26 3.4.3 Migrants characteristics to HCMC 27 3.4.3.1 Migrants origins 27 3.4.3.2 Educational attainment 28 3.4.3.3 Age differentials 29 3.4.3.4 Gender 30 3.4.3.5 Occupational differentials and income 31 3.5 Migration impacts 33 3.3.1 Impacts on urban areas 33 3.3.2 Impacts on rural areas 34 CHAPTER4:METHODOLOGY 4.1 Definition of migrant 36 4.2 The objects of surveys 38 4.3 Data collection methods 38 4.3.1 Secondary data 39 4.3.2 Primary data 39 4.4 Regression model of rural to urban migration 40 4.5 Description variables in the model 40 CHAPTER 5: HOUSEHOLD CHARACTERISTICS IMPACT ON RURAL TO URBAN MIGRATION 5.1 Descriptive analysis 44 5.1.1 The Socio-economic situation in Quang Ngai Province 44 5.1.2 Statistical Analysis 45 5.2 Regression results 48 5.2.1 Estimate coefficients 48 2.2 Model test 49 2.3 Determinants of migration probability 49 5.3 Comment on results 50 CHAPTER 6: CONCLUTIONS AND RECOMMENDATIONS 6.1 Introduction 53 6.2 Conclusions 53 6.3 Recommendations 54 6.3.1 Recommendations for original authorities 55 6.3.2 Recommendations for urban authorities 56 6.4 Limits and suggestion for further studies 56 Appendix ! 57 Appendix 58 Appendix 59 _ Questionnaire 60 References 63 LIST OF TABLES Table 1: Annual average population growth rate of the city(%) 26 Table 2: Migration from regions to HCM city 1999-2004 28 Table 3: Population rate in school by ages (%) 29 Table 4: Population structure by age group (%) 30 Table 5: Migrants gender index by ages 31 Table 6: Distribution of employment of migrants in 2004 by economic sectors 31 Table 7: Monthly average migrant income (VND) in 2004 32 Table 8: Description variables used in the model 43 Table 9: Household characteristics in survey Table 10: Regression result of unrestricted model (7 variables) 48 Table 11: Regression result of restricted model (5 variables) 49 Table 12: Influent levels of variables to household migration probability 50 CHAPTER I INTRODUCTION • 1.1 Problem statement: Rural to urban migration is the natural process in economic development m all countries, especially in less development countries Migration is not a consequence of economic development progress, but the economic growth has promoted migrants from rural regions to economic centers and urbanization areas The booming in economic growth in Vietnam for two decades has changed remarkably economic structure The production rate in agriculture has decreased strongly in total GDP, from 27.17 % in 1995 to 24.37% in 2000 and 21.76% in 2004 1• In rural areas, beside traditional agricultural sector, service and industrial production have been expanded Therefore, income rate earnings from farming fields are decreasing in total household income However, high natural population growth rate and new jobs created slowly in rural areas have made the pressure on labor force in agriculture Moreover, the productivity in agriculture is still low, agricultural lands are reduced due to demands of development industrial zones, urbanization and environmental pollution caused by industry These factors lead an increasing in working -off time and an expanding working pressure in rural areas In this situation, migration is considered an acceptable solution for labor force in rural areas Rural to urban migration has played a very important role m industrialization and urbanization process, migration labor force has met the large labor demand in economic development progress The migration also decreases jobs pressure on jobs in agriculture and contributes poverty reduction by remittances to households On the other hand, an uncontrolled migration progress overloads the urban infrastructure, affects on the social fundamentals and may become a disadvantage factor to economic development progress in urban areas GSO 2005, available at www.gso.gov.vn Being a less development country, Vietnam is still in the initial stage of urbanization, given by about 70% population still lives in rural areas and tendency of rural to urban migration will occur strongly in coming years, especially in large city such as HCM city, Ha Noi, Da Nang The thesis also • identifies the socio- economic consequences not only to rural but also to urban areas The research is to identify the household characteristics that effect on rural to urban migration in Quang Ngai province by reviewing the different approaches to migration decision The study also allows us determine which are the main factors affecting on migration decision of household in rural areas 1.2 Research objectives The main objective of this exploratory study is to get a better understanding of rural to urban migration through out the features of households in rural The migration process of any individual in a family are considered as an allocation the jobs amongst members of a household, therefore, in this research individual's migration is also considered a migratory household Although many previous researches have mentioned about rural to urban migration under the other viewpoints, but almost have focused on migrants' characteristics and other socio- economic, geographic factors that affects on the migration This research focuses on the households characteristics to migration decisions of members in the family Moreover, this research implicates the policy framework for rural to urban migration Therefore, households in rural regions are main objects in the research and are central units of migration survey More specifically, the study focuses on the following objectives: • To analyze the households characteristics in the rural areas that have influences to migration possibility, and identify what the factors have affected on migration possibility most strongly • To understand the migrants' livelihoods in destination places and the features of migration labored force in recent years • To analyze public response and government policies to deal with the increasing migration flows from rural to urban areas 1.3 Research questions To realize our main objective mentioned in the previous paragraph, the following central question is formulated "How the household characteristics affect on rural to urban migration in Quang Ngai province?" From the central question the following sub-questions are formulated, which are addressed in the subsequent chapters in the study as follows: What are the household characteristics in original place? What are the initial conditions for rural to urban migration? What the features of household affect on migration probability? 1.4 Methodology Descriptive analysis will be use to show an overview of rural to urban migration based on secondary data analysis from Vietnam migration census 2004 _ conducted by General Statistics Office, mid-term population survey 2004 conducted by HCM city statistics department, the previous migration study in Vietnam, and data extracted from various papers, magazines, books, and many other related books Data for regression are collected from a migration survey in three communes of Quang Ngai province with 102 observations And a logistic model will be run to analyze variables affecting on migration decision, regression results will be used for explaining effective level by each variable 1.5 Organization of the thesis The thesis is organized into five chapters Following the chapter of introduction, the second chapter overviewed development theories, which link to migration progress and some implications exacted from these theories Next chapter describes briefly historical development of migration in Vietnam, natural element and social context speeding up migration, this chapter also shows Influent levels of variables to household migration probability are illustrated in table 11, which reflects the fluctuation of household migration probability response of increasing by one unit of a variable and at a given probability, in term of holding other variables constant • Table 12: Influent levels of variables to household migration probability PMIG SIZE increases by one member EDUHH increases by one educational level HIGHRATE increases by FARMRATE increases by DIFFRATE increases by 1% 1% 1% 0.1 0.27936 0.24751 0.10600 0.10597 0.09566 0.2 0.46587 0.42531 0.21059 0.21054 0.19225 0.3 0.59923 0.55922 0.31381 0.31374 0.28977 0.4 0.69933 0.66370 0.41568 0.41560 0.38826 0.5 0.77723 0.74749 0.51623 0.51614 0.48771 0.6 0.83957 0.81619 0.61548 0.61540 0.58814 0.7 0.89060 0.87353 0.71346 0.71339 0.68957 0.8 0.93313 0.92212 0.86491 0.81014 0.79201 (Source: Author's calculation from regression result in appendix 2) 5.3 Comment on results Effect of household size (SIZE variable) It is clear that household size affects on migration probability positively and strongly The more member household has the higher migration probability of household In fact, household has favorite conditions for sharing works between the members if household size is larger Addition, a large household size made up more pressure on its expenditures in term of limited production resources in rural areas will push migration possibility For example, an increasing more one member in the family leads an increasing household migration probability from 50% to 77.62 %, holding other variables constant 50 ~ - - Effect of household head's education (EDUHH variable) The regression results confirm the positive relationship between educational level of household head and household migration probability This • relation can be explained by two approaches, the first is influence of household head spreading over all members, leading a high level of education that is usually directly relation to migration probability The second concern to the household head's opened -mind, people with high educational level is often broad-minded person, which easily encourages the members' movement for getting a better job The regression results state that educational level of household head obtains more one level will increase household migration probability from 50% to 74.47% in term of constant other variables The result implies that in the survey location, well educational or skilled people have higher tendency in movement population, and influence of education degree on migration decision is rather strongly Effect of labor structure (HIGHRATE Variable) The labored force in group 18 -40 have the highest productivity than other ages group, and this group is also the best dynamic labors force in economy The historical data shown that people in this group are the highest migration possibility because they can get a job in urban easier that other age groups Therefore, in condition of limited jobs in rural areas, this labored force is considered as main component in migration waves to urban areas Similarly, with the initial probability of 50%, the labored force in ages of 18 - 40 increases 1% will raise migration possibility of household to 51.62% Effect of agricultural income rate (FARMRATE variable) It is not surprising that the ratio of earnings from agriculture to total • income has positive impact on migration probability, the more dependent on earnings from farming activities, the lower household total income, leads to a higher migration probability Moreover, in condition of working opportunities are limited in rural areas, a high rate of agricultural income over total income means that household's free time is long Hence, the members of the household have 51 tendency of movement for finding other jobs in the free time The influent level of proportion of income from agricultural activities to migration probability is closed with ratio of labor in ages of 18 - 40 over household size If rate of agricultural earnings increases 1%, migration probability will increases slightly from 50% to 51.61%, meanwhile this probability rise 51.62% from 50% if ratio of labor in ages of 18 - 40 increases 1% Effect of different income (DIFFRATE variable) Conversely, once labor's income in rural areas increases quickly and closet with income that a migrant can getting in rural, there is no any economics reasons to encourage the farmers' movement to urban areas In other words, when the different income between a labor in rural and migratory labor in urban decreases, rural to urban migration is also decreases As regression result from table 11, income ratio between a labor in rural and a migratory labor rises up 1% will decrease migration probability down 48.85% from 50% Effect of lands Abnormally, the average square of agricultural lands per capita has less impact on migration decision and this variable is not statistical significantly Although the relation between LAND variable and migration probability is negative as expected, but the effective level is very low far away as expected It can be explained that average square of lands per capita is small and income from cultivation not contributes a large percentage in total income of household Therefore, an increasing one unit of land (100 m2) does not increase much earnings as well as waste time of rural labors Moreover, in location survey, the average square of agricultural lands per capita is not large as in Mekong Delta or High Lands area The difference average square of agricultural lands per capita between group of migration and non-migration is rather small (1.254 m vs.l.322 m2) As resulted, the lands are not considered the dominant factor to keep the • farmers living together with their traditional occupation 52 CHAPTER VI CONCLUSIONS AND RECOMMENDATIONS • 6.1 Introduction This chapter is to summarize the main results of study, including history of migration waves in Vietnam, migrants' characteristics into urban areas, research methodologies and major findings Based on results from statistics analysis and regressiOn results from chapter 5, policy implications will be given for optimal migration Finally, the shortcomings of studies as well as suggestion for further researches are also presented in the last section this chapter 6.2 Conclusions It is needed to confirm that rural to urban migration is a natural progress in urbanization and economic growth in all developing countries Migration process in Vietnam had occurred for thousands year accompany with the social growth Many migration flows in the past moved to the delta areas along the fertilized rivers for cultivation, other population movements moved to destinations with better natural resources In general, the great population growths, the inequality of living standards, the imbalance in distribution of agricultural lands and natural resources have stimulated the great streams of migration people in the past Rural to urban migration is indispensable result of industrialization and urbanization, and a large scale of migration ware has started when Vietnamese economy has significantly developed since 1990s Hanoi, Da Nang, Hochiminh city and other big cities have become the idea destination for migrants from rural areas Amongst economic centers, Hochiminh city is the best choice by high rate in economic growth, highest income per capita in whole country and getting a job easily Most of migrants are in working ages, healthy and better education than people who still stay at rural areas Although the unemployment rate in urban areas is rather high but migrants' unemployment rate is very low, because a 53 significant ratio of migrants work in informal sector and they accept simple jobs or heavy works with low incomes as mentioned in chapter 3.5 Besides, migrant remittances to their family are considered as investment • capital in rural areas for improvement living standard, innovating production materials for increasing productivities However, the imbalance in migration structure in sex and labor force maybe brings about bad affects on both rural and urban regions in the future The regression result confirms the positive relation between differentials income rural -urban and migration possibility Besides that, educational level of household head has active influence on migration decision of the family members Because the family owner's education reflects the members' educational background, therefore, a well educational level allows people integrating to new life environment easier, that means higher migration probability The household size is also strongly affects on migration, in terms of limited production resources in rural, the larger household size, the higher • migration probability (table 5.3) Although lands is considered as one of the mainly production materials in agriculture, but the research has no impact of average square of production lands on migration probability, it shown that lands is no longer important factor to keep farmers' life together with their homelands 6.3 Recommendations As discussed above, the mam reason of rural to urban migration is economic, therefore, the policies to deal with rural to urban migration issues should based on the economic viewpoints and control of migration process should be taken by the economic measures replacing administrational ones The policies makers not only create a friendly environment for migrants to stay and • have stable lives in urban areas, but also set up the effective measures to control the flows of migration rather than to be restricted as current regulations are 54 6.3.1 Recommendation for original authorities Regression results of the model and statistical data state that household size has effected strongly to household migration probability Therefore, • maintaining the household size by implementing the policies of family panning and reducing birth rate are the first policies In fact, these policies have conducted successfully for more three decades whole country However, in some rural regions, the natural population growth rate is still high, it makes the pressure on new jobs in local and promotes movement waves Keeping a low birth rate is the long-term process with many obstacles, it is not only for reducing migration but also for economic growth in local Household head' educational level can be seen as represented for family education level Although educational degree is positive relative with migration probability, but in converse, statistics data states that high educational level can reduce remarkably birth rate Besides that, skilled or educational degree labors are usually movement for finding the jobs, which are not available in local, release the pressure on jobs in local However, the local governments should focus on training vocational schools or practical training courses because the schools can help farmer in reducing the dependence on agricultural income, creating the non-farm works at the local Such as, farmers can expand production and improving their income without migration to urban areas The local authorities also should review the credit policies to create more non-farm jobs in rural, support credits to household farming projects and intensive labor plans, especially unskilled labor in local In fact, the extra income from the household' small projects have good affects on decreasing number of migrants to urban than any macro strategies implementing by the authorities A very important problem in agricultural production is affected by weather and natural risks, leading to the regular fluctuation in farmers' income Therefore, the essential small credits can assist farmers overcoming difficulties and reducing migration ware temporarily 55 6.3.2 Recommendation for urban authorities Although migratory labored force have contributed considerably in economic growth of urban areas, but urban authorities should review the economic strategies based on using labor force the best effectively Is the rational • if the cities focus on important industries using skilled technical labors and capital intensive? For labor-intensive industrials, the urban governments should control closely Some people think that the growth of labor-intensive industrials will reduce the unemployment rate in urban But in fact, migratory labor work in these industries contributed a large rate, and most of them are unskilled or low educated labors, their contribution to urban economic growth is very limited 6.4 Limits and suggestion for further studies The surveyed data collected in the thesis covering only two communes in Quang Ngai province with 102 households, so the sample could not be seen as typical sample represent for 200.000 households in the province Moreover, the surveyed data comes from mainly responses of household heads and the accurate appraisals in the answers are still limited • The further studies should focus on more the average lands square as well as effect in using lands in various communes to identify the effective level of lands to migration decision I think that if the lands factor is analysed in more details and more observation, lands will have impacted on rural to urban migration decision 56 APPENDIX! Method: ML- Binary Logit (Quadratic hill climbing) Date: 07/16/07 Time: 08:31 Sample: 102 Included observations: 102 Convergence achieved after iterations Covariance matrix computed using second derivatives Coefficient Std Error z-Statistic Pro b SIZE 1.243855 0.381779 3.258048 0.0011 EDU 1.070524 0.498772 2.146321 0.0318 DEPEND -0.002491 0.022286 -0.11177 0.911 HIGH RATE 0.064304 0.0241 2.668179 0.0076 LAND -0.009941 0.026888 -0.369718 0.7116 FARM RATE 0.066885 0.020358 3.285496 0.001 DIFFRATE -0.045997 0.019671 -2.338328 0.0194 c -11.84601 2.99256 -3.958488 0.0001 Variable Mean dependent var 0.54902 S.D.dependentvar 0.500049 S.E of regression 0.34633 Akaike info criterion 0.887338 Sum squared resid 11.27476 Schwarz criterion 1.093218 0.970706 ) Log likelihood -37.25424 Hannan-Quinn criter Restr log likelihood -70.21003 Avg log likelihood • LR statistic (7 df) 65.91159 Probability(LR stat) 9.86E-12 Obs with Dep=O 46 Obs with Dep=1 56 McFadden A-squared Total obs 57 -0.365238 0.469389 102 APPENDIX2 Dependent Variable: PMIG Method: ML- Binary Logit (Quadratic hill climbing) • Date: 07/16/07 Time: 08:31 Sample: 102 Included observations: 102 Convergence achieved after iterations Covariance matrix computed using second derivatives Coefficient Std Error z-Statistic Pro b SIZE 1.249573 0.372603 3.353635 0.0008 EDU 1.085278 0.487523 2.226104 0.0260 HIGHRATE 0.06495 0.023401 2.775557 0.0055 FARM RATE 0.064594 0.019214 3.361865 0.0008 DIFFRATE -0.049182 0.014495 -3.393085 0.0007 c -11.90912 3.007512 -3.959793 0.0001 Variable Mean dependent var ' ~ 0.54902 S.D.dependentvar 0.500049 S.E of regression 0.341929 Akaike info criterion 0.849526 Sum squared resid 11.22386 Schwarz criterion 1.003937 Log likelihood -37.32585 Hannan-Quinn criter 0.912052 Restr log likelihood -70.21003 Avg log likelihood -0.36594 LR statistic (5 df) Probability(LR stat) 65.76836 7.76E-13 McFadden A-squared 0.468369 Obs with Dep=O 46 Obs with Dep=1 56 Total obs ' 58 102 APPENDIX3 Because the probabilities of DEPEND and LAND variables which are not statistically • significant difference from zero, we formulate the restricted version of the model subject to the joint hypothesis Ho: ~2 =0 and ~3 =0 Hl: at least ~ or ~ i- or both ~2 and ~3 i- We use a chi- square statistic with two degrees of freedom based on the likelihood ratio, LRILu X2k =- 2ln(LR/LuR) =-2(1nLR -lnLuR) =-2(-37.32585 + 37.2542436833) =0.14322 We have lo.os,2 = chiinv(0.05,2) = 5.99 Since l k < l o.os => we accept the null hypothesis - ' • 59 A ? ? ? , BANG CAU HOI KHAO SAT QUESTIONNAIRE ;; • Chung t6i la sinh vien dang nghien coo va"n de di dan tu nong th6n vao ph6 Anh (chi) vui long giup chung t6i b~ng each tni lOi cac diu hoi du'oi day: We are students and studying the migration from rural to urban area Would you please an.nver the following questions II THONG TIN HQ GIA DINH/ FAMILY BACKGROUND HQ ten chii h9: Household Owner name Tinh: Province xa Commune l Huyl$n: District Thon Hamlet Thong tin v~ cac vien h() Information t~f household members Gidi tinh Quanh~ Stt Hten No Full name vdi chii hQ Relation with household o~vner Tu6i ARes Sex Nam A-tale Ncr Female HQc va"n Edu-leFel Khii nang lao d()ng In working - s6 nguoi d() tu6i 18 -40 : nguoi persons Nwnber ofpeople in 18- 40 years old nguoi Persons - s6 nguoi c6 kha nang lao d()ng Number of people is in working Di~n tich da't san xua't (Production lands) o E>a"t tr6ng lua: (lands for rice) mz E>a"t cho chan nuoi (lands for breeding) mz E>a"t nong nghi~p khac (lands for other cultivation) ········· m Thu nhi)p hang nam ciia gia dinn t~i dja phttdngLI_ _ _ _ _.I d6ng (VND) (Ammalfamily income in local) o Til cay lua Kg, tudng duong (From rice) (Equal to) o Tu chan nuoi d6ng (VND; d~ong (From breeding) (VND) o Tu cay tr6ng khac d6ng (From other cultivation) * (VND) T6ng thu nh~p nong nghi~p (Income fi-mn agricultural activities) o Tu cong vi~c khac (From works (?ff-farm activities) o Luong, tr